rust/tantivy
1
0
Fork 0
mirror of https://github.com/quickwit-oss/tantivy.git synced 2025-12-28 04:52:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: rust:fix-bench
Branches Tags
rust:main rust:stuhood.upstream-erased-sort-computer rust:trinity.pointard/ff-query rust:bucket_id_agg rust:paradedb/fix-agg-validation rust:flatheadmill-geo rust:paradedb-paradedb/fix-overflow-issue rust:low_card_optimization rust:paul.masurel/clippy rust:fix_size_hint rust:qw-airmail-20250916 rust:unit-test-aggregation-sum-other rust:clippy-and-cleanups rust:trinity.pointard/opti-sstable-sorted-ords-to-term rust:release_0.24 rust:dependabot/cargo/rand-0.9.1 rust:dependabot/cargo/lru-0.14.0 rust:dependabot/cargo/fs4-0.13.1 rust:tagged-user-input-ast rust:1686a/parse-agg-res rust:dependabot/cargo/rand_distr-0.5.1 rust:issue/2411-adding-panic-handler-to-rayon-merge-thread-pool rust:dependabot/cargo/nom-8 rust:upgrade_deps rust:test_parser rust:dependabot/github_actions/codecov/codecov-action-5 rust:binggan-0.12.0 rust:agg_format rust:path_collector rust:test_order_bug rust:test_columnar_multi rust:columnar_optionalmask_in_multivalues rust:prefix-phrase-query-optim rust:tantivy-with-oneshot rust:update_example rust:store-reader-enumerate rust:document-deserialize-seed rust:check_doc_mem rust:experimental-do-not-merge rust:expdonotmerge rust:cherry-pick-vec-instead-of-btreemap rust:validate_length rust:raphael_bench_avx_filter rust:raphael_optim_memset2 rust:query-field-name-avoid-alloc rust:check_bug rust:missing_test rust:list_fields rust:block-cache-control rust:trinity--multilayer-sstable rust:tracing-tantivy2 rust:tracing-tantivy rust:index_writer_err rust:default_fast_field_tokenizer rust:fmassot/add-preserve-original-ascii-folding rust:numeric_with_str_col rust:straightforward-simplify rust:fmassot/fake-test-branch rust:fmassot/dynamic-tokenizer-poc-follow-up rust:fmassot/add-box-token-filter-and-refactor rust:dynamic-tokenizer-proof-of-concept rust:fmassot/add-box-tokenizer-and-token-filter rust:tokenizer-api-0.1.1 rust:addconversion rust:removed-extra-enum-alternative-for-datetime-precision rust:issue/1981-reproduce-bm25-bug rust:tokenizer_docs rust:merge-proptests rust:columnar-todo rust:quickwit-0.5 rust:remove-byteorder rust:remove_dyn rust:pr/alexcole/1855-inject-bm25-statistics rust:evance-expose-build-query-from-user-input-ast rust:missing-specialized-method-arcdyncolumn rust:fast-u64-range-query rust:0.19.2 rust:columnar-cli2 rust:trinity--pub-term-as-slice rust:typed-column3 rust:columnar-merge3 rust:columnar-merge-exploring-fn rust:use_columnar rust:typed-column2 rust:term_agg_bucket_limit rust:range rust:typed-column rust:columnar-main rust:forced-types rust:u128-columnar rust:use_stats rust:truncation_comment rust:0.19.1 rust:trinity/yoke-documents rust:column-handle rust:columnar-cli rust:columnar rust:revert-1711-sparse_dense_index rust:commit-changes rust:sparse_dense_index_headroom rust:optional_column rust:no-column-for-multivalued rust:fmassot--bench-hdfs-with-array rust:bugfix-position-broken rust:float rust:simplify rust:debug_time rust:sparse_codecs rust:fix_estimate rust:column-reader rust:column-reader-b rust:columnreader-simple rust:col-trait-refact rust:ip_fastfield rust:column-trait-2 rust:refact-dyn-column rust:column-trait rust:refact-codec-trait rust:pr/PSeitz/1485-1 rust:fastfieldcodec rust:termmap_per_field rust:fastfield-codecs-tests rust:rayon-indexing-writers rust:issue/1251 rust:index-bench rust:fix_open_bytes_index rust:warming rust:revert-1222-issue/1195 rust:add-random-bench-for-fast-field rust:wasm-friendly-exploratory rust:revert-1159-patch-1 rust:debug-position-panic-0.15.3 rust:fix-issue-1111 rust:remove_rand rust:unused-deps rust:hfb/phrase_match_slop rust:0.15.3 rust:expose-min-max-on-multi-fast-field-reader rust:revert-1094-expose-min-max-on-multi-fast-field-reader rust:hotfix-1088 rust:owned-bytes-instead-of-rawdoc rust:rihardsk-date-ranges rust:issue/997 rust:0.13.3 rust:fail-fast-store-checkpoint rust:debugging-store rust:issue/reproduce-897 rust:issue/969-reproduce rust:troublescooter-static rust:intermediate-mergeable-troublescooter rust:barrotsteindev-filter-collector-tpredicatevalue rust:issue/938b rust:issue/938 rust:barrotsteindev-filter-collector rust:issue/922 rust:issue/896 rust:quickfix-explain rust:slog rust:limit-random-seeks rust:0.13.1 rust:issue/866b rust:bugfix-unittest-macos rust:githubactions rust:fieldnorm_readers rust:blockwand rust:nodeffeatfails rust:removedaliveiterator rust:robyoung-introduce-offset-to-top-docs rust:specialization-of-foreach-for-union-scorer rust:elshize-bmw rust:cutt rust:nrt rust:storewriter-out rust:segment_writer_in_ram2 rust:polling rust:audunhalland-query-boost rust:poll-meta rust:segment_writer_in_ram rust:bundle rust:petr-tik-n662_footer_is_incompatible_impl rust:race-condition-in-tests rust:hotfix-0.10.3-issue681 rust:remove-tru rust:kkoziara-pre-tokenized-text rust:issue/681 rust:issue/680 rust:refactoring/field rust:updating/uuid rust:fix-bench rust:hotfix-656 rust:readd-macos-ci rust:incremental-search rust:crate-division rust:streamer-with-state rust:criterion rust:python-binding-changes rust:bump-version rust:gh-pages rust:issue/554 rust:perf/experimental-union-for_each rust:issue/526b rust:softcommits rust:0.9.1-hotfix rust:revert-521-fix-non-english-stemmers rust:issue/500-hotfix rust:hotfix/0.8.2 rust:issue/483 rust:bug/merge-deted rust:issue/407 rust:issue/457 rust:segment_fruits rust:doc rust:dds/lenient rust:issue/webasm rust:issue/396 rust:broken_collector_with_result rust:dss/automaton-weight-suggestion rust:wasm rust:common-crawl rust:tantivy-imhotep
rust:0.25.0 rust:0.22.1 rust:0.24.2 rust:0.24.1 rust:tantivy-query-grammar-v0.24.0 rust:tantivy-columnar-v0.5.0 rust:tantivy-bitpacker-v0.8.0 rust:ownedbytes-v0.9.0 rust:tantivy-stacker-v0.5.0 rust:tantivy-sstable-v0.5.0 rust:tantivy-common-v0.9.0 rust:tantivy-tokenizer-api-v0.5.0 rust:0.24 rust:qw-airmail-20250224-hotfix-merge-panic rust:qw-airmail-20250219-hotfix-merge-panic rust:qw-airmail-202406 rust:qw-airmail-20240611 rust:qw-airmail-20240606 rust:quickwit-rev rust:0.22.0 rust:tantivy-stacker-v0.3.0 rust:ownedbytes-v0.7.0 rust:tantivy-tokenizer-api-v0.3.0 rust:tantivy-columnar-v0.3.0 rust:tantivy-common-v0.7.0 rust:tantivy-query-grammar-v0.22.0 rust:tantivy-sstable-v0.3.0 rust:tantivy-bitpacker-v0.6.0 rust:0.21.1 rust:0.21 rust:0.20.2 rust:0.20.1 rust:0.20 rust:quickwit-0.5-rev rust:0.19.2 rust:0.19.1 rust:0.19 rust:0.18.1 rust:0.18 rust:0.17 rust:0.16.1 rust:0.15.3 rust:0.15.2 rust:0.15.1 rust:0.15 rust:0.14 rust:0.13.3 rust:0.13.2 rust:0.13.1 rust:0.13 rust:0.12 rust:0.11.3 rust:0.11.1 rust:0.11 rust:0.10.3 rust:0.1 rust:0.10.2 rust:0.10.1 rust:0.10.0 rust:0.9.1 rust:0.9 rust:0.8.2 rust:0.8.2b rust:0.8.1 rust:0.8.0 rust:0.7.2 rust:0.7.1 rust:0.7.0 rust:0.6.1 rust:0.6.0 rust:0.5.2 rust:0.5.1 rust:0.5.0 rust:0.4.3 rust:0.4.2 rust:0.4.0 rust:0.3.1 rust:0.3.0 rust:0.2.0
..
compare: rust:issue/396
Branches Tags
rust:stuhood.upstream-erased-sort-computer rust:main rust:trinity.pointard/ff-query rust:bucket_id_agg rust:paradedb/fix-agg-validation rust:flatheadmill-geo rust:paradedb-paradedb/fix-overflow-issue rust:low_card_optimization rust:paul.masurel/clippy rust:fix_size_hint rust:qw-airmail-20250916 rust:unit-test-aggregation-sum-other rust:clippy-and-cleanups rust:trinity.pointard/opti-sstable-sorted-ords-to-term rust:release_0.24 rust:dependabot/cargo/rand-0.9.1 rust:dependabot/cargo/lru-0.14.0 rust:dependabot/cargo/fs4-0.13.1 rust:tagged-user-input-ast rust:1686a/parse-agg-res rust:dependabot/cargo/rand_distr-0.5.1 rust:issue/2411-adding-panic-handler-to-rayon-merge-thread-pool rust:dependabot/cargo/nom-8 rust:upgrade_deps rust:test_parser rust:dependabot/github_actions/codecov/codecov-action-5 rust:binggan-0.12.0 rust:agg_format rust:path_collector rust:test_order_bug rust:test_columnar_multi rust:columnar_optionalmask_in_multivalues rust:prefix-phrase-query-optim rust:tantivy-with-oneshot rust:update_example rust:store-reader-enumerate rust:document-deserialize-seed rust:check_doc_mem rust:experimental-do-not-merge rust:expdonotmerge rust:cherry-pick-vec-instead-of-btreemap rust:validate_length rust:raphael_bench_avx_filter rust:raphael_optim_memset2 rust:query-field-name-avoid-alloc rust:check_bug rust:missing_test rust:list_fields rust:block-cache-control rust:trinity--multilayer-sstable rust:tracing-tantivy2 rust:tracing-tantivy rust:index_writer_err rust:default_fast_field_tokenizer rust:fmassot/add-preserve-original-ascii-folding rust:numeric_with_str_col rust:straightforward-simplify rust:fmassot/fake-test-branch rust:fmassot/dynamic-tokenizer-poc-follow-up rust:fmassot/add-box-token-filter-and-refactor rust:dynamic-tokenizer-proof-of-concept rust:fmassot/add-box-tokenizer-and-token-filter rust:tokenizer-api-0.1.1 rust:addconversion rust:removed-extra-enum-alternative-for-datetime-precision rust:issue/1981-reproduce-bm25-bug rust:tokenizer_docs rust:merge-proptests rust:columnar-todo rust:quickwit-0.5 rust:remove-byteorder rust:remove_dyn rust:pr/alexcole/1855-inject-bm25-statistics rust:evance-expose-build-query-from-user-input-ast rust:missing-specialized-method-arcdyncolumn rust:fast-u64-range-query rust:0.19.2 rust:columnar-cli2 rust:trinity--pub-term-as-slice rust:typed-column3 rust:columnar-merge3 rust:columnar-merge-exploring-fn rust:use_columnar rust:typed-column2 rust:term_agg_bucket_limit rust:range rust:typed-column rust:columnar-main rust:forced-types rust:u128-columnar rust:use_stats rust:truncation_comment rust:0.19.1 rust:trinity/yoke-documents rust:column-handle rust:columnar-cli rust:columnar rust:revert-1711-sparse_dense_index rust:commit-changes rust:sparse_dense_index_headroom rust:optional_column rust:no-column-for-multivalued rust:fmassot--bench-hdfs-with-array rust:bugfix-position-broken rust:float rust:simplify rust:debug_time rust:sparse_codecs rust:fix_estimate rust:column-reader rust:column-reader-b rust:columnreader-simple rust:col-trait-refact rust:ip_fastfield rust:column-trait-2 rust:refact-dyn-column rust:column-trait rust:refact-codec-trait rust:pr/PSeitz/1485-1 rust:fastfieldcodec rust:termmap_per_field rust:fastfield-codecs-tests rust:rayon-indexing-writers rust:issue/1251 rust:index-bench rust:fix_open_bytes_index rust:warming rust:revert-1222-issue/1195 rust:add-random-bench-for-fast-field rust:wasm-friendly-exploratory rust:revert-1159-patch-1 rust:debug-position-panic-0.15.3 rust:fix-issue-1111 rust:remove_rand rust:unused-deps rust:hfb/phrase_match_slop rust:0.15.3 rust:expose-min-max-on-multi-fast-field-reader rust:revert-1094-expose-min-max-on-multi-fast-field-reader rust:hotfix-1088 rust:owned-bytes-instead-of-rawdoc rust:rihardsk-date-ranges rust:issue/997 rust:0.13.3 rust:fail-fast-store-checkpoint rust:debugging-store rust:issue/reproduce-897 rust:issue/969-reproduce rust:troublescooter-static rust:intermediate-mergeable-troublescooter rust:barrotsteindev-filter-collector-tpredicatevalue rust:issue/938b rust:issue/938 rust:barrotsteindev-filter-collector rust:issue/922 rust:issue/896 rust:quickfix-explain rust:slog rust:limit-random-seeks rust:0.13.1 rust:issue/866b rust:bugfix-unittest-macos rust:githubactions rust:fieldnorm_readers rust:blockwand rust:nodeffeatfails rust:removedaliveiterator rust:robyoung-introduce-offset-to-top-docs rust:specialization-of-foreach-for-union-scorer rust:elshize-bmw rust:cutt rust:nrt rust:storewriter-out rust:segment_writer_in_ram2 rust:polling rust:audunhalland-query-boost rust:poll-meta rust:segment_writer_in_ram rust:bundle rust:petr-tik-n662_footer_is_incompatible_impl rust:race-condition-in-tests rust:hotfix-0.10.3-issue681 rust:remove-tru rust:kkoziara-pre-tokenized-text rust:issue/681 rust:issue/680 rust:refactoring/field rust:updating/uuid rust:fix-bench rust:hotfix-656 rust:readd-macos-ci rust:incremental-search rust:crate-division rust:streamer-with-state rust:criterion rust:python-binding-changes rust:bump-version rust:gh-pages rust:issue/554 rust:perf/experimental-union-for_each rust:issue/526b rust:softcommits rust:0.9.1-hotfix rust:revert-521-fix-non-english-stemmers rust:issue/500-hotfix rust:hotfix/0.8.2 rust:issue/483 rust:bug/merge-deted rust:issue/407 rust:issue/457 rust:segment_fruits rust:doc rust:dds/lenient rust:issue/webasm rust:issue/396 rust:broken_collector_with_result rust:dss/automaton-weight-suggestion rust:wasm rust:common-crawl rust:tantivy-imhotep
rust:0.25.0 rust:0.22.1 rust:0.24.2 rust:0.24.1 rust:tantivy-query-grammar-v0.24.0 rust:tantivy-columnar-v0.5.0 rust:tantivy-bitpacker-v0.8.0 rust:ownedbytes-v0.9.0 rust:tantivy-stacker-v0.5.0 rust:tantivy-sstable-v0.5.0 rust:tantivy-common-v0.9.0 rust:tantivy-tokenizer-api-v0.5.0 rust:0.24 rust:qw-airmail-20250224-hotfix-merge-panic rust:qw-airmail-20250219-hotfix-merge-panic rust:qw-airmail-202406 rust:qw-airmail-20240611 rust:qw-airmail-20240606 rust:quickwit-rev rust:0.22.0 rust:tantivy-stacker-v0.3.0 rust:ownedbytes-v0.7.0 rust:tantivy-tokenizer-api-v0.3.0 rust:tantivy-columnar-v0.3.0 rust:tantivy-common-v0.7.0 rust:tantivy-query-grammar-v0.22.0 rust:tantivy-sstable-v0.3.0 rust:tantivy-bitpacker-v0.6.0 rust:0.21.1 rust:0.21 rust:0.20.2 rust:0.20.1 rust:0.20 rust:quickwit-0.5-rev rust:0.19.2 rust:0.19.1 rust:0.19 rust:0.18.1 rust:0.18 rust:0.17 rust:0.16.1 rust:0.15.3 rust:0.15.2 rust:0.15.1 rust:0.15 rust:0.14 rust:0.13.3 rust:0.13.2 rust:0.13.1 rust:0.13 rust:0.12 rust:0.11.3 rust:0.11.1 rust:0.11 rust:0.10.3 rust:0.1 rust:0.10.2 rust:0.10.1 rust:0.10.0 rust:0.9.1 rust:0.9 rust:0.8.2 rust:0.8.2b rust:0.8.1 rust:0.8.0 rust:0.7.2 rust:0.7.1 rust:0.7.0 rust:0.6.1 rust:0.6.0 rust:0.5.2 rust:0.5.1 rust:0.5.0 rust:0.4.3 rust:0.4.2 rust:0.4.0 rust:0.3.1 rust:0.3.0 rust:0.2.0

1 Commits
fix-bench ... issue/396

Author SHA1 Message Date
Paul Masurel
4cbcc59e8f First stab at #396 2018-08-28 10:01:19 +09:00
223 changed files with 7710 additions and 20240 deletions
Expand all files Collapse all files

1
.gitignore vendored
View File

@@ -1,4 +1,3 @@
tantivy.iml
*.swp
target
target/debug

32
.travis.yml
View File

@@ -10,7 +10,7 @@ env:
global:
- CRATE_NAME=tantivy
- TRAVIS_CARGO_NIGHTLY_FEATURE=""
# - secure: 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
- secure: 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
addons:
apt:
@@ -38,21 +38,20 @@ matrix:
# Linux
#- env: TARGET=aarch64-unknown-linux-gnu
#- env: TARGET=i686-unknown-linux-gnu
- env: TARGET=x86_64-unknown-linux-gnu CODECOV=1 #UPLOAD_DOCS=1
- env: TARGET=x86_64-unknown-linux-gnu CODECOV=1
# - env: TARGET=x86_64-unknown-linux-musl CODECOV=1
# OSX
#- env: TARGET=x86_64-apple-darwin
# os: osx
- env: TARGET=x86_64-apple-darwin
os: osx
before_install:
- set -e
- rustup self update
- rustup component add rustfmt
install:
- sh ci/install.sh
- source ~/.cargo/env || true
- env | grep "TRAVIS"
before_script:
- export PATH=$HOME/.cargo/bin:$PATH
@@ -61,25 +60,14 @@ before_script:
script:
- bash ci/script.sh
- cargo fmt --all -- --check
before_deploy:
- sh ci/before_deploy.sh
after_success:
# Needs GH_TOKEN env var to be set in travis settings
- if [[ -v GH_TOKEN ]]; then echo "GH TOKEN IS SET"; else echo "GH TOKEN NOT SET"; fi
- if [[ -v UPLOAD_DOCS ]]; then cargo doc; cargo doc-upload; else echo "doc upload disabled."; fi
#cache: cargo
#before_cache:
# # Travis can't cache files that are not readable by "others"
# - chmod -R a+r $HOME/.cargo
# - find ./target/debug -type f -maxdepth 1 -delete
# - rm -f ./target/.rustc_info.json
# - rm -fr ./target/debug/{deps,.fingerprint}/tantivy*
# - rm -r target/debug/examples/
# - ls -1 examples/ | sed -e 's/\.rs$//' | xargs -I "{}" find target/* -name "*{}*" -type f -delete
cache: cargo
before_cache:
# Travis can't cache files that are not readable by "others"
- chmod -R a+r $HOME/.cargo
#branches:
# only:
@@ -89,4 +77,4 @@ after_success:
notifications:
email:
on_success: never
on_success: never

156
CHANGELOG.md
View File

@@ -1,164 +1,10 @@
Tantivy 0.11.0
=====================
- Added f64 field. Internally reuse u64 code the same way i64 does (@fdb-hiroshima)
- Various bugfixes in the query parser.
- Better handling of hyphens in query parser. (#609)
- Better handling of whitespaces.
- Closes #498 - add support for Elastic-style unbounded range queries for alphanumeric types eg. "title:>hello", "weight:>=70.5", "height:<200" (@petr-tik)
- API change around `Box<BoxableTokenizer>`. See detail in #629
- Avoid rebuilding Regex automaton whenever a regex query is reused. #639 (@brainlock)
- Add footer with some metadata to index files. #605 (@fdb-hiroshima)
## How to update?
- `Box<dyn BoxableTokenizer>` has been replaced by a `BoxedTokenizer` struct.
- Regex are now compiled when the `RegexQuery` instance is built. As a result, it can now return
an error and handling the `Result` is required.
Tantivy 0.10.2
=====================
- Closes #656. Solving memory leak.
Tantivy 0.10.1
=====================
- Closes #544. A few users experienced problems with the directory watching system.
Avoid watching the mmap directory until someone effectively creates a reader that uses
this functionality.
Tantivy 0.10.0
=====================
*Tantivy 0.10.0 index format is compatible with the index format in 0.9.0.*
- Added an API to easily tweak or entirely replace the
default score. See `TopDocs::tweak_score`and `TopScore::custom_score` (@pmasurel)
- Added an ASCII folding filter (@drusellers)
- Bugfix in `query.count` in presence of deletes (@pmasurel)
- Added `.explain(...)` in `Query` and `Weight` to (@pmasurel)
- Added an efficient way to `delete_all_documents` in `IndexWriter` (@petr-tik).
All segments are simply removed.
Minor
---------
- Switched to Rust 2018 (@uvd)
- Small simplification of the code.
Calling .freq() or .doc() when .advance() has never been called
on segment postings should panic from now on.
- Tokens exceeding `u16::max_value() - 4` chars are discarded silently instead of panicking.
- Fast fields are now preloaded when the `SegmentReader` is created.
- `IndexMeta` is now public. (@hntd187)
- `IndexWriter` `add_document`, `delete_term`. `IndexWriter` is `Sync`, making it possible to use it with a `
Arc<RwLock<IndexWriter>>`. `add_document` and `delete_term` can
only require a read lock. (@pmasurel)
- Introducing `Opstamp` as an expressive type alias for `u64`. (@petr-tik)
- Stamper now relies on `AtomicU64` on all platforms (@petr-tik)
- Bugfix - Files get deleted slightly earlier
- Compilation resources improved (@fdb-hiroshima)
## How to update?
Your program should be usable as is.
### Fast fields
Fast fields used to be accessed directly from the `SegmentReader`.
The API changed, you are now required to acquire your fast field reader via the
`segment_reader.fast_fields()`, and use one of the typed method:
- `.u64()`, `.i64()` if your field is single-valued ;
- `.u64s()`, `.i64s()` if your field is multi-valued ;
- `.bytes()` if your field is bytes fast field.
Tantivy 0.9.0
=====================
*0.9.0 index format is not compatible with the
previous index format.*
- MAJOR BUGFIX :
Some `Mmap` objects were being leaked, and would never get released. (@fulmicoton)
- Removed most unsafe (@fulmicoton)
- Indexer memory footprint improved. (VInt comp, inlining the first block. (@fulmicoton)
- Stemming in other language possible (@pentlander)
- Segments with no docs are deleted earlier (@barrotsteindev)
- Added grouped add and delete operations.
They are guaranteed to happen together (i.e. they cannot be split by a commit).
In addition, adds are guaranteed to happen on the same segment. (@elbow-jason)
- Removed `INT_STORED` and `INT_INDEXED`. It is now possible to use `STORED` and `INDEXED`
for int fields. (@fulmicoton)
- Added DateTime field (@barrotsteindev)
- Added IndexReader. By default, index is reloaded automatically upon new commits (@fulmicoton)
- SIMD linear search within blocks (@fulmicoton)
## How to update ?
tantivy 0.9 brought some API breaking change.
To update from tantivy 0.8, you will need to go through the following steps.
- `schema::INT_INDEXED` and `schema::INT_STORED` should be replaced by `schema::INDEXED` and `schema::INT_STORED`.
- The index now does not hold the pool of searcher anymore. You are required to create an intermediary object called
`IndexReader` for this.
```rust
// create the reader. You typically need to create 1 reader for the entire
// lifetime of you program.
let reader = index.reader()?;
// Acquire a searcher (previously `index.searcher()`) is now written:
let searcher = reader.searcher();
// With the default setting of the reader, you are not required to
// call `index.load_searchers()` anymore.
//
// The IndexReader will pick up that change automatically, regardless
// of whether the update was done in a different process or not.
// If this behavior is not wanted, you can create your reader with
// the `ReloadPolicy::Manual`, and manually decide when to reload the index
// by calling `reader.reload()?`.
```
Tantivy 0.8.2
=====================
Fixing build for x86_64 platforms. (#496)
No need to update from 0.8.1 if tantivy
is building on your platform.
Tantivy 0.8.1
=====================
Hotfix of #476.
Merge was reflecting deletes before commit was passed.
Thanks @barrotsteindev for reporting the bug.
Tantivy 0.8.0
=====================
*No change in the index format*
- API Breaking change in the collector API. (@jwolfe, @fulmicoton)
- Multithreaded search (@jwolfe, @fulmicoton)
Tantivy 0.7.1
=====================
*No change in the index format*
- Bugfix: NGramTokenizer panics on non ascii chars
- Added a space usage API
Tantivy 0.7
=====================
- Skip data for doc ids and positions (@fulmicoton),
greatly improving performance
- Tantivy error now rely on the failure crate (@drusellers)
- Added support for `AND`, `OR`, `NOT` syntax in addition to the `+`,`-` syntax
- Added a snippet generator with highlight (@vigneshsarma, @fulmicoton)
- Added a `TopFieldCollector` (@pentlander)
Tantivy 0.6.1
=========================

77
Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "tantivy"
version = "0.11.0"
version = "0.7.0-dev"
authors = ["Paul Masurel <paul.masurel@gmail.com>"]
license = "MIT"
categories = ["database-implementations", "data-structures"]
@@ -10,95 +10,62 @@ homepage = "https://github.com/tantivy-search/tantivy"
repository = "https://github.com/tantivy-search/tantivy"
readme = "README.md"
keywords = ["search", "information", "retrieval"]
edition = "2018"
[dependencies]
base64 = "0.10.0"
base64 = "0.9.1"
byteorder = "1.0"
crc32fast = "1.2.0"
once_cell = "1.0"
regex ={version = "1.3.0", default-features = false, features = ["std"]}
tantivy-fst = "0.1"
memmap = {version = "0.7", optional=true}
lazy_static = "1"
tinysegmenter = "0.1.0"
regex = "1.0"
fst = {version="0.3", default-features=false}
fst-regex = { version="0.2" }
lz4 = {version="1.20", optional=true}
snap = {version="0.2"}
atomicwrites = {version="0.2.2", optional=true}
tempfile = "3.0"
log = "0.4"
combine = "3"
tempdir = "0.3"
serde = "1.0"
serde_derive = "1.0"
serde_json = "1.0"
num_cpus = "1.2"
fs2={version="0.4", optional=true}
itertools = "0.8"
itertools = "0.7"
levenshtein_automata = {version="0.1", features=["fst_automaton"]}
notify = {version="4", optional=true}
bit-set = "0.5"
uuid = { version = "0.7.2", features = ["v4", "serde"] }
crossbeam = "0.7"
uuid = { version = "0.6", features = ["v4", "serde"] }
crossbeam = "0.4"
crossbeam-channel = "0.2"
futures = "0.1"
futures-cpupool = "0.1"
owning_ref = "0.4"
stable_deref_trait = "1.0.0"
rust-stemmers = "1.1"
downcast-rs = { version="1.0" }
tantivy-query-grammar = { path="./query-grammar" }
bitpacking = {version="0.8", default-features = false, features=["bitpacker4x"]}
census = "0.2"
rust-stemmers = "1"
downcast = { version="0.9" }
matches = "0.1"
bitpacking = "0.5"
census = "0.1"
fnv = "1.0.6"
owned-read = "0.4"
failure = "0.1"
htmlescape = "0.3.1"
fail = "0.3"
scoped-pool = "1.0"
murmurhash32 = "0.2"
chrono = "0.4"
smallvec = "0.6"
[target.'cfg(windows)'.dependencies]
winapi = "0.3"
winapi = "0.2"
[dev-dependencies]
rand = "0.7"
maplit = "1"
matches = "0.1.8"
time = "0.1.42"
rand = "0.5"
[profile.release]
opt-level = 3
debug = false
lto = true
debug-assertions = false
[profile.test]
debug-assertions = true
overflow-checks = true
[features]
default = ["mmap"]
mmap = ["atomicwrites", "fs2", "memmap", "notify"]
mmap = ["fst/mmap", "atomicwrites"]
lz4-compression = ["lz4"]
failpoints = ["fail/failpoints"]
unstable = [] # useful for benches.
wasm-bindgen = ["uuid/wasm-bindgen"]
[workspace]
members = ["query-grammar"]
[badges]
travis-ci = { repository = "tantivy-search/tantivy" }
[dev-dependencies.fail]
version = "0.3"
features = ["failpoints"]
# Following the "fail" crate best practises, we isolate
# tests that define specific behavior in fail check points
# in a different binary.
#
# We do that because, fail rely on a global definition of
# failpoints behavior and hence, it is incompatible with
# multithreading.
[[test]]
name = "failpoints"
path = "tests/failpoints/mod.rs"
required-features = ["fail/failpoints"]

3
Makefile
View File

@@ -1,3 +0,0 @@
test:
echo "Run test only... No examples."
cargo test --tests --lib

87
README.md
View File

@@ -4,8 +4,6 @@
[![Join the chat at https://gitter.im/tantivy-search/tantivy](https://badges.gitter.im/tantivy-search/tantivy.svg)](https://gitter.im/tantivy-search/tantivy?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
[![Build status](https://ci.appveyor.com/api/projects/status/r7nb13kj23u8m9pj/branch/master?svg=true)](https://ci.appveyor.com/project/fulmicoton/tantivy/branch/master)
[![Crates.io](https://img.shields.io/crates/v/tantivy.svg)](https://crates.io/crates/tantivy)
[![Say Thanks!](https://img.shields.io/badge/Say%20Thanks-!-1EAEDB.svg)](https://saythanks.io/to/fulmicoton)
![Tantivy](https://tantivy-search.github.io/logo/tantivy-logo.png)
@@ -18,52 +16,40 @@
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/6)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/6)
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/7)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/7)
[![Become a patron](https://c5.patreon.com/external/logo/become_a_patron_button.png)](https://www.patreon.com/fulmicoton)
**Tantivy** is a **full text search engine library** written in rust.
It is closer to [Apache Lucene](https://lucene.apache.org/) than to [Elasticsearch](https://www.elastic.co/products/elasticsearch) and [Apache Solr](https://lucene.apache.org/solr/) in the sense it is not
It is closer to Lucene than to Elastic Search and Solr in the sense it is not
an off-the-shelf search engine server, but rather a crate that can be used
to build such a search engine.
Tantivy is, in fact, strongly inspired by Lucene's design.
# Benchmark
Tantivy is typically faster than Lucene, but the results will depend on
the nature of the queries in your workload.
The following [benchmark](https://tantivy-search.github.io/bench/) break downs
performance for different type of queries / collection.
# Features
- Full-text search
- Configurable tokenizer. (stemming available for 17 latin languages. Third party support for Chinese ([tantivy-jieba](https://crates.io/crates/tantivy-jieba) and [cang-jie](https://crates.io/crates/cang-jie)) and [Japanese](https://crates.io/crates/tantivy-tokenizer-tiny-segmenter)
- Fast (check out the :racehorse: :sparkles: [benchmark](https://tantivy-search.github.io/bench/) :sparkles: :racehorse:)
- Tiny startup time (<10ms), perfect for command line tools
- BM25 scoring (the same as lucene)
- Natural query language `(michael AND jackson) OR "king of pop"`
- Phrase queries search (`"michael jackson"`)
- Basic query language (`+michael +jackson`)
- Phrase queries search (\"michael jackson\"`)
- Incremental indexing
- Multithreaded indexing (indexing English Wikipedia takes < 3 minutes on my desktop)
- Mmap directory
- SIMD integer compression when the platform/CPU includes the SSE2 instruction set.
- Single valued and multivalued u64, i64 and f64 fast fields (equivalent of doc values in Lucene)
- Single valued and multivalued u64 and i64 fast fields (equivalent of doc values in Lucene)
- `&[u8]` fast fields
- Text, i64, u64, f64, dates and hierarchical facet fields
- LZ4 compressed document store
- Range queries
- Faceted search
- Configurable indexing (optional term frequency and position indexing)
- Configurable indexing (optional term frequency and position indexing
- Cheesy logo with a horse
# Non-features
- Distributed search is out of the scope of tantivy. That being said, tantivy is meant as a
library upon which one could build a distributed search. Serializable/mergeable collector state for instance,
are within the scope of tantivy.
- Distributed search and will not be in the scope of tantivy.
# Supported OS and compiler
@@ -71,68 +57,27 @@ Tantivy works on stable rust (>= 1.27) and supports Linux, MacOS and Windows.
# Getting started
- [tantivy's simple search example](https://tantivy-search.github.io/examples/basic_search.html)
- [tantivy's simple search example](http://fulmicoton.com/tantivy-examples/simple_search.html)
- [tantivy-cli and its tutorial](https://github.com/tantivy-search/tantivy-cli).
`tantivy-cli` is an actual command line interface that makes it easy for you to create a search engine,
index documents and search via the CLI or a small server with a REST API.
It will walk you through getting a wikipedia search engine up and running in a few minutes.
- [reference doc for the last released version](https://docs.rs/tantivy/)
- [reference doc]
- [For the last released version](https://docs.rs/tantivy/)
- [For the last master branch](https://tantivy-search.github.io/tantivy/tantivy/index.html)
# How can I support this project?
# Compiling
There are many ways to support this project.
- Use tantivy and tell us about your experience on [gitter](https://gitter.im/tantivy-search/tantivy) or by email (paul.masurel@gmail.com)
- Report bugs
- Write a blog post
- Help with documentation by asking questions or submitting PRs
- Contribute code (you can join [our gitter](https://gitter.im/tantivy-search/tantivy) )
- Talk about tantivy around you
- Drop a word on on [![Say Thanks!](https://img.shields.io/badge/Say%20Thanks-!-1EAEDB.svg)](https://saythanks.io/to/fulmicoton) or even [![Become a patron](https://c5.patreon.com/external/logo/become_a_patron_button.png)](https://www.patreon.com/fulmicoton)
# Contributing code
We use the GitHub Pull Request workflow - reference a GitHub ticket and/or include a comprehensive commit message when opening a PR.
## Clone and build locally
## Development
Tantivy compiles on stable rust but requires `Rust >= 1.27`.
To check out and run tests, you can simply run :
```bash
git clone https://github.com/tantivy-search/tantivy.git
git clone git@github.com:tantivy-search/tantivy.git
cd tantivy
cargo build
```
## Run tests
Some tests will not run with just `cargo test` because of `fail-rs`.
To run the tests exhaustively, run `./run-tests.sh`
# Contribute
## Debug
You might find it useful to step through the programme with a debugger.
### A failing test
Make sure you haven't run `cargo clean` after the most recent `cargo test` or `cargo build` to guarantee that `target/` dir exists. Use this bash script to find the most name of the most recent debug build of tantivy and run it under rust-gdb.
```bash
find target/debug/ -maxdepth 1 -executable -type f -name "tantivy*" -printf '%TY-%Tm-%Td %TT %p\n' | sort -r | cut -d " " -f 3 | xargs -I RECENT_DBG_TANTIVY rust-gdb RECENT_DBG_TANTIVY
```
Now that you are in rust-gdb, you can set breakpoints on lines and methods that match your source-code and run the debug executable with flags that you normally pass to `cargo test` to like this
```bash
$gdb run --test-threads 1 --test $NAME_OF_TEST
```
### An example
By default, rustc compiles everything in the `examples/` dir in debug mode. This makes it easy for you to make examples to reproduce bugs.
```bash
rust-gdb target/debug/examples/$EXAMPLE_NAME
$ gdb run
```
Send me an email (paul.masurel at gmail.com) if you want to contribute to tantivy.

4
appveyor.yml
View File

@@ -18,5 +18,5 @@ install:
build: false
test_script:
- REM SET RUST_LOG=tantivy,test & cargo test --verbose --no-default-features --features mmap
- REM SET RUST_BACKTRACE=1 & cargo build --examples
- REM SET RUST_LOG=tantivy,test & cargo test --verbose
- REM SET RUST_BACKTRACE=1 & cargo build --examples

6
ci/script.sh
View File

@@ -7,16 +7,16 @@ set -ex
main() {
if [ ! -z $CODECOV ]; then
echo "Codecov"
cargo build --verbose && cargo coverage --verbose --all && bash <(curl -s https://codecov.io/bash) -s target/kcov
cargo build --verbose && cargo coverage --verbose && bash <(curl -s https://codecov.io/bash) -s target/kcov
else
echo "Build"
cross build --target $TARGET
cross build --target $TARGET --release
if [ ! -z $DISABLE_TESTS ]; then
return
fi
echo "Test"
cross test --target $TARGET --no-default-features --features mmap
cross test --target $TARGET --no-default-features --features mmap query-grammar
cross test --target $TARGET
fi
for example in $(ls examples/*.rs)
do

1
doc/src/SUMMARY.md
View File

@@ -9,7 +9,6 @@
- [Facetting](./facetting.md)
- [Innerworkings](./innerworkings.md)
- [Inverted index](./inverted_index.md)
- [Best practise](./inverted_index.md)
[Frequently Asked Questions](./faq.md)
[Examples](./examples.md)

29
doc/src/avant-propos.md
View File

@@ -2,8 +2,8 @@
> Tantivy is a **search** engine **library** for Rust.
If you are familiar with Lucene, it's an excellent approximation to consider tantivy as Lucene for rust. tantivy is heavily inspired by Lucene's design and
they both have the same scope and targetted use cases.
If you are familiar with Lucene, tantivy is heavily inspired by Lucene's design and
they both have the same scope and targetted users.
If you are not familiar with Lucene, let's break down our little tagline.
@@ -17,18 +17,15 @@ relevancy, collapsing, highlighting, spatial search.
experience. But keep in mind this is just a toolbox.
Which bring us to the second keyword...
- **Library** means that you will have to write code. tantivy is not an *all-in-one* server solution like elastic search for instance.
Sometimes a functionality will not be available in tantivy because it is too
specific to your use case. By design, tantivy should make it possible to extend
the available set of features using the existing rock-solid datastructures.
Most frequently this will mean writing your own `Collector`, your own `Scorer` or your own
`TokenFilter`... Some of your requirements may also be related to
something closer to architecture or operations. For instance, you may
want to build a large corpus on Hadoop, fine-tune the merge policy to keep your
index sharded in a time-wise fashion, or you may want to convert and existing
index from a different format.
Tantivy exposes a lot of low level API to do all of these things.
- **Library** means that you will have to write code. tantivy is not an *all-in-one* server solution.
Sometimes a functionality will not be available in tantivy because it is too specific to your use case. By design, tantivy should make it possible to extend
the available set of features using the existing rock-solid datastructures.
Most frequently this will mean writing your own `Collector`, your own `Scorer` or your own
`Tokenizer/TokenFilter`... But some of your requirement may also be related to
architecture or operations. For instance, you may want to build a large corpus on Hadoop,
fine-tune the merge policy to keep your index sharded in a time-wise fashion, or you may want
to convert and existing index from a different format.
Tantivy exposes its API to do all of these things.

81
doc/src/basis.md
View File

@@ -2,76 +2,47 @@
## Straight from disk
Tantivy accesses its data using an abstracting trait called `Directory`.
In theory, one can come and override the data access logic. In practise, the
trait somewhat assumes that your data can be mapped to memory, and tantivy
seems deeply married to using `mmap` for its io [^1], and the only persisting
directory shipped with tantivy is the `MmapDirectory`.
By default, tantivy accesses its data using its `MMapDirectory`.
While this design has some downsides, this greatly simplifies the source code of tantivy,
and entirely delegates the caching to the OS.
While this design has some downsides, this greatly simplifies the source code of
tantivy. Caching is also entirely delegated to the OS.
`tantivy` works entirely (or almost) by directly reading the datastructures as they are layed on disk.
As a result, the act of opening an indexing does not involve loading different datastructures
from the disk into random access memory : starting a process, opening an index, and performing a query
can typically be done in a matter of milliseconds.
`tantivy` works entirely (or almost) by directly reading the datastructures as they are layed on disk. As a result, the act of opening an indexing does not involve loading different datastructures from the disk into random access memory : starting a process, opening an index, and performing your first query can typically be done in a matter of milliseconds.
This is an interesting property for a command line search engine, or for some multi-tenant log search engine : spawning a new process for each new query can be a perfectly sensible solution in some use case.
This is an interesting property for a command line search engine, or for some multi-tenant log search engine.
Spawning a new process for each new query can be a perfectly sensible solution in some use case.
In later chapters, we will discuss tantivy's inverted index data layout.
One key take away is that to achieve great performance, search indexes are extremely compact.
One key take away is that to achieve great performance, search indexes are extremely compact.
Of course this is crucial to reduce IO, and ensure that as much of our index can sit in RAM.
Also, whenever possible its data is accessed sequentially. Of course, this is an amazing property when tantivy needs to access the data from your spinning hard disk, but this is also
critical for performance, if your data is read from and an `SSD` or even already in your pagecache.
Also, whenever possible the data is accessed sequentially. Of course, this is an amazing property when tantivy needs to access
the data from your spinning hard disk, but this is also a great property when working with `SSD` or `RAM`,
as it makes our read patterns very predictable for the CPU.
## Segments, and the log method
That kind of compact layout comes at one cost: it prevents our datastructures from being dynamic.
In fact, the `Directory` trait does not even allow you to modify part of a file.
That kind compact layout comes at one cost: it prevents our datastructures from being dynamic.
In fact, a trait called `Directory` is in charge of abstracting all of tantivy's data access
and its API does not even allow editing these file once they are written.
To allow the addition / deletion of documents, and create the illusion that
your index is dynamic (i.e.: adding and deleting documents), tantivy uses a common database trick sometimes referred to as the *log method*.
your index is dynamic (i.e.: adding and deleting documents), tantivy uses a common database trick sometimes
referred to as the *log method*.
Let's forget about deletes for a moment.
As you add documents, these documents are processed and stored in a dedicated datastructure, in a `RAM` buffer. This datastructure is not ready for search, but it is useful to receive your data and rearrange it very rapidly.
As you add documents, this buffer will reach its capacity and tantivy will transparently stop adding document to it and start converting this datastructure to its final read-only format on disk. Once written, an brand empty buffer is available to resume adding documents.
Let's forget about deletes for a moment. As you add documents, these documents are processed and stored in
a dedicated datastructure, in a `RAM` buffer. This datastructure is designed to be dynamic but
cannot be accessed for search. As you add documents, this buffer will reach its capacity and tantivy will
transparently stop adding document to it and start converting this datastructure to its final
read-only format on disk. Once written, an brand empty buffer is available to resume adding documents.
The resulting chunk of index obtained after this serialization is called a `Segment`.
> A segment is a self-contained atomic piece of index. It is identified with a UUID, and all of its files are identified using the naming scheme : `<UUID>.*`.
Which brings us to the nature of a tantivy `Index`.
> A tantivy `Index` is a collection of `Segments`.
Physically, this really just means and index is a bunch of segment files in a given `Directory`,
linked together by a `meta.json` file. This transparency can become extremely handy
to get tantivy to fit your use case:
*Example 1* You could for instance use hadoop to build a very large search index in a timely manner, copy all of the resulting segment files in the same directory and edit the `meta.json` to get a functional index.[^2]
*Example 2* You could also disable your merge policy and enforce daily segments. Removing data after one week can then be done very efficiently by just editing the `meta.json` and deleting the files associated to segment `D-7`.
> A segment is a self-contained atomic piece of index. It is identified with a UUID, and all of its files
are identified using the naming scheme : `<UUID>.*`.
# Merging
As you index more and more data, your index will accumulate more and more segments.
Having a lot of small segments is not really optimal. There is a bit of redundancy in having
all these term dictionary. Also when searching, we will need to do term lookups as many times as we have segments. It can hurt search performance a bit.
That's where merging or compacting comes into place. Tantivy will continuously consider merge
opportunities and start merging segments in the background.
# Indexing throughput, number of indexing threads
[^1]: This may eventually change.
[^2]: Be careful however. By default these files will not be considered as *managed* by tantivy. This means they will never be garbage collected by tantivy, regardless of whether they become obsolete or not.
> A tantivy `Index` is a collection of `Segments`.

0
doc/src/best_practise.md.rs
View File

2
doc/src/examples.md
View File

@@ -1,3 +1 @@
# Examples
- [Basic search](/examples/basic_search.html)

136
examples/basic_search.rs
View File

@@ -5,23 +5,27 @@
//
// We will :
// - define our schema
// - create an index in a directory
// - index a few documents into our index
// - search for the best document matching a basic query
// - retrieve the best document's original content.
// = create an index in a directory
// - index few documents in our index
// - search for the best document matchings "sea whale"
// - retrieve the best document original content.
extern crate tempdir;
// ---
// Importing tantivy...
use tantivy::collector::TopDocs;
#[macro_use]
extern crate tantivy;
use tantivy::collector::TopCollector;
use tantivy::query::QueryParser;
use tantivy::schema::*;
use tantivy::{doc, Index, ReloadPolicy};
use tempfile::TempDir;
use tantivy::Index;
fn main() -> tantivy::Result<()> {
// Let's create a temporary directory for the
// sake of this example
let index_path = TempDir::new()?;
let index_path = TempDir::new("tantivy_example_dir")?;
// # Defining the schema
//
@@ -30,8 +34,8 @@ fn main() -> tantivy::Result<()> {
// and for each field, its type and "the way it should
// be indexed".
// First we need to define a schema ...
let mut schema_builder = Schema::builder();
// first we need to define a schema ...
let mut schema_builder = SchemaBuilder::default();
// Our first field is title.
// We want full-text search for it, and we also want
@@ -45,7 +49,7 @@ fn main() -> tantivy::Result<()> {
//
// `STORED` means that the field will also be saved
// in a compressed, row-oriented key-value store.
// This store is useful for reconstructing the
// This store is useful to reconstruct the
// documents that were selected during the search phase.
schema_builder.add_text_field("title", TEXT | STORED);
@@ -54,7 +58,8 @@ fn main() -> tantivy::Result<()> {
// need to be able to be able to retrieve it
// for our application.
//
// We can make our index lighter by omitting the `STORED` flag.
// We can make our index lighter and
// by omitting `STORED` flag.
schema_builder.add_text_field("body", TEXT);
let schema = schema_builder.build();
@@ -67,7 +72,7 @@ fn main() -> tantivy::Result<()> {
// with our schema in the directory.
let index = Index::create_in_dir(&index_path, schema.clone())?;
// To insert a document we will need an index writer.
// To insert document we need an index writer.
// There must be only one writer at a time.
// This single `IndexWriter` is already
// multithreaded.
@@ -101,37 +106,37 @@ fn main() -> tantivy::Result<()> {
// For convenience, tantivy also comes with a macro to
// reduce the boilerplate above.
index_writer.add_document(doc!(
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
));
index_writer.add_document(doc!(
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
));
// Multivalued field just need to be repeated.
index_writer.add_document(doc!(
title => "Frankenstein",
title => "The Modern Prometheus",
body => "You will rejoice to hear that no disaster has accompanied the commencement of an \
enterprise which you have regarded with such evil forebodings. I arrived here \
yesterday, and my first task is to assure my dear sister of my welfare and \
increasing confidence in the success of my undertaking."
title => "Frankenstein",
title => "The Modern Prometheus",
body => "You will rejoice to hear that no disaster has accompanied the commencement of an \
enterprise which you have regarded with such evil forebodings. I arrived here \
yesterday, and my first task is to assure my dear sister of my welfare and \
increasing confidence in the success of my undertaking."
));
// This is an example, so we will only index 3 documents
@@ -145,8 +150,8 @@ fn main() -> tantivy::Result<()> {
// At this point our documents are not searchable.
//
//
// We need to call `.commit()` explicitly to force the
// `index_writer` to finish processing the documents in the queue,
// We need to call .commit() explicitly to force the
// index_writer to finish processing the documents in the queue,
// flush the current index to the disk, and advertise
// the existence of new documents.
//
@@ -158,40 +163,31 @@ fn main() -> tantivy::Result<()> {
// persistently indexed.
//
// In the scenario of a crash or a power failure,
// tantivy behaves as if it has rolled back to its last
// tantivy behaves as if has rolled back to its last
// commit.
// # Searching
//
// ### Searcher
//
// A reader is required first in order to search an index.
// It acts as a `Searcher` pool that reloads itself,
// depending on a `ReloadPolicy`.
//
// For a search server you will typically create one reader for the entire lifetime of your
// program, and acquire a new searcher for every single request.
//
// In the code below, we rely on the 'ON_COMMIT' policy: the reader
// will reload the index automatically after each commit.
let reader = index
.reader_builder()
.reload_policy(ReloadPolicy::OnCommit)
.try_into()?;
// Let's search our index. Start by reloading
// searchers in the index. This should be done
// after every `commit()`.
index.load_searchers()?;
// We now need to acquire a searcher.
//
// A searcher points to a snapshotted, immutable version of the index.
//
// Some search experience might require more than
// one query. Using the same searcher ensures that all of these queries will run on the
// same version of the index.
// one query.
//
// The searcher ensure that we get to work
// with a consistent version of the index.
//
// Acquiring a `searcher` is very cheap.
//
// You should acquire a searcher every time you start processing a request and
// You should acquire a searcher every time you
// start processing a request and
// and release it right after your query is finished.
let searcher = reader.searcher();
let searcher = index.searcher();
// ### Query
@@ -201,7 +197,7 @@ fn main() -> tantivy::Result<()> {
// in both title and body.
let query_parser = QueryParser::for_index(&index, vec![title, body]);
// `QueryParser` may fail if the query is not in the right
// QueryParser may fail if the query is not in the right
// format. For user facing applications, this can be a problem.
// A ticket has been opened regarding this problem.
let query = query_parser.parse_query("sea whale")?;
@@ -217,10 +213,15 @@ fn main() -> tantivy::Result<()> {
//
// We are not interested in all of the documents but
// only in the top 10. Keeping track of our top 10 best documents
// is the role of the `TopDocs` collector.
// is the role of the TopCollector.
let mut top_collector = TopCollector::with_limit(10);
// We can now perform our query.
let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
searcher.search(&*query, &mut top_collector)?;
// Our top collector now contains the 10
// most relevant doc ids...
let doc_addresses = top_collector.docs();
// The actual documents still need to be
// retrieved from Tantivy's store.
@@ -228,10 +229,15 @@ fn main() -> tantivy::Result<()> {
// Since the body field was not configured as stored,
// the document returned will only contain
// a title.
for (_score, doc_address) in top_docs {
let retrieved_doc = searcher.doc(doc_address)?;
for doc_address in doc_addresses {
let retrieved_doc = searcher.doc(&doc_address)?;
println!("{}", schema.to_json(&retrieved_doc));
}
Ok(())
}
use tempdir::TempDir;

191
examples/custom_collector.rs
View File

@@ -1,191 +0,0 @@
// # Custom collector example
//
// This example shows how you can implement your own
// collector. As an example, we will compute a collector
// that computes the standard deviation of a given fast field.
//
// Of course, you can have a look at the tantivy's built-in collectors
// such as the `CountCollector` for more examples.
// ---
// Importing tantivy...
use tantivy::collector::{Collector, SegmentCollector};
use tantivy::fastfield::FastFieldReader;
use tantivy::query::QueryParser;
use tantivy::schema::Field;
use tantivy::schema::{Schema, FAST, INDEXED, TEXT};
use tantivy::{doc, Index, SegmentReader, TantivyError};
#[derive(Default)]
struct Stats {
count: usize,
sum: f64,
squared_sum: f64,
}
impl Stats {
pub fn count(&self) -> usize {
self.count
}
pub fn mean(&self) -> f64 {
self.sum / (self.count as f64)
}
fn square_mean(&self) -> f64 {
self.squared_sum / (self.count as f64)
}
pub fn standard_deviation(&self) -> f64 {
let mean = self.mean();
(self.square_mean() - mean * mean).sqrt()
}
fn non_zero_count(self) -> Option<Stats> {
if self.count == 0 {
None
} else {
Some(self)
}
}
}
struct StatsCollector {
field: Field,
}
impl StatsCollector {
fn with_field(field: Field) -> StatsCollector {
StatsCollector { field }
}
}
impl Collector for StatsCollector {
// That's the type of our result.
// Our standard deviation will be a float.
type Fruit = Option<Stats>;
type Child = StatsSegmentCollector;
fn for_segment(
&self,
_segment_local_id: u32,
segment_reader: &SegmentReader,
) -> tantivy::Result<StatsSegmentCollector> {
let fast_field_reader = segment_reader
.fast_fields()
.u64(self.field)
.ok_or_else(|| {
let field_name = segment_reader.schema().get_field_name(self.field);
TantivyError::SchemaError(format!(
"Field {:?} is not a u64 fast field.",
field_name
))
})?;
Ok(StatsSegmentCollector {
fast_field_reader,
stats: Stats::default(),
})
}
fn requires_scoring(&self) -> bool {
// this collector does not care about score.
false
}
fn merge_fruits(&self, segment_stats: Vec<Option<Stats>>) -> tantivy::Result<Option<Stats>> {
let mut stats = Stats::default();
for segment_stats_opt in segment_stats {
if let Some(segment_stats) = segment_stats_opt {
stats.count += segment_stats.count;
stats.sum += segment_stats.sum;
stats.squared_sum += segment_stats.squared_sum;
}
}
Ok(stats.non_zero_count())
}
}
struct StatsSegmentCollector {
fast_field_reader: FastFieldReader<u64>,
stats: Stats,
}
impl SegmentCollector for StatsSegmentCollector {
type Fruit = Option<Stats>;
fn collect(&mut self, doc: u32, _score: f32) {
let value = self.fast_field_reader.get(doc) as f64;
self.stats.count += 1;
self.stats.sum += value;
self.stats.squared_sum += value * value;
}
fn harvest(self) -> <Self as SegmentCollector>::Fruit {
self.stats.non_zero_count()
}
}
fn main() -> tantivy::Result<()> {
// # Defining the schema
//
// The Tantivy index requires a very strict schema.
// The schema declares which fields are in the index,
// and for each field, its type and "the way it should
// be indexed".
// first we need to define a schema ...
let mut schema_builder = Schema::builder();
// We'll assume a fictional index containing
// products, and with a name, a description, and a price.
let product_name = schema_builder.add_text_field("name", TEXT);
let product_description = schema_builder.add_text_field("description", TEXT);
let price = schema_builder.add_u64_field("price", INDEXED | FAST);
let schema = schema_builder.build();
// # Indexing documents
//
// Lets index a bunch of fake documents for the sake of
// this example.
let index = Index::create_in_ram(schema.clone());
let mut index_writer = index.writer(50_000_000)?;
index_writer.add_document(doc!(
product_name => "Super Broom 2000",
product_description => "While it is ok for short distance travel, this broom \
was designed quiditch. It will up your game.",
price => 30_200u64
));
index_writer.add_document(doc!(
product_name => "Turbulobroom",
product_description => "You might have heard of this broom before : it is the sponsor of the Wales team.\
You'll enjoy its sharp turns, and rapid acceleration",
price => 29_240u64
));
index_writer.add_document(doc!(
product_name => "Broomio",
product_description => "Great value for the price. This broom is a market favorite",
price => 21_240u64
));
index_writer.add_document(doc!(
product_name => "Whack a Mole",
product_description => "Prime quality bat.",
price => 5_200u64
));
index_writer.commit()?;
let reader = index.reader()?;
let searcher = reader.searcher();
let query_parser = QueryParser::for_index(&index, vec![product_name, product_description]);
// here we want to get a hit on the 'ken' in Frankenstein
let query = query_parser.parse_query("broom")?;
if let Some(stats) = searcher.search(&query, &StatsCollector::with_field(price))? {
println!("count: {}", stats.count());
println!("mean: {}", stats.mean());
println!("standard deviation: {}", stats.standard_deviation());
}
Ok(())
}

33
examples/custom_tokenizer.rs
View File

@@ -2,11 +2,16 @@
//
// In this example, we'll see how to define a tokenizer pipeline
// by aligning a bunch of `TokenFilter`.
use tantivy::collector::TopDocs;
#[macro_use]
extern crate tantivy;
use tantivy::collector::TopCollector;
use tantivy::query::QueryParser;
use tantivy::schema::*;
use tantivy::tokenizer::NgramTokenizer;
use tantivy::{doc, Index};
use tantivy::Index;
fn main() -> tantivy::Result<()> {
// # Defining the schema
@@ -17,7 +22,7 @@ fn main() -> tantivy::Result<()> {
// be indexed".
// first we need to define a schema ...
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
// Our first field is title.
// In this example we want to use NGram searching
@@ -65,12 +70,12 @@ fn main() -> tantivy::Result<()> {
// heap for the indexer can increase its throughput.
let mut index_writer = index.writer(50_000_000)?;
index_writer.add_document(doc!(
title => "The Old Man and the Sea",
body => "He was an old man who fished alone in a skiff in the Gulf Stream and \
he had gone eighty-four days now without taking a fish."
title => "The Old Man and the Sea",
body => "He was an old man who fished alone in a skiff in the Gulf Stream and \
he had gone eighty-four days now without taking a fish."
));
index_writer.add_document(doc!(
title => "Of Mice and Men",
title => "Of Mice and Men",
body => r#"A few miles south of Soledad, the Salinas River drops in close to the hillside
bank and runs deep and green. The water is warm too, for it has slipped twinkling
over the yellow sands in the sunlight before reaching the narrow pool. On one
@@ -81,16 +86,16 @@ fn main() -> tantivy::Result<()> {
limbs and branches that arch over the pool"#
));
index_writer.add_document(doc!(
title => "Frankenstein",
title => "Frankenstein",
body => r#"You will rejoice to hear that no disaster has accompanied the commencement of an
enterprise which you have regarded with such evil forebodings. I arrived here
yesterday, and my first task is to assure my dear sister of my welfare and
increasing confidence in the success of my undertaking."#
));
index_writer.commit()?;
index.load_searchers()?;
let reader = index.reader()?;
let searcher = reader.searcher();
let searcher = index.searcher();
// The query parser can interpret human queries.
// Here, if the user does not specify which
@@ -101,10 +106,12 @@ fn main() -> tantivy::Result<()> {
// here we want to get a hit on the 'ken' in Frankenstein
let query = query_parser.parse_query("ken")?;
let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
let mut top_collector = TopCollector::with_limit(10);
searcher.search(&*query, &mut top_collector)?;
for (_, doc_address) in top_docs {
let retrieved_doc = searcher.doc(doc_address)?;
let doc_addresses = top_collector.docs();
for doc_address in doc_addresses {
let retrieved_doc = searcher.doc(&doc_address)?;
println!("{}", schema.to_json(&retrieved_doc));
}

37
examples/deleting_updating_documents.rs
View File

@@ -8,19 +8,19 @@
//
// ---
// Importing tantivy...
use tantivy::collector::TopDocs;
use tantivy::query::TermQuery;
#[macro_use]
extern crate tantivy;
use tantivy::collector::TopCollector;
use tantivy::schema::*;
use tantivy::{doc, Index, IndexReader};
use tantivy::Index;
use tantivy::query::TermQuery;
// A simple helper function to fetch a single document
// given its id from our index.
// It will be helpful to check our work.
fn extract_doc_given_isbn(
reader: &IndexReader,
isbn_term: &Term,
) -> tantivy::Result<Option<Document>> {
let searcher = reader.searcher();
fn extract_doc_given_isbn(index: &Index, isbn_term: &Term) -> tantivy::Result<Option<Document>> {
let searcher = index.searcher();
// This is the simplest query you can think of.
// It matches all of the documents containing a specific term.
@@ -28,10 +28,11 @@ fn extract_doc_given_isbn(
// The second argument is here to tell we don't care about decoding positions,
// or term frequencies.
let term_query = TermQuery::new(isbn_term.clone(), IndexRecordOption::Basic);
let top_docs = searcher.search(&term_query, &TopDocs::with_limit(1))?;
let mut top_collector = TopCollector::with_limit(1);
searcher.search(&term_query, &mut top_collector)?;
if let Some((_score, doc_address)) = top_docs.first() {
let doc = searcher.doc(*doc_address)?;
if let Some(doc_address) = top_collector.docs().first() {
let doc = searcher.doc(doc_address)?;
Ok(Some(doc))
} else {
// no doc matching this ID.
@@ -40,11 +41,12 @@ fn extract_doc_given_isbn(
}
fn main() -> tantivy::Result<()> {
// # Defining the schema
//
// Check out the *basic_search* example if this makes
// small sense to you.
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
// Tantivy does not really have a notion of primary id.
// This may change in the future.
@@ -86,12 +88,12 @@ fn main() -> tantivy::Result<()> {
isbn => "978-9176370711",
));
index_writer.commit()?;
let reader = index.reader()?;
index.load_searchers()?;
let frankenstein_isbn = Term::from_field_text(isbn, "978-9176370711");
// Oops our frankenstein doc seems mispelled
let frankenstein_doc_misspelled = extract_doc_given_isbn(&reader, &frankenstein_isbn)?.unwrap();
let frankenstein_doc_misspelled = extract_doc_given_isbn(&index, &frankenstein_isbn)?.unwrap();
assert_eq!(
schema.to_json(&frankenstein_doc_misspelled),
r#"{"isbn":["978-9176370711"],"title":["Frankentein"]}"#,
@@ -124,20 +126,21 @@ fn main() -> tantivy::Result<()> {
isbn => "978-9176370711",
));
// You are guaranteed that your clients will only observe your index in
// the state it was in after a commit.
// In this example, your search engine will at no point be missing the *Frankenstein* document.
// Everything happened as if the document was updated.
index_writer.commit()?;
// We reload our searcher to make our change available to clients.
reader.reload()?;
index.load_searchers()?;
// No more typo!
let frankenstein_new_doc = extract_doc_given_isbn(&reader, &frankenstein_isbn)?.unwrap();
let frankenstein_new_doc = extract_doc_given_isbn(&index, &frankenstein_isbn)?.unwrap();
assert_eq!(
schema.to_json(&frankenstein_new_doc),
r#"{"isbn":["978-9176370711"],"title":["Frankenstein"]}"#,
);
Ok(())
}
}

74
examples/faceted_search.rs
View File

@@ -10,66 +10,72 @@
// - search for the best document matchings "sea whale"
// - retrieve the best document original content.
extern crate tempdir;
// ---
// Importing tantivy...
#[macro_use]
extern crate tantivy;
use tantivy::collector::FacetCollector;
use tantivy::query::AllQuery;
use tantivy::schema::*;
use tantivy::{doc, Index};
use tempfile::TempDir;
use tantivy::Index;
fn main() -> tantivy::Result<()> {
// Let's create a temporary directory for the
// sake of this example
let index_path = TempDir::new()?;
let mut schema_builder = Schema::builder();
// Let's create a temporary directory for the
// sake of this example
let index_path = TempDir::new("tantivy_facet_example_dir")?;
let mut schema_builder = SchemaBuilder::default();
schema_builder.add_text_field("name", TEXT | STORED);
schema_builder.add_text_field("name", TEXT | STORED);
// this is our faceted field
schema_builder.add_facet_field("tags");
// this is our faceted field
schema_builder.add_facet_field("tags");
let schema = schema_builder.build();
let schema = schema_builder.build();
let index = Index::create_in_dir(&index_path, schema.clone())?;
let index = Index::create_in_dir(&index_path, schema.clone())?;
let mut index_writer = index.writer(50_000_000)?;
let mut index_writer = index.writer(50_000_000)?;
let name = schema.get_field("name").unwrap();
let tags = schema.get_field("tags").unwrap();
let name = schema.get_field("name").unwrap();
let tags = schema.get_field("tags").unwrap();
// For convenience, tantivy also comes with a macro to
// reduce the boilerplate above.
index_writer.add_document(doc!(
// For convenience, tantivy also comes with a macro to
// reduce the boilerplate above.
index_writer.add_document(doc!(
name => "the ditch",
tags => Facet::from("/pools/north")
));
index_writer.add_document(doc!(
index_writer.add_document(doc!(
name => "little stacey",
tags => Facet::from("/pools/south")
));
index_writer.commit()?;
index_writer.commit()?;
let reader = index.reader()?;
index.load_searchers()?;
let searcher = reader.searcher();
let searcher = index.searcher();
let mut facet_collector = FacetCollector::for_field(tags);
facet_collector.add_facet("/pools");
let mut facet_collector = FacetCollector::for_field(tags);
facet_collector.add_facet("/pools");
let facet_counts = searcher.search(&AllQuery, &facet_collector).unwrap();
searcher.search(&AllQuery, &mut facet_collector).unwrap();
// This lists all of the facet counts
let facets: Vec<(&Facet, u64)> = facet_counts.get("/pools").collect();
assert_eq!(
facets,
vec![
(&Facet::from("/pools/north"), 1),
(&Facet::from("/pools/south"), 1),
]
);
let counts = facet_collector.harvest();
// This lists all of the facet counts
let facets: Vec<(&Facet, u64)> = counts.get("/pools").collect();
assert_eq!(
facets,
vec![
(&Facet::from("/pools/north"), 1),
(&Facet::from("/pools/south"), 1)
]
);
Ok(())
Ok(())
}
use tempdir::TempDir;

39
examples/integer_range_search.rs
View File

@@ -1,39 +0,0 @@
// # Searching a range on an indexed int field.
//
// Below is an example of creating an indexed integer field in your schema
// You can use RangeQuery to get a Count of all occurrences in a given range.
use tantivy::collector::Count;
use tantivy::query::RangeQuery;
use tantivy::schema::{Schema, INDEXED};
use tantivy::{doc, Index, Result};
fn run() -> Result<()> {
// For the sake of simplicity, this schema will only have 1 field
let mut schema_builder = Schema::builder();
// `INDEXED` is a short-hand to indicate that our field should be "searchable".
let year_field = schema_builder.add_u64_field("year", INDEXED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let reader = index.reader()?;
{
let mut index_writer = index.writer_with_num_threads(1, 6_000_000)?;
for year in 1950u64..2019u64 {
index_writer.add_document(doc!(year_field => year));
}
index_writer.commit()?;
// The index will be a range of years
}
reader.reload()?;
let searcher = reader.searcher();
// The end is excluded i.e. here we are searching up to 1969
let docs_in_the_sixties = RangeQuery::new_u64(year_field, 1960..1970);
// Uses a Count collector to sum the total number of docs in the range
let num_60s_books = searcher.search(&docs_in_the_sixties, &Count)?;
assert_eq!(num_60s_books, 10);
Ok(())
}
fn main() {
run().unwrap()
}

29
examples/iterating_docs_and_positions.rs
View File

@@ -7,15 +7,21 @@
// the list of documents containing a term, getting
// its term frequency, and accessing its positions.
// ---
// Importing tantivy...
#[macro_use]
extern crate tantivy;
use tantivy::schema::*;
use tantivy::{doc, DocId, DocSet, Index, Postings};
use tantivy::Index;
use tantivy::{DocSet, DocId, Postings};
fn main() -> tantivy::Result<()> {
// We first create a schema for the sake of the
// example. Check the `basic_search` example for more information.
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
// For this example, we need to make sure to index positions for our title
// field. `TEXT` precisely does this.
@@ -30,9 +36,9 @@ fn main() -> tantivy::Result<()> {
index_writer.add_document(doc!(title => "The modern Promotheus"));
index_writer.commit()?;
let reader = index.reader()?;
index.load_searchers()?;
let searcher = reader.searcher();
let searcher = index.searcher();
// A tantivy index is actually a collection of segments.
// Similarly, a searcher just wraps a list `segment_reader`.
@@ -41,6 +47,7 @@ fn main() -> tantivy::Result<()> {
// there is actually only one segment here, but let's iterate through the list
// anyway)
for segment_reader in searcher.segment_readers() {
// A segment contains different data structure.
// Inverted index stands for the combination of
// - the term dictionary
@@ -51,18 +58,19 @@ fn main() -> tantivy::Result<()> {
// Let's go through all docs containing the term `title:the` and access their position
let term_the = Term::from_field_text(title, "the");
// This segment posting object is like a cursor over the documents matching the term.
// The `IndexRecordOption` arguments tells tantivy we will be interested in both term frequencies
// and positions.
//
// If you don't need all this information, you may get better performance by decompressing less
// information.
if let Some(mut segment_postings) =
inverted_index.read_postings(&term_the, IndexRecordOption::WithFreqsAndPositions)
{
if let Some(mut segment_postings) = inverted_index.read_postings(&term_the, IndexRecordOption::WithFreqsAndPositions) {
// this buffer will be used to request for positions
let mut positions: Vec<u32> = Vec::with_capacity(100);
while segment_postings.advance() {
// the number of time the term appears in the document.
let doc_id: DocId = segment_postings.doc(); //< do not try to access this before calling advance once.
@@ -90,6 +98,7 @@ fn main() -> tantivy::Result<()> {
}
}
// A `Term` is a text token associated with a field.
// Let's go through all docs containing the term `title:the` and access their position
let term_the = Term::from_field_text(title, "the");
@@ -102,6 +111,7 @@ fn main() -> tantivy::Result<()> {
// Also, for some VERY specific high performance use case like an OLAP analysis of logs,
// you can get better performance by accessing directly the blocks of doc ids.
for segment_reader in searcher.segment_readers() {
// A segment contains different data structure.
// Inverted index stands for the combination of
// - the term dictionary
@@ -114,9 +124,7 @@ fn main() -> tantivy::Result<()> {
//
// If you don't need all this information, you may get better performance by decompressing less
// information.
if let Some(mut block_segment_postings) =
inverted_index.read_block_postings(&term_the, IndexRecordOption::Basic)
{
if let Some(mut block_segment_postings) = inverted_index.read_block_postings(&term_the, IndexRecordOption::Basic) {
while block_segment_postings.advance() {
// Once again these docs MAY contains deleted documents as well.
let docs = block_segment_postings.docs();
@@ -128,3 +136,4 @@ fn main() -> tantivy::Result<()> {
Ok(())
}

100
examples/multiple_producer.rs
View File

@@ -1,100 +0,0 @@
// # Indexing from different threads.
//
// It is fairly common to have to index from different threads.
// Tantivy forbids to create more than one `IndexWriter` at a time.
//
// This `IndexWriter` itself has its own multithreaded layer, so managing your own
// indexing threads will not help. However, it can still be useful for some applications.
//
// For instance, if preparing documents to send to tantivy before indexing is the bottleneck of
// your application, it is reasonable to have multiple threads.
//
// Another very common reason to want to index from multiple threads, is implementing a webserver
// with CRUD capabilities. The server framework will most likely handle request from
// different threads.
//
// The recommended way to address both of these use case is to wrap your `IndexWriter` into a
// `Arc<RwLock<IndexWriter>>`.
//
// While this is counterintuitive, adding and deleting documents do not require mutability
// over the `IndexWriter`, so several threads will be able to do this operation concurrently.
//
// The example below does not represent an actual real-life use case (who would spawn thread to
// index a single document?), but aims at demonstrating the mechanism that makes indexing
// from several threads possible.
// ---
// Importing tantivy...
use std::sync::{Arc, RwLock};
use std::thread;
use std::time::Duration;
use tantivy::schema::{Schema, STORED, TEXT};
use tantivy::{doc, Index, IndexWriter, Opstamp};
fn main() -> tantivy::Result<()> {
// # Defining the schema
let mut schema_builder = Schema::builder();
let title = schema_builder.add_text_field("title", TEXT | STORED);
let body = schema_builder.add_text_field("body", TEXT);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let index_writer: Arc<RwLock<IndexWriter>> = Arc::new(RwLock::new(index.writer(50_000_000)?));
// # First indexing thread.
let index_writer_clone_1 = index_writer.clone();
thread::spawn(move || {
// we index 100 times the document... for the sake of the example.
for i in 0..100 {
let opstamp = index_writer_clone_1
.read().unwrap() //< A read lock is sufficient here.
.add_document(
doc!(
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
));
println!("add doc {} from thread 1 - opstamp {}", i, opstamp);
thread::sleep(Duration::from_millis(20));
}
});
// # Second indexing thread.
let index_writer_clone_2 = index_writer.clone();
// For convenience, tantivy also comes with a macro to
// reduce the boilerplate above.
thread::spawn(move || {
// we index 100 times the document... for the sake of the example.
for i in 0..100 {
// A read lock is sufficient here.
let opstamp = {
let index_writer_rlock = index_writer_clone_2.read().unwrap();
index_writer_rlock.add_document(doc!(
title => "Manufacturing consent",
body => "Some great book description..."
))
};
println!("add doc {} from thread 2 - opstamp {}", i, opstamp);
thread::sleep(Duration::from_millis(10));
}
});
// # In the main thread, we commit 10 times, once every 500ms.
for _ in 0..10 {
let opstamp: Opstamp = {
// Committing or rollbacking on the other hand requires write lock. This will block other threads.
let mut index_writer_wlock = index_writer.write().unwrap();
index_writer_wlock.commit().unwrap()
};
println!("committed with opstamp {}", opstamp);
thread::sleep(Duration::from_millis(500));
}
Ok(())
}

82
examples/snippet.rs
View File

@@ -1,82 +0,0 @@
// # Snippet example
//
// This example shows how to return a representative snippet of
// your hit result.
// Snippet are an extracted of a target document, and returned in HTML format.
// The keyword searched by the user are highlighted with a `<b>` tag.
// ---
// Importing tantivy...
use tantivy::collector::TopDocs;
use tantivy::query::QueryParser;
use tantivy::schema::*;
use tantivy::{doc, Index, Snippet, SnippetGenerator};
use tempfile::TempDir;
fn main() -> tantivy::Result<()> {
// Let's create a temporary directory for the
// sake of this example
let index_path = TempDir::new()?;
// # Defining the schema
let mut schema_builder = Schema::builder();
let title = schema_builder.add_text_field("title", TEXT | STORED);
let body = schema_builder.add_text_field("body", TEXT | STORED);
let schema = schema_builder.build();
// # Indexing documents
let index = Index::create_in_dir(&index_path, schema.clone())?;
let mut index_writer = index.writer(50_000_000)?;
// we'll only need one doc for this example.
index_writer.add_document(doc!(
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
));
// ...
index_writer.commit()?;
let reader = index.reader()?;
let searcher = reader.searcher();
let query_parser = QueryParser::for_index(&index, vec![title, body]);
let query = query_parser.parse_query("sycamore spring")?;
let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
let snippet_generator = SnippetGenerator::create(&searcher, &*query, body)?;
for (score, doc_address) in top_docs {
let doc = searcher.doc(doc_address)?;
let snippet = snippet_generator.snippet_from_doc(&doc);
println!("Document score {}:", score);
println!("title: {}", doc.get_first(title).unwrap().text().unwrap());
println!("snippet: {}", snippet.to_html());
println!("custom highlighting: {}", highlight(snippet));
}
Ok(())
}
fn highlight(snippet: Snippet) -> String {
let mut result = String::new();
let mut start_from = 0;
for (start, end) in snippet.highlighted().iter().map(|h| h.bounds()) {
result.push_str(&snippet.fragments()[start_from..start]);
result.push_str(" --> ");
result.push_str(&snippet.fragments()[start..end]);
result.push_str(" <-- ");
start_from = end;
}
result.push_str(&snippet.fragments()[start_from..]);
result
}

156
examples/stop_words.rs
View File

@@ -9,105 +9,121 @@
// - add a few stop words
// - index few documents in our index
extern crate tempdir;
// ---
// Importing tantivy...
use tantivy::collector::TopDocs;
#[macro_use]
extern crate tantivy;
use tantivy::collector::TopCollector;
use tantivy::query::QueryParser;
use tantivy::schema::*;
use tantivy::tokenizer::*;
use tantivy::{doc, Index};
use tantivy::Index;
fn main() -> tantivy::Result<()> {
// this example assumes you understand the content in `basic_search`
let mut schema_builder = Schema::builder();
// this example assumes you understand the content in `basic_search`
let index_path = TempDir::new("tantivy_stopwords_example_dir")?;
let mut schema_builder = SchemaBuilder::default();
// This configures your custom options for how tantivy will
// store and process your content in the index; The key
// to note is that we are setting the tokenizer to `stoppy`
// which will be defined and registered below.
let text_field_indexing = TextFieldIndexing::default()
.set_tokenizer("stoppy")
.set_index_option(IndexRecordOption::WithFreqsAndPositions);
let text_options = TextOptions::default()
.set_indexing_options(text_field_indexing)
.set_stored();
// This configures your custom options for how tantivy will
// store and process your content in the index; The key
// to note is that we are setting the tokenizer to `stoppy`
// which will be defined and registered below.
let text_field_indexing = TextFieldIndexing::default()
.set_tokenizer("stoppy")
.set_index_option(IndexRecordOption::WithFreqsAndPositions);
let text_options = TextOptions::default()
.set_indexing_options(text_field_indexing)
.set_stored();
// Our first field is title.
schema_builder.add_text_field("title", text_options);
// Our first field is title.
schema_builder.add_text_field("title", text_options);
// Our second field is body.
let text_field_indexing = TextFieldIndexing::default()
.set_tokenizer("stoppy")
.set_index_option(IndexRecordOption::WithFreqsAndPositions);
let text_options = TextOptions::default()
.set_indexing_options(text_field_indexing)
.set_stored();
schema_builder.add_text_field("body", text_options);
// Our second field is body.
let text_field_indexing = TextFieldIndexing::default()
.set_tokenizer("stoppy")
.set_index_option(IndexRecordOption::WithFreqsAndPositions);
let text_options = TextOptions::default()
.set_indexing_options(text_field_indexing)
.set_stored();
schema_builder.add_text_field("body", text_options);
let schema = schema_builder.build();
let schema = schema_builder.build();
let index = Index::create_in_ram(schema.clone());
let index = Index::create_in_dir(&index_path, schema.clone())?;
// This tokenizer lowers all of the text (to help with stop word matching)
// then removes all instances of `the` and `and` from the corpus
let tokenizer = SimpleTokenizer
.filter(LowerCaser)
.filter(StopWordFilter::remove(vec![
"the".to_string(),
"and".to_string(),
]));
// This tokenizer lowers all of the text (to help with stop word matching)
// then removes all instances of `the` and `and` from the corpus
let tokenizer = SimpleTokenizer
.filter(LowerCaser)
.filter(StopWordFilter::remove(vec![
"the".to_string(),
"and".to_string(),
]));
index.tokenizers().register("stoppy", tokenizer);
index.tokenizers().register("stoppy", tokenizer);
let mut index_writer = index.writer(50_000_000)?;
let mut index_writer = index.writer(50_000_000)?;
let title = schema.get_field("title").unwrap();
let body = schema.get_field("body").unwrap();
let title = schema.get_field("title").unwrap();
let body = schema.get_field("body").unwrap();
index_writer.add_document(doc!(
index_writer.add_document(doc!(
title => "The Old Man and the Sea",
body => "He was an old man who fished alone in a skiff in the Gulf Stream and \
he had gone eighty-four days now without taking a fish."
));
index_writer.add_document(doc!(
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
));
index_writer.add_document(doc!(
title => "Of Mice and Men",
body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
bank and runs deep and green. The water is warm too, for it has slipped twinkling \
over the yellow sands in the sunlight before reaching the narrow pool. On one \
side of the river the golden foothill slopes curve up to the strong and rocky \
Gabilan Mountains, but on the valley side the water is lined with trees—willows \
fresh and green with every spring, carrying in their lower leaf junctures the \
debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
limbs and branches that arch over the pool"
index_writer.add_document(doc!(
title => "Frankenstein",
body => "You will rejoice to hear that no disaster has accompanied the commencement of an \
enterprise which you have regarded with such evil forebodings. I arrived here \
yesterday, and my first task is to assure my dear sister of my welfare and \
increasing confidence in the success of my undertaking."
));
index_writer.add_document(doc!(
title => "Frankenstein",
body => "You will rejoice to hear that no disaster has accompanied the commencement of an \
enterprise which you have regarded with such evil forebodings. I arrived here \
yesterday, and my first task is to assure my dear sister of my welfare and \
increasing confidence in the success of my undertaking."
));
index_writer.commit()?;
index_writer.commit()?;
index.load_searchers()?;
let reader = index.reader()?;
let searcher = index.searcher();
let searcher = reader.searcher();
let query_parser = QueryParser::for_index(&index, vec![title, body]);
let query_parser = QueryParser::for_index(&index, vec![title, body]);
// this will have NO hits because it was filtered out
// because the query is run through the analyzer you
// actually will get an error here because the query becomes
// empty
assert!(query_parser.parse_query("the").is_err());
// stop words are applied on the query as well.
// The following will be equivalent to `title:frankenstein`
let query = query_parser.parse_query("title:\"the Frankenstein\"")?;
let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
// this will have hits
let query = query_parser.parse_query("is")?;
for (score, doc_address) in top_docs {
let retrieved_doc = searcher.doc(doc_address)?;
println!("\n==\nDocument score {}:", score);
println!("{}", schema.to_json(&retrieved_doc));
}
let mut top_collector = TopCollector::with_limit(10);
Ok(())
searcher.search(&*query, &mut top_collector)?;
let doc_addresses = top_collector.docs();
for doc_address in doc_addresses {
let retrieved_doc = searcher.doc(&doc_address)?;
println!("{}", schema.to_json(&retrieved_doc));
}
Ok(())
}
use tempdir::TempDir;

6
examples/working_with_json.rs
View File

@@ -1,4 +1,4 @@
use tantivy;
extern crate tantivy;
use tantivy::schema::*;
// # Document from json
@@ -9,10 +9,10 @@ fn main() -> tantivy::Result<()> {
// Check out the basic example if this is confusing to you.
//
// first we need to define a schema ...
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
schema_builder.add_text_field("title", TEXT | STORED);
schema_builder.add_text_field("body", TEXT);
schema_builder.add_u64_field("year", INDEXED);
schema_builder.add_u64_field("year", INT_INDEXED);
let schema = schema_builder.build();
// Let's assume we have a json-serialized document.

16
query-grammar/Cargo.toml
View File

@@ -1,16 +0,0 @@
[package]
name = "tantivy-query-grammar"
version = "0.11.0"
authors = ["Paul Masurel <paul.masurel@gmail.com>"]
license = "MIT"
categories = ["database-implementations", "data-structures"]
description = """Search engine library"""
documentation = "https://tantivy-search.github.io/tantivy/tantivy/index.html"
homepage = "https://github.com/tantivy-search/tantivy"
repository = "https://github.com/tantivy-search/tantivy"
readme = "README.md"
keywords = ["search", "information", "retrieval"]
edition = "2018"
[dependencies]
combine = ">=3.6.0,<4.0.0"

17
query-grammar/src/lib.rs
View File

@@ -1,17 +0,0 @@
#![recursion_limit = "100"]
mod occur;
mod query_grammar;
mod user_input_ast;
use combine::parser::Parser;
pub use crate::occur::Occur;
use crate::query_grammar::parse_to_ast;
pub use crate::user_input_ast::{UserInputAST, UserInputBound, UserInputLeaf, UserInputLiteral};
pub struct Error;
pub fn parse_query(query: &str) -> Result<UserInputAST, Error> {
let (user_input_ast, _remaining) = parse_to_ast().parse(query).map_err(|_| Error)?;
Ok(user_input_ast)
}

58
query-grammar/src/occur.rs
View File

@@ -1,58 +0,0 @@
use std::fmt;
use std::fmt::Write;
/// Defines whether a term in a query must be present,
/// should be present or must not be present.
#[derive(Debug, Clone, Hash, Copy, Eq, PartialEq)]
pub enum Occur {
/// For a given document to be considered for scoring,
/// at least one of the document with the Should or the Must
/// Occur constraint must be within the document.
Should,
/// Document without the term are excluded from the search.
Must,
/// Document that contain the term are excluded from the
/// search.
MustNot,
}
impl Occur {
/// Returns the one-char prefix symbol for this `Occur`.
/// - `Should` => '?',
/// - `Must` => '+'
/// - `Not` => '-'
fn to_char(self) -> char {
match self {
Occur::Should => '?',
Occur::Must => '+',
Occur::MustNot => '-',
}
}
/// Compose two occur values.
pub fn compose(left: Occur, right: Occur) -> Occur {
match left {
Occur::Should => right,
Occur::Must => {
if right == Occur::MustNot {
Occur::MustNot
} else {
Occur::Must
}
}
Occur::MustNot => {
if right == Occur::MustNot {
Occur::Must
} else {
Occur::MustNot
}
}
}
}
}
impl fmt::Display for Occur {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.write_char(self.to_char())
}
}

380
query-grammar/src/query_grammar.rs
View File

@@ -1,380 +0,0 @@
use super::user_input_ast::*;
use crate::Occur;
use combine::char::*;
use combine::error::StreamError;
use combine::stream::StreamErrorFor;
use combine::*;
parser! {
fn field[I]()(I) -> String
where [I: Stream<Item = char>] {
(
letter(),
many(satisfy(|c: char| c.is_alphanumeric() || c == '_')),
).skip(char(':')).map(|(s1, s2): (char, String)| format!("{}{}", s1, s2))
}
}
parser! {
fn word[I]()(I) -> String
where [I: Stream<Item = char>] {
(
satisfy(|c: char| !c.is_whitespace() && !['-', '`', ':', '{', '}', '"', '[', ']', '(',')'].contains(&c) ),
many(satisfy(|c: char| !c.is_whitespace() && ![':', '{', '}', '"', '[', ']', '(',')'].contains(&c)))
)
.map(|(s1, s2): (char, String)| format!("{}{}", s1, s2))
.and_then(|s: String|
match s.as_str() {
"OR" => Err(StreamErrorFor::<I>::unexpected_static_message("OR")),
"AND" => Err(StreamErrorFor::<I>::unexpected_static_message("AND")),
"NOT" => Err(StreamErrorFor::<I>::unexpected_static_message("NOT")),
_ => Ok(s)
})
}
}
parser! {
fn literal[I]()(I) -> UserInputLeaf
where [I: Stream<Item = char>]
{
let term_val = || {
let phrase = char('"').with(many1(satisfy(|c| c != '"'))).skip(char('"'));
phrase.or(word())
};
let term_val_with_field = negative_number().or(term_val());
let term_query =
(field(), term_val_with_field)
.map(|(field_name, phrase)| UserInputLiteral {
field_name: Some(field_name),
phrase,
});
let term_default_field = term_val().map(|phrase| UserInputLiteral {
field_name: None,
phrase,
});
attempt(term_query)
.or(term_default_field)
.map(UserInputLeaf::from)
}
}
parser! {
fn negative_number[I]()(I) -> String
where [I: Stream<Item = char>]
{
(char('-'), many1(satisfy(char::is_numeric)),
optional((char('.'), many1(satisfy(char::is_numeric)))))
.map(|(s1, s2, s3): (char, String, Option<(char, String)>)| {
if let Some(('.', s3)) = s3 {
format!("{}{}.{}", s1, s2, s3)
} else {
format!("{}{}", s1, s2)
}
})
}
}
parser! {
fn spaces1[I]()(I) -> ()
where [I: Stream<Item = char>] {
skip_many1(space())
}
}
parser! {
/// Function that parses a range out of a Stream
/// Supports ranges like:
/// [5 TO 10], {5 TO 10}, [* TO 10], [10 TO *], {10 TO *], >5, <=10
/// [a TO *], [a TO c], [abc TO bcd}
fn range[I]()(I) -> UserInputLeaf
where [I: Stream<Item = char>] {
let range_term_val = || {
word().or(negative_number()).or(char('*').with(value("*".to_string())))
};
// check for unbounded range in the form of <5, <=10, >5, >=5
let elastic_unbounded_range = (choice([attempt(string(">=")),
attempt(string("<=")),
attempt(string("<")),
attempt(string(">"))])
.skip(spaces()),
range_term_val()).
map(|(comparison_sign, bound): (&str, String)|
match comparison_sign {
">=" => (UserInputBound::Inclusive(bound), UserInputBound::Unbounded),
"<=" => (UserInputBound::Unbounded, UserInputBound::Inclusive(bound)),
"<" => (UserInputBound::Unbounded, UserInputBound::Exclusive(bound)),
">" => (UserInputBound::Exclusive(bound), UserInputBound::Unbounded),
// default case
_ => (UserInputBound::Unbounded, UserInputBound::Unbounded)
});
let lower_bound = (one_of("{[".chars()), range_term_val())
.map(|(boundary_char, lower_bound): (char, String)|
if lower_bound == "*" {
UserInputBound::Unbounded
} else if boundary_char == '{' {
UserInputBound::Exclusive(lower_bound)
} else {
UserInputBound::Inclusive(lower_bound)
});
let upper_bound = (range_term_val(), one_of("}]".chars()))
.map(|(higher_bound, boundary_char): (String, char)|
if higher_bound == "*" {
UserInputBound::Unbounded
} else if boundary_char == '}' {
UserInputBound::Exclusive(higher_bound)
} else {
UserInputBound::Inclusive(higher_bound)
});
// return only lower and upper
let lower_to_upper = (lower_bound.
skip((spaces(),
string("TO"),
spaces())),
upper_bound);
(optional(field()).skip(spaces()),
// try elastic first, if it matches, the range is unbounded
attempt(elastic_unbounded_range).or(lower_to_upper))
.map(|(field, (lower, upper))|
// Construct the leaf from extracted field (optional)
// and bounds
UserInputLeaf::Range {
field,
lower,
upper
})
}
}
fn negate(expr: UserInputAST) -> UserInputAST {
expr.unary(Occur::MustNot)
}
fn must(expr: UserInputAST) -> UserInputAST {
expr.unary(Occur::Must)
}
parser! {
fn leaf[I]()(I) -> UserInputAST
where [I: Stream<Item = char>] {
char('-').with(leaf()).map(negate)
.or(char('+').with(leaf()).map(must))
.or(char('(').with(ast()).skip(char(')')))
.or(char('*').map(|_| UserInputAST::from(UserInputLeaf::All)))
.or(attempt(string("NOT").skip(spaces1()).with(leaf()).map(negate)))
.or(attempt(range().map(UserInputAST::from)))
.or(literal().map(UserInputAST::from))
}
}
#[derive(Clone, Copy)]
enum BinaryOperand {
Or,
And,
}
parser! {
fn binary_operand[I]()(I) -> BinaryOperand
where [I: Stream<Item = char>]
{
string("AND").with(value(BinaryOperand::And))
.or(string("OR").with(value(BinaryOperand::Or)))
}
}
fn aggregate_binary_expressions(
left: UserInputAST,
others: Vec<(BinaryOperand, UserInputAST)>,
) -> UserInputAST {
let mut dnf: Vec<Vec<UserInputAST>> = vec![vec![left]];
for (operator, operand_ast) in others {
match operator {
BinaryOperand::And => {
if let Some(last) = dnf.last_mut() {
last.push(operand_ast);
}
}
BinaryOperand::Or => {
dnf.push(vec![operand_ast]);
}
}
}
if dnf.len() == 1 {
UserInputAST::and(dnf.into_iter().next().unwrap()) //< safe
} else {
let conjunctions = dnf.into_iter().map(UserInputAST::and).collect();
UserInputAST::or(conjunctions)
}
}
parser! {
pub fn ast[I]()(I) -> UserInputAST
where [I: Stream<Item = char>]
{
let operand_leaf = (binary_operand().skip(spaces()), leaf().skip(spaces()));
let boolean_expr = (leaf().skip(spaces().silent()), many1(operand_leaf)).map(
|(left, right)| aggregate_binary_expressions(left,right));
let whitespace_separated_leaves = many1(leaf().skip(spaces().silent()))
.map(|subqueries: Vec<UserInputAST>|
if subqueries.len() == 1 {
subqueries.into_iter().next().unwrap()
} else {
UserInputAST::Clause(subqueries.into_iter().collect())
});
let expr = attempt(boolean_expr).or(whitespace_separated_leaves);
spaces().with(expr).skip(spaces())
}
}
parser! {
pub fn parse_to_ast[I]()(I) -> UserInputAST
where [I: Stream<Item = char>]
{
spaces().with(optional(ast()).skip(eof())).map(|opt_ast| opt_ast.unwrap_or_else(UserInputAST::empty_query))
}
}
#[cfg(test)]
mod test {
use super::*;
fn test_parse_query_to_ast_helper(query: &str, expected: &str) {
let query = parse_to_ast().parse(query).unwrap().0;
let query_str = format!("{:?}", query);
assert_eq!(query_str, expected);
}
fn test_is_parse_err(query: &str) {
assert!(parse_to_ast().parse(query).is_err());
}
#[test]
fn test_parse_empty_to_ast() {
test_parse_query_to_ast_helper("", "<emptyclause>");
}
#[test]
fn test_parse_query_to_ast_hyphen() {
test_parse_query_to_ast_helper("\"www-form-encoded\"", "\"www-form-encoded\"");
test_parse_query_to_ast_helper("www-form-encoded", "\"www-form-encoded\"");
test_parse_query_to_ast_helper("www-form-encoded", "\"www-form-encoded\"");
}
#[test]
fn test_parse_query_to_ast_not_op() {
assert_eq!(
format!("{:?}", parse_to_ast().parse("NOT")),
"Err(UnexpectedParse)"
);
test_parse_query_to_ast_helper("NOTa", "\"NOTa\"");
test_parse_query_to_ast_helper("NOT a", "-(\"a\")");
}
#[test]
fn test_parse_query_to_ast_binary_op() {
test_parse_query_to_ast_helper("a AND b", "(+(\"a\") +(\"b\"))");
test_parse_query_to_ast_helper("a OR b", "(?(\"a\") ?(\"b\"))");
test_parse_query_to_ast_helper("a OR b AND c", "(?(\"a\") ?((+(\"b\") +(\"c\"))))");
test_parse_query_to_ast_helper("a AND b AND c", "(+(\"a\") +(\"b\") +(\"c\"))");
assert_eq!(
format!("{:?}", parse_to_ast().parse("a OR b aaa")),
"Err(UnexpectedParse)"
);
assert_eq!(
format!("{:?}", parse_to_ast().parse("a AND b aaa")),
"Err(UnexpectedParse)"
);
assert_eq!(
format!("{:?}", parse_to_ast().parse("aaa a OR b ")),
"Err(UnexpectedParse)"
);
assert_eq!(
format!("{:?}", parse_to_ast().parse("aaa ccc a OR b ")),
"Err(UnexpectedParse)"
);
}
#[test]
fn test_parse_elastic_query_ranges() {
test_parse_query_to_ast_helper("title: >a", "title:{\"a\" TO \"*\"}");
test_parse_query_to_ast_helper("title:>=a", "title:[\"a\" TO \"*\"}");
test_parse_query_to_ast_helper("title: <a", "title:{\"*\" TO \"a\"}");
test_parse_query_to_ast_helper("title:<=a", "title:{\"*\" TO \"a\"]");
test_parse_query_to_ast_helper("title:<=bsd", "title:{\"*\" TO \"bsd\"]");
test_parse_query_to_ast_helper("weight: >70", "weight:{\"70\" TO \"*\"}");
test_parse_query_to_ast_helper("weight:>=70", "weight:[\"70\" TO \"*\"}");
test_parse_query_to_ast_helper("weight: <70", "weight:{\"*\" TO \"70\"}");
test_parse_query_to_ast_helper("weight:<=70", "weight:{\"*\" TO \"70\"]");
test_parse_query_to_ast_helper("weight: >60.7", "weight:{\"60.7\" TO \"*\"}");
test_parse_query_to_ast_helper("weight: <= 70", "weight:{\"*\" TO \"70\"]");
test_parse_query_to_ast_helper("weight: <= 70.5", "weight:{\"*\" TO \"70.5\"]");
}
#[test]
fn test_range_parser() {
// testing the range() parser separately
let res = range().parse("title: <hello").unwrap().0;
let expected = UserInputLeaf::Range {
field: Some("title".to_string()),
lower: UserInputBound::Unbounded,
upper: UserInputBound::Exclusive("hello".to_string()),
};
let res2 = range().parse("title:{* TO hello}").unwrap().0;
assert_eq!(res, expected);
assert_eq!(res2, expected);
let expected_weight = UserInputLeaf::Range {
field: Some("weight".to_string()),
lower: UserInputBound::Inclusive("71.2".to_string()),
upper: UserInputBound::Unbounded,
};
let res3 = range().parse("weight: >=71.2").unwrap().0;
let res4 = range().parse("weight:[71.2 TO *}").unwrap().0;
assert_eq!(res3, expected_weight);
assert_eq!(res4, expected_weight);
}
#[test]
fn test_parse_query_to_triming_spaces() {
test_parse_query_to_ast_helper(" abc", "\"abc\"");
test_parse_query_to_ast_helper("abc ", "\"abc\"");
test_parse_query_to_ast_helper("( a OR abc)", "(?(\"a\") ?(\"abc\"))");
test_parse_query_to_ast_helper("(a OR abc)", "(?(\"a\") ?(\"abc\"))");
test_parse_query_to_ast_helper("(a OR abc)", "(?(\"a\") ?(\"abc\"))");
test_parse_query_to_ast_helper("a OR abc ", "(?(\"a\") ?(\"abc\"))");
test_parse_query_to_ast_helper("(a OR abc )", "(?(\"a\") ?(\"abc\"))");
test_parse_query_to_ast_helper("(a OR abc) ", "(?(\"a\") ?(\"abc\"))");
}
#[test]
fn test_parse_query_to_ast() {
test_parse_query_to_ast_helper("abc", "\"abc\"");
test_parse_query_to_ast_helper("a b", "(\"a\" \"b\")");
test_parse_query_to_ast_helper("+(a b)", "+((\"a\" \"b\"))");
test_parse_query_to_ast_helper("+d", "+(\"d\")");
test_parse_query_to_ast_helper("+(a b) +d", "(+((\"a\" \"b\")) +(\"d\"))");
test_parse_query_to_ast_helper("(+a +b) d", "((+(\"a\") +(\"b\")) \"d\")");
test_parse_query_to_ast_helper("(+a)", "+(\"a\")");
test_parse_query_to_ast_helper("(+a +b)", "(+(\"a\") +(\"b\"))");
test_parse_query_to_ast_helper("abc:toto", "abc:\"toto\"");
test_parse_query_to_ast_helper("abc:1.1", "abc:\"1.1\"");
test_parse_query_to_ast_helper("+abc:toto", "+(abc:\"toto\")");
test_parse_query_to_ast_helper("(+abc:toto -titi)", "(+(abc:\"toto\") -(\"titi\"))");
test_parse_query_to_ast_helper("-abc:toto", "-(abc:\"toto\")");
test_parse_query_to_ast_helper("abc:a b", "(abc:\"a\" \"b\")");
test_parse_query_to_ast_helper("abc:\"a b\"", "abc:\"a b\"");
test_parse_query_to_ast_helper("foo:[1 TO 5]", "foo:[\"1\" TO \"5\"]");
test_parse_query_to_ast_helper("[1 TO 5]", "[\"1\" TO \"5\"]");
test_parse_query_to_ast_helper("foo:{a TO z}", "foo:{\"a\" TO \"z\"}");
test_parse_query_to_ast_helper("foo:[1 TO toto}", "foo:[\"1\" TO \"toto\"}");
test_parse_query_to_ast_helper("foo:[* TO toto}", "foo:{\"*\" TO \"toto\"}");
test_parse_query_to_ast_helper("foo:[1 TO *}", "foo:[\"1\" TO \"*\"}");
test_parse_query_to_ast_helper("foo:[1.1 TO *}", "foo:[\"1.1\" TO \"*\"}");
test_is_parse_err("abc + ");
}
}

159
query-grammar/src/user_input_ast.rs
View File

@@ -1,159 +0,0 @@
use std::fmt;
use std::fmt::{Debug, Formatter};
use crate::Occur;
#[derive(PartialEq)]
pub enum UserInputLeaf {
Literal(UserInputLiteral),
All,
Range {
field: Option<String>,
lower: UserInputBound,
upper: UserInputBound,
},
}
impl Debug for UserInputLeaf {
fn fmt(&self, formatter: &mut Formatter<'_>) -> Result<(), fmt::Error> {
match self {
UserInputLeaf::Literal(literal) => literal.fmt(formatter),
UserInputLeaf::Range {
ref field,
ref lower,
ref upper,
} => {
if let Some(ref field) = field {
write!(formatter, "{}:", field)?;
}
lower.display_lower(formatter)?;
write!(formatter, " TO ")?;
upper.display_upper(formatter)?;
Ok(())
}
UserInputLeaf::All => write!(formatter, "*"),
}
}
}
#[derive(PartialEq)]
pub struct UserInputLiteral {
pub field_name: Option<String>,
pub phrase: String,
}
impl fmt::Debug for UserInputLiteral {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
match self.field_name {
Some(ref field_name) => write!(formatter, "{}:\"{}\"", field_name, self.phrase),
None => write!(formatter, "\"{}\"", self.phrase),
}
}
}
#[derive(PartialEq)]
pub enum UserInputBound {
Inclusive(String),
Exclusive(String),
Unbounded,
}
impl UserInputBound {
fn display_lower(&self, formatter: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
match *self {
UserInputBound::Inclusive(ref word) => write!(formatter, "[\"{}\"", word),
UserInputBound::Exclusive(ref word) => write!(formatter, "{{\"{}\"", word),
UserInputBound::Unbounded => write!(formatter, "{{\"*\""),
}
}
fn display_upper(&self, formatter: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
match *self {
UserInputBound::Inclusive(ref word) => write!(formatter, "\"{}\"]", word),
UserInputBound::Exclusive(ref word) => write!(formatter, "\"{}\"}}", word),
UserInputBound::Unbounded => write!(formatter, "\"*\"}}"),
}
}
pub fn term_str(&self) -> &str {
match *self {
UserInputBound::Inclusive(ref contents) => contents,
UserInputBound::Exclusive(ref contents) => contents,
UserInputBound::Unbounded => &"*",
}
}
}
pub enum UserInputAST {
Clause(Vec<UserInputAST>),
Unary(Occur, Box<UserInputAST>),
Leaf(Box<UserInputLeaf>),
}
impl UserInputAST {
pub fn unary(self, occur: Occur) -> UserInputAST {
UserInputAST::Unary(occur, Box::new(self))
}
fn compose(occur: Occur, asts: Vec<UserInputAST>) -> UserInputAST {
assert_ne!(occur, Occur::MustNot);
assert!(!asts.is_empty());
if asts.len() == 1 {
asts.into_iter().next().unwrap() //< safe
} else {
UserInputAST::Clause(
asts.into_iter()
.map(|ast: UserInputAST| ast.unary(occur))
.collect::<Vec<_>>(),
)
}
}
pub fn empty_query() -> UserInputAST {
UserInputAST::Clause(Vec::default())
}
pub fn and(asts: Vec<UserInputAST>) -> UserInputAST {
UserInputAST::compose(Occur::Must, asts)
}
pub fn or(asts: Vec<UserInputAST>) -> UserInputAST {
UserInputAST::compose(Occur::Should, asts)
}
}
impl From<UserInputLiteral> for UserInputLeaf {
fn from(literal: UserInputLiteral) -> UserInputLeaf {
UserInputLeaf::Literal(literal)
}
}
impl From<UserInputLeaf> for UserInputAST {
fn from(leaf: UserInputLeaf) -> UserInputAST {
UserInputAST::Leaf(Box::new(leaf))
}
}
impl fmt::Debug for UserInputAST {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
match *self {
UserInputAST::Clause(ref subqueries) => {
if subqueries.is_empty() {
write!(formatter, "<emptyclause>")?;
} else {
write!(formatter, "(")?;
write!(formatter, "{:?}", &subqueries[0])?;
for subquery in &subqueries[1..] {
write!(formatter, " {:?}", subquery)?;
}
write!(formatter, ")")?;
}
Ok(())
}
UserInputAST::Unary(ref occur, ref subquery) => {
write!(formatter, "{}({:?})", occur, subquery)
}
UserInputAST::Leaf(ref subquery) => write!(formatter, "{:?}", subquery),
}
}
}

2
run-tests.sh
View File

@@ -1,2 +0,0 @@
#!/bin/bash
cargo test

142
src/collector/chained_collector.rs Normal file
View File

@@ -0,0 +1,142 @@
use collector::Collector;
use DocId;
use Result;
use Score;
use SegmentLocalId;
use SegmentReader;
/// Collector that does nothing.
/// This is used in the chain Collector and will hopefully
/// be optimized away by the compiler.
pub struct DoNothingCollector;
impl Collector for DoNothingCollector {
#[inline]
fn set_segment(&mut self, _: SegmentLocalId, _: &SegmentReader) -> Result<()> {
Ok(())
}
#[inline]
fn collect(&mut self, _doc: DocId, _score: Score) {}
#[inline]
fn requires_scoring(&self) -> bool {
false
}
}
/// Zero-cost abstraction used to collect on multiple collectors.
/// This contraption is only usable if the type of your collectors
/// are known at compile time.
///
/// ```rust
/// #[macro_use]
/// extern crate tantivy;
/// use tantivy::schema::{SchemaBuilder, TEXT};
/// use tantivy::{Index, Result};
/// use tantivy::collector::{CountCollector, TopCollector, chain};
/// use tantivy::query::QueryParser;
///
/// # fn main() { example().unwrap(); }
/// fn example() -> Result<()> {
/// let mut schema_builder = SchemaBuilder::new();
/// let title = schema_builder.add_text_field("title", TEXT);
/// let schema = schema_builder.build();
/// let index = Index::create_in_ram(schema);
/// {
/// let mut index_writer = index.writer(3_000_000)?;
/// index_writer.add_document(doc!(
/// title => "The Name of the Wind",
/// ));
/// index_writer.add_document(doc!(
/// title => "The Diary of Muadib",
/// ));
/// index_writer.add_document(doc!(
/// title => "A Dairy Cow",
/// ));
/// index_writer.add_document(doc!(
/// title => "The Diary of a Young Girl",
/// ));
/// index_writer.commit().unwrap();
/// }
///
/// index.load_searchers()?;
/// let searcher = index.searcher();
///
/// {
/// let mut top_collector = TopCollector::with_limit(2);
/// let mut count_collector = CountCollector::default();
/// {
/// let mut collectors = chain().push(&mut top_collector).push(&mut count_collector);
/// let query_parser = QueryParser::for_index(&index, vec![title]);
/// let query = query_parser.parse_query("diary")?;
/// searcher.search(&*query, &mut collectors).unwrap();
/// }
/// assert_eq!(count_collector.count(), 2);
/// assert!(top_collector.at_capacity());
/// }
///
/// Ok(())
/// }
/// ```
pub struct ChainedCollector<Left: Collector, Right: Collector> {
left: Left,
right: Right,
}
impl<Left: Collector, Right: Collector> ChainedCollector<Left, Right> {
/// Adds a collector
pub fn push<C: Collector>(self, new_collector: &mut C) -> ChainedCollector<Self, &mut C> {
ChainedCollector {
left: self,
right: new_collector,
}
}
}
impl<Left: Collector, Right: Collector> Collector for ChainedCollector<Left, Right> {
fn set_segment(
&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader,
) -> Result<()> {
self.left.set_segment(segment_local_id, segment)?;
self.right.set_segment(segment_local_id, segment)?;
Ok(())
}
fn collect(&mut self, doc: DocId, score: Score) {
self.left.collect(doc, score);
self.right.collect(doc, score);
}
fn requires_scoring(&self) -> bool {
self.left.requires_scoring() || self.right.requires_scoring()
}
}
/// Creates a `ChainedCollector`
pub fn chain() -> ChainedCollector<DoNothingCollector, DoNothingCollector> {
ChainedCollector {
left: DoNothingCollector,
right: DoNothingCollector,
}
}
#[cfg(test)]
mod tests {
use super::*;
use collector::{Collector, CountCollector, TopCollector};
#[test]
fn test_chained_collector() {
let mut top_collector = TopCollector::with_limit(2);
let mut count_collector = CountCollector::default();
{
let mut collectors = chain().push(&mut top_collector).push(&mut count_collector);
collectors.collect(1, 0.2);
collectors.collect(2, 0.1);
collectors.collect(3, 0.5);
}
assert_eq!(count_collector.count(), 3);
assert!(top_collector.at_capacity());
}
}

109
src/collector/count_collector.rs
View File

@@ -1,23 +1,24 @@
use super::Collector;
use crate::collector::SegmentCollector;
use crate::DocId;
use crate::Result;
use crate::Score;
use crate::SegmentLocalId;
use crate::SegmentReader;
use DocId;
use Result;
use Score;
use SegmentLocalId;
use SegmentReader;
/// `CountCollector` collector only counts how many
/// documents match the query.
///
/// ```rust
/// use tantivy::collector::Count;
/// #[macro_use]
/// extern crate tantivy;
/// use tantivy::schema::{SchemaBuilder, TEXT};
/// use tantivy::{Index, Result};
/// use tantivy::collector::CountCollector;
/// use tantivy::query::QueryParser;
/// use tantivy::schema::{Schema, TEXT};
/// use tantivy::{doc, Index, Result};
///
/// # fn main() { example().unwrap(); }
/// fn example() -> Result<()> {
/// let mut schema_builder = Schema::builder();
/// let mut schema_builder = SchemaBuilder::new();
/// let title = schema_builder.add_text_field("title", TEXT);
/// let schema = schema_builder.build();
/// let index = Index::create_in_ram(schema);
@@ -38,89 +39,63 @@ use crate::SegmentReader;
/// index_writer.commit().unwrap();
/// }
///
/// let reader = index.reader()?;
/// let searcher = reader.searcher();
/// index.load_searchers()?;
/// let searcher = index.searcher();
///
/// {
/// let mut count_collector = CountCollector::default();
/// let query_parser = QueryParser::for_index(&index, vec![title]);
/// let query = query_parser.parse_query("diary")?;
/// let count = searcher.search(&query, &Count).unwrap();
/// searcher.search(&*query, &mut count_collector).unwrap();
///
/// assert_eq!(count, 2);
/// assert_eq!(count_collector.count(), 2);
/// }
///
/// Ok(())
/// }
/// ```
pub struct Count;
impl Collector for Count {
type Fruit = usize;
type Child = SegmentCountCollector;
fn for_segment(&self, _: SegmentLocalId, _: &SegmentReader) -> Result<SegmentCountCollector> {
Ok(SegmentCountCollector::default())
}
fn requires_scoring(&self) -> bool {
false
}
fn merge_fruits(&self, segment_counts: Vec<usize>) -> Result<usize> {
Ok(segment_counts.into_iter().sum())
}
}
#[derive(Default)]
pub struct SegmentCountCollector {
pub struct CountCollector {
count: usize,
}
impl SegmentCollector for SegmentCountCollector {
type Fruit = usize;
impl CountCollector {
/// Returns the count of documents that were
/// collected.
pub fn count(&self) -> usize {
self.count
}
}
impl Collector for CountCollector {
fn set_segment(&mut self, _: SegmentLocalId, _: &SegmentReader) -> Result<()> {
Ok(())
}
fn collect(&mut self, _: DocId, _: Score) {
self.count += 1;
}
fn harvest(self) -> usize {
self.count
fn requires_scoring(&self) -> bool {
false
}
}
#[cfg(test)]
mod tests {
use super::{Count, SegmentCountCollector};
use crate::collector::Collector;
use crate::collector::SegmentCollector;
use collector::{Collector, CountCollector};
#[test]
fn test_count_collect_does_not_requires_scoring() {
assert!(!Count.requires_scoring());
fn test_count_collector() {
let mut count_collector = CountCollector::default();
assert_eq!(count_collector.count(), 0);
count_collector.collect(0u32, 1f32);
assert_eq!(count_collector.count(), 1);
assert_eq!(count_collector.count(), 1);
count_collector.collect(1u32, 1f32);
assert_eq!(count_collector.count(), 2);
assert!(!count_collector.requires_scoring());
}
#[test]
fn test_segment_count_collector() {
{
let count_collector = SegmentCountCollector::default();
assert_eq!(count_collector.harvest(), 0);
}
{
let mut count_collector = SegmentCountCollector::default();
count_collector.collect(0u32, 1f32);
assert_eq!(count_collector.harvest(), 1);
}
{
let mut count_collector = SegmentCountCollector::default();
count_collector.collect(0u32, 1f32);
assert_eq!(count_collector.harvest(), 1);
}
{
let mut count_collector = SegmentCountCollector::default();
count_collector.collect(0u32, 1f32);
count_collector.collect(1u32, 1f32);
assert_eq!(count_collector.harvest(), 2);
}
}
}

126
src/collector/custom_score_top_collector.rs
View File

@@ -1,126 +0,0 @@
use crate::collector::top_collector::{TopCollector, TopSegmentCollector};
use crate::collector::{Collector, SegmentCollector};
use crate::Result;
use crate::{DocAddress, DocId, Score, SegmentReader};
pub(crate) struct CustomScoreTopCollector<TCustomScorer, TScore = Score> {
custom_scorer: TCustomScorer,
collector: TopCollector<TScore>,
}
impl<TCustomScorer, TScore> CustomScoreTopCollector<TCustomScorer, TScore>
where
TScore: Clone + PartialOrd,
{
pub fn new(
custom_scorer: TCustomScorer,
limit: usize,
) -> CustomScoreTopCollector<TCustomScorer, TScore> {
CustomScoreTopCollector {
custom_scorer,
collector: TopCollector::with_limit(limit),
}
}
}
/// A custom segment scorer makes it possible to define any kind of score
/// for a given document belonging to a specific segment.
///
/// It is the segment local version of the [`CustomScorer`](./trait.CustomScorer.html).
pub trait CustomSegmentScorer<TScore>: 'static {
/// Computes the score of a specific `doc`.
fn score(&self, doc: DocId) -> TScore;
}
/// `CustomScorer` makes it possible to define any kind of score.
///
/// The `CustomerScorer` itself does not make much of the computation itself.
/// Instead, it helps constructing `Self::Child` instances that will compute
/// the score at a segment scale.
pub trait CustomScorer<TScore>: Sync {
/// Type of the associated [`CustomSegmentScorer`](./trait.CustomSegmentScorer.html).
type Child: CustomSegmentScorer<TScore>;
/// Builds a child scorer for a specific segment. The child scorer is associated to
/// a specific segment.
fn segment_scorer(&self, segment_reader: &SegmentReader) -> Result<Self::Child>;
}
impl<TCustomScorer, TScore> Collector for CustomScoreTopCollector<TCustomScorer, TScore>
where
TCustomScorer: CustomScorer<TScore>,
TScore: 'static + PartialOrd + Clone + Send + Sync,
{
type Fruit = Vec<(TScore, DocAddress)>;
type Child = CustomScoreTopSegmentCollector<TCustomScorer::Child, TScore>;
fn for_segment(
&self,
segment_local_id: u32,
segment_reader: &SegmentReader,
) -> Result<Self::Child> {
let segment_scorer = self.custom_scorer.segment_scorer(segment_reader)?;
let segment_collector = self
.collector
.for_segment(segment_local_id, segment_reader)?;
Ok(CustomScoreTopSegmentCollector {
segment_collector,
segment_scorer,
})
}
fn requires_scoring(&self) -> bool {
false
}
fn merge_fruits(&self, segment_fruits: Vec<Self::Fruit>) -> Result<Self::Fruit> {
self.collector.merge_fruits(segment_fruits)
}
}
pub struct CustomScoreTopSegmentCollector<T, TScore>
where
TScore: 'static + PartialOrd + Clone + Send + Sync + Sized,
T: CustomSegmentScorer<TScore>,
{
segment_collector: TopSegmentCollector<TScore>,
segment_scorer: T,
}
impl<T, TScore> SegmentCollector for CustomScoreTopSegmentCollector<T, TScore>
where
TScore: 'static + PartialOrd + Clone + Send + Sync,
T: 'static + CustomSegmentScorer<TScore>,
{
type Fruit = Vec<(TScore, DocAddress)>;
fn collect(&mut self, doc: DocId, _score: Score) {
let score = self.segment_scorer.score(doc);
self.segment_collector.collect(doc, score);
}
fn harvest(self) -> Vec<(TScore, DocAddress)> {
self.segment_collector.harvest()
}
}
impl<F, TScore, T> CustomScorer<TScore> for F
where
F: 'static + Send + Sync + Fn(&SegmentReader) -> T,
T: CustomSegmentScorer<TScore>,
{
type Child = T;
fn segment_scorer(&self, segment_reader: &SegmentReader) -> Result<Self::Child> {
Ok((self)(segment_reader))
}
}
impl<F, TScore> CustomSegmentScorer<TScore> for F
where
F: 'static + Sync + Send + Fn(DocId) -> TScore,
{
fn score(&self, doc: DocId) -> TScore {
(self)(doc)
}
}

414
src/collector/facet_collector.rs
View File

@@ -1,23 +1,25 @@
use crate::collector::Collector;
use crate::collector::SegmentCollector;
use crate::docset::SkipResult;
use crate::fastfield::FacetReader;
use crate::schema::Facet;
use crate::schema::Field;
use crate::DocId;
use crate::Result;
use crate::Score;
use crate::SegmentLocalId;
use crate::SegmentReader;
use crate::TantivyError;
use std::cmp::Ordering;
use collector::Collector;
use docset::SkipResult;
use fastfield::FacetReader;
use schema::Facet;
use schema::Field;
use std::cell::UnsafeCell;
use std::collections::btree_map;
use std::collections::BTreeMap;
use std::collections::BTreeSet;
use std::collections::BinaryHeap;
use std::collections::Bound;
use std::iter::Peekable;
use std::mem;
use std::{u64, usize};
use termdict::TermMerger;
use std::cmp::Ordering;
use DocId;
use Result;
use Score;
use SegmentLocalId;
use SegmentReader;
struct Hit<'a> {
count: u64,
@@ -27,13 +29,13 @@ struct Hit<'a> {
impl<'a> Eq for Hit<'a> {}
impl<'a> PartialEq<Hit<'a>> for Hit<'a> {
fn eq(&self, other: &Hit<'_>) -> bool {
fn eq(&self, other: &Hit) -> bool {
self.count == other.count
}
}
impl<'a> PartialOrd<Hit<'a>> for Hit<'a> {
fn partial_cmp(&self, other: &Hit<'_>) -> Option<Ordering> {
fn partial_cmp(&self, other: &Hit) -> Option<Ordering> {
Some(self.cmp(other))
}
}
@@ -44,6 +46,12 @@ impl<'a> Ord for Hit<'a> {
}
}
struct SegmentFacetCounter {
pub facet_reader: FacetReader,
pub facet_ords: Vec<u64>,
pub facet_counts: Vec<u64>,
}
fn facet_depth(facet_bytes: &[u8]) -> usize {
if facet_bytes.is_empty() {
0
@@ -81,14 +89,16 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
///
///
/// ```rust
/// #[macro_use]
/// extern crate tantivy;
/// use tantivy::schema::{Facet, SchemaBuilder, TEXT};
/// use tantivy::{Index, Result};
/// use tantivy::collector::FacetCollector;
/// use tantivy::query::AllQuery;
/// use tantivy::schema::{Facet, Schema, TEXT};
/// use tantivy::{doc, Index, Result};
///
/// # fn main() { example().unwrap(); }
/// fn example() -> Result<()> {
/// let mut schema_builder = Schema::builder();
/// let mut schema_builder = SchemaBuilder::new();
///
/// // Facet have their own specific type.
/// // It is not a bad practise to put all of your
@@ -121,19 +131,23 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
/// facet => Facet::from("/lang/en"),
/// facet => Facet::from("/category/biography")
/// ));
/// index_writer.commit()?;
/// index_writer.commit().unwrap();
/// }
/// let reader = index.reader()?;
/// let searcher = reader.searcher();
///
/// index.load_searchers()?;
/// let searcher = index.searcher();
///
/// {
/// let mut facet_collector = FacetCollector::for_field(facet);
/// facet_collector.add_facet("/lang");
/// facet_collector.add_facet("/category");
/// let facet_counts = searcher.search(&AllQuery, &facet_collector)?;
/// searcher.search(&AllQuery, &mut facet_collector).unwrap();
///
/// // this object contains count aggregate for all of the facets.
/// let counts = facet_collector.harvest();
///
/// // This lists all of the facet counts
/// let facets: Vec<(&Facet, u64)> = facet_counts
/// let facets: Vec<(&Facet, u64)> = counts
/// .get("/category")
/// .collect();
/// assert_eq!(facets, vec![
@@ -145,10 +159,13 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
/// {
/// let mut facet_collector = FacetCollector::for_field(facet);
/// facet_collector.add_facet("/category/fiction");
/// let facet_counts = searcher.search(&AllQuery, &facet_collector)?;
/// searcher.search(&AllQuery, &mut facet_collector).unwrap();
///
/// // this object contains count aggregate for all of the facets.
/// let counts = facet_collector.harvest();
///
/// // This lists all of the facet counts
/// let facets: Vec<(&Facet, u64)> = facet_counts
/// let facets: Vec<(&Facet, u64)> = counts
/// .get("/category/fiction")
/// .collect();
/// assert_eq!(facets, vec![
@@ -161,10 +178,13 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
/// {
/// let mut facet_collector = FacetCollector::for_field(facet);
/// facet_collector.add_facet("/category/fiction");
/// let facet_counts = searcher.search(&AllQuery, &facet_collector)?;
/// searcher.search(&AllQuery, &mut facet_collector).unwrap();
///
/// // this object contains count aggregate for all of the facets.
/// let counts = facet_collector.harvest();
///
/// // This lists all of the facet counts
/// let facets: Vec<(&Facet, u64)> = facet_counts.top_k("/category/fiction", 1);
/// let facets: Vec<(&Facet, u64)> = counts.top_k("/category/fiction", 1);
/// assert_eq!(facets, vec![
/// (&Facet::from("/category/fiction/fantasy"), 2)
/// ]);
@@ -174,19 +194,19 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
/// }
/// ```
pub struct FacetCollector {
facet_ords: Vec<u64>,
field: Field,
facets: BTreeSet<Facet>,
}
ff_reader: Option<UnsafeCell<FacetReader>>,
segment_counters: Vec<SegmentFacetCounter>,
pub struct FacetSegmentCollector {
reader: FacetReader,
facet_ords_buf: Vec<u64>,
// facet_ord -> collapse facet_id
collapse_mapping: Vec<usize>,
current_segment_collapse_mapping: Vec<usize>,
// collapse facet_id -> count
counts: Vec<u64>,
current_segment_counts: Vec<u64>,
// collapse facet_id -> facet_ord
collapse_facet_ords: Vec<u64>,
current_collapse_facet_ords: Vec<u64>,
facets: BTreeSet<Facet>,
}
fn skip<'a, I: Iterator<Item = &'a Facet>>(
@@ -195,7 +215,7 @@ fn skip<'a, I: Iterator<Item = &'a Facet>>(
) -> SkipResult {
loop {
match collapse_it.peek() {
Some(facet_bytes) => match facet_bytes.encoded_str().as_bytes().cmp(target) {
Some(facet_bytes) => match facet_bytes.encoded_bytes().cmp(target) {
Ordering::Less => {}
Ordering::Greater => {
return SkipResult::OverStep;
@@ -220,8 +240,15 @@ impl FacetCollector {
/// is of the proper type.
pub fn for_field(field: Field) -> FacetCollector {
FacetCollector {
facet_ords: Vec::with_capacity(255),
segment_counters: Vec::new(),
field,
facets: BTreeSet::default(),
ff_reader: None,
facets: BTreeSet::new(),
current_segment_collapse_mapping: Vec::new(),
current_collapse_facet_ords: Vec::new(),
current_segment_counts: Vec::new(),
}
}
@@ -251,103 +278,141 @@ impl FacetCollector {
}
self.facets.insert(facet);
}
}
impl Collector for FacetCollector {
type Fruit = FacetCounts;
type Child = FacetSegmentCollector;
fn for_segment(
&self,
_: SegmentLocalId,
reader: &SegmentReader,
) -> Result<FacetSegmentCollector> {
let field_name = reader.schema().get_field_name(self.field);
let facet_reader = reader.facet_reader(self.field).ok_or_else(|| {
TantivyError::SchemaError(format!("Field {:?} is not a facet field.", field_name))
})?;
let mut collapse_mapping = Vec::new();
let mut counts = Vec::new();
let mut collapse_facet_ords = Vec::new();
fn set_collapse_mapping(&mut self, facet_reader: &FacetReader) {
self.current_segment_collapse_mapping.clear();
self.current_collapse_facet_ords.clear();
self.current_segment_counts.clear();
let mut collapse_facet_it = self.facets.iter().peekable();
collapse_facet_ords.push(0);
{
let mut facet_streamer = facet_reader.facet_dict().range().into_stream();
if facet_streamer.advance() {
'outer: loop {
// at the begining of this loop, facet_streamer
// is positionned on a term that has not been processed yet.
let skip_result = skip(facet_streamer.key(), &mut collapse_facet_it);
match skip_result {
SkipResult::Reached => {
// we reach a facet we decided to collapse.
let collapse_depth = facet_depth(facet_streamer.key());
let mut collapsed_id = 0;
collapse_mapping.push(0);
while facet_streamer.advance() {
let depth = facet_depth(facet_streamer.key());
if depth <= collapse_depth {
continue 'outer;
}
if depth == collapse_depth + 1 {
collapsed_id = collapse_facet_ords.len();
collapse_facet_ords.push(facet_streamer.term_ord());
collapse_mapping.push(collapsed_id);
} else {
collapse_mapping.push(collapsed_id);
}
}
break;
self.current_collapse_facet_ords.push(0);
let mut facet_streamer = facet_reader.facet_dict().range().into_stream();
if !facet_streamer.advance() {
return;
}
'outer: loop {
// at the begining of this loop, facet_streamer
// is positionned on a term that has not been processed yet.
let skip_result = skip(facet_streamer.key(), &mut collapse_facet_it);
match skip_result {
SkipResult::Reached => {
// we reach a facet we decided to collapse.
let collapse_depth = facet_depth(facet_streamer.key());
let mut collapsed_id = 0;
self.current_segment_collapse_mapping.push(0);
while facet_streamer.advance() {
let depth = facet_depth(facet_streamer.key());
if depth <= collapse_depth {
continue 'outer;
}
SkipResult::End | SkipResult::OverStep => {
collapse_mapping.push(0);
if !facet_streamer.advance() {
break;
}
if depth == collapse_depth + 1 {
collapsed_id = self.current_collapse_facet_ords.len();
self.current_collapse_facet_ords
.push(facet_streamer.term_ord());
self.current_segment_collapse_mapping.push(collapsed_id);
} else {
self.current_segment_collapse_mapping.push(collapsed_id);
}
}
break;
}
SkipResult::End | SkipResult::OverStep => {
self.current_segment_collapse_mapping.push(0);
if !facet_streamer.advance() {
break;
}
}
}
}
counts.resize(collapse_facet_ords.len(), 0);
Ok(FacetSegmentCollector {
reader: facet_reader,
facet_ords_buf: Vec::with_capacity(255),
collapse_mapping,
counts,
collapse_facet_ords,
})
}
fn requires_scoring(&self) -> bool {
false
fn finalize_segment(&mut self) {
if self.ff_reader.is_some() {
self.segment_counters.push(SegmentFacetCounter {
facet_reader: self.ff_reader.take().unwrap().into_inner(),
facet_ords: mem::replace(&mut self.current_collapse_facet_ords, Vec::new()),
facet_counts: mem::replace(&mut self.current_segment_counts, Vec::new()),
});
}
}
fn merge_fruits(&self, segments_facet_counts: Vec<FacetCounts>) -> Result<FacetCounts> {
let mut facet_counts: BTreeMap<Facet, u64> = BTreeMap::new();
for segment_facet_counts in segments_facet_counts {
for (facet, count) in segment_facet_counts.facet_counts {
*(facet_counts.entry(facet).or_insert(0)) += count;
/// Returns the results of the collection.
///
/// This method does not just return the counters,
/// it also translates the facet ordinals of the last segment.
pub fn harvest(mut self) -> FacetCounts {
self.finalize_segment();
let collapsed_facet_ords: Vec<&[u64]> = self.segment_counters
.iter()
.map(|segment_counter| &segment_counter.facet_ords[..])
.collect();
let collapsed_facet_counts: Vec<&[u64]> = self.segment_counters
.iter()
.map(|segment_counter| &segment_counter.facet_counts[..])
.collect();
let facet_streams = self.segment_counters
.iter()
.map(|seg_counts| seg_counts.facet_reader.facet_dict().range().into_stream())
.collect::<Vec<_>>();
let mut facet_merger = TermMerger::new(facet_streams);
let mut facet_counts = BTreeMap::new();
while facet_merger.advance() {
let count = facet_merger
.current_kvs()
.iter()
.map(|it| {
let seg_ord = it.segment_ord;
let term_ord = it.streamer.term_ord();
collapsed_facet_ords[seg_ord]
.binary_search(&term_ord)
.map(|collapsed_term_id| {
if collapsed_term_id == 0 {
0
} else {
collapsed_facet_counts[seg_ord][collapsed_term_id]
}
})
.unwrap_or(0)
})
.sum();
if count > 0u64 {
let bytes: Vec<u8> = facet_merger.key().to_owned();
// may create an corrupted facet if the term dicitonary is corrupted
let facet = unsafe { Facet::from_encoded(bytes) };
facet_counts.insert(facet, count);
}
}
Ok(FacetCounts { facet_counts })
FacetCounts { facet_counts }
}
}
impl SegmentCollector for FacetSegmentCollector {
type Fruit = FacetCounts;
impl Collector for FacetCollector {
fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
self.finalize_segment();
let facet_reader = reader.facet_reader(self.field)?;
self.set_collapse_mapping(&facet_reader);
self.current_segment_counts
.resize(self.current_collapse_facet_ords.len(), 0);
self.ff_reader = Some(UnsafeCell::new(facet_reader));
Ok(())
}
fn collect(&mut self, doc: DocId, _: Score) {
self.reader.facet_ords(doc, &mut self.facet_ords_buf);
let facet_reader: &mut FacetReader = unsafe {
&mut *self.ff_reader
.as_ref()
.expect("collect() was called before set_segment. This should never happen.")
.get()
};
facet_reader.facet_ords(doc, &mut self.facet_ords);
let mut previous_collapsed_ord: usize = usize::MAX;
for &facet_ord in &self.facet_ords_buf {
let collapsed_ord = self.collapse_mapping[facet_ord as usize];
self.counts[collapsed_ord] += if collapsed_ord == previous_collapsed_ord {
for &facet_ord in &self.facet_ords {
let collapsed_ord = self.current_segment_collapse_mapping[facet_ord as usize];
self.current_segment_counts[collapsed_ord] += if collapsed_ord == previous_collapsed_ord
{
0
} else {
1
@@ -356,24 +421,8 @@ impl SegmentCollector for FacetSegmentCollector {
}
}
/// Returns the results of the collection.
///
/// This method does not just return the counters,
/// it also translates the facet ordinals of the last segment.
fn harvest(self) -> FacetCounts {
let mut facet_counts = BTreeMap::new();
let facet_dict = self.reader.facet_dict();
for (collapsed_facet_ord, count) in self.counts.iter().cloned().enumerate() {
if count == 0 {
continue;
}
let mut facet = vec![];
let facet_ord = self.collapse_facet_ords[collapsed_facet_ord];
facet_dict.ord_to_term(facet_ord as u64, &mut facet);
// TODO
facet_counts.insert(Facet::from_encoded(facet).unwrap(), count);
}
FacetCounts { facet_counts }
fn requires_scoring(&self) -> bool {
false
}
}
@@ -396,7 +445,7 @@ impl<'a> Iterator for FacetChildIterator<'a> {
}
impl FacetCounts {
pub fn get<T>(&self, facet_from: T) -> FacetChildIterator<'_>
pub fn get<T>(&self, facet_from: T) -> FacetChildIterator
where
Facet: From<T>,
{
@@ -405,13 +454,12 @@ impl FacetCounts {
let right_bound = if facet.is_root() {
Bound::Unbounded
} else {
let mut facet_after_bytes: String = facet.encoded_str().to_owned();
facet_after_bytes.push('\u{1}');
let facet_after = Facet::from_encoded_string(facet_after_bytes);
let mut facet_after_bytes: Vec<u8> = facet.encoded_bytes().to_owned();
facet_after_bytes.push(1u8);
let facet_after = unsafe { Facet::from_encoded(facet_after_bytes) }; // ok logic
Bound::Excluded(facet_after)
};
let underlying: btree_map::Range<'_, _, _> =
self.facet_counts.range((left_bound, right_bound));
let underlying: btree_map::Range<_, _> = self.facet_counts.range((left_bound, right_bound));
FacetChildIterator { underlying }
}
@@ -428,8 +476,9 @@ impl FacetCounts {
heap.push(Hit { count, facet });
}
let mut lowest_count: u64 = heap.peek().map(|hit| hit.count).unwrap_or(u64::MIN); //< the `unwrap_or` case may be triggered but the value
// is never used in that case.
let mut lowest_count: u64 = heap.peek().map(|hit| hit.count)
.unwrap_or(u64::MIN); //< the `unwrap_or` case may be triggered but the value
// is never used in that case.
for (facet, count) in it {
if count > lowest_count {
@@ -452,17 +501,17 @@ impl FacetCounts {
#[cfg(test)]
mod tests {
use super::{FacetCollector, FacetCounts};
use crate::core::Index;
use crate::query::AllQuery;
use crate::schema::{Document, Facet, Field, Schema};
use core::Index;
use query::AllQuery;
use rand::distributions::Uniform;
use rand::prelude::SliceRandom;
use rand::{thread_rng, Rng};
use schema::Field;
use schema::{Document, Facet, SchemaBuilder};
use std::iter;
#[test]
fn test_facet_collector_drilldown() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
let facet_field = schema_builder.add_facet_field("facet");
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -485,12 +534,14 @@ mod tests {
index_writer.add_document(doc);
}
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
index.load_searchers().unwrap();
let searcher = index.searcher();
let mut facet_collector = FacetCollector::for_field(facet_field);
facet_collector.add_facet(Facet::from("/top1"));
let counts = searcher.search(&AllQuery, &facet_collector).unwrap();
searcher.search(&AllQuery, &mut facet_collector).unwrap();
let counts: FacetCounts = facet_collector.harvest();
{
let facets: Vec<(String, u64)> = counts
.get("/top1")
@@ -503,17 +554,18 @@ mod tests {
("/top1/mid1", 50),
("/top1/mid2", 50),
("/top1/mid3", 50),
]
.iter()
.map(|&(facet_str, count)| (String::from(facet_str), count))
.collect::<Vec<_>>()
].iter()
.map(|&(facet_str, count)| (String::from(facet_str), count))
.collect::<Vec<_>>()
);
}
}
#[test]
#[should_panic(expected = "Tried to add a facet which is a descendant of \
an already added facet.")]
#[should_panic(
expected = "Tried to add a facet which is a descendant of \
an already added facet."
)]
fn test_misused_facet_collector() {
let mut facet_collector = FacetCollector::for_field(Field(0));
facet_collector.add_facet(Facet::from("/country"));
@@ -522,7 +574,7 @@ mod tests {
#[test]
fn test_doc_unsorted_multifacet() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
let facet_field = schema_builder.add_facet_field("facets");
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -534,12 +586,13 @@ mod tests {
facet_field => Facet::from_text(&"/subjects/B/b"),
));
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
index.load_searchers().unwrap();
let searcher = index.searcher();
assert_eq!(searcher.num_docs(), 1);
let mut facet_collector = FacetCollector::for_field(facet_field);
facet_collector.add_facet("/subjects");
let counts = searcher.search(&AllQuery, &facet_collector).unwrap();
searcher.search(&AllQuery, &mut facet_collector).unwrap();
let counts = facet_collector.harvest();
let facets: Vec<(&Facet, u64)> = counts.get("/subjects").collect();
assert_eq!(facets[0].1, 1);
}
@@ -553,7 +606,7 @@ mod tests {
#[test]
fn test_facet_collector_topk() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
let facet_field = schema_builder.add_facet_field("facet");
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -566,27 +619,24 @@ mod tests {
let doc = doc!(facet_field => facet);
iter::repeat(doc).take(count)
})
.map(|mut doc| {
doc.add_facet(
facet_field,
&format!("/facet/{}", thread_rng().sample(&uniform)),
);
doc
})
.map(|mut doc| { doc.add_facet(facet_field, &format!("/facet/{}", thread_rng().sample(&uniform) )); doc})
.collect();
docs[..].shuffle(&mut thread_rng());
thread_rng().shuffle(&mut docs[..]);
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
for doc in docs {
index_writer.add_document(doc);
}
index_writer.commit().unwrap();
let searcher = index.reader().unwrap().searcher();
index.load_searchers().unwrap();
let searcher = index.searcher();
let mut facet_collector = FacetCollector::for_field(facet_field);
facet_collector.add_facet("/facet");
let counts: FacetCounts = searcher.search(&AllQuery, &facet_collector).unwrap();
searcher.search(&AllQuery, &mut facet_collector).unwrap();
let counts: FacetCounts = facet_collector.harvest();
{
let facets: Vec<(&Facet, u64)> = counts.top_k("/facet", 3);
assert_eq!(
@@ -599,22 +649,23 @@ mod tests {
);
}
}
}
#[cfg(all(test, feature = "unstable"))]
mod bench {
use crate::collector::FacetCollector;
use crate::query::AllQuery;
use crate::schema::{Facet, Schema};
use crate::Index;
use rand::seq::SliceRandom;
use rand::thread_rng;
use collector::FacetCollector;
use query::AllQuery;
use rand::{thread_rng, Rng};
use schema::Facet;
use schema::SchemaBuilder;
use test::Bencher;
use Index;
#[bench]
fn bench_facet_collector(b: &mut Bencher) {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
let facet_field = schema_builder.add_facet_field("facet");
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -627,18 +678,19 @@ mod bench {
}
}
// 40425 docs
docs[..].shuffle(&mut thread_rng());
thread_rng().shuffle(&mut docs[..]);
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
for doc in docs {
index_writer.add_document(doc);
}
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
index.load_searchers().unwrap();
b.iter(|| {
let searcher = reader.searcher();
let facet_collector = FacetCollector::for_field(facet_field);
searcher.search(&AllQuery, &facet_collector).unwrap();
let searcher = index.searcher();
let mut facet_collector = FacetCollector::for_field(facet_field);
searcher.search(&AllQuery, &mut facet_collector).unwrap();
});
}
}

14
src/collector/int_facet_collector.rs
View File

@@ -79,23 +79,21 @@ mod tests {
// make sure we have facet counters correctly filled
fn test_facet_collector_results() {
let mut schema_builder = schema::Schema::builder();
let mut schema_builder = schema::SchemaBuilder::new();
let num_field_i64 = schema_builder.add_i64_field("num_i64", FAST);
let num_field_u64 = schema_builder.add_u64_field("num_u64", FAST);
let num_field_f64 = schema_builder.add_f64_field("num_f64", FAST);
let text_field = schema_builder.add_text_field("text", STRING);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema.clone());
{
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
{
for i in 0u64..10u64 {
index_writer.add_document(doc!(
num_field_i64 => ((i as i64) % 3i64) as i64,
num_field_u64 => (i % 2u64) as u64,
num_field_f64 => (i % 4u64) as f64,
text_field => "text"
));
}
@@ -103,14 +101,14 @@ mod tests {
assert_eq!(index_writer.commit().unwrap(), 10u64);
}
let searcher = index.reader().searcher();
index.load_searchers().unwrap();
let searcher = index.searcher();
let mut ffvf_i64: IntFacetCollector<I64FastFieldReader> = IntFacetCollector::new(num_field_i64);
let mut ffvf_u64: IntFacetCollector<U64FastFieldReader> = IntFacetCollector::new(num_field_u64);
let mut ffvf_f64: IntFacetCollector<F64FastFieldReader> = IntFacetCollector::new(num_field_f64);
{
// perform the query
let mut facet_collectors = chain().push(&mut ffvf_i64).push(&mut ffvf_u64).push(&mut ffvf_f64);
let mut facet_collectors = chain().push(&mut ffvf_i64).push(&mut ffvf_u64);
let mut query_parser = QueryParser::for_index(index, vec![text_field]);
let query = query_parser.parse_query("text:text").unwrap();
query.search(&searcher, &mut facet_collectors).unwrap();
@@ -120,8 +118,6 @@ mod tests {
assert_eq!(ffvf_u64.counters[&1], 5);
assert_eq!(ffvf_i64.counters[&0], 4);
assert_eq!(ffvf_i64.counters[&1], 3);
assert_eq!(ffvf_f64.counters[&0.0], 3);
assert_eq!(ffvf_f64.counters[&2.0], 2);
}
}

526
src/collector/mod.rs
View File

@@ -1,366 +1,246 @@
/*!
# Collectors
Collectors define the information you want to extract from the documents matching the queries.
In tantivy jargon, we call this information your search "fruit".
Your fruit could for instance be :
- [the count of matching documents](./struct.Count.html)
- [the top 10 documents, by relevancy or by a fast field](./struct.TopDocs.html)
- [facet counts](./struct.FacetCollector.html)
At one point in your code, you will trigger the actual search operation by calling
[the `search(...)` method of your `Searcher` object](../struct.Searcher.html#method.search).
This call will look like this.
```verbatim
let fruit = searcher.search(&query, &collector)?;
```
Here the type of fruit is actually determined as an associated type of the collector (`Collector::Fruit`).
# Combining several collectors
A rich search experience often requires to run several collectors on your search query.
For instance,
- selecting the top-K products matching your query
- counting the matching documents
- computing several facets
- computing statistics about the matching product prices
A simple and efficient way to do that is to pass your collectors as one tuple.
The resulting `Fruit` will then be a typed tuple with each collector's original fruits
in their respective position.
```rust
# use tantivy::schema::*;
# use tantivy::*;
# use tantivy::query::*;
use tantivy::collector::{Count, TopDocs};
#
# fn main() -> tantivy::Result<()> {
# let mut schema_builder = Schema::builder();
# let title = schema_builder.add_text_field("title", TEXT);
# let schema = schema_builder.build();
# let index = Index::create_in_ram(schema);
# let mut index_writer = index.writer(3_000_000)?;
# index_writer.add_document(doc!(
# title => "The Name of the Wind",
# ));
# index_writer.add_document(doc!(
# title => "The Diary of Muadib",
# ));
# index_writer.commit()?;
# let reader = index.reader()?;
# let searcher = reader.searcher();
# let query_parser = QueryParser::for_index(&index, vec![title]);
# let query = query_parser.parse_query("diary")?;
let (doc_count, top_docs): (usize, Vec<(Score, DocAddress)>) =
searcher.search(&query, &(Count, TopDocs::with_limit(2)))?;
# Ok(())
# }
```
The `Collector` trait is implemented for up to 4 collectors.
If you have more than 4 collectors, you can either group them into
tuples of tuples `(a,(b,(c,d)))`, or rely on [`MultiCollector`](./struct.MultiCollector.html).
# Combining several collectors dynamically
Combining collectors into a tuple is a zero-cost abstraction: everything
happens as if you had manually implemented a single collector
combining all of our features.
Unfortunately it requires you to know at compile time your collector types.
If on the other hand, the collectors depend on some query parameter,
you can rely on `MultiCollector`'s.
# Implementing your own collectors.
See the `custom_collector` example.
Defines how the documents matching a search query should be processed.
*/
use crate::DocId;
use crate::Result;
use crate::Score;
use crate::SegmentLocalId;
use crate::SegmentReader;
use downcast_rs::impl_downcast;
use DocId;
use Result;
use Score;
use SegmentLocalId;
use SegmentReader;
mod count_collector;
pub use self::count_collector::Count;
pub use self::count_collector::CountCollector;
mod multi_collector;
pub use self::multi_collector::MultiCollector;
mod top_collector;
mod top_score_collector;
pub use self::top_score_collector::TopDocs;
mod custom_score_top_collector;
pub use self::custom_score_top_collector::{CustomScorer, CustomSegmentScorer};
mod tweak_score_top_collector;
pub use self::tweak_score_top_collector::{ScoreSegmentTweaker, ScoreTweaker};
pub use self::top_collector::TopCollector;
mod facet_collector;
pub use self::facet_collector::FacetCollector;
/// `Fruit` is the type for the result of our collection.
/// e.g. `usize` for the `Count` collector.
pub trait Fruit: Send + downcast_rs::Downcast {}
impl<T> Fruit for T where T: Send + downcast_rs::Downcast {}
mod chained_collector;
pub use self::chained_collector::{chain, ChainedCollector};
/// Collectors are in charge of collecting and retaining relevant
/// information from the document found and scored by the query.
///
///
/// For instance,
///
/// - keeping track of the top 10 best documents
/// - computing a breakdown over a fast field
/// - computing the number of documents matching the query
///
/// Our search index is in fact a collection of segments, so
/// a `Collector` trait is actually more of a factory to instance
/// `SegmentCollector`s for each segments.
/// Queries are in charge of pushing the `DocSet` to the collector.
///
/// The collection logic itself is in the `SegmentCollector`.
/// As they work on multiple segments, they first inform
/// the collector of a change in a segment and then
/// call the `collect` method to push the document to the collector.
///
/// Temporally, our collector will receive calls
/// - `.set_segment(0, segment_reader_0)`
/// - `.collect(doc0_of_segment_0)`
/// - `.collect(...)`
/// - `.collect(last_doc_of_segment_0)`
/// - `.set_segment(1, segment_reader_1)`
/// - `.collect(doc0_of_segment_1)`
/// - `.collect(...)`
/// - `.collect(last_doc_of_segment_1)`
/// - `...`
/// - `.collect(last_doc_of_last_segment)`
///
/// Segments are not guaranteed to be visited in any specific order.
pub trait Collector: Sync {
/// `Fruit` is the type for the result of our collection.
/// e.g. `usize` for the `Count` collector.
type Fruit: Fruit;
/// Type of the `SegmentCollector` associated to this collector.
type Child: SegmentCollector<Fruit = Self::Fruit>;
pub trait Collector {
/// `set_segment` is called before beginning to enumerate
/// on this segment.
fn for_segment(
&self,
fn set_segment(
&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader,
) -> Result<Self::Child>;
/// Returns true iff the collector requires to compute scores for documents.
fn requires_scoring(&self) -> bool;
/// Combines the fruit associated to the collection of each segments
/// into one fruit.
fn merge_fruits(&self, segment_fruits: Vec<Self::Fruit>) -> Result<Self::Fruit>;
}
/// The `SegmentCollector` is the trait in charge of defining the
/// collect operation at the scale of the segment.
///
/// `.collect(doc, score)` will be called for every documents
/// matching the query.
pub trait SegmentCollector: 'static {
/// `Fruit` is the type for the result of our collection.
/// e.g. `usize` for the `Count` collector.
type Fruit: Fruit;
) -> Result<()>;
/// The query pushes the scored document to the collector via this method.
fn collect(&mut self, doc: DocId, score: Score);
/// Extract the fruit of the collection from the `SegmentCollector`.
fn harvest(self) -> Self::Fruit;
/// Returns true iff the collector requires to compute scores for documents.
fn requires_scoring(&self) -> bool;
}
// -----------------------------------------------
// Tuple implementations.
impl<Left, Right> Collector for (Left, Right)
where
Left: Collector,
Right: Collector,
{
type Fruit = (Left::Fruit, Right::Fruit);
type Child = (Left::Child, Right::Child);
fn for_segment(&self, segment_local_id: u32, segment: &SegmentReader) -> Result<Self::Child> {
let left = self.0.for_segment(segment_local_id, segment)?;
let right = self.1.for_segment(segment_local_id, segment)?;
Ok((left, right))
impl<'a, C: Collector> Collector for &'a mut C {
fn set_segment(
&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader,
) -> Result<()> {
(*self).set_segment(segment_local_id, segment)
}
/// The query pushes the scored document to the collector via this method.
fn collect(&mut self, doc: DocId, score: Score) {
C::collect(self, doc, score)
}
fn requires_scoring(&self) -> bool {
self.0.requires_scoring() || self.1.requires_scoring()
}
fn merge_fruits(
&self,
children: Vec<(Left::Fruit, Right::Fruit)>,
) -> Result<(Left::Fruit, Right::Fruit)> {
let mut left_fruits = vec![];
let mut right_fruits = vec![];
for (left_fruit, right_fruit) in children {
left_fruits.push(left_fruit);
right_fruits.push(right_fruit);
}
Ok((
self.0.merge_fruits(left_fruits)?,
self.1.merge_fruits(right_fruits)?,
))
C::requires_scoring(self)
}
}
impl<Left, Right> SegmentCollector for (Left, Right)
where
Left: SegmentCollector,
Right: SegmentCollector,
{
type Fruit = (Left::Fruit, Right::Fruit);
fn collect(&mut self, doc: DocId, score: Score) {
self.0.collect(doc, score);
self.1.collect(doc, score);
}
fn harvest(self) -> <Self as SegmentCollector>::Fruit {
(self.0.harvest(), self.1.harvest())
}
}
// 3-Tuple
impl<One, Two, Three> Collector for (One, Two, Three)
where
One: Collector,
Two: Collector,
Three: Collector,
{
type Fruit = (One::Fruit, Two::Fruit, Three::Fruit);
type Child = (One::Child, Two::Child, Three::Child);
fn for_segment(&self, segment_local_id: u32, segment: &SegmentReader) -> Result<Self::Child> {
let one = self.0.for_segment(segment_local_id, segment)?;
let two = self.1.for_segment(segment_local_id, segment)?;
let three = self.2.for_segment(segment_local_id, segment)?;
Ok((one, two, three))
}
fn requires_scoring(&self) -> bool {
self.0.requires_scoring() || self.1.requires_scoring() || self.2.requires_scoring()
}
fn merge_fruits(&self, children: Vec<Self::Fruit>) -> Result<Self::Fruit> {
let mut one_fruits = vec![];
let mut two_fruits = vec![];
let mut three_fruits = vec![];
for (one_fruit, two_fruit, three_fruit) in children {
one_fruits.push(one_fruit);
two_fruits.push(two_fruit);
three_fruits.push(three_fruit);
}
Ok((
self.0.merge_fruits(one_fruits)?,
self.1.merge_fruits(two_fruits)?,
self.2.merge_fruits(three_fruits)?,
))
}
}
impl<One, Two, Three> SegmentCollector for (One, Two, Three)
where
One: SegmentCollector,
Two: SegmentCollector,
Three: SegmentCollector,
{
type Fruit = (One::Fruit, Two::Fruit, Three::Fruit);
fn collect(&mut self, doc: DocId, score: Score) {
self.0.collect(doc, score);
self.1.collect(doc, score);
self.2.collect(doc, score);
}
fn harvest(self) -> <Self as SegmentCollector>::Fruit {
(self.0.harvest(), self.1.harvest(), self.2.harvest())
}
}
// 4-Tuple
impl<One, Two, Three, Four> Collector for (One, Two, Three, Four)
where
One: Collector,
Two: Collector,
Three: Collector,
Four: Collector,
{
type Fruit = (One::Fruit, Two::Fruit, Three::Fruit, Four::Fruit);
type Child = (One::Child, Two::Child, Three::Child, Four::Child);
fn for_segment(&self, segment_local_id: u32, segment: &SegmentReader) -> Result<Self::Child> {
let one = self.0.for_segment(segment_local_id, segment)?;
let two = self.1.for_segment(segment_local_id, segment)?;
let three = self.2.for_segment(segment_local_id, segment)?;
let four = self.3.for_segment(segment_local_id, segment)?;
Ok((one, two, three, four))
}
fn requires_scoring(&self) -> bool {
self.0.requires_scoring()
|| self.1.requires_scoring()
|| self.2.requires_scoring()
|| self.3.requires_scoring()
}
fn merge_fruits(&self, children: Vec<Self::Fruit>) -> Result<Self::Fruit> {
let mut one_fruits = vec![];
let mut two_fruits = vec![];
let mut three_fruits = vec![];
let mut four_fruits = vec![];
for (one_fruit, two_fruit, three_fruit, four_fruit) in children {
one_fruits.push(one_fruit);
two_fruits.push(two_fruit);
three_fruits.push(three_fruit);
four_fruits.push(four_fruit);
}
Ok((
self.0.merge_fruits(one_fruits)?,
self.1.merge_fruits(two_fruits)?,
self.2.merge_fruits(three_fruits)?,
self.3.merge_fruits(four_fruits)?,
))
}
}
impl<One, Two, Three, Four> SegmentCollector for (One, Two, Three, Four)
where
One: SegmentCollector,
Two: SegmentCollector,
Three: SegmentCollector,
Four: SegmentCollector,
{
type Fruit = (One::Fruit, Two::Fruit, Three::Fruit, Four::Fruit);
fn collect(&mut self, doc: DocId, score: Score) {
self.0.collect(doc, score);
self.1.collect(doc, score);
self.2.collect(doc, score);
self.3.collect(doc, score);
}
fn harvest(self) -> <Self as SegmentCollector>::Fruit {
(
self.0.harvest(),
self.1.harvest(),
self.2.harvest(),
self.3.harvest(),
)
}
}
impl_downcast!(Fruit);
#[cfg(test)]
pub mod tests;
pub mod tests {
use super::*;
use core::SegmentReader;
use fastfield::BytesFastFieldReader;
use fastfield::FastFieldReader;
use schema::Field;
use DocId;
use Score;
use SegmentLocalId;
/// Stores all of the doc ids.
/// This collector is only used for tests.
/// It is unusable in practise, as it does not store
/// the segment ordinals
pub struct TestCollector {
offset: DocId,
segment_max_doc: DocId,
docs: Vec<DocId>,
scores: Vec<Score>,
}
impl TestCollector {
/// Return the exhalist of documents.
pub fn docs(self) -> Vec<DocId> {
self.docs
}
pub fn scores(self) -> Vec<Score> {
self.scores
}
}
impl Default for TestCollector {
fn default() -> TestCollector {
TestCollector {
offset: 0,
segment_max_doc: 0,
docs: Vec::new(),
scores: Vec::new(),
}
}
}
impl Collector for TestCollector {
fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
self.offset += self.segment_max_doc;
self.segment_max_doc = reader.max_doc();
Ok(())
}
fn collect(&mut self, doc: DocId, score: Score) {
self.docs.push(doc + self.offset);
self.scores.push(score);
}
fn requires_scoring(&self) -> bool {
true
}
}
/// Collects in order all of the fast fields for all of the
/// doc in the `DocSet`
///
/// This collector is mainly useful for tests.
pub struct FastFieldTestCollector {
vals: Vec<u64>,
field: Field,
ff_reader: Option<FastFieldReader<u64>>,
}
impl FastFieldTestCollector {
pub fn for_field(field: Field) -> FastFieldTestCollector {
FastFieldTestCollector {
vals: Vec::new(),
field,
ff_reader: None,
}
}
pub fn vals(self) -> Vec<u64> {
self.vals
}
}
impl Collector for FastFieldTestCollector {
fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
self.ff_reader = Some(reader.fast_field_reader(self.field)?);
Ok(())
}
fn collect(&mut self, doc: DocId, _score: Score) {
let val = self.ff_reader.as_ref().unwrap().get(doc);
self.vals.push(val);
}
fn requires_scoring(&self) -> bool {
false
}
}
/// Collects in order all of the fast field bytes for all of the
/// docs in the `DocSet`
///
/// This collector is mainly useful for tests.
pub struct BytesFastFieldTestCollector {
vals: Vec<u8>,
field: Field,
ff_reader: Option<BytesFastFieldReader>,
}
impl BytesFastFieldTestCollector {
pub fn for_field(field: Field) -> BytesFastFieldTestCollector {
BytesFastFieldTestCollector {
vals: Vec::new(),
field,
ff_reader: None,
}
}
pub fn vals(self) -> Vec<u8> {
self.vals
}
}
impl Collector for BytesFastFieldTestCollector {
fn set_segment(&mut self, _segment_local_id: u32, segment: &SegmentReader) -> Result<()> {
self.ff_reader = Some(segment.bytes_fast_field_reader(self.field)?);
Ok(())
}
fn collect(&mut self, doc: u32, _score: f32) {
let val = self.ff_reader.as_ref().unwrap().get_val(doc);
self.vals.extend(val);
}
fn requires_scoring(&self) -> bool {
false
}
}
}
#[cfg(all(test, feature = "unstable"))]
mod bench {
use collector::{Collector, CountCollector};
use test::Bencher;
#[bench]
fn build_collector(b: &mut Bencher) {
b.iter(|| {
let mut count_collector = CountCollector::default();
let docs: Vec<u32> = (0..1_000_000).collect();
for doc in docs {
count_collector.collect(doc, 1f32);
}
count_collector.count()
});
}
}

283
src/collector/multi_collector.rs
View File

@@ -1,118 +1,26 @@
use super::Collector;
use super::SegmentCollector;
use crate::collector::Fruit;
use crate::DocId;
use crate::Result;
use crate::Score;
use crate::SegmentLocalId;
use crate::SegmentReader;
use crate::TantivyError;
use std::marker::PhantomData;
use std::ops::Deref;
pub struct MultiFruit {
sub_fruits: Vec<Option<Box<dyn Fruit>>>,
}
pub struct CollectorWrapper<TCollector: Collector>(TCollector);
impl<TCollector: Collector> Collector for CollectorWrapper<TCollector> {
type Fruit = Box<dyn Fruit>;
type Child = Box<dyn BoxableSegmentCollector>;
fn for_segment(
&self,
segment_local_id: u32,
reader: &SegmentReader,
) -> Result<Box<dyn BoxableSegmentCollector>> {
let child = self.0.for_segment(segment_local_id, reader)?;
Ok(Box::new(SegmentCollectorWrapper(child)))
}
fn requires_scoring(&self) -> bool {
self.0.requires_scoring()
}
fn merge_fruits(&self, children: Vec<<Self as Collector>::Fruit>) -> Result<Box<dyn Fruit>> {
let typed_fruit: Vec<TCollector::Fruit> = children
.into_iter()
.map(|untyped_fruit| {
untyped_fruit
.downcast::<TCollector::Fruit>()
.map(|boxed_but_typed| *boxed_but_typed)
.map_err(|_| {
TantivyError::InvalidArgument("Failed to cast child fruit.".to_string())
})
})
.collect::<Result<_>>()?;
let merged_fruit = self.0.merge_fruits(typed_fruit)?;
Ok(Box::new(merged_fruit))
}
}
impl SegmentCollector for Box<dyn BoxableSegmentCollector> {
type Fruit = Box<dyn Fruit>;
fn collect(&mut self, doc: u32, score: f32) {
self.as_mut().collect(doc, score);
}
fn harvest(self) -> Box<dyn Fruit> {
BoxableSegmentCollector::harvest_from_box(self)
}
}
pub trait BoxableSegmentCollector {
fn collect(&mut self, doc: u32, score: f32);
fn harvest_from_box(self: Box<Self>) -> Box<dyn Fruit>;
}
pub struct SegmentCollectorWrapper<TSegmentCollector: SegmentCollector>(TSegmentCollector);
impl<TSegmentCollector: SegmentCollector> BoxableSegmentCollector
for SegmentCollectorWrapper<TSegmentCollector>
{
fn collect(&mut self, doc: u32, score: f32) {
self.0.collect(doc, score);
}
fn harvest_from_box(self: Box<Self>) -> Box<dyn Fruit> {
Box::new(self.0.harvest())
}
}
pub struct FruitHandle<TFruit: Fruit> {
pos: usize,
_phantom: PhantomData<TFruit>,
}
impl<TFruit: Fruit> FruitHandle<TFruit> {
pub fn extract(self, fruits: &mut MultiFruit) -> TFruit {
let boxed_fruit = fruits.sub_fruits[self.pos].take().expect("");
*boxed_fruit
.downcast::<TFruit>()
.map_err(|_| ())
.expect("Failed to downcast collector fruit.")
}
}
use DocId;
use Result;
use Score;
use SegmentLocalId;
use SegmentReader;
/// Multicollector makes it possible to collect on more than one collector.
/// It should only be used for use cases where the Collector types is unknown
/// at compile time.
///
/// If the type of the collectors is known, you can just group yours collectors
/// in a tuple. See the
/// [Combining several collectors section of the collector documentation](./index.html#combining-several-collectors).
/// If the type of the collectors is known, you should prefer to use `ChainedCollector`.
///
/// ```rust
/// use tantivy::collector::{Count, TopDocs, MultiCollector};
/// #[macro_use]
/// extern crate tantivy;
/// use tantivy::schema::{SchemaBuilder, TEXT};
/// use tantivy::{Index, Result};
/// use tantivy::collector::{CountCollector, TopCollector, MultiCollector};
/// use tantivy::query::QueryParser;
/// use tantivy::schema::{Schema, TEXT};
/// use tantivy::{doc, Index, Result};
///
/// # fn main() { example().unwrap(); }
/// fn example() -> Result<()> {
/// let mut schema_builder = Schema::builder();
/// let mut schema_builder = SchemaBuilder::new();
/// let title = schema_builder.add_text_field("title", TEXT);
/// let schema = schema_builder.build();
/// let index = Index::create_in_ram(schema);
@@ -133,122 +41,58 @@ impl<TFruit: Fruit> FruitHandle<TFruit> {
/// index_writer.commit().unwrap();
/// }
///
/// let reader = index.reader()?;
/// let searcher = reader.searcher();
/// index.load_searchers()?;
/// let searcher = index.searcher();
///
/// let mut collectors = MultiCollector::new();
/// let top_docs_handle = collectors.add_collector(TopDocs::with_limit(2));
/// let count_handle = collectors.add_collector(Count);
/// let query_parser = QueryParser::for_index(&index, vec![title]);
/// let query = query_parser.parse_query("diary")?;
/// let mut multi_fruit = searcher.search(&query, &collectors)?;
///
/// let count = count_handle.extract(&mut multi_fruit);
/// let top_docs = top_docs_handle.extract(&mut multi_fruit);
///
/// # assert_eq!(count, 2);
/// # assert_eq!(top_docs.len(), 2);
/// {
/// let mut top_collector = TopCollector::with_limit(2);
/// let mut count_collector = CountCollector::default();
/// {
/// let mut collectors =
/// MultiCollector::from(vec![&mut top_collector, &mut count_collector]);
/// let query_parser = QueryParser::for_index(&index, vec![title]);
/// let query = query_parser.parse_query("diary")?;
/// searcher.search(&*query, &mut collectors).unwrap();
/// }
/// assert_eq!(count_collector.count(), 2);
/// assert!(top_collector.at_capacity());
/// }
///
/// Ok(())
/// }
/// ```
#[allow(clippy::type_complexity)]
#[derive(Default)]
pub struct MultiCollector<'a> {
collector_wrappers: Vec<
Box<dyn Collector<Child = Box<dyn BoxableSegmentCollector>, Fruit = Box<dyn Fruit>> + 'a>,
>,
collectors: Vec<&'a mut Collector>,
}
impl<'a> MultiCollector<'a> {
/// Create a new `MultiCollector`
pub fn new() -> Self {
Default::default()
}
/// Add a new collector to our `MultiCollector`.
pub fn add_collector<'b: 'a, TCollector: Collector + 'b>(
&mut self,
collector: TCollector,
) -> FruitHandle<TCollector::Fruit> {
let pos = self.collector_wrappers.len();
self.collector_wrappers
.push(Box::new(CollectorWrapper(collector)));
FruitHandle {
pos,
_phantom: PhantomData,
}
/// Constructor
pub fn from(collectors: Vec<&'a mut Collector>) -> MultiCollector {
MultiCollector { collectors }
}
}
impl<'a> Collector for MultiCollector<'a> {
type Fruit = MultiFruit;
type Child = MultiCollectorChild;
fn for_segment(
&self,
fn set_segment(
&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader,
) -> Result<MultiCollectorChild> {
let children = self
.collector_wrappers
.iter()
.map(|collector_wrapper| collector_wrapper.for_segment(segment_local_id, segment))
.collect::<Result<Vec<_>>>()?;
Ok(MultiCollectorChild { children })
}
fn requires_scoring(&self) -> bool {
self.collector_wrappers
.iter()
.map(Deref::deref)
.any(Collector::requires_scoring)
}
fn merge_fruits(&self, segments_multifruits: Vec<MultiFruit>) -> Result<MultiFruit> {
let mut segment_fruits_list: Vec<Vec<Box<dyn Fruit>>> = (0..self.collector_wrappers.len())
.map(|_| Vec::with_capacity(segments_multifruits.len()))
.collect::<Vec<_>>();
for segment_multifruit in segments_multifruits {
for (idx, segment_fruit_opt) in segment_multifruit.sub_fruits.into_iter().enumerate() {
if let Some(segment_fruit) = segment_fruit_opt {
segment_fruits_list[idx].push(segment_fruit);
}
}
) -> Result<()> {
for collector in &mut self.collectors {
collector.set_segment(segment_local_id, segment)?;
}
let sub_fruits = self
.collector_wrappers
.iter()
.zip(segment_fruits_list)
.map(|(child_collector, segment_fruits)| {
Ok(Some(child_collector.merge_fruits(segment_fruits)?))
})
.collect::<Result<_>>()?;
Ok(MultiFruit { sub_fruits })
Ok(())
}
}
pub struct MultiCollectorChild {
children: Vec<Box<dyn BoxableSegmentCollector>>,
}
impl SegmentCollector for MultiCollectorChild {
type Fruit = MultiFruit;
fn collect(&mut self, doc: DocId, score: Score) {
for child in &mut self.children {
child.collect(doc, score);
for collector in &mut self.collectors {
collector.collect(doc, score);
}
}
fn harvest(self) -> MultiFruit {
MultiFruit {
sub_fruits: self
.children
.into_iter()
.map(|child| Some(child.harvest()))
.collect(),
}
fn requires_scoring(&self) -> bool {
self.collectors
.iter()
.any(|collector| collector.requires_scoring())
}
}
@@ -256,41 +100,20 @@ impl SegmentCollector for MultiCollectorChild {
mod tests {
use super::*;
use crate::collector::{Count, TopDocs};
use crate::query::TermQuery;
use crate::schema::IndexRecordOption;
use crate::schema::{Schema, TEXT};
use crate::Index;
use crate::Term;
use collector::{Collector, CountCollector, TopCollector};
#[test]
fn test_multi_collector() {
let mut schema_builder = Schema::builder();
let text = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut top_collector = TopCollector::with_limit(2);
let mut count_collector = CountCollector::default();
{
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
index_writer.add_document(doc!(text=>"abc"));
index_writer.add_document(doc!(text=>"abc abc abc"));
index_writer.add_document(doc!(text=>"abc abc"));
index_writer.commit().unwrap();
index_writer.add_document(doc!(text=>""));
index_writer.add_document(doc!(text=>"abc abc abc abc"));
index_writer.add_document(doc!(text=>"abc"));
index_writer.commit().unwrap();
let mut collectors =
MultiCollector::from(vec![&mut top_collector, &mut count_collector]);
collectors.collect(1, 0.2);
collectors.collect(2, 0.1);
collectors.collect(3, 0.5);
}
let searcher = index.reader().unwrap().searcher();
let term = Term::from_field_text(text, "abc");
let query = TermQuery::new(term, IndexRecordOption::Basic);
let mut collectors = MultiCollector::new();
let topdocs_handler = collectors.add_collector(TopDocs::with_limit(2));
let count_handler = collectors.add_collector(Count);
let mut multifruits = searcher.search(&query, &mut collectors).unwrap();
assert_eq!(count_handler.extract(&mut multifruits), 5);
assert_eq!(topdocs_handler.extract(&mut multifruits).len(), 2);
assert_eq!(count_collector.count(), 3);
assert!(top_collector.at_capacity());
}
}

217
src/collector/tests.rs
View File

@@ -1,217 +0,0 @@
use super::*;
use crate::core::SegmentReader;
use crate::fastfield::BytesFastFieldReader;
use crate::fastfield::FastFieldReader;
use crate::schema::Field;
use crate::DocAddress;
use crate::DocId;
use crate::Score;
use crate::SegmentLocalId;
pub const TEST_COLLECTOR_WITH_SCORE: TestCollector = TestCollector {
compute_score: true,
};
pub const TEST_COLLECTOR_WITHOUT_SCORE: TestCollector = TestCollector {
compute_score: true,
};
/// Stores all of the doc ids.
/// This collector is only used for tests.
/// It is unusable in pr
///
/// actise, as it does not store
/// the segment ordinals
pub struct TestCollector {
pub compute_score: bool,
}
pub struct TestSegmentCollector {
segment_id: SegmentLocalId,
fruit: TestFruit,
}
#[derive(Default)]
pub struct TestFruit {
docs: Vec<DocAddress>,
scores: Vec<Score>,
}
impl TestFruit {
/// Return the list of matching documents exhaustively.
pub fn docs(&self) -> &[DocAddress] {
&self.docs[..]
}
pub fn scores(&self) -> &[Score] {
&self.scores[..]
}
}
impl Collector for TestCollector {
type Fruit = TestFruit;
type Child = TestSegmentCollector;
fn for_segment(
&self,
segment_id: SegmentLocalId,
_reader: &SegmentReader,
) -> Result<TestSegmentCollector> {
Ok(TestSegmentCollector {
segment_id,
fruit: TestFruit::default(),
})
}
fn requires_scoring(&self) -> bool {
self.compute_score
}
fn merge_fruits(&self, mut children: Vec<TestFruit>) -> Result<TestFruit> {
children.sort_by_key(|fruit| {
if fruit.docs().is_empty() {
0
} else {
fruit.docs()[0].segment_ord()
}
});
let mut docs = vec![];
let mut scores = vec![];
for child in children {
docs.extend(child.docs());
scores.extend(child.scores);
}
Ok(TestFruit { docs, scores })
}
}
impl SegmentCollector for TestSegmentCollector {
type Fruit = TestFruit;
fn collect(&mut self, doc: DocId, score: Score) {
self.fruit.docs.push(DocAddress(self.segment_id, doc));
self.fruit.scores.push(score);
}
fn harvest(self) -> <Self as SegmentCollector>::Fruit {
self.fruit
}
}
/// Collects in order all of the fast fields for all of the
/// doc in the `DocSet`
///
/// This collector is mainly useful for tests.
pub struct FastFieldTestCollector {
field: Field,
}
pub struct FastFieldSegmentCollector {
vals: Vec<u64>,
reader: FastFieldReader<u64>,
}
impl FastFieldTestCollector {
pub fn for_field(field: Field) -> FastFieldTestCollector {
FastFieldTestCollector { field }
}
}
impl Collector for FastFieldTestCollector {
type Fruit = Vec<u64>;
type Child = FastFieldSegmentCollector;
fn for_segment(
&self,
_: SegmentLocalId,
segment_reader: &SegmentReader,
) -> Result<FastFieldSegmentCollector> {
let reader = segment_reader
.fast_fields()
.u64(self.field)
.expect("Requested field is not a fast field.");
Ok(FastFieldSegmentCollector {
vals: Vec::new(),
reader,
})
}
fn requires_scoring(&self) -> bool {
false
}
fn merge_fruits(&self, children: Vec<Vec<u64>>) -> Result<Vec<u64>> {
Ok(children.into_iter().flat_map(|v| v.into_iter()).collect())
}
}
impl SegmentCollector for FastFieldSegmentCollector {
type Fruit = Vec<u64>;
fn collect(&mut self, doc: DocId, _score: Score) {
let val = self.reader.get(doc);
self.vals.push(val);
}
fn harvest(self) -> Vec<u64> {
self.vals
}
}
/// Collects in order all of the fast field bytes for all of the
/// docs in the `DocSet`
///
/// This collector is mainly useful for tests.
pub struct BytesFastFieldTestCollector {
field: Field,
}
pub struct BytesFastFieldSegmentCollector {
vals: Vec<u8>,
reader: BytesFastFieldReader,
}
impl BytesFastFieldTestCollector {
pub fn for_field(field: Field) -> BytesFastFieldTestCollector {
BytesFastFieldTestCollector { field }
}
}
impl Collector for BytesFastFieldTestCollector {
type Fruit = Vec<u8>;
type Child = BytesFastFieldSegmentCollector;
fn for_segment(
&self,
_segment_local_id: u32,
segment_reader: &SegmentReader,
) -> Result<BytesFastFieldSegmentCollector> {
Ok(BytesFastFieldSegmentCollector {
vals: Vec::new(),
reader: segment_reader
.fast_fields()
.bytes(self.field)
.expect("Field is not a bytes fast field."),
})
}
fn requires_scoring(&self) -> bool {
false
}
fn merge_fruits(&self, children: Vec<Vec<u8>>) -> Result<Vec<u8>> {
Ok(children.into_iter().flat_map(|c| c.into_iter()).collect())
}
}
impl SegmentCollector for BytesFastFieldSegmentCollector {
type Fruit = Vec<u8>;
fn collect(&mut self, doc: u32, _score: f32) {
let data = self.reader.get_bytes(doc);
self.vals.extend(data);
}
fn harvest(self) -> <Self as SegmentCollector>::Fruit {
self.vals
}
}

319
src/collector/top_collector.rs
View File

@@ -1,217 +1,244 @@
use crate::DocAddress;
use crate::DocId;
use crate::Result;
use crate::SegmentLocalId;
use crate::SegmentReader;
use serde::export::PhantomData;
use super::Collector;
use std::cmp::Ordering;
use std::collections::BinaryHeap;
use DocAddress;
use DocId;
use Result;
use Score;
use SegmentLocalId;
use SegmentReader;
/// Contains a feature (field, score, etc.) of a document along with the document address.
///
/// It has a custom implementation of `PartialOrd` that reverses the order. This is because the
/// default Rust heap is a max heap, whereas a min heap is needed.
///
/// WARNING: equality is not what you would expect here.
/// Two elements are equal if their feature is equal, and regardless of whether `doc`
/// is equal. This should be perfectly fine for this usage, but let's make sure this
/// struct is never public.
struct ComparableDoc<T, D> {
feature: T,
doc: D,
// Rust heap is a max-heap and we need a min heap.
#[derive(Clone, Copy)]
struct GlobalScoredDoc {
score: Score,
doc_address: DocAddress,
}
impl<T: PartialOrd, D> PartialOrd for ComparableDoc<T, D> {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
impl PartialOrd for GlobalScoredDoc {
fn partial_cmp(&self, other: &GlobalScoredDoc) -> Option<Ordering> {
Some(self.cmp(other))
}
}
impl<T: PartialOrd, D> Ord for ComparableDoc<T, D> {
impl Ord for GlobalScoredDoc {
#[inline]
fn cmp(&self, other: &Self) -> Ordering {
fn cmp(&self, other: &GlobalScoredDoc) -> Ordering {
other
.feature
.partial_cmp(&self.feature)
.unwrap_or_else(|| Ordering::Equal)
.score
.partial_cmp(&self.score)
.unwrap_or_else(|| other.doc_address.cmp(&self.doc_address))
}
}
impl<T: PartialOrd, D> PartialEq for ComparableDoc<T, D> {
fn eq(&self, other: &Self) -> bool {
impl PartialEq for GlobalScoredDoc {
fn eq(&self, other: &GlobalScoredDoc) -> bool {
self.cmp(other) == Ordering::Equal
}
}
impl<T: PartialOrd, D> Eq for ComparableDoc<T, D> {}
impl Eq for GlobalScoredDoc {}
pub(crate) struct TopCollector<T> {
/// The Top Collector keeps track of the K documents
/// with the best scores.
///
/// The implementation is based on a `BinaryHeap`.
/// The theorical complexity for collecting the top `K` out of `n` documents
/// is `O(n log K)`.
///
/// ```rust
/// #[macro_use]
/// extern crate tantivy;
/// use tantivy::schema::{SchemaBuilder, TEXT};
/// use tantivy::{Index, Result, DocId, Score};
/// use tantivy::collector::TopCollector;
/// use tantivy::query::QueryParser;
///
/// # fn main() { example().unwrap(); }
/// fn example() -> Result<()> {
/// let mut schema_builder = SchemaBuilder::new();
/// let title = schema_builder.add_text_field("title", TEXT);
/// let schema = schema_builder.build();
/// let index = Index::create_in_ram(schema);
/// {
/// let mut index_writer = index.writer_with_num_threads(1, 3_000_000)?;
/// index_writer.add_document(doc!(
/// title => "The Name of the Wind",
/// ));
/// index_writer.add_document(doc!(
/// title => "The Diary of Muadib",
/// ));
/// index_writer.add_document(doc!(
/// title => "A Dairy Cow",
/// ));
/// index_writer.add_document(doc!(
/// title => "The Diary of a Young Girl",
/// ));
/// index_writer.commit().unwrap();
/// }
///
/// index.load_searchers()?;
/// let searcher = index.searcher();
///
/// {
/// let mut top_collector = TopCollector::with_limit(2);
/// let query_parser = QueryParser::for_index(&index, vec![title]);
/// let query = query_parser.parse_query("diary")?;
/// searcher.search(&*query, &mut top_collector).unwrap();
///
/// let score_docs: Vec<(Score, DocId)> = top_collector
/// .score_docs()
/// .into_iter()
/// .map(|(score, doc_address)| (score, doc_address.doc()))
/// .collect();
///
/// assert_eq!(score_docs, vec![(0.7261542, 1), (0.6099695, 3)]);
/// }
///
/// Ok(())
/// }
/// ```
pub struct TopCollector {
limit: usize,
_marker: PhantomData<T>,
heap: BinaryHeap<GlobalScoredDoc>,
segment_id: u32,
}
impl<T> TopCollector<T>
where
T: PartialOrd + Clone,
{
impl TopCollector {
/// Creates a top collector, with a number of documents equal to "limit".
///
/// # Panics
/// The method panics if limit is 0
pub fn with_limit(limit: usize) -> TopCollector<T> {
pub fn with_limit(limit: usize) -> TopCollector {
if limit < 1 {
panic!("Limit must be strictly greater than 0.");
}
TopCollector {
limit,
_marker: PhantomData,
}
}
pub fn limit(&self) -> usize {
self.limit
}
pub fn merge_fruits(
&self,
children: Vec<Vec<(T, DocAddress)>>,
) -> Result<Vec<(T, DocAddress)>> {
if self.limit == 0 {
return Ok(Vec::new());
}
let mut top_collector = BinaryHeap::new();
for child_fruit in children {
for (feature, doc) in child_fruit {
if top_collector.len() < self.limit {
top_collector.push(ComparableDoc { feature, doc });
} else if let Some(mut head) = top_collector.peek_mut() {
if head.feature < feature {
*head = ComparableDoc { feature, doc };
}
}
}
}
Ok(top_collector
.into_sorted_vec()
.into_iter()
.map(|cdoc| (cdoc.feature, cdoc.doc))
.collect())
}
pub(crate) fn for_segment<F: PartialOrd>(
&self,
segment_id: SegmentLocalId,
_: &SegmentReader,
) -> Result<TopSegmentCollector<F>> {
Ok(TopSegmentCollector::new(segment_id, self.limit))
}
}
/// The Top Collector keeps track of the K documents
/// sorted by type `T`.
///
/// The implementation is based on a `BinaryHeap`.
/// The theorical complexity for collecting the top `K` out of `n` documents
/// is `O(n log K)`.
pub(crate) struct TopSegmentCollector<T> {
limit: usize,
heap: BinaryHeap<ComparableDoc<T, DocId>>,
segment_id: u32,
}
impl<T: PartialOrd> TopSegmentCollector<T> {
fn new(segment_id: SegmentLocalId, limit: usize) -> TopSegmentCollector<T> {
TopSegmentCollector {
limit,
heap: BinaryHeap::with_capacity(limit),
segment_id,
segment_id: 0,
}
}
}
impl<T: PartialOrd + Clone> TopSegmentCollector<T> {
pub fn harvest(self) -> Vec<(T, DocAddress)> {
let segment_id = self.segment_id;
self.heap
.into_sorted_vec()
/// Returns K best documents sorted in decreasing order.
///
/// Calling this method triggers the sort.
/// The result of the sort is not cached.
pub fn docs(&self) -> Vec<DocAddress> {
self.score_docs()
.into_iter()
.map(|comparable_doc| {
(
comparable_doc.feature,
DocAddress(segment_id, comparable_doc.doc),
)
})
.map(|score_doc| score_doc.1)
.collect()
}
/// Returns K best ScoredDocument sorted in decreasing order.
///
/// Calling this method triggers the sort.
/// The result of the sort is not cached.
pub fn score_docs(&self) -> Vec<(Score, DocAddress)> {
let mut scored_docs: Vec<GlobalScoredDoc> = self.heap.iter().cloned().collect();
scored_docs.sort();
scored_docs
.into_iter()
.map(|GlobalScoredDoc { score, doc_address }| (score, doc_address))
.collect()
}
/// Return true iff at least K documents have gone through
/// the collector.
#[inline(always)]
pub(crate) fn at_capacity(&self) -> bool {
#[inline]
pub fn at_capacity(&self) -> bool {
self.heap.len() >= self.limit
}
}
/// Collects a document scored by the given feature
///
/// It collects documents until it has reached the max capacity. Once it reaches capacity, it
/// will compare the lowest scoring item with the given one and keep whichever is greater.
#[inline(always)]
pub fn collect(&mut self, doc: DocId, feature: T) {
impl Collector for TopCollector {
fn set_segment(&mut self, segment_id: SegmentLocalId, _: &SegmentReader) -> Result<()> {
self.segment_id = segment_id;
Ok(())
}
fn collect(&mut self, doc: DocId, score: Score) {
if self.at_capacity() {
// It's ok to unwrap as long as a limit of 0 is forbidden.
if let Some(limit_feature) = self.heap.peek().map(|head| head.feature.clone()) {
if limit_feature < feature {
if let Some(mut head) = self.heap.peek_mut() {
head.feature = feature;
head.doc = doc;
}
}
let limit_doc: GlobalScoredDoc = *self.heap
.peek()
.expect("Top collector with size 0 is forbidden");
if limit_doc.score < score {
let mut mut_head = self.heap
.peek_mut()
.expect("Top collector with size 0 is forbidden");
mut_head.score = score;
mut_head.doc_address = DocAddress(self.segment_id, doc);
}
} else {
// we have not reached capacity yet, so we can just push the
// element.
self.heap.push(ComparableDoc { feature, doc });
let wrapped_doc = GlobalScoredDoc {
score,
doc_address: DocAddress(self.segment_id, doc),
};
self.heap.push(wrapped_doc);
}
}
fn requires_scoring(&self) -> bool {
true
}
}
#[cfg(test)]
mod tests {
use super::TopSegmentCollector;
use crate::DocAddress;
use super::*;
use collector::Collector;
use DocId;
use Score;
#[test]
fn test_top_collector_not_at_capacity() {
let mut top_collector = TopSegmentCollector::new(0, 4);
let mut top_collector = TopCollector::with_limit(4);
top_collector.collect(1, 0.8);
top_collector.collect(3, 0.2);
top_collector.collect(5, 0.3);
assert_eq!(
top_collector.harvest(),
vec![
(0.8, DocAddress(0, 1)),
(0.3, DocAddress(0, 5)),
(0.2, DocAddress(0, 3))
]
);
assert!(!top_collector.at_capacity());
let score_docs: Vec<(Score, DocId)> = top_collector
.score_docs()
.into_iter()
.map(|(score, doc_address)| (score, doc_address.doc()))
.collect();
assert_eq!(score_docs, vec![(0.8, 1), (0.3, 5), (0.2, 3)]);
}
#[test]
fn test_top_collector_at_capacity() {
let mut top_collector = TopSegmentCollector::new(0, 4);
let mut top_collector = TopCollector::with_limit(4);
top_collector.collect(1, 0.8);
top_collector.collect(3, 0.2);
top_collector.collect(5, 0.3);
top_collector.collect(7, 0.9);
top_collector.collect(9, -0.2);
assert_eq!(
top_collector.harvest(),
vec![
(0.9, DocAddress(0, 7)),
(0.8, DocAddress(0, 1)),
(0.3, DocAddress(0, 5)),
(0.2, DocAddress(0, 3))
]
);
assert!(top_collector.at_capacity());
{
let score_docs: Vec<(Score, DocId)> = top_collector
.score_docs()
.into_iter()
.map(|(score, doc_address)| (score, doc_address.doc()))
.collect();
assert_eq!(score_docs, vec![(0.9, 7), (0.8, 1), (0.3, 5), (0.2, 3)]);
}
{
let docs: Vec<DocId> = top_collector
.docs()
.into_iter()
.map(|doc_address| doc_address.doc())
.collect();
assert_eq!(docs, vec![7, 1, 5, 3]);
}
}
#[test]
#[should_panic]
fn test_top_0() {
TopCollector::with_limit(0);
}
}

595
src/collector/top_score_collector.rs
View File

@@ -1,595 +0,0 @@
use super::Collector;
use crate::collector::custom_score_top_collector::CustomScoreTopCollector;
use crate::collector::top_collector::TopCollector;
use crate::collector::top_collector::TopSegmentCollector;
use crate::collector::tweak_score_top_collector::TweakedScoreTopCollector;
use crate::collector::{
CustomScorer, CustomSegmentScorer, ScoreSegmentTweaker, ScoreTweaker, SegmentCollector,
};
use crate::schema::Field;
use crate::DocAddress;
use crate::DocId;
use crate::Result;
use crate::Score;
use crate::SegmentLocalId;
use crate::SegmentReader;
use std::fmt;
/// The Top Score Collector keeps track of the K documents
/// sorted by their score.
///
/// The implementation is based on a `BinaryHeap`.
/// The theorical complexity for collecting the top `K` out of `n` documents
/// is `O(n log K)`.
///
/// ```rust
/// use tantivy::collector::TopDocs;
/// use tantivy::query::QueryParser;
/// use tantivy::schema::{Schema, TEXT};
/// use tantivy::{doc, DocAddress, Index, Result};
///
/// # fn main() { example().unwrap(); }
/// fn example() -> Result<()> {
/// let mut schema_builder = Schema::builder();
/// let title = schema_builder.add_text_field("title", TEXT);
/// let schema = schema_builder.build();
/// let index = Index::create_in_ram(schema);
/// {
/// let mut index_writer = index.writer_with_num_threads(1, 3_000_000)?;
/// index_writer.add_document(doc!(
/// title => "The Name of the Wind",
/// ));
/// index_writer.add_document(doc!(
/// title => "The Diary of Muadib",
/// ));
/// index_writer.add_document(doc!(
/// title => "A Dairy Cow",
/// ));
/// index_writer.add_document(doc!(
/// title => "The Diary of a Young Girl",
/// ));
/// index_writer.commit().unwrap();
/// }
///
/// let reader = index.reader()?;
/// let searcher = reader.searcher();
///
/// let query_parser = QueryParser::for_index(&index, vec![title]);
/// let query = query_parser.parse_query("diary")?;
/// let top_docs = searcher.search(&query, &TopDocs::with_limit(2))?;
///
/// assert_eq!(&top_docs[0], &(0.7261542, DocAddress(0, 1)));
/// assert_eq!(&top_docs[1], &(0.6099695, DocAddress(0, 3)));
///
/// Ok(())
/// }
/// ```
pub struct TopDocs(TopCollector<Score>);
impl fmt::Debug for TopDocs {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "TopDocs({})", self.0.limit())
}
}
impl TopDocs {
/// Creates a top score collector, with a number of documents equal to "limit".
///
/// # Panics
/// The method panics if limit is 0
pub fn with_limit(limit: usize) -> TopDocs {
TopDocs(TopCollector::with_limit(limit))
}
/// Set top-K to rank documents by a given fast field.
///
/// ```rust
/// # use tantivy::schema::{Schema, FAST, TEXT};
/// # use tantivy::{doc, Index, Result, DocAddress};
/// # use tantivy::query::{Query, QueryParser};
/// use tantivy::Searcher;
/// use tantivy::collector::TopDocs;
/// use tantivy::schema::Field;
///
/// # fn main() -> tantivy::Result<()> {
/// # let mut schema_builder = Schema::builder();
/// # let title = schema_builder.add_text_field("title", TEXT);
/// # let rating = schema_builder.add_u64_field("rating", FAST);
/// # let schema = schema_builder.build();
/// #
/// # let index = Index::create_in_ram(schema);
/// # let mut index_writer = index.writer_with_num_threads(1, 3_000_000)?;
/// # index_writer.add_document(doc!(
/// # title => "The Name of the Wind",
/// # rating => 92u64,
/// # ));
/// # index_writer.add_document(doc!(title => "The Diary of Muadib", rating => 97u64));
/// # index_writer.add_document(doc!(title => "A Dairy Cow", rating => 63u64));
/// # index_writer.add_document(doc!(title => "The Diary of a Young Girl", rating => 80u64));
/// # index_writer.commit()?;
/// # let reader = index.reader()?;
/// # let query = QueryParser::for_index(&index, vec![title]).parse_query("diary")?;
/// # let top_docs = docs_sorted_by_rating(&reader.searcher(), &query, rating)?;
/// # assert_eq!(top_docs,
/// # vec![(97u64, DocAddress(0u32, 1)),
/// # (80u64, DocAddress(0u32, 3))]);
/// # Ok(())
/// # }
///
///
/// /// Searches the document matching the given query, and
/// /// collects the top 10 documents, order by the u64-`field`
/// /// given in argument.
/// ///
/// /// `field` is required to be a FAST field.
/// fn docs_sorted_by_rating(searcher: &Searcher,
/// query: &dyn Query,
/// sort_by_field: Field)
/// -> Result<Vec<(u64, DocAddress)>> {
///
/// // This is where we build our topdocs collector
/// //
/// // Note the generics parameter that needs to match the
/// // type `sort_by_field`.
/// let top_docs_by_rating = TopDocs
/// ::with_limit(10)
/// .order_by_u64_field(sort_by_field);
///
/// // ... and here are our documents. Note this is a simple vec.
/// // The `u64` in the pair is the value of our fast field for
/// // each documents.
/// //
/// // The vec is sorted decreasingly by `sort_by_field`, and has a
/// // length of 10, or less if not enough documents matched the
/// // query.
/// let resulting_docs: Vec<(u64, DocAddress)> =
/// searcher.search(query, &top_docs_by_rating)?;
///
/// Ok(resulting_docs)
/// }
/// ```
///
/// # Panics
///
/// May panic if the field requested is not a fast field.
///
pub fn order_by_u64_field(
self,
field: Field,
) -> impl Collector<Fruit = Vec<(u64, DocAddress)>> {
self.custom_score(move |segment_reader: &SegmentReader| {
let ff_reader = segment_reader
.fast_fields()
.u64(field)
.expect("Field requested is not a i64/u64 fast field.");
//TODO error message missmatch actual behavior for i64
move |doc: DocId| ff_reader.get(doc)
})
}
/// Ranks the documents using a custom score.
///
/// This method offers a convenient way to tweak or replace
/// the documents score. As suggested by the prototype you can
/// manually define your own [`ScoreTweaker`](./trait.ScoreTweaker.html)
/// and pass it as an argument, but there is a much simpler way to
/// tweak your score: you can use a closure as in the following
/// example.
///
/// # Example
///
/// Typically, you will want to rely on one or more fast fields,
/// to alter the original relevance `Score`.
///
/// For instance, in the following, we assume that we are implementing
/// an e-commerce website that has a fast field called `popularity`
/// that rates whether a product is typically often bought by users.
///
/// In the following example will will tweak our ranking a bit by
/// boosting popular products a notch.
///
/// In more serious application, this tweaking could involved running a
/// learning-to-rank model over various features
///
/// ```rust
/// # use tantivy::schema::{Schema, FAST, TEXT};
/// # use tantivy::{doc, Index, DocAddress, DocId, Score};
/// # use tantivy::query::QueryParser;
/// use tantivy::SegmentReader;
/// use tantivy::collector::TopDocs;
/// use tantivy::schema::Field;
///
/// # fn create_schema() -> Schema {
/// # let mut schema_builder = Schema::builder();
/// # schema_builder.add_text_field("product_name", TEXT);
/// # schema_builder.add_u64_field("popularity", FAST);
/// # schema_builder.build()
/// # }
/// #
/// # fn main() -> tantivy::Result<()> {
/// # let schema = create_schema();
/// # let index = Index::create_in_ram(schema);
/// # let mut index_writer = index.writer_with_num_threads(1, 3_000_000)?;
/// # let product_name = index.schema().get_field("product_name").unwrap();
/// #
/// let popularity: Field = index.schema().get_field("popularity").unwrap();
/// # index_writer.add_document(doc!(product_name => "The Diary of Muadib", popularity => 1u64));
/// # index_writer.add_document(doc!(product_name => "A Dairy Cow", popularity => 10u64));
/// # index_writer.add_document(doc!(product_name => "The Diary of a Young Girl", popularity => 15u64));
/// # index_writer.commit()?;
/// // ...
/// # let user_query = "diary";
/// # let query = QueryParser::for_index(&index, vec![product_name]).parse_query(user_query)?;
///
/// // This is where we build our collector with our custom score.
/// let top_docs_by_custom_score = TopDocs
/// ::with_limit(10)
/// .tweak_score(move |segment_reader: &SegmentReader| {
/// // The argument is a function that returns our scoring
/// // function.
/// //
/// // The point of this "mother" function is to gather all
/// // of the segment level information we need for scoring.
/// // Typically, fast_fields.
/// //
/// // In our case, we will get a reader for the popularity
/// // fast field.
/// let popularity_reader =
/// segment_reader.fast_fields().u64(popularity).unwrap();
///
/// // We can now define our actual scoring function
/// move |doc: DocId, original_score: Score| {
/// let popularity: u64 = popularity_reader.get(doc);
/// // Well.. For the sake of the example we use a simple logarithm
/// // function.
/// let popularity_boost_score = ((2u64 + popularity) as f32).log2();
/// popularity_boost_score * original_score
/// }
/// });
/// # let reader = index.reader()?;
/// # let searcher = reader.searcher();
/// // ... and here are our documents. Note this is a simple vec.
/// // The `Score` in the pair is our tweaked score.
/// let resulting_docs: Vec<(Score, DocAddress)> =
/// searcher.search(&*query, &top_docs_by_custom_score)?;
///
/// # Ok(())
/// # }
/// ```
///
/// # See also
/// [custom_score(...)](#method.custom_score).
pub fn tweak_score<TScore, TScoreSegmentTweaker, TScoreTweaker>(
self,
score_tweaker: TScoreTweaker,
) -> impl Collector<Fruit = Vec<(TScore, DocAddress)>>
where
TScore: 'static + Send + Sync + Clone + PartialOrd,
TScoreSegmentTweaker: ScoreSegmentTweaker<TScore> + 'static,
TScoreTweaker: ScoreTweaker<TScore, Child = TScoreSegmentTweaker>,
{
TweakedScoreTopCollector::new(score_tweaker, self.0.limit())
}
/// Ranks the documents using a custom score.
///
/// This method offers a convenient way to use a different score.
///
/// As suggested by the prototype you can manually define your
/// own [`CustomScorer`](./trait.CustomScorer.html)
/// and pass it as an argument, but there is a much simpler way to
/// tweak your score: you can use a closure as in the following
/// example.
///
/// # Limitation
///
/// This method only makes it possible to compute the score from a given
/// `DocId`, fastfield values for the doc and any information you could
/// have precomputed beforehands. It does not make it possible for instance
/// to compute something like TfIdf as it does not have access to the list of query
/// terms present in the document, nor the term frequencies for the different terms.
///
/// It can be used if your search engine relies on a learning-to-rank model for instance,
/// which does not rely on the term frequencies or positions as features.
///
/// # Example
///
/// ```rust
/// # use tantivy::schema::{Schema, FAST, TEXT};
/// # use tantivy::{doc, Index, DocAddress, DocId};
/// # use tantivy::query::QueryParser;
/// use tantivy::SegmentReader;
/// use tantivy::collector::TopDocs;
/// use tantivy::schema::Field;
///
/// # fn create_schema() -> Schema {
/// # let mut schema_builder = Schema::builder();
/// # schema_builder.add_text_field("product_name", TEXT);
/// # schema_builder.add_u64_field("popularity", FAST);
/// # schema_builder.add_u64_field("boosted", FAST);
/// # schema_builder.build()
/// # }
/// #
/// # fn main() -> tantivy::Result<()> {
/// # let schema = create_schema();
/// # let index = Index::create_in_ram(schema);
/// # let mut index_writer = index.writer_with_num_threads(1, 3_000_000)?;
/// # let product_name = index.schema().get_field("product_name").unwrap();
/// #
/// let popularity: Field = index.schema().get_field("popularity").unwrap();
/// let boosted: Field = index.schema().get_field("boosted").unwrap();
/// # index_writer.add_document(doc!(boosted=>1u64, product_name => "The Diary of Muadib", popularity => 1u64));
/// # index_writer.add_document(doc!(boosted=>0u64, product_name => "A Dairy Cow", popularity => 10u64));
/// # index_writer.add_document(doc!(boosted=>0u64, product_name => "The Diary of a Young Girl", popularity => 15u64));
/// # index_writer.commit()?;
/// // ...
/// # let user_query = "diary";
/// # let query = QueryParser::for_index(&index, vec![product_name]).parse_query(user_query)?;
///
/// // This is where we build our collector with our custom score.
/// let top_docs_by_custom_score = TopDocs
/// ::with_limit(10)
/// .custom_score(move |segment_reader: &SegmentReader| {
/// // The argument is a function that returns our scoring
/// // function.
/// //
/// // The point of this "mother" function is to gather all
/// // of the segment level information we need for scoring.
/// // Typically, fast_fields.
/// //
/// // In our case, we will get a reader for the popularity
/// // fast field and a boosted field.
/// //
/// // We want to get boosted items score, and when we get
/// // a tie, return the item with the highest popularity.
/// //
/// // Note that this is implemented by using a `(u64, u64)`
/// // as a score.
/// let popularity_reader =
/// segment_reader.fast_fields().u64(popularity).unwrap();
/// let boosted_reader =
/// segment_reader.fast_fields().u64(boosted).unwrap();
///
/// // We can now define our actual scoring function
/// move |doc: DocId| {
/// let popularity: u64 = popularity_reader.get(doc);
/// let boosted: u64 = boosted_reader.get(doc);
/// // Score do not have to be `f64` in tantivy.
/// // Here we return a couple to get lexicographical order
/// // for free.
/// (boosted, popularity)
/// }
/// });
/// # let reader = index.reader()?;
/// # let searcher = reader.searcher();
/// // ... and here are our documents. Note this is a simple vec.
/// // The `Score` in the pair is our tweaked score.
/// let resulting_docs: Vec<((u64, u64), DocAddress)> =
/// searcher.search(&*query, &top_docs_by_custom_score)?;
///
/// # Ok(())
/// # }
/// ```
///
/// # See also
/// [tweak_score(...)](#method.tweak_score).
pub fn custom_score<TScore, TCustomSegmentScorer, TCustomScorer>(
self,
custom_score: TCustomScorer,
) -> impl Collector<Fruit = Vec<(TScore, DocAddress)>>
where
TScore: 'static + Send + Sync + Clone + PartialOrd,
TCustomSegmentScorer: CustomSegmentScorer<TScore> + 'static,
TCustomScorer: CustomScorer<TScore, Child = TCustomSegmentScorer>,
{
CustomScoreTopCollector::new(custom_score, self.0.limit())
}
}
impl Collector for TopDocs {
type Fruit = Vec<(Score, DocAddress)>;
type Child = TopScoreSegmentCollector;
fn for_segment(
&self,
segment_local_id: SegmentLocalId,
reader: &SegmentReader,
) -> Result<Self::Child> {
let collector = self.0.for_segment(segment_local_id, reader)?;
Ok(TopScoreSegmentCollector(collector))
}
fn requires_scoring(&self) -> bool {
true
}
fn merge_fruits(&self, child_fruits: Vec<Vec<(Score, DocAddress)>>) -> Result<Self::Fruit> {
self.0.merge_fruits(child_fruits)
}
}
/// Segment Collector associated to `TopDocs`.
pub struct TopScoreSegmentCollector(TopSegmentCollector<Score>);
impl SegmentCollector for TopScoreSegmentCollector {
type Fruit = Vec<(Score, DocAddress)>;
fn collect(&mut self, doc: DocId, score: Score) {
self.0.collect(doc, score)
}
fn harvest(self) -> Vec<(Score, DocAddress)> {
self.0.harvest()
}
}
#[cfg(test)]
mod tests {
use super::TopDocs;
use crate::collector::Collector;
use crate::query::{Query, QueryParser};
use crate::schema::{Field, Schema, FAST, STORED, TEXT};
use crate::DocAddress;
use crate::Index;
use crate::IndexWriter;
use crate::Score;
fn make_index() -> Index {
let mut schema_builder = Schema::builder();
let text_field = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
{
// writing the segment
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
index_writer.add_document(doc!(text_field=>"Hello happy tax payer."));
index_writer.add_document(doc!(text_field=>"Droopy says hello happy tax payer"));
index_writer.add_document(doc!(text_field=>"I like Droopy"));
assert!(index_writer.commit().is_ok());
}
index
}
#[test]
fn test_top_collector_not_at_capacity() {
let index = make_index();
let field = index.schema().get_field("text").unwrap();
let query_parser = QueryParser::for_index(&index, vec![field]);
let text_query = query_parser.parse_query("droopy tax").unwrap();
let score_docs: Vec<(Score, DocAddress)> = index
.reader()
.unwrap()
.searcher()
.search(&text_query, &TopDocs::with_limit(4))
.unwrap();
assert_eq!(
score_docs,
vec![
(0.81221175, DocAddress(0u32, 1)),
(0.5376842, DocAddress(0u32, 2)),
(0.48527452, DocAddress(0, 0))
]
);
}
#[test]
fn test_top_collector_at_capacity() {
let index = make_index();
let field = index.schema().get_field("text").unwrap();
let query_parser = QueryParser::for_index(&index, vec![field]);
let text_query = query_parser.parse_query("droopy tax").unwrap();
let score_docs: Vec<(Score, DocAddress)> = index
.reader()
.unwrap()
.searcher()
.search(&text_query, &TopDocs::with_limit(2))
.unwrap();
assert_eq!(
score_docs,
vec![
(0.81221175, DocAddress(0u32, 1)),
(0.5376842, DocAddress(0u32, 2)),
]
);
}
#[test]
#[should_panic]
fn test_top_0() {
TopDocs::with_limit(0);
}
const TITLE: &str = "title";
const SIZE: &str = "size";
#[test]
fn test_top_field_collector_not_at_capacity() {
let mut schema_builder = Schema::builder();
let title = schema_builder.add_text_field(TITLE, TEXT);
let size = schema_builder.add_u64_field(SIZE, FAST);
let schema = schema_builder.build();
let (index, query) = index("beer", title, schema, |index_writer| {
index_writer.add_document(doc!(
title => "bottle of beer",
size => 12u64,
));
index_writer.add_document(doc!(
title => "growler of beer",
size => 64u64,
));
index_writer.add_document(doc!(
title => "pint of beer",
size => 16u64,
));
});
let searcher = index.reader().unwrap().searcher();
let top_collector = TopDocs::with_limit(4).order_by_u64_field(size);
let top_docs: Vec<(u64, DocAddress)> = searcher.search(&query, &top_collector).unwrap();
assert_eq!(
top_docs,
vec![
(64, DocAddress(0, 1)),
(16, DocAddress(0, 2)),
(12, DocAddress(0, 0))
]
);
}
#[test]
#[should_panic]
fn test_field_does_not_exist() {
let mut schema_builder = Schema::builder();
let title = schema_builder.add_text_field(TITLE, TEXT);
let size = schema_builder.add_u64_field(SIZE, FAST);
let schema = schema_builder.build();
let (index, _) = index("beer", title, schema, |index_writer| {
index_writer.add_document(doc!(
title => "bottle of beer",
size => 12u64,
));
});
let searcher = index.reader().unwrap().searcher();
let top_collector = TopDocs::with_limit(4).order_by_u64_field(Field(2));
let segment_reader = searcher.segment_reader(0u32);
top_collector
.for_segment(0, segment_reader)
.expect("should panic");
}
#[test]
#[should_panic(expected = "Field requested is not a i64/u64 fast field")]
fn test_field_not_fast_field() {
let mut schema_builder = Schema::builder();
let title = schema_builder.add_text_field(TITLE, TEXT);
let size = schema_builder.add_u64_field(SIZE, STORED);
let schema = schema_builder.build();
let (index, _) = index("beer", title, schema, |index_writer| {
index_writer.add_document(doc!(
title => "bottle of beer",
size => 12u64,
));
});
let searcher = index.reader().unwrap().searcher();
let segment = searcher.segment_reader(0);
let top_collector = TopDocs::with_limit(4).order_by_u64_field(size);
assert!(top_collector.for_segment(0, segment).is_ok());
}
fn index(
query: &str,
query_field: Field,
schema: Schema,
mut doc_adder: impl FnMut(&mut IndexWriter) -> (),
) -> (Index, Box<dyn Query>) {
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
doc_adder(&mut index_writer);
index_writer.commit().unwrap();
let query_parser = QueryParser::for_index(&index, vec![query_field]);
let query = query_parser.parse_query(query).unwrap();
(index, query)
}
}

129
src/collector/tweak_score_top_collector.rs
View File

@@ -1,129 +0,0 @@
use crate::collector::top_collector::{TopCollector, TopSegmentCollector};
use crate::collector::{Collector, SegmentCollector};
use crate::DocAddress;
use crate::{DocId, Result, Score, SegmentReader};
pub(crate) struct TweakedScoreTopCollector<TScoreTweaker, TScore = Score> {
score_tweaker: TScoreTweaker,
collector: TopCollector<TScore>,
}
impl<TScoreTweaker, TScore> TweakedScoreTopCollector<TScoreTweaker, TScore>
where
TScore: Clone + PartialOrd,
{
pub fn new(
score_tweaker: TScoreTweaker,
limit: usize,
) -> TweakedScoreTopCollector<TScoreTweaker, TScore> {
TweakedScoreTopCollector {
score_tweaker,
collector: TopCollector::with_limit(limit),
}
}
}
/// A `ScoreSegmentTweaker` makes it possible to modify the default score
/// for a given document belonging to a specific segment.
///
/// It is the segment local version of the [`ScoreTweaker`](./trait.ScoreTweaker.html).
pub trait ScoreSegmentTweaker<TScore>: 'static {
/// Tweak the given `score` for the document `doc`.
fn score(&self, doc: DocId, score: Score) -> TScore;
}
/// `ScoreTweaker` makes it possible to tweak the score
/// emitted by the scorer into another one.
///
/// The `ScoreTweaker` itself does not make much of the computation itself.
/// Instead, it helps constructing `Self::Child` instances that will compute
/// the score at a segment scale.
pub trait ScoreTweaker<TScore>: Sync {
/// Type of the associated [`ScoreSegmentTweaker`](./trait.ScoreSegmentTweaker.html).
type Child: ScoreSegmentTweaker<TScore>;
/// Builds a child tweaker for a specific segment. The child scorer is associated to
/// a specific segment.
fn segment_tweaker(&self, segment_reader: &SegmentReader) -> Result<Self::Child>;
}
impl<TScoreTweaker, TScore> Collector for TweakedScoreTopCollector<TScoreTweaker, TScore>
where
TScoreTweaker: ScoreTweaker<TScore>,
TScore: 'static + PartialOrd + Clone + Send + Sync,
{
type Fruit = Vec<(TScore, DocAddress)>;
type Child = TopTweakedScoreSegmentCollector<TScoreTweaker::Child, TScore>;
fn for_segment(
&self,
segment_local_id: u32,
segment_reader: &SegmentReader,
) -> Result<Self::Child> {
let segment_scorer = self.score_tweaker.segment_tweaker(segment_reader)?;
let segment_collector = self
.collector
.for_segment(segment_local_id, segment_reader)?;
Ok(TopTweakedScoreSegmentCollector {
segment_collector,
segment_scorer,
})
}
fn requires_scoring(&self) -> bool {
true
}
fn merge_fruits(&self, segment_fruits: Vec<Self::Fruit>) -> Result<Self::Fruit> {
self.collector.merge_fruits(segment_fruits)
}
}
pub struct TopTweakedScoreSegmentCollector<TSegmentScoreTweaker, TScore>
where
TScore: 'static + PartialOrd + Clone + Send + Sync + Sized,
TSegmentScoreTweaker: ScoreSegmentTweaker<TScore>,
{
segment_collector: TopSegmentCollector<TScore>,
segment_scorer: TSegmentScoreTweaker,
}
impl<TSegmentScoreTweaker, TScore> SegmentCollector
for TopTweakedScoreSegmentCollector<TSegmentScoreTweaker, TScore>
where
TScore: 'static + PartialOrd + Clone + Send + Sync,
TSegmentScoreTweaker: 'static + ScoreSegmentTweaker<TScore>,
{
type Fruit = Vec<(TScore, DocAddress)>;
fn collect(&mut self, doc: DocId, score: Score) {
let score = self.segment_scorer.score(doc, score);
self.segment_collector.collect(doc, score);
}
fn harvest(self) -> Vec<(TScore, DocAddress)> {
self.segment_collector.harvest()
}
}
impl<F, TScore, TSegmentScoreTweaker> ScoreTweaker<TScore> for F
where
F: 'static + Send + Sync + Fn(&SegmentReader) -> TSegmentScoreTweaker,
TSegmentScoreTweaker: ScoreSegmentTweaker<TScore>,
{
type Child = TSegmentScoreTweaker;
fn segment_tweaker(&self, segment_reader: &SegmentReader) -> Result<Self::Child> {
Ok((self)(segment_reader))
}
}
impl<F, TScore> ScoreSegmentTweaker<TScore> for F
where
F: 'static + Sync + Send + Fn(DocId, Score) -> TScore,
{
fn score(&self, doc: DocId, score: Score) -> TScore {
(self)(doc, score)
}
}

70
src/common/bitpacker.rs
View File

@@ -1,6 +1,9 @@
use byteorder::{ByteOrder, LittleEndian, WriteBytesExt};
use common::serialize::BinarySerializable;
use std::io;
use std::io::Write;
use std::mem;
use std::ops::Deref;
use std::ptr;
pub(crate) struct BitPacker {
mini_buffer: u64,
@@ -15,7 +18,7 @@ impl BitPacker {
}
}
pub fn write<TWrite: io::Write>(
pub fn write<TWrite: Write>(
&mut self,
val: u64,
num_bits: u8,
@@ -25,14 +28,14 @@ impl BitPacker {
let num_bits = num_bits as usize;
if self.mini_buffer_written + num_bits > 64 {
self.mini_buffer |= val_u64.wrapping_shl(self.mini_buffer_written as u32);
output.write_u64::<LittleEndian>(self.mini_buffer)?;
self.mini_buffer.serialize(output)?;
self.mini_buffer = val_u64.wrapping_shr((64 - self.mini_buffer_written) as u32);
self.mini_buffer_written = self.mini_buffer_written + num_bits - 64;
} else {
self.mini_buffer |= val_u64 << self.mini_buffer_written;
self.mini_buffer_written += num_bits;
if self.mini_buffer_written == 64 {
output.write_u64::<LittleEndian>(self.mini_buffer)?;
self.mini_buffer.serialize(output)?;
self.mini_buffer_written = 0;
self.mini_buffer = 0u64;
}
@@ -40,18 +43,17 @@ impl BitPacker {
Ok(())
}
pub fn flush<TWrite: io::Write>(&mut self, output: &mut TWrite) -> io::Result<()> {
pub fn flush<TWrite: Write>(&mut self, output: &mut TWrite) -> io::Result<()> {
if self.mini_buffer_written > 0 {
let num_bytes = (self.mini_buffer_written + 7) / 8;
let mut arr: [u8; 8] = [0u8; 8];
LittleEndian::write_u64(&mut arr, self.mini_buffer);
let arr: [u8; 8] = unsafe { mem::transmute::<u64, [u8; 8]>(self.mini_buffer.to_le()) };
output.write_all(&arr[..num_bytes])?;
self.mini_buffer_written = 0;
}
Ok(())
}
pub fn close<TWrite: io::Write>(&mut self, output: &mut TWrite) -> io::Result<()> {
pub fn close<TWrite: Write>(&mut self, output: &mut TWrite) -> io::Result<()> {
self.flush(output)?;
// Padding the write file to simplify reads.
output.write_all(&[0u8; 7])?;
@@ -64,7 +66,7 @@ pub struct BitUnpacker<Data>
where
Data: Deref<Target = [u8]>,
{
num_bits: u64,
num_bits: usize,
mask: u64,
data: Data,
}
@@ -80,13 +82,13 @@ where
(1u64 << num_bits) - 1u64
};
BitUnpacker {
num_bits: u64::from(num_bits),
num_bits: num_bits as usize,
mask,
data,
}
}
pub fn get(&self, idx: u64) -> u64 {
pub fn get(&self, idx: usize) -> u64 {
if self.num_bits == 0 {
return 0u64;
}
@@ -97,13 +99,40 @@ where
let addr = addr_in_bits >> 3;
let bit_shift = addr_in_bits & 7;
debug_assert!(
addr + 8 <= data.len() as u64,
addr + 8 <= data.len(),
"The fast field field should have been padded with 7 bytes."
);
let val_unshifted_unmasked: u64 = LittleEndian::read_u64(&data[(addr as usize)..]);
let val_unshifted_unmasked: u64 =
u64::from_le(unsafe { ptr::read_unaligned(data[addr..].as_ptr() as *const u64) });
let val_shifted = (val_unshifted_unmasked >> bit_shift) as u64;
val_shifted & mask
}
/// Reads a range of values from the fast field.
///
/// The range of values read is from
/// `[start..start + output.len()[`
pub fn get_range(&self, start: u32, output: &mut [u64]) {
if self.num_bits == 0 {
for val in output.iter_mut() {
*val = 0u64;
}
} else {
let data: &[u8] = &*self.data;
let num_bits = self.num_bits;
let mask = self.mask;
let mut addr_in_bits = (start as usize) * num_bits;
for output_val in output.iter_mut() {
let addr = addr_in_bits >> 3;
let bit_shift = addr_in_bits & 7;
let val_unshifted_unmasked: u64 =
unsafe { ptr::read_unaligned(data[addr..].as_ptr() as *const u64) };
let val_shifted = (val_unshifted_unmasked >> bit_shift) as u64;
*output_val = val_shifted & mask;
addr_in_bits += num_bits;
}
}
}
}
#[cfg(test)]
@@ -129,7 +158,7 @@ mod test {
fn test_bitpacker_util(len: usize, num_bits: u8) {
let (bitunpacker, vals) = create_fastfield_bitpacker(len, num_bits);
for (i, val) in vals.iter().enumerate() {
assert_eq!(bitunpacker.get(i as u64), *val);
assert_eq!(bitunpacker.get(i), *val);
}
}
@@ -141,4 +170,17 @@ mod test {
test_bitpacker_util(6, 14);
test_bitpacker_util(1000, 14);
}
#[test]
fn test_bitpacker_range() {
let (bitunpacker, vals) = create_fastfield_bitpacker(100_000, 12);
let buffer_len = 100;
let mut buffer = vec![0u64; buffer_len];
for start in vec![0, 10, 20, 100, 1_000] {
bitunpacker.get_range(start as u32, &mut buffer[..]);
for i in 0..buffer_len {
assert_eq!(buffer[i], vals[start + i]);
}
}
}
}

26
src/common/bitset.rs
View File

@@ -5,7 +5,7 @@ use std::u64;
pub(crate) struct TinySet(u64);
impl fmt::Debug for TinySet {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.into_iter().collect::<Vec<u32>>().fmt(f)
}
}
@@ -34,17 +34,17 @@ impl TinySet {
}
/// Returns the complement of the set in `[0, 64[`.
fn complement(self) -> TinySet {
fn complement(&self) -> TinySet {
TinySet(!self.0)
}
/// Returns true iff the `TinySet` contains the element `el`.
pub fn contains(self, el: u32) -> bool {
pub fn contains(&self, el: u32) -> bool {
!self.intersect(TinySet::singleton(el)).is_empty()
}
/// Returns the intersection of `self` and `other`
pub fn intersect(self, other: TinySet) -> TinySet {
pub fn intersect(&self, other: TinySet) -> TinySet {
TinySet(self.0 & other.0)
}
@@ -77,7 +77,7 @@ impl TinySet {
/// Returns true iff the `TinySet` is empty.
#[inline(always)]
pub fn is_empty(self) -> bool {
pub fn is_empty(&self) -> bool {
self.0 == 0u64
}
@@ -114,7 +114,7 @@ impl TinySet {
self.0 = 0u64;
}
pub fn len(self) -> u32 {
pub fn len(&self) -> u32 {
self.0.count_ones()
}
}
@@ -204,12 +204,12 @@ mod tests {
use super::BitSet;
use super::TinySet;
use crate::docset::DocSet;
use crate::query::BitSetDocSet;
use crate::tests;
use crate::tests::generate_nonunique_unsorted;
use docset::DocSet;
use query::BitSetDocSet;
use std::collections::BTreeSet;
use std::collections::HashSet;
use tests;
use tests::generate_nonunique_unsorted;
#[test]
fn test_tiny_set() {
@@ -266,14 +266,14 @@ mod tests {
#[test]
fn test_bitset_large() {
let arr = generate_nonunique_unsorted(100_000, 5_000);
let arr = generate_nonunique_unsorted(1_000_000, 50_000);
let mut btreeset: BTreeSet<u32> = BTreeSet::new();
let mut bitset = BitSet::with_max_value(100_000);
let mut bitset = BitSet::with_max_value(1_000_000);
for el in arr {
btreeset.insert(el);
bitset.insert(el);
}
for i in 0..100_000 {
for i in 0..1_000_000 {
assert_eq!(btreeset.contains(&i), bitset.contains(i));
}
assert_eq!(btreeset.len(), bitset.len());

44
src/common/composite_file.rs
View File

@@ -1,11 +1,9 @@
use crate::common::BinarySerializable;
use crate::common::CountingWriter;
use crate::common::VInt;
use crate::directory::ReadOnlySource;
use crate::directory::{TerminatingWrite, WritePtr};
use crate::schema::Field;
use crate::space_usage::FieldUsage;
use crate::space_usage::PerFieldSpaceUsage;
use common::BinarySerializable;
use common::CountingWriter;
use common::VInt;
use directory::ReadOnlySource;
use directory::WritePtr;
use schema::Field;
use std::collections::HashMap;
use std::io::Write;
use std::io::{self, Read};
@@ -39,10 +37,10 @@ impl BinarySerializable for FileAddr {
/// A `CompositeWrite` is used to write a `CompositeFile`.
pub struct CompositeWrite<W = WritePtr> {
write: CountingWriter<W>,
offsets: HashMap<FileAddr, u64>,
offsets: HashMap<FileAddr, usize>,
}
impl<W: TerminatingWrite + Write> CompositeWrite<W> {
impl<W: Write> CompositeWrite<W> {
/// Crate a new API writer that writes a composite file
/// in a given write.
pub fn wrap(w: W) -> CompositeWrite<W> {
@@ -74,8 +72,7 @@ impl<W: TerminatingWrite + Write> CompositeWrite<W> {
let footer_offset = self.write.written_bytes();
VInt(self.offsets.len() as u64).serialize(&mut self.write)?;
let mut offset_fields: Vec<_> = self
.offsets
let mut offset_fields: Vec<_> = self.offsets
.iter()
.map(|(file_addr, offset)| (*offset, *file_addr))
.collect();
@@ -91,7 +88,8 @@ impl<W: TerminatingWrite + Write> CompositeWrite<W> {
let footer_len = (self.write.written_bytes() - footer_offset) as u32;
footer_len.serialize(&mut self.write)?;
self.write.terminate()
self.write.flush()?;
Ok(())
}
}
@@ -167,27 +165,16 @@ impl CompositeFile {
.get(&FileAddr { field, idx })
.map(|&(from, to)| self.data.slice(from, to))
}
pub fn space_usage(&self) -> PerFieldSpaceUsage {
let mut fields = HashMap::new();
for (&field_addr, &(start, end)) in self.offsets_index.iter() {
fields
.entry(field_addr.field)
.or_insert_with(|| FieldUsage::empty(field_addr.field))
.add_field_idx(field_addr.idx, end - start);
}
PerFieldSpaceUsage::new(fields)
}
}
#[cfg(test)]
mod test {
use super::{CompositeFile, CompositeWrite};
use crate::common::BinarySerializable;
use crate::common::VInt;
use crate::directory::{Directory, RAMDirectory};
use crate::schema::Field;
use common::BinarySerializable;
use common::VInt;
use directory::{Directory, RAMDirectory};
use schema::Field;
use std::io::Write;
use std::path::Path;
@@ -230,4 +217,5 @@ mod test {
}
}
}
}

27
src/common/counting_writer.rs
View File

@@ -1,11 +1,9 @@
use crate::directory::AntiCallToken;
use crate::directory::TerminatingWrite;
use std::io;
use std::io::Write;
pub struct CountingWriter<W> {
underlying: W,
written_bytes: u64,
written_bytes: usize,
}
impl<W: Write> CountingWriter<W> {
@@ -16,11 +14,11 @@ impl<W: Write> CountingWriter<W> {
}
}
pub fn written_bytes(&self) -> u64 {
pub fn written_bytes(&self) -> usize {
self.written_bytes
}
pub fn finish(mut self) -> io::Result<(W, u64)> {
pub fn finish(mut self) -> io::Result<(W, usize)> {
self.flush()?;
Ok((self.underlying, self.written_bytes))
}
@@ -29,28 +27,15 @@ impl<W: Write> CountingWriter<W> {
impl<W: Write> Write for CountingWriter<W> {
fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
let written_size = self.underlying.write(buf)?;
self.written_bytes += written_size as u64;
self.written_bytes += written_size;
Ok(written_size)
}
fn write_all(&mut self, buf: &[u8]) -> io::Result<()> {
self.underlying.write_all(buf)?;
self.written_bytes += buf.len() as u64;
Ok(())
}
fn flush(&mut self) -> io::Result<()> {
self.underlying.flush()
}
}
impl<W: TerminatingWrite> TerminatingWrite for CountingWriter<W> {
fn terminate_ref(&mut self, token: AntiCallToken) -> io::Result<()> {
self.flush()?;
self.underlying.terminate_ref(token)
}
}
#[cfg(test)]
mod test {
@@ -63,8 +48,8 @@ mod test {
let mut counting_writer = CountingWriter::wrap(buffer);
let bytes = (0u8..10u8).collect::<Vec<u8>>();
counting_writer.write_all(&bytes).unwrap();
let (w, len): (Vec<u8>, u64) = counting_writer.finish().unwrap();
assert_eq!(len, 10u64);
let (w, len): (Vec<u8>, usize) = counting_writer.finish().unwrap();
assert_eq!(len, 10);
assert_eq!(w.len(), 10);
}
}

81
src/common/mod.rs
View File

@@ -10,13 +10,10 @@ pub(crate) use self::bitset::TinySet;
pub(crate) use self::composite_file::{CompositeFile, CompositeWrite};
pub use self::counting_writer::CountingWriter;
pub use self::serialize::{BinarySerializable, FixedSize};
pub use self::vint::{read_u32_vint, serialize_vint_u32, write_u32_vint, VInt};
pub use self::vint::VInt;
pub use byteorder::LittleEndian as Endianness;
/// Segment's max doc must be `< MAX_DOC_LIMIT`.
///
/// We do not allow segments with more than
pub const MAX_DOC_LIMIT: u32 = 1 << 31;
use std::io;
/// Computes the number of bits that will be used for bitpacking.
///
@@ -55,6 +52,11 @@ pub(crate) fn is_power_of_2(n: usize) -> bool {
(n > 0) && (n & (n - 1) == 0)
}
/// Create a default io error given a string.
pub(crate) fn make_io_err(msg: String) -> io::Error {
io::Error::new(io::ErrorKind::Other, msg)
}
/// Has length trait
pub trait HasLen {
/// Return length
@@ -99,53 +101,16 @@ pub fn u64_to_i64(val: u64) -> i64 {
(val ^ HIGHEST_BIT) as i64
}
/// Maps a `f64` to `u64`
///
/// For simplicity, tantivy internally handles `f64` as `u64`.
/// The mapping is defined by this function.
///
/// Maps `f64` to `u64` so that lexical order is preserved.
///
/// This is more suited than simply casting (`val as u64`)
/// which would truncate the result
///
/// # See also
/// The [reverse mapping is `u64_to_f64`](./fn.u64_to_f64.html).
#[inline(always)]
pub fn f64_to_u64(val: f64) -> u64 {
let bits = val.to_bits();
if val.is_sign_positive() {
bits ^ HIGHEST_BIT
} else {
!bits
}
}
/// Reverse the mapping given by [`i64_to_u64`](./fn.i64_to_u64.html).
#[inline(always)]
pub fn u64_to_f64(val: u64) -> f64 {
f64::from_bits(if val & HIGHEST_BIT != 0 {
val ^ HIGHEST_BIT
} else {
!val
})
}
#[cfg(test)]
pub(crate) mod test {
pub use super::serialize::test::fixed_size_test;
use super::{compute_num_bits, f64_to_u64, i64_to_u64, u64_to_f64, u64_to_i64};
use std::f64;
use super::{compute_num_bits, i64_to_u64, u64_to_i64};
fn test_i64_converter_helper(val: i64) {
assert_eq!(u64_to_i64(i64_to_u64(val)), val);
}
fn test_f64_converter_helper(val: f64) {
assert_eq!(u64_to_f64(f64_to_u64(val)), val);
}
#[test]
fn test_i64_converter() {
assert_eq!(i64_to_u64(i64::min_value()), u64::min_value());
@@ -158,29 +123,6 @@ pub(crate) mod test {
}
}
#[test]
fn test_f64_converter() {
test_f64_converter_helper(f64::INFINITY);
test_f64_converter_helper(f64::NEG_INFINITY);
test_f64_converter_helper(0.0);
test_f64_converter_helper(-0.0);
test_f64_converter_helper(1.0);
test_f64_converter_helper(-1.0);
}
#[test]
fn test_f64_order() {
assert!(!(f64_to_u64(f64::NEG_INFINITY)..f64_to_u64(f64::INFINITY))
.contains(&f64_to_u64(f64::NAN))); //nan is not a number
assert!(f64_to_u64(1.5) > f64_to_u64(1.0)); //same exponent, different mantissa
assert!(f64_to_u64(2.0) > f64_to_u64(1.0)); //same mantissa, different exponent
assert!(f64_to_u64(2.0) > f64_to_u64(1.5)); //different exponent and mantissa
assert!(f64_to_u64(1.0) > f64_to_u64(-1.0)); // pos > neg
assert!(f64_to_u64(-1.5) < f64_to_u64(-1.0));
assert!(f64_to_u64(-2.0) < f64_to_u64(1.0));
assert!(f64_to_u64(-2.0) < f64_to_u64(-1.5));
}
#[test]
fn test_compute_num_bits() {
assert_eq!(compute_num_bits(1), 1u8);
@@ -192,11 +134,4 @@ pub(crate) mod test {
assert_eq!(compute_num_bits(256), 9u8);
assert_eq!(compute_num_bits(5_000_000_000), 33u8);
}
#[test]
fn test_max_doc() {
// this is the first time I write a unit test for a constant.
assert!(((super::MAX_DOC_LIMIT - 1) as i32) >= 0);
assert!((super::MAX_DOC_LIMIT as i32) < 0);
}
}

29
src/common/serialize.rs
View File

@@ -1,6 +1,6 @@
use crate::common::Endianness;
use crate::common::VInt;
use byteorder::{ReadBytesExt, WriteBytesExt};
use common::Endianness;
use common::VInt;
use std::fmt;
use std::io;
use std::io::Read;
@@ -102,19 +102,6 @@ impl FixedSize for i64 {
const SIZE_IN_BYTES: usize = 8;
}
impl BinarySerializable for f64 {
fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
writer.write_f64::<Endianness>(*self)
}
fn deserialize<R: Read>(reader: &mut R) -> io::Result<Self> {
reader.read_f64::<Endianness>()
}
}
impl FixedSize for f64 {
const SIZE_IN_BYTES: usize = 8;
}
impl BinarySerializable for u8 {
fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
writer.write_u8(*self)
@@ -149,7 +136,7 @@ impl BinarySerializable for String {
pub mod test {
use super::*;
use crate::common::VInt;
use common::VInt;
pub fn fixed_size_test<O: BinarySerializable + FixedSize + Default>() {
let mut buffer = Vec::new();
@@ -185,11 +172,6 @@ pub mod test {
fixed_size_test::<i64>();
}
#[test]
fn test_serialize_f64() {
fixed_size_test::<f64>();
}
#[test]
fn test_serialize_u64() {
fixed_size_test::<u64>();
@@ -199,7 +181,10 @@ pub mod test {
fn test_serialize_string() {
assert_eq!(serialize_test(String::from("")), 1);
assert_eq!(serialize_test(String::from("ぽよぽよ")), 1 + 3 * 4);
assert_eq!(serialize_test(String::from("富士さん見える。")), 1 + 3 * 8);
assert_eq!(
serialize_test(String::from("富士さん見える。")),
1 + 3 * 8
);
}
#[test]

143
src/common/vint.rs
View File

@@ -1,5 +1,4 @@
use super::BinarySerializable;
use byteorder::{ByteOrder, LittleEndian};
use std::io;
use std::io::Read;
use std::io::Write;
@@ -10,101 +9,9 @@ pub struct VInt(pub u64);
const STOP_BIT: u8 = 128;
pub fn serialize_vint_u32(val: u32) -> (u64, usize) {
const START_2: u64 = 1 << 7;
const START_3: u64 = 1 << 14;
const START_4: u64 = 1 << 21;
const START_5: u64 = 1 << 28;
const STOP_1: u64 = START_2 - 1;
const STOP_2: u64 = START_3 - 1;
const STOP_3: u64 = START_4 - 1;
const STOP_4: u64 = START_5 - 1;
const MASK_1: u64 = 127;
const MASK_2: u64 = MASK_1 << 7;
const MASK_3: u64 = MASK_2 << 7;
const MASK_4: u64 = MASK_3 << 7;
const MASK_5: u64 = MASK_4 << 7;
let val = u64::from(val);
const STOP_BIT: u64 = 128u64;
match val {
0..=STOP_1 => (val | STOP_BIT, 1),
START_2..=STOP_2 => (
(val & MASK_1) | ((val & MASK_2) << 1) | (STOP_BIT << (8)),
2,
),
START_3..=STOP_3 => (
(val & MASK_1) | ((val & MASK_2) << 1) | ((val & MASK_3) << 2) | (STOP_BIT << (8 * 2)),
3,
),
START_4..=STOP_4 => (
(val & MASK_1)
| ((val & MASK_2) << 1)
| ((val & MASK_3) << 2)
| ((val & MASK_4) << 3)
| (STOP_BIT << (8 * 3)),
4,
),
_ => (
(val & MASK_1)
| ((val & MASK_2) << 1)
| ((val & MASK_3) << 2)
| ((val & MASK_4) << 3)
| ((val & MASK_5) << 4)
| (STOP_BIT << (8 * 4)),
5,
),
}
}
/// Returns the number of bytes covered by a
/// serialized vint `u32`.
///
/// Expects a buffer data that starts
/// by the serialized `vint`, scans at most 5 bytes ahead until
/// it finds the vint final byte.
///
/// # May Panic
/// If the payload does not start by a valid `vint`
fn vint_len(data: &[u8]) -> usize {
for (i, &val) in data.iter().enumerate().take(5) {
if val >= STOP_BIT {
return i + 1;
}
}
panic!("Corrupted data. Invalid VInt 32");
}
/// Reads a vint `u32` from a buffer, and
/// consumes its payload data.
///
/// # Panics
///
/// If the buffer does not start by a valid
/// vint payload
pub fn read_u32_vint(data: &mut &[u8]) -> u32 {
let vlen = vint_len(*data);
let mut result = 0u32;
let mut shift = 0u64;
for &b in &data[..vlen] {
result |= u32::from(b & 127u8) << shift;
shift += 7;
}
*data = &data[vlen..];
result
}
/// Write a `u32` as a vint payload.
pub fn write_u32_vint<W: io::Write>(val: u32, writer: &mut W) -> io::Result<()> {
let (val, num_bytes) = serialize_vint_u32(val);
let mut buffer = [0u8; 8];
LittleEndian::write_u64(&mut buffer, val);
writer.write_all(&buffer[..num_bytes])
}
impl VInt {
pub fn val(&self) -> u64 {
self.0
}
@@ -113,13 +20,14 @@ impl VInt {
VInt::deserialize(reader).map(|vint| vint.0)
}
pub fn serialize_into_vec(&self, output: &mut Vec<u8>) {
pub fn serialize_into_vec(&self, output: &mut Vec<u8>){
let mut buffer = [0u8; 10];
let num_bytes = self.serialize_into(&mut buffer);
output.extend(&buffer[0..num_bytes]);
}
pub fn serialize_into(&self, buffer: &mut [u8; 10]) -> usize {
fn serialize_into(&self, buffer: &mut [u8; 10]) -> usize {
let mut remaining = self.0;
for (i, b) in buffer.iter_mut().enumerate() {
let next_byte: u8 = (remaining % 128u64) as u8;
@@ -159,20 +67,19 @@ impl BinarySerializable for VInt {
return Err(io::Error::new(
io::ErrorKind::InvalidData,
"Reach end of buffer while reading VInt",
));
))
}
}
}
}
}
#[cfg(test)]
mod tests {
use super::serialize_vint_u32;
use super::VInt;
use crate::common::BinarySerializable;
use byteorder::{ByteOrder, LittleEndian};
use common::BinarySerializable;
fn aux_test_vint(val: u64) {
let mut v = [14u8; 10];
@@ -182,10 +89,10 @@ mod tests {
}
assert!(num_bytes > 0);
if num_bytes < 10 {
assert!(1u64 << (7 * num_bytes) > val);
assert!(1u64 << (7*num_bytes) > val);
}
if num_bytes > 1 {
assert!(1u64 << (7 * (num_bytes - 1)) <= val);
assert!(1u64 << (7*(num_bytes-1)) <= val);
}
let serdeser_val = VInt::deserialize(&mut &v[..]).unwrap();
assert_eq!(val, serdeser_val.0);
@@ -198,35 +105,11 @@ mod tests {
aux_test_vint(5);
aux_test_vint(u64::max_value());
for i in 1..9 {
let power_of_128 = 1u64 << (7 * i);
let power_of_128 = 1u64 << (7*i);
aux_test_vint(power_of_128 - 1u64);
aux_test_vint(power_of_128);
aux_test_vint(power_of_128 );
aux_test_vint(power_of_128 + 1u64);
}
aux_test_vint(10);
}
fn aux_test_serialize_vint_u32(val: u32) {
let mut buffer = [0u8; 10];
let mut buffer2 = [0u8; 10];
let len_vint = VInt(val as u64).serialize_into(&mut buffer);
let (vint, len) = serialize_vint_u32(val);
assert_eq!(len, len_vint, "len wrong for val {}", val);
LittleEndian::write_u64(&mut buffer2, vint);
assert_eq!(&buffer[..len], &buffer2[..len], "array wrong for {}", val);
}
#[test]
fn test_vint_u32() {
aux_test_serialize_vint_u32(0);
aux_test_serialize_vint_u32(1);
aux_test_serialize_vint_u32(5);
for i in 1..3 {
let power_of_128 = 1u32 << (7 * i);
aux_test_serialize_vint_u32(power_of_128 - 1u32);
aux_test_serialize_vint_u32(power_of_128);
aux_test_serialize_vint_u32(power_of_128 + 1u32);
}
aux_test_serialize_vint_u32(u32::max_value());
}
}
}

136
src/core/executor.rs
View File

@@ -1,136 +0,0 @@
use crate::Result;
use crossbeam::channel;
use scoped_pool::{Pool, ThreadConfig};
/// Search executor whether search request are single thread or multithread.
///
/// We don't expose Rayon thread pool directly here for several reasons.
///
/// First dependency hell. It is not a good idea to expose the
/// API of a dependency, knowing it might conflict with a different version
/// used by the client. Second, we may stop using rayon in the future.
pub enum Executor {
SingleThread,
ThreadPool(Pool),
}
impl Executor {
/// Creates an Executor that performs all task in the caller thread.
pub fn single_thread() -> Executor {
Executor::SingleThread
}
// Creates an Executor that dispatches the tasks in a thread pool.
pub fn multi_thread(num_threads: usize, prefix: &'static str) -> Executor {
let thread_config = ThreadConfig::new().prefix(prefix);
let pool = Pool::with_thread_config(num_threads, thread_config);
Executor::ThreadPool(pool)
}
// Perform a map in the thread pool.
//
// Regardless of the executor (`SingleThread` or `ThreadPool`), panics in the task
// will propagate to the caller.
pub fn map<
A: Send,
R: Send,
AIterator: Iterator<Item = A>,
F: Sized + Sync + Fn(A) -> Result<R>,
>(
&self,
f: F,
args: AIterator,
) -> Result<Vec<R>> {
match self {
Executor::SingleThread => args.map(f).collect::<Result<_>>(),
Executor::ThreadPool(pool) => {
let args_with_indices: Vec<(usize, A)> = args.enumerate().collect();
let num_fruits = args_with_indices.len();
let fruit_receiver = {
let (fruit_sender, fruit_receiver) = channel::unbounded();
pool.scoped(|scope| {
for arg_with_idx in args_with_indices {
scope.execute(|| {
let (idx, arg) = arg_with_idx;
let fruit = f(arg);
if let Err(err) = fruit_sender.send((idx, fruit)) {
error!("Failed to send search task. It probably means all search threads have panicked. {:?}", err);
}
});
}
});
fruit_receiver
// This ends the scope of fruit_sender.
// This is important as it makes it possible for the fruit_receiver iteration to
// terminate.
};
// This is lame, but safe.
let mut results_with_position = Vec::with_capacity(num_fruits);
for (pos, fruit_res) in fruit_receiver {
let fruit = fruit_res?;
results_with_position.push((pos, fruit));
}
results_with_position.sort_by_key(|(pos, _)| *pos);
assert_eq!(results_with_position.len(), num_fruits);
Ok(results_with_position
.into_iter()
.map(|(_, fruit)| fruit)
.collect::<Vec<_>>())
}
}
}
}
#[cfg(test)]
mod tests {
use super::Executor;
#[test]
#[should_panic(expected = "panic should propagate")]
fn test_panic_propagates_single_thread() {
let _result: Vec<usize> = Executor::single_thread()
.map(
|_| {
panic!("panic should propagate");
},
vec![0].into_iter(),
)
.unwrap();
}
#[test]
#[should_panic] //< unfortunately the panic message is not propagated
fn test_panic_propagates_multi_thread() {
let _result: Vec<usize> = Executor::multi_thread(1, "search-test")
.map(
|_| {
panic!("panic should propagate");
},
vec![0].into_iter(),
)
.unwrap();
}
#[test]
fn test_map_singlethread() {
let result: Vec<usize> = Executor::single_thread()
.map(|i| Ok(i * 2), 0..1_000)
.unwrap();
assert_eq!(result.len(), 1_000);
for i in 0..1_000 {
assert_eq!(result[i], i * 2);
}
}
#[test]
fn test_map_multithread() {
let result: Vec<usize> = Executor::multi_thread(3, "search-test")
.map(|i| Ok(i * 2), 0..10)
.unwrap();
assert_eq!(result.len(), 10);
for i in 0..10 {
assert_eq!(result[i], i * 2);
}
}
}

513
src/core/index.rs
View File

@@ -1,89 +1,52 @@
use core::SegmentId;
use error::TantivyError;
use schema::Schema;
use serde_json;
use std::borrow::BorrowMut;
use std::fmt;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::sync::Arc;
use Result;
use super::pool::LeasedItem;
use super::pool::Pool;
use super::segment::create_segment;
use super::segment::Segment;
use crate::core::Executor;
use crate::core::IndexMeta;
use crate::core::SegmentId;
use crate::core::SegmentMeta;
use crate::core::SegmentMetaInventory;
use crate::core::META_FILEPATH;
use crate::directory::ManagedDirectory;
use core::searcher::Searcher;
use core::IndexMeta;
use core::SegmentMeta;
use core::SegmentReader;
use core::META_FILEPATH;
#[cfg(feature = "mmap")]
use crate::directory::MmapDirectory;
use crate::directory::INDEX_WRITER_LOCK;
use crate::directory::{Directory, RAMDirectory};
use crate::error::DataCorruption;
use crate::error::TantivyError;
use crate::indexer::index_writer::HEAP_SIZE_MIN;
use crate::indexer::segment_updater::save_new_metas;
use crate::reader::IndexReader;
use crate::reader::IndexReaderBuilder;
use crate::schema::Field;
use crate::schema::FieldType;
use crate::schema::Schema;
use crate::tokenizer::BoxedTokenizer;
use crate::tokenizer::TokenizerManager;
use crate::IndexWriter;
use crate::Result;
use directory::MmapDirectory;
use directory::{Directory, RAMDirectory};
use directory::{DirectoryClone, ManagedDirectory};
use indexer::index_writer::open_index_writer;
use indexer::index_writer::HEAP_SIZE_MIN;
use indexer::segment_updater::save_new_metas;
use indexer::DirectoryLock;
use num_cpus;
use std::borrow::BorrowMut;
use std::collections::HashSet;
use std::fmt;
#[cfg(feature = "mmap")]
use std::path::{Path, PathBuf};
use std::sync::Arc;
use std::path::Path;
use tokenizer::TokenizerManager;
use IndexWriter;
fn load_metas(directory: &dyn Directory, inventory: &SegmentMetaInventory) -> Result<IndexMeta> {
fn load_metas(directory: &Directory) -> Result<IndexMeta> {
let meta_data = directory.atomic_read(&META_FILEPATH)?;
let meta_string = String::from_utf8_lossy(&meta_data);
IndexMeta::deserialize(&meta_string, &inventory)
.map_err(|e| {
DataCorruption::new(
META_FILEPATH.to_path_buf(),
format!("Meta file cannot be deserialized. {:?}.", e),
)
})
.map_err(From::from)
serde_json::from_str(&meta_string)
.map_err(|_| TantivyError::CorruptedFile(META_FILEPATH.clone()))
}
/// Search Index
#[derive(Clone)]
pub struct Index {
directory: ManagedDirectory,
schema: Schema,
executor: Arc<Executor>,
num_searchers: Arc<AtomicUsize>,
searcher_pool: Arc<Pool<Searcher>>,
tokenizers: TokenizerManager,
inventory: SegmentMetaInventory,
}
impl Index {
/// Examines the director to see if it contains an index
pub fn exists<Dir: Directory>(dir: &Dir) -> bool {
dir.exists(&META_FILEPATH)
}
/// Accessor to the search executor.
///
/// This pool is used by default when calling `searcher.search(...)`
/// to perform search on the individual segments.
///
/// By default the executor is single thread, and simply runs in the calling thread.
pub fn search_executor(&self) -> &Executor {
self.executor.as_ref()
}
/// Replace the default single thread search executor pool
/// by a thread pool with a given number of threads.
pub fn set_multithread_executor(&mut self, num_threads: usize) {
self.executor = Arc::new(Executor::multi_thread(num_threads, "thrd-tantivy-search-"));
}
/// Replace the default single thread search executor pool
/// by a thread pool with a given number of threads.
pub fn set_default_multithread_executor(&mut self) {
let default_num_threads = num_cpus::get();
self.set_multithread_executor(default_num_threads);
}
/// Creates a new index using the `RAMDirectory`.
///
/// The index will be allocated in anonymous memory.
@@ -100,29 +63,9 @@ impl Index {
#[cfg(feature = "mmap")]
pub fn create_in_dir<P: AsRef<Path>>(directory_path: P, schema: Schema) -> Result<Index> {
let mmap_directory = MmapDirectory::open(directory_path)?;
if Index::exists(&mmap_directory) {
return Err(TantivyError::IndexAlreadyExists);
}
Index::create(mmap_directory, schema)
}
/// Opens or creates a new index in the provided directory
pub fn open_or_create<Dir: Directory>(dir: Dir, schema: Schema) -> Result<Index> {
if Index::exists(&dir) {
let index = Index::open(dir)?;
if index.schema() == schema {
Ok(index)
} else {
Err(TantivyError::SchemaError(
"An index exists but the schema does not match.".to_string(),
))
}
} else {
Index::create(dir, schema)
}
}
/// Creates a new index in a temp directory.
///
/// The index will use the `MMapDirectory` in a newly created directory.
@@ -139,33 +82,30 @@ impl Index {
/// Creates a new index given an implementation of the trait `Directory`
pub fn create<Dir: Directory>(dir: Dir, schema: Schema) -> Result<Index> {
let directory = ManagedDirectory::wrap(dir)?;
let directory = ManagedDirectory::new(dir)?;
Index::from_directory(directory, schema)
}
/// Create a new index from a directory.
///
/// This will overwrite existing meta.json
fn from_directory(mut directory: ManagedDirectory, schema: Schema) -> Result<Index> {
save_new_metas(schema.clone(), directory.borrow_mut())?;
save_new_metas(schema.clone(), 0, directory.borrow_mut())?;
let metas = IndexMeta::with_schema(schema);
Index::create_from_metas(directory, &metas, SegmentMetaInventory::default())
Index::create_from_metas(directory, &metas)
}
/// Creates a new index given a directory and an `IndexMeta`.
fn create_from_metas(
directory: ManagedDirectory,
metas: &IndexMeta,
inventory: SegmentMetaInventory,
) -> Result<Index> {
fn create_from_metas(directory: ManagedDirectory, metas: &IndexMeta) -> Result<Index> {
let schema = metas.schema.clone();
Ok(Index {
let n_cpus = num_cpus::get();
let index = Index {
directory,
schema,
num_searchers: Arc::new(AtomicUsize::new(n_cpus)),
searcher_pool: Arc::new(Pool::new()),
tokenizers: TokenizerManager::default(),
executor: Arc::new(Executor::single_thread()),
inventory,
})
};
index.load_searchers()?;
Ok(index)
}
/// Accessor for the tokenizer manager.
@@ -173,43 +113,6 @@ impl Index {
&self.tokenizers
}
/// Helper to access the tokenizer associated to a specific field.
pub fn tokenizer_for_field(&self, field: Field) -> Result<BoxedTokenizer> {
let field_entry = self.schema.get_field_entry(field);
let field_type = field_entry.field_type();
let tokenizer_manager: &TokenizerManager = self.tokenizers();
let tokenizer_name_opt: Option<BoxedTokenizer> = match field_type {
FieldType::Str(text_options) => text_options
.get_indexing_options()
.map(|text_indexing_options| text_indexing_options.tokenizer().to_string())
.and_then(|tokenizer_name| tokenizer_manager.get(&tokenizer_name)),
_ => None,
};
match tokenizer_name_opt {
Some(tokenizer) => Ok(tokenizer),
None => Err(TantivyError::SchemaError(format!(
"{:?} is not a text field.",
field_entry.name()
))),
}
}
/// Create a default `IndexReader` for the given index.
///
/// See [`Index.reader_builder()`](#method.reader_builder).
pub fn reader(&self) -> Result<IndexReader> {
self.reader_builder().try_into()
}
/// Create a `IndexReader` for the given index.
///
/// Most project should create at most one reader for a given index.
/// This method is typically called only once per `Index` instance,
/// over the lifetime of most problem.
pub fn reader_builder(&self) -> IndexReaderBuilder {
IndexReaderBuilder::new(self.clone())
}
/// Opens a new directory from an index path.
#[cfg(feature = "mmap")]
pub fn open_in_dir<P: AsRef<Path>>(directory_path: P) -> Result<Index> {
@@ -217,35 +120,16 @@ impl Index {
Index::open(mmap_directory)
}
/// Returns the list of the segment metas tracked by the index.
///
/// Such segments can of course be part of the index,
/// but also they could be segments being currently built or in the middle of a merge
/// operation.
pub fn list_all_segment_metas(&self) -> Vec<SegmentMeta> {
self.inventory.all()
}
/// Creates a new segment_meta (Advanced user only).
///
/// As long as the `SegmentMeta` lives, the files associated with the
/// `SegmentMeta` are guaranteed to not be garbage collected, regardless of
/// whether the segment is recorded as part of the index or not.
pub fn new_segment_meta(&self, segment_id: SegmentId, max_doc: u32) -> SegmentMeta {
self.inventory.new_segment_meta(segment_id, max_doc)
}
/// Open the index using the provided directory
pub fn open<D: Directory>(directory: D) -> Result<Index> {
let directory = ManagedDirectory::wrap(directory)?;
let inventory = SegmentMetaInventory::default();
let metas = load_metas(&directory, &inventory)?;
Index::create_from_metas(directory, &metas, inventory)
let directory = ManagedDirectory::new(directory)?;
let metas = load_metas(&directory)?;
Index::create_from_metas(directory, &metas)
}
/// Reads the index meta file from the directory.
pub fn load_metas(&self) -> Result<IndexMeta> {
load_metas(self.directory(), &self.inventory)
load_metas(self.directory())
}
/// Open a new index writer. Attempts to acquire a lockfile.
@@ -264,8 +148,7 @@ impl Index {
/// Each thread will receive a budget of `overall_heap_size_in_bytes / num_threads`.
///
/// # Errors
/// If the lockfile already exists, returns `Error::DirectoryLockBusy` or an `Error::IOError`.
///
/// If the lockfile already exists, returns `Error::FileAlreadyExists`.
/// # Panics
/// If the heap size per thread is too small, panics.
pub fn writer_with_num_threads(
@@ -273,23 +156,9 @@ impl Index {
num_threads: usize,
overall_heap_size_in_bytes: usize,
) -> Result<IndexWriter> {
let directory_lock = self
.directory
.acquire_lock(&INDEX_WRITER_LOCK)
.map_err(|err| {
TantivyError::LockFailure(
err,
Some(
"Failed to acquire index lock. If you are using\
a regular directory, this means there is already an \
`IndexWriter` working on this `Directory`, in this process \
or in a different process."
.to_string(),
),
)
})?;
let directory_lock = DirectoryLock::lock(self.directory().box_clone())?;
let heap_size_in_bytes_per_thread = overall_heap_size_in_bytes / num_threads;
IndexWriter::new(
open_index_writer(
self,
num_threads,
heap_size_in_bytes_per_thread,
@@ -325,8 +194,7 @@ impl Index {
/// Returns the list of segments that are searchable
pub fn searchable_segments(&self) -> Result<Vec<Segment>> {
Ok(self
.searchable_segment_metas()?
Ok(self.searchable_segment_metas()?
.into_iter()
.map(|segment_meta| self.segment(segment_meta))
.collect())
@@ -339,9 +207,7 @@ impl Index {
/// Creates a new segment.
pub fn new_segment(&self) -> Segment {
let segment_meta = self
.inventory
.new_segment_meta(SegmentId::generate_random(), 0);
let segment_meta = SegmentMeta::new(SegmentId::generate_random(), 0);
self.segment(segment_meta)
}
@@ -363,242 +229,69 @@ impl Index {
/// Returns the list of segment ids that are searchable.
pub fn searchable_segment_ids(&self) -> Result<Vec<SegmentId>> {
Ok(self
.searchable_segment_metas()?
Ok(self.searchable_segment_metas()?
.iter()
.map(SegmentMeta::id)
.map(|segment_meta| segment_meta.id())
.collect())
}
/// Returns the set of corrupted files
pub fn validate_checksum(&self) -> Result<HashSet<PathBuf>> {
self.directory.list_damaged().map_err(Into::into)
/// Sets the number of searchers to use
///
/// Only works after the next call to `load_searchers`
pub fn set_num_searchers(&mut self, num_searchers: usize) {
self.num_searchers.store(num_searchers, Ordering::Release);
}
/// Creates a new generation of searchers after
/// a change of the set of searchable indexes.
///
/// This needs to be called when a new segment has been
/// published or after a merge.
pub fn load_searchers(&self) -> Result<()> {
let searchable_segments = self.searchable_segments()?;
let segment_readers: Vec<SegmentReader> = searchable_segments
.iter()
.map(SegmentReader::open)
.collect::<Result<_>>()?;
let schema = self.schema();
let num_searchers: usize = self.num_searchers.load(Ordering::Acquire);
let searchers = (0..num_searchers)
.map(|_| Searcher::new(schema.clone(), segment_readers.clone()))
.collect();
self.searcher_pool.publish_new_generation(searchers);
Ok(())
}
/// Returns a searcher
///
/// This method should be called every single time a search
/// query is performed.
/// The searchers are taken from a pool of `num_searchers` searchers.
/// If no searcher is available
/// this may block.
///
/// The same searcher must be used for a given query, as it ensures
/// the use of a consistent segment set.
pub fn searcher(&self) -> LeasedItem<Searcher> {
self.searcher_pool.acquire()
}
}
impl fmt::Debug for Index {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "Index({:?})", self.directory)
}
}
#[cfg(test)]
mod tests {
use crate::directory::RAMDirectory;
use crate::schema::Field;
use crate::schema::{Schema, INDEXED, TEXT};
use crate::Index;
use crate::IndexReader;
use crate::IndexWriter;
use crate::ReloadPolicy;
use std::thread;
use std::time::Duration;
#[test]
fn test_indexer_for_field() {
let mut schema_builder = Schema::builder();
let num_likes_field = schema_builder.add_u64_field("num_likes", INDEXED);
let body_field = schema_builder.add_text_field("body", TEXT);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
assert!(index.tokenizer_for_field(body_field).is_ok());
assert_eq!(
format!("{:?}", index.tokenizer_for_field(num_likes_field).err()),
"Some(SchemaError(\"\\\"num_likes\\\" is not a text field.\"))"
);
}
#[test]
fn test_index_exists() {
let directory = RAMDirectory::create();
assert!(!Index::exists(&directory));
assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
assert!(Index::exists(&directory));
}
#[test]
fn open_or_create_should_create() {
let directory = RAMDirectory::create();
assert!(!Index::exists(&directory));
assert!(Index::open_or_create(directory.clone(), throw_away_schema()).is_ok());
assert!(Index::exists(&directory));
}
#[test]
fn open_or_create_should_open() {
let directory = RAMDirectory::create();
assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
assert!(Index::exists(&directory));
assert!(Index::open_or_create(directory, throw_away_schema()).is_ok());
}
#[test]
fn create_should_wipeoff_existing() {
let directory = RAMDirectory::create();
assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
assert!(Index::exists(&directory));
assert!(Index::create(directory.clone(), Schema::builder().build()).is_ok());
}
#[test]
fn open_or_create_exists_but_schema_does_not_match() {
let directory = RAMDirectory::create();
assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
assert!(Index::exists(&directory));
assert!(Index::open_or_create(directory.clone(), throw_away_schema()).is_ok());
let err = Index::open_or_create(directory, Schema::builder().build());
assert_eq!(
format!("{:?}", err.unwrap_err()),
"SchemaError(\"An index exists but the schema does not match.\")"
);
}
fn throw_away_schema() -> Schema {
let mut schema_builder = Schema::builder();
let _ = schema_builder.add_u64_field("num_likes", INDEXED);
schema_builder.build()
}
#[test]
fn test_index_on_commit_reload_policy() {
let schema = throw_away_schema();
let field = schema.get_field("num_likes").unwrap();
let index = Index::create_in_ram(schema);
let reader = index
.reader_builder()
.reload_policy(ReloadPolicy::OnCommit)
.try_into()
.unwrap();
assert_eq!(reader.searcher().num_docs(), 0);
let mut writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
test_index_on_commit_reload_policy_aux(field, &mut writer, &reader);
}
#[cfg(feature = "mmap")]
mod mmap_specific {
use super::*;
use std::path::PathBuf;
use tempfile::TempDir;
#[test]
fn test_index_on_commit_reload_policy_mmap() {
let schema = throw_away_schema();
let field = schema.get_field("num_likes").unwrap();
let tempdir = TempDir::new().unwrap();
let tempdir_path = PathBuf::from(tempdir.path());
let index = Index::create_in_dir(&tempdir_path, schema).unwrap();
let mut writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
writer.commit().unwrap();
let reader = index
.reader_builder()
.reload_policy(ReloadPolicy::OnCommit)
.try_into()
.unwrap();
assert_eq!(reader.searcher().num_docs(), 0);
test_index_on_commit_reload_policy_aux(field, &mut writer, &reader);
impl Clone for Index {
fn clone(&self) -> Index {
Index {
directory: self.directory.clone(),
schema: self.schema.clone(),
num_searchers: Arc::clone(&self.num_searchers),
searcher_pool: Arc::clone(&self.searcher_pool),
tokenizers: self.tokenizers.clone(),
}
#[test]
fn test_index_manual_policy_mmap() {
let schema = throw_away_schema();
let field = schema.get_field("num_likes").unwrap();
let index = Index::create_from_tempdir(schema).unwrap();
let mut writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
writer.commit().unwrap();
let reader = index
.reader_builder()
.reload_policy(ReloadPolicy::Manual)
.try_into()
.unwrap();
assert_eq!(reader.searcher().num_docs(), 0);
writer.add_document(doc!(field=>1u64));
writer.commit().unwrap();
thread::sleep(Duration::from_millis(500));
assert_eq!(reader.searcher().num_docs(), 0);
reader.reload().unwrap();
assert_eq!(reader.searcher().num_docs(), 1);
}
#[test]
fn test_index_on_commit_reload_policy_different_directories() {
let schema = throw_away_schema();
let field = schema.get_field("num_likes").unwrap();
let tempdir = TempDir::new().unwrap();
let tempdir_path = PathBuf::from(tempdir.path());
let write_index = Index::create_in_dir(&tempdir_path, schema).unwrap();
let read_index = Index::open_in_dir(&tempdir_path).unwrap();
let reader = read_index
.reader_builder()
.reload_policy(ReloadPolicy::OnCommit)
.try_into()
.unwrap();
assert_eq!(reader.searcher().num_docs(), 0);
let mut writer = write_index.writer_with_num_threads(1, 3_000_000).unwrap();
test_index_on_commit_reload_policy_aux(field, &mut writer, &reader);
}
}
fn test_index_on_commit_reload_policy_aux(
field: Field,
writer: &mut IndexWriter,
reader: &IndexReader,
) {
assert_eq!(reader.searcher().num_docs(), 0);
writer.add_document(doc!(field=>1u64));
writer.commit().unwrap();
let mut count = 0;
for _ in 0..100 {
count = reader.searcher().num_docs();
if count > 0 {
break;
}
thread::sleep(Duration::from_millis(100));
}
assert_eq!(count, 1);
writer.add_document(doc!(field=>2u64));
writer.commit().unwrap();
let mut count = 0;
for _ in 0..10 {
count = reader.searcher().num_docs();
if count > 1 {
break;
}
thread::sleep(Duration::from_millis(100));
}
assert_eq!(count, 2);
}
// This test will not pass on windows, because windows
// prevent deleting files that are MMapped.
#[cfg(not(target_os = "windows"))]
#[test]
fn garbage_collect_works_as_intended() {
let directory = RAMDirectory::create();
let schema = throw_away_schema();
let field = schema.get_field("num_likes").unwrap();
let index = Index::create(directory.clone(), schema).unwrap();
let mut writer = index.writer_with_num_threads(8, 24_000_000).unwrap();
for i in 0u64..8_000u64 {
writer.add_document(doc!(field => i));
}
writer.commit().unwrap();
let mem_right_after_commit = directory.total_mem_usage();
thread::sleep(Duration::from_millis(1_000));
let reader = index
.reader_builder()
.reload_policy(ReloadPolicy::Manual)
.try_into()
.unwrap();
assert_eq!(reader.searcher().num_docs(), 8_000);
writer.wait_merging_threads().unwrap();
let mem_right_after_merge_finished = directory.total_mem_usage();
reader.reload().unwrap();
let searcher = reader.searcher();
assert_eq!(searcher.num_docs(), 8_000);
assert!(mem_right_after_merge_finished < mem_right_after_commit);
}
}

238
src/core/index_meta.rs
View File

@@ -1,184 +1,7 @@
use super::SegmentComponent;
use crate::core::SegmentId;
use crate::schema::Schema;
use crate::Opstamp;
use census::{Inventory, TrackedObject};
use serde;
use core::SegmentMeta;
use schema::Schema;
use serde_json;
use std::collections::HashSet;
use std::fmt;
use std::path::PathBuf;
#[derive(Clone, Debug, Serialize, Deserialize)]
struct DeleteMeta {
num_deleted_docs: u32,
opstamp: Opstamp,
}
#[derive(Clone, Default)]
pub struct SegmentMetaInventory {
inventory: Inventory<InnerSegmentMeta>,
}
impl SegmentMetaInventory {
/// Lists all living `SegmentMeta` object at the time of the call.
pub fn all(&self) -> Vec<SegmentMeta> {
self.inventory
.list()
.into_iter()
.map(SegmentMeta::from)
.collect::<Vec<_>>()
}
pub fn new_segment_meta(&self, segment_id: SegmentId, max_doc: u32) -> SegmentMeta {
let inner = InnerSegmentMeta {
segment_id,
max_doc,
deletes: None,
};
SegmentMeta::from(self.inventory.track(inner))
}
}
/// `SegmentMeta` contains simple meta information about a segment.
///
/// For instance the number of docs it contains,
/// how many are deleted, etc.
#[derive(Clone)]
pub struct SegmentMeta {
tracked: TrackedObject<InnerSegmentMeta>,
}
impl fmt::Debug for SegmentMeta {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
self.tracked.fmt(f)
}
}
impl serde::Serialize for SegmentMeta {
fn serialize<S>(
&self,
serializer: S,
) -> Result<<S as serde::Serializer>::Ok, <S as serde::Serializer>::Error>
where
S: serde::Serializer,
{
self.tracked.serialize(serializer)
}
}
impl From<TrackedObject<InnerSegmentMeta>> for SegmentMeta {
fn from(tracked: TrackedObject<InnerSegmentMeta>) -> SegmentMeta {
SegmentMeta { tracked }
}
}
impl SegmentMeta {
// Creates a new `SegmentMeta` object.
/// Returns the segment id.
pub fn id(&self) -> SegmentId {
self.tracked.segment_id
}
/// Returns the number of deleted documents.
pub fn num_deleted_docs(&self) -> u32 {
self.tracked
.deletes
.as_ref()
.map(|delete_meta| delete_meta.num_deleted_docs)
.unwrap_or(0u32)
}
/// Returns the list of files that
/// are required for the segment meta.
///
/// This is useful as the way tantivy removes files
/// is by removing all files that have been created by tantivy
/// and are not used by any segment anymore.
pub fn list_files(&self) -> HashSet<PathBuf> {
SegmentComponent::iterator()
.map(|component| self.relative_path(*component))
.collect::<HashSet<PathBuf>>()
}
/// Returns the relative path of a component of our segment.
///
/// It just joins the segment id with the extension
/// associated to a segment component.
pub fn relative_path(&self, component: SegmentComponent) -> PathBuf {
let mut path = self.id().uuid_string();
path.push_str(&*match component {
SegmentComponent::POSTINGS => ".idx".to_string(),
SegmentComponent::POSITIONS => ".pos".to_string(),
SegmentComponent::POSITIONSSKIP => ".posidx".to_string(),
SegmentComponent::TERMS => ".term".to_string(),
SegmentComponent::STORE => ".store".to_string(),
SegmentComponent::FASTFIELDS => ".fast".to_string(),
SegmentComponent::FIELDNORMS => ".fieldnorm".to_string(),
SegmentComponent::DELETE => format!(".{}.del", self.delete_opstamp().unwrap_or(0)),
});
PathBuf::from(path)
}
/// Return the highest doc id + 1
///
/// If there are no deletes, then num_docs = max_docs
/// and all the doc ids contains in this segment
/// are exactly (0..max_doc).
pub fn max_doc(&self) -> u32 {
self.tracked.max_doc
}
/// Return the number of documents in the segment.
pub fn num_docs(&self) -> u32 {
self.max_doc() - self.num_deleted_docs()
}
/// Returns the `Opstamp` of the last delete operation
/// taken in account in this segment.
pub fn delete_opstamp(&self) -> Option<Opstamp> {
self.tracked
.deletes
.as_ref()
.map(|delete_meta| delete_meta.opstamp)
}
/// Returns true iff the segment meta contains
/// delete information.
pub fn has_deletes(&self) -> bool {
self.num_deleted_docs() > 0
}
#[doc(hidden)]
pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> SegmentMeta {
let delete_meta = DeleteMeta {
num_deleted_docs,
opstamp,
};
let tracked = self.tracked.map(move |inner_meta| InnerSegmentMeta {
segment_id: inner_meta.segment_id,
max_doc: inner_meta.max_doc,
deletes: Some(delete_meta),
});
SegmentMeta { tracked }
}
}
#[derive(Debug, Clone, Serialize, Deserialize)]
struct InnerSegmentMeta {
segment_id: SegmentId,
max_doc: u32,
deletes: Option<DeleteMeta>,
}
impl InnerSegmentMeta {
pub fn track(self, inventory: &SegmentMetaInventory) -> SegmentMeta {
SegmentMeta {
tracked: inventory.inventory.track(self),
}
}
}
/// Meta information about the `Index`.
///
@@ -188,53 +11,16 @@ impl InnerSegmentMeta {
/// * the index `docstamp`
/// * the schema
///
#[derive(Clone, Serialize)]
#[derive(Clone, Serialize, Deserialize)]
pub struct IndexMeta {
/// List of `SegmentMeta` informations associated to each finalized segment of the index.
pub segments: Vec<SegmentMeta>,
/// Index `Schema`
pub schema: Schema,
/// Opstamp associated to the last `commit` operation.
pub opstamp: Opstamp,
#[serde(skip_serializing_if = "Option::is_none")]
/// Payload associated to the last commit.
///
/// Upon commit, clients can optionally add a small `Striing` payload to their commit
/// to help identify this commit.
/// This payload is entirely unused by tantivy.
pub payload: Option<String>,
}
#[derive(Deserialize)]
struct UntrackedIndexMeta {
pub segments: Vec<InnerSegmentMeta>,
pub schema: Schema,
pub opstamp: Opstamp,
pub opstamp: u64,
#[serde(skip_serializing_if = "Option::is_none")]
pub payload: Option<String>,
}
impl UntrackedIndexMeta {
pub fn track(self, inventory: &SegmentMetaInventory) -> IndexMeta {
IndexMeta {
segments: self
.segments
.into_iter()
.map(|inner_seg_meta| inner_seg_meta.track(inventory))
.collect::<Vec<SegmentMeta>>(),
schema: self.schema,
opstamp: self.opstamp,
payload: self.payload,
}
}
}
impl IndexMeta {
/// Create an `IndexMeta` object representing a brand new `Index`
/// with the given index.
///
/// This new index does not contains any segments.
/// Opstamp will the value `0u64`.
pub fn with_schema(schema: Schema) -> IndexMeta {
IndexMeta {
segments: vec![],
@@ -243,18 +29,10 @@ impl IndexMeta {
payload: None,
}
}
pub(crate) fn deserialize(
meta_json: &str,
inventory: &SegmentMetaInventory,
) -> serde_json::Result<IndexMeta> {
let untracked_meta_json: UntrackedIndexMeta = serde_json::from_str(meta_json)?;
Ok(untracked_meta_json.track(inventory))
}
}
impl fmt::Debug for IndexMeta {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(
f,
"{}",
@@ -268,19 +46,19 @@ impl fmt::Debug for IndexMeta {
mod tests {
use super::IndexMeta;
use crate::schema::{Schema, TEXT};
use schema::{SchemaBuilder, TEXT};
use serde_json;
#[test]
fn test_serialize_metas() {
let schema = {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
schema_builder.add_text_field("text", TEXT);
schema_builder.build()
};
let index_metas = IndexMeta {
segments: Vec::new(),
schema,
schema: schema,
opstamp: 0u64,
payload: None,
};

29
src/core/inverted_index_reader.rs
View File

@@ -1,13 +1,13 @@
use crate::common::BinarySerializable;
use crate::directory::ReadOnlySource;
use crate::positions::PositionReader;
use crate::postings::TermInfo;
use crate::postings::{BlockSegmentPostings, SegmentPostings};
use crate::schema::FieldType;
use crate::schema::IndexRecordOption;
use crate::schema::Term;
use crate::termdict::TermDictionary;
use common::BinarySerializable;
use directory::ReadOnlySource;
use postings::TermInfo;
use postings::{BlockSegmentPostings, SegmentPostings};
use schema::FieldType;
use schema::IndexRecordOption;
use schema::Term;
use termdict::TermDictionary;
use owned_read::OwnedRead;
use positions::PositionReader;
/// The inverted index reader is in charge of accessing
/// the inverted index associated to a specific field.
@@ -32,7 +32,6 @@ pub struct InvertedIndexReader {
}
impl InvertedIndexReader {
#[cfg_attr(feature = "cargo-clippy", allow(clippy::needless_pass_by_value))] // for symmetry
pub(crate) fn new(
termdict: TermDictionary,
postings_source: ReadOnlySource,
@@ -55,12 +54,12 @@ impl InvertedIndexReader {
/// Creates an empty `InvertedIndexReader` object, which
/// contains no terms at all.
pub fn empty(field_type: &FieldType) -> InvertedIndexReader {
pub fn empty(field_type: FieldType) -> InvertedIndexReader {
let record_option = field_type
.get_index_record_option()
.unwrap_or(IndexRecordOption::Basic);
InvertedIndexReader {
termdict: TermDictionary::empty(&field_type),
termdict: TermDictionary::empty(field_type),
postings_source: ReadOnlySource::empty(),
positions_source: ReadOnlySource::empty(),
positions_idx_source: ReadOnlySource::empty(),
@@ -101,6 +100,7 @@ impl InvertedIndexReader {
block_postings.reset(term_info.doc_freq, postings_reader);
}
/// Returns a block postings given a `Term`.
/// This method is for an advanced usage only.
///
@@ -111,7 +111,7 @@ impl InvertedIndexReader {
option: IndexRecordOption,
) -> Option<BlockSegmentPostings> {
self.get_term_info(term)
.map(move |term_info| self.read_block_postings_from_terminfo(&term_info, option))
.map(move|term_info| self.read_block_postings_from_terminfo(&term_info, option))
}
/// Returns a block postings given a `term_info`.
@@ -147,8 +147,7 @@ impl InvertedIndexReader {
if option.has_positions() {
let position_reader = self.positions_source.clone();
let skip_reader = self.positions_idx_source.clone();
let position_reader =
PositionReader::new(position_reader, skip_reader, term_info.positions_idx);
let position_reader = PositionReader::new(position_reader, skip_reader, term_info.positions_idx);
Some(position_reader)
} else {
None

36
src/core/mod.rs
View File

@@ -1,34 +1,42 @@
mod executor;
pub mod index;
mod index_meta;
mod inverted_index_reader;
mod pool;
pub mod searcher;
mod segment;
mod segment_component;
mod segment_id;
mod segment_meta;
mod segment_reader;
pub use self::executor::Executor;
pub use self::index::Index;
pub use self::index_meta::{IndexMeta, SegmentMeta, SegmentMetaInventory};
pub use self::index_meta::IndexMeta;
pub use self::inverted_index_reader::InvertedIndexReader;
pub use self::searcher::Searcher;
pub use self::segment::Segment;
pub use self::segment::SerializableSegment;
pub use self::segment_component::SegmentComponent;
pub use self::segment_id::SegmentId;
pub use self::segment_meta::SegmentMeta;
pub use self::segment_reader::SegmentReader;
use once_cell::sync::Lazy;
use std::path::Path;
use std::path::PathBuf;
/// The meta file contains all the information about the list of segments and the schema
/// of the index.
pub static META_FILEPATH: Lazy<&'static Path> = Lazy::new(|| Path::new("meta.json"));
lazy_static! {
/// The meta file contains all the information about the list of segments and the schema
/// of the index.
pub static ref META_FILEPATH: PathBuf = PathBuf::from("meta.json");
/// The managed file contains a list of files that were created by the tantivy
/// and will therefore be garbage collected when they are deemed useless by tantivy.
///
/// Removing this file is safe, but will prevent the garbage collection of all of the file that
/// are currently in the directory
pub static MANAGED_FILEPATH: Lazy<&'static Path> = Lazy::new(|| Path::new(".managed.json"));
/// The managed file contains a list of files that were created by the tantivy
/// and will therefore be garbage collected when they are deemed useless by tantivy.
///
/// Removing this file is safe, but will prevent the garbage collection of all of the file that
/// are currently in the directory
pub static ref MANAGED_FILEPATH: PathBuf = PathBuf::from(".managed.json");
/// Only one process should be able to write tantivy's index at a time.
/// This file, when present, is in charge of preventing other processes to open an IndexWriter.
///
/// If the process is killed and this file remains, it is safe to remove it manually.
pub static ref LOCKFILE_FILEPATH: PathBuf = PathBuf::from(".tantivy-indexer.lock");
}

134
src/core/pool.rs Normal file
View File

@@ -0,0 +1,134 @@
use crossbeam::queue::MsQueue;
use std::mem;
use std::ops::{Deref, DerefMut};
use std::sync::atomic::AtomicUsize;
use std::sync::atomic::Ordering;
use std::sync::Arc;
pub struct GenerationItem<T> {
generation: usize,
item: T,
}
pub struct Pool<T> {
queue: Arc<MsQueue<GenerationItem<T>>>,
freshest_generation: AtomicUsize,
next_generation: AtomicUsize,
}
impl<T> Pool<T> {
pub fn new() -> Pool<T> {
let queue = Arc::new(MsQueue::new());
Pool {
queue,
freshest_generation: AtomicUsize::default(),
next_generation: AtomicUsize::default(),
}
}
pub fn publish_new_generation(&self, items: Vec<T>) {
let next_generation = self.next_generation.fetch_add(1, Ordering::SeqCst) + 1;
for item in items {
let gen_item = GenerationItem {
item,
generation: next_generation,
};
self.queue.push(gen_item);
}
self.advertise_generation(next_generation);
}
/// At the exit of this method,
/// - freshest_generation has a value greater or equal than generation
/// - freshest_generation has a value that has been advertised
/// - freshest_generation has)
fn advertise_generation(&self, generation: usize) {
// not optimal at all but the easiest to read proof.
loop {
let former_generation = self.freshest_generation.load(Ordering::Acquire);
if former_generation >= generation {
break;
}
self.freshest_generation.compare_and_swap(
former_generation,
generation,
Ordering::SeqCst,
);
}
}
fn generation(&self) -> usize {
self.freshest_generation.load(Ordering::Acquire)
}
pub fn acquire(&self) -> LeasedItem<T> {
let generation = self.generation();
loop {
let gen_item = self.queue.pop();
if gen_item.generation >= generation {
return LeasedItem {
gen_item: Some(gen_item),
recycle_queue: Arc::clone(&self.queue),
};
} else {
// this searcher is obsolete,
// removing it from the pool.
}
}
}
}
pub struct LeasedItem<T> {
gen_item: Option<GenerationItem<T>>,
recycle_queue: Arc<MsQueue<GenerationItem<T>>>,
}
impl<T> Deref for LeasedItem<T> {
type Target = T;
fn deref(&self) -> &T {
&self.gen_item
.as_ref()
.expect("Unwrapping a leased item should never fail")
.item // unwrap is safe here
}
}
impl<T> DerefMut for LeasedItem<T> {
fn deref_mut(&mut self) -> &mut T {
&mut self.gen_item
.as_mut()
.expect("Unwrapping a mut leased item should never fail")
.item // unwrap is safe here
}
}
impl<T> Drop for LeasedItem<T> {
fn drop(&mut self) {
let gen_item: GenerationItem<T> = mem::replace(&mut self.gen_item, None)
.expect("Unwrapping a leased item should never fail");
self.recycle_queue.push(gen_item);
}
}
#[cfg(test)]
mod tests {
use super::Pool;
use std::iter;
#[test]
fn test_pool() {
let items10: Vec<usize> = iter::repeat(10).take(10).collect();
let pool = Pool::new();
pool.publish_new_generation(items10);
for _ in 0..20 {
assert_eq!(*pool.acquire(), 10);
}
let items11: Vec<usize> = iter::repeat(11).take(10).collect();
pool.publish_new_generation(items11);
for _ in 0..20 {
assert_eq!(*pool.acquire(), 11);
}
}
}

160
src/core/searcher.rs
View File

@@ -1,42 +1,15 @@
use crate::collector::Collector;
use crate::collector::SegmentCollector;
use crate::core::Executor;
use crate::core::InvertedIndexReader;
use crate::core::SegmentReader;
use crate::query::Query;
use crate::query::Scorer;
use crate::query::Weight;
use crate::schema::Document;
use crate::schema::Schema;
use crate::schema::{Field, Term};
use crate::space_usage::SearcherSpaceUsage;
use crate::store::StoreReader;
use crate::termdict::TermMerger;
use crate::DocAddress;
use crate::Index;
use crate::Result;
use collector::Collector;
use core::InvertedIndexReader;
use core::SegmentReader;
use query::Query;
use schema::Document;
use schema::Schema;
use schema::{Field, Term};
use std::fmt;
use std::sync::Arc;
fn collect_segment<C: Collector>(
collector: &C,
weight: &dyn Weight,
segment_ord: u32,
segment_reader: &SegmentReader,
) -> Result<C::Fruit> {
let mut scorer = weight.scorer(segment_reader)?;
let mut segment_collector = collector.for_segment(segment_ord as u32, segment_reader)?;
if let Some(delete_bitset) = segment_reader.delete_bitset() {
scorer.for_each(&mut |doc, score| {
if delete_bitset.is_alive(doc) {
segment_collector.collect(doc, score);
}
});
} else {
scorer.for_each(&mut |doc, score| segment_collector.collect(doc, score));
}
Ok(segment_collector.harvest())
}
use termdict::TermMerger;
use DocAddress;
use Result;
/// Holds a list of `SegmentReader`s ready for search.
///
@@ -45,43 +18,25 @@ fn collect_segment<C: Collector>(
///
pub struct Searcher {
schema: Schema,
index: Index,
segment_readers: Vec<SegmentReader>,
store_readers: Vec<StoreReader>,
}
impl Searcher {
/// Creates a new `Searcher`
pub(crate) fn new(
schema: Schema,
index: Index,
segment_readers: Vec<SegmentReader>,
) -> Searcher {
let store_readers = segment_readers
.iter()
.map(SegmentReader::get_store_reader)
.collect();
pub(crate) fn new(schema: Schema, segment_readers: Vec<SegmentReader>) -> Searcher {
Searcher {
schema,
index,
segment_readers,
store_readers,
}
}
/// Returns the `Index` associated to the `Searcher`
pub fn index(&self) -> &Index {
&self.index
}
/// Fetches a document from tantivy's store given a `DocAddress`.
///
/// The searcher uses the segment ordinal to route the
/// the request to the right `Segment`.
pub fn doc(&self, doc_address: DocAddress) -> Result<Document> {
let DocAddress(segment_local_id, doc_id) = doc_address;
let store_reader = &self.store_readers[segment_local_id as usize];
store_reader.get(doc_id)
pub fn doc(&self, doc_address: &DocAddress) -> Result<Document> {
let DocAddress(segment_local_id, doc_id) = *doc_address;
let segment_reader = &self.segment_readers[segment_local_id as usize];
segment_reader.doc(doc_id)
}
/// Access the schema associated to the index of this searcher.
@@ -93,7 +48,7 @@ impl Searcher {
pub fn num_docs(&self) -> u64 {
self.segment_readers
.iter()
.map(|segment_reader| u64::from(segment_reader.num_docs()))
.map(|segment_reader| segment_reader.num_docs() as u64)
.sum::<u64>()
}
@@ -102,9 +57,7 @@ impl Searcher {
pub fn doc_freq(&self, term: &Term) -> u64 {
self.segment_readers
.iter()
.map(|segment_reader| {
u64::from(segment_reader.inverted_index(term.field()).doc_freq(term))
})
.map(|segment_reader| segment_reader.inverted_index(term.field()).doc_freq(term) as u64)
.sum::<u64>()
}
@@ -118,78 +71,19 @@ impl Searcher {
&self.segment_readers[segment_ord as usize]
}
/// Runs a query on the segment readers wrapped by the searcher.
///
/// Search works as follows :
///
/// First the weight object associated to the query is created.
///
/// Then, the query loops over the segments and for each segment :
/// - setup the collector and informs it that the segment being processed has changed.
/// - creates a SegmentCollector for collecting documents associated to the segment
/// - creates a `Scorer` object associated for this segment
/// - iterate through the matched documents and push them to the segment collector.
///
/// Finally, the Collector merges each of the child collectors into itself for result usability
/// by the caller.
pub fn search<C: Collector>(&self, query: &dyn Query, collector: &C) -> Result<C::Fruit> {
let executor = self.index.search_executor();
self.search_with_executor(query, collector, executor)
}
/// Same as [`search(...)`](#method.search) but multithreaded.
///
/// The current implementation is rather naive :
/// multithreading is by splitting search into as many task
/// as there are segments.
///
/// It is powerless at making search faster if your index consists in
/// one large segment.
///
/// Also, keep in my multithreading a single query on several
/// threads will not improve your throughput. It can actually
/// hurt it. It will however, decrease the average response time.
pub fn search_with_executor<C: Collector>(
&self,
query: &dyn Query,
collector: &C,
executor: &Executor,
) -> Result<C::Fruit> {
let scoring_enabled = collector.requires_scoring();
let weight = query.weight(self, scoring_enabled)?;
let segment_readers = self.segment_readers();
let fruits = executor.map(
|(segment_ord, segment_reader)| {
collect_segment(
collector,
weight.as_ref(),
segment_ord as u32,
segment_reader,
)
},
segment_readers.iter().enumerate(),
)?;
collector.merge_fruits(fruits)
/// Runs a query on the segment readers wrapped by the searcher
pub fn search<C: Collector>(&self, query: &Query, collector: &mut C) -> Result<()> {
query.search(self, collector)
}
/// Return the field searcher associated to a `Field`.
pub fn field(&self, field: Field) -> FieldSearcher {
let inv_index_readers = self
.segment_readers
let inv_index_readers = self.segment_readers
.iter()
.map(|segment_reader| segment_reader.inverted_index(field))
.collect::<Vec<_>>();
FieldSearcher::new(inv_index_readers)
}
/// Summarize total space usage of this searcher.
pub fn space_usage(&self) -> SearcherSpaceUsage {
let mut space_usage = SearcherSpaceUsage::new();
for segment_reader in self.segment_readers.iter() {
space_usage.add_segment(segment_reader.space_usage());
}
space_usage
}
}
pub struct FieldSearcher {
@@ -203,9 +97,8 @@ impl FieldSearcher {
/// Returns a Stream over all of the sorted unique terms of
/// for the given field.
pub fn terms(&self) -> TermMerger<'_> {
let term_streamers: Vec<_> = self
.inv_index_readers
pub fn terms(&self) -> TermMerger {
let term_streamers: Vec<_> = self.inv_index_readers
.iter()
.map(|inverted_index| inverted_index.terms().stream())
.collect();
@@ -214,11 +107,10 @@ impl FieldSearcher {
}
impl fmt::Debug for Searcher {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
let segment_ids = self
.segment_readers
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
let segment_ids = self.segment_readers
.iter()
.map(SegmentReader::segment_id)
.map(|segment_reader| segment_reader.segment_id())
.collect::<Vec<_>>();
write!(f, "Searcher({:?})", segment_ids)
}

23
src/core/segment.rs
View File

@@ -1,17 +1,16 @@
use super::SegmentComponent;
use crate::core::Index;
use crate::core::SegmentId;
use crate::core::SegmentMeta;
use crate::directory::error::{OpenReadError, OpenWriteError};
use crate::directory::Directory;
use crate::directory::{ReadOnlySource, WritePtr};
use crate::indexer::segment_serializer::SegmentSerializer;
use crate::schema::Schema;
use crate::Opstamp;
use crate::Result;
use core::Index;
use core::SegmentId;
use core::SegmentMeta;
use directory::error::{OpenReadError, OpenWriteError};
use directory::Directory;
use directory::{ReadOnlySource, WritePtr};
use indexer::segment_serializer::SegmentSerializer;
use schema::Schema;
use std::fmt;
use std::path::PathBuf;
use std::result;
use Result;
/// A segment is a piece of the index.
#[derive(Clone)]
@@ -21,7 +20,7 @@ pub struct Segment {
}
impl fmt::Debug for Segment {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "Segment({:?})", self.id().uuid_string())
}
}
@@ -51,7 +50,7 @@ impl Segment {
}
#[doc(hidden)]
pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> Segment {
pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: u64) -> Segment {
Segment {
index: self.index,
meta: self.meta.with_delete_meta(num_deleted_docs, opstamp),

2
src/core/segment_component.rs
View File

@@ -41,6 +41,6 @@ impl SegmentComponent {
SegmentComponent::STORE,
SegmentComponent::DELETE,
];
SEGMENT_COMPONENTS.iter()
SEGMENT_COMPONENTS.into_iter()
}
}

73
src/core/segment_id.rs
View File

@@ -2,10 +2,6 @@ use std::cmp::{Ord, Ordering};
use std::fmt;
use uuid::Uuid;
#[cfg(test)]
use once_cell::sync::Lazy;
use std::error::Error;
use std::str::FromStr;
#[cfg(test)]
use std::sync::atomic;
@@ -21,10 +17,10 @@ use std::sync::atomic;
pub struct SegmentId(Uuid);
#[cfg(test)]
static AUTO_INC_COUNTER: Lazy<atomic::AtomicUsize> = Lazy::new(|| atomic::AtomicUsize::default());
#[cfg(test)]
const ZERO_ARRAY: [u8; 8] = [0u8; 8];
lazy_static! {
static ref AUTO_INC_COUNTER: atomic::AtomicUsize = atomic::AtomicUsize::default();
static ref EMPTY_ARR: [u8; 8] = [0u8; 8];
}
// During tests, we generate the segment id in a autoincrement manner
// for consistency of segment id between run.
@@ -34,7 +30,7 @@ const ZERO_ARRAY: [u8; 8] = [0u8; 8];
#[cfg(test)]
fn create_uuid() -> Uuid {
let new_auto_inc_id = (*AUTO_INC_COUNTER).fetch_add(1, atomic::Ordering::SeqCst);
Uuid::from_fields(new_auto_inc_id as u32, 0, 0, &ZERO_ARRAY).unwrap()
Uuid::from_fields(new_auto_inc_id as u32, 0, 0, &*EMPTY_ARR).unwrap()
}
#[cfg(not(test))]
@@ -54,55 +50,19 @@ impl SegmentId {
/// and the rest is random.
///
/// Picking the first 8 chars is ok to identify
/// segments in a display message (e.g. a5c4dfcb).
/// segments in a display message.
pub fn short_uuid_string(&self) -> String {
(&self.0.to_simple_ref().to_string()[..8]).to_string()
(&self.0.simple().to_string()[..8]).to_string()
}
/// Returns a segment uuid string.
///
/// It consists in 32 lowercase hexadecimal chars
/// (e.g. a5c4dfcbdfe645089129e308e26d5523)
pub fn uuid_string(&self) -> String {
self.0.to_simple_ref().to_string()
}
/// Build a `SegmentId` string from the full uuid string.
///
/// E.g. "a5c4dfcbdfe645089129e308e26d5523"
pub fn from_uuid_string(uuid_string: &str) -> Result<SegmentId, SegmentIdParseError> {
FromStr::from_str(uuid_string)
}
}
/// Error type used when parsing a `SegmentId` from a string fails.
pub struct SegmentIdParseError(uuid::parser::ParseError);
impl Error for SegmentIdParseError {}
impl fmt::Debug for SegmentIdParseError {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.0.fmt(f)
}
}
impl fmt::Display for SegmentIdParseError {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.0.fmt(f)
}
}
impl FromStr for SegmentId {
type Err = SegmentIdParseError;
fn from_str(uuid_string: &str) -> Result<Self, SegmentIdParseError> {
let uuid = Uuid::parse_str(uuid_string).map_err(SegmentIdParseError)?;
Ok(SegmentId(uuid))
self.0.simple().to_string()
}
}
impl fmt::Debug for SegmentId {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "Seg({:?})", self.short_uuid_string())
}
}
@@ -118,18 +78,3 @@ impl Ord for SegmentId {
self.0.as_bytes().cmp(other.0.as_bytes())
}
}
#[cfg(test)]
mod tests {
use super::SegmentId;
#[test]
fn test_to_uuid_string() {
let full_uuid = "a5c4dfcbdfe645089129e308e26d5523";
let segment_id = SegmentId::from_uuid_string(full_uuid).unwrap();
assert_eq!(segment_id.uuid_string(), full_uuid);
assert_eq!(segment_id.short_uuid_string(), "a5c4dfcb");
// one extra char
assert!(SegmentId::from_uuid_string("a5c4dfcbdfe645089129e308e26d5523b").is_err());
}
}

174
src/core/segment_meta.rs Normal file
View File

@@ -0,0 +1,174 @@
use super::SegmentComponent;
use census::{Inventory, TrackedObject};
use core::SegmentId;
use serde;
use std::collections::HashSet;
use std::fmt;
use std::path::PathBuf;
lazy_static! {
static ref INVENTORY: Inventory<InnerSegmentMeta> = { Inventory::new() };
}
#[derive(Clone, Debug, Serialize, Deserialize)]
struct DeleteMeta {
num_deleted_docs: u32,
opstamp: u64,
}
/// `SegmentMeta` contains simple meta information about a segment.
///
/// For instance the number of docs it contains,
/// how many are deleted, etc.
#[derive(Clone)]
pub struct SegmentMeta {
tracked: TrackedObject<InnerSegmentMeta>,
}
impl fmt::Debug for SegmentMeta {
fn fmt(&self, f: &mut fmt::Formatter) -> Result<(), fmt::Error> {
self.tracked.fmt(f)
}
}
impl serde::Serialize for SegmentMeta {
fn serialize<S>(
&self,
serializer: S,
) -> Result<<S as serde::Serializer>::Ok, <S as serde::Serializer>::Error>
where
S: serde::Serializer,
{
self.tracked.serialize(serializer)
}
}
impl<'a> serde::Deserialize<'a> for SegmentMeta {
fn deserialize<D>(deserializer: D) -> Result<Self, <D as serde::Deserializer<'a>>::Error>
where
D: serde::Deserializer<'a>,
{
let inner = InnerSegmentMeta::deserialize(deserializer)?;
let tracked = INVENTORY.track(inner);
Ok(SegmentMeta { tracked: tracked })
}
}
impl SegmentMeta {
/// Lists all living `SegmentMeta` object at the time of the call.
pub fn all() -> Vec<SegmentMeta> {
INVENTORY
.list()
.into_iter()
.map(|inner| SegmentMeta { tracked: inner })
.collect::<Vec<_>>()
}
/// Creates a new `SegmentMeta` object.
#[doc(hidden)]
pub fn new(segment_id: SegmentId, max_doc: u32) -> SegmentMeta {
let inner = InnerSegmentMeta {
segment_id,
max_doc,
deletes: None,
};
SegmentMeta {
tracked: INVENTORY.track(inner),
}
}
/// Returns the segment id.
pub fn id(&self) -> SegmentId {
self.tracked.segment_id
}
/// Returns the number of deleted documents.
pub fn num_deleted_docs(&self) -> u32 {
self.tracked
.deletes
.as_ref()
.map(|delete_meta| delete_meta.num_deleted_docs)
.unwrap_or(0u32)
}
/// Returns the list of files that
/// are required for the segment meta.
///
/// This is useful as the way tantivy removes files
/// is by removing all files that have been created by tantivy
/// and are not used by any segment anymore.
pub fn list_files(&self) -> HashSet<PathBuf> {
SegmentComponent::iterator()
.map(|component| self.relative_path(*component))
.collect::<HashSet<PathBuf>>()
}
/// Returns the relative path of a component of our segment.
///
/// It just joins the segment id with the extension
/// associated to a segment component.
pub fn relative_path(&self, component: SegmentComponent) -> PathBuf {
let mut path = self.id().uuid_string();
path.push_str(&*match component {
SegmentComponent::POSTINGS => ".idx".to_string(),
SegmentComponent::POSITIONS => ".pos".to_string(),
SegmentComponent::POSITIONSSKIP => ".posidx".to_string(),
SegmentComponent::TERMS => ".term".to_string(),
SegmentComponent::STORE => ".store".to_string(),
SegmentComponent::FASTFIELDS => ".fast".to_string(),
SegmentComponent::FIELDNORMS => ".fieldnorm".to_string(),
SegmentComponent::DELETE => format!(".{}.del", self.delete_opstamp().unwrap_or(0)),
});
PathBuf::from(path)
}
/// Return the highest doc id + 1
///
/// If there are no deletes, then num_docs = max_docs
/// and all the doc ids contains in this segment
/// are exactly (0..max_doc).
pub fn max_doc(&self) -> u32 {
self.tracked.max_doc
}
/// Return the number of documents in the segment.
pub fn num_docs(&self) -> u32 {
self.max_doc() - self.num_deleted_docs()
}
/// Returns the opstamp of the last delete operation
/// taken in account in this segment.
pub fn delete_opstamp(&self) -> Option<u64> {
self.tracked
.deletes
.as_ref()
.map(|delete_meta| delete_meta.opstamp)
}
/// Returns true iff the segment meta contains
/// delete information.
pub fn has_deletes(&self) -> bool {
self.num_deleted_docs() > 0
}
#[doc(hidden)]
pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: u64) -> SegmentMeta {
let delete_meta = DeleteMeta {
num_deleted_docs,
opstamp,
};
let tracked = self.tracked.map(move |inner_meta| InnerSegmentMeta {
segment_id: inner_meta.segment_id,
max_doc: inner_meta.max_doc,
deletes: Some(delete_meta),
});
SegmentMeta { tracked }
}
}
#[derive(Clone, Debug, Serialize, Deserialize)]
struct InnerSegmentMeta {
segment_id: SegmentId,
max_doc: u32,
deletes: Option<DeleteMeta>,
}

225
src/core/segment_reader.rs
View File

@@ -1,27 +1,30 @@
use crate::common::CompositeFile;
use crate::common::HasLen;
use crate::core::InvertedIndexReader;
use crate::core::Segment;
use crate::core::SegmentComponent;
use crate::core::SegmentId;
use crate::directory::ReadOnlySource;
use crate::fastfield::DeleteBitSet;
use crate::fastfield::FacetReader;
use crate::fastfield::FastFieldReaders;
use crate::fieldnorm::FieldNormReader;
use crate::schema::Field;
use crate::schema::FieldType;
use crate::schema::Schema;
use crate::space_usage::SegmentSpaceUsage;
use crate::store::StoreReader;
use crate::termdict::TermDictionary;
use crate::DocId;
use crate::Result;
use fail::fail_point;
use common::CompositeFile;
use common::HasLen;
use core::InvertedIndexReader;
use core::Segment;
use core::SegmentComponent;
use core::SegmentId;
use core::SegmentMeta;
use error::TantivyError;
use fastfield::DeleteBitSet;
use fastfield::FacetReader;
use fastfield::FastFieldReader;
use fastfield::{self, FastFieldNotAvailableError};
use fastfield::{BytesFastFieldReader, FastValue, MultiValueIntFastFieldReader};
use fieldnorm::FieldNormReader;
use schema::Cardinality;
use schema::Document;
use schema::Field;
use schema::FieldType;
use schema::Schema;
use std::collections::HashMap;
use std::fmt;
use std::sync::Arc;
use std::sync::RwLock;
use store::StoreReader;
use termdict::TermDictionary;
use DocId;
use Result;
/// Entry point to access all of the datastructures of the `Segment`
///
@@ -41,17 +44,16 @@ pub struct SegmentReader {
inv_idx_reader_cache: Arc<RwLock<HashMap<Field, Arc<InvertedIndexReader>>>>,
segment_id: SegmentId,
max_doc: DocId,
num_docs: DocId,
segment_meta: SegmentMeta,
termdict_composite: CompositeFile,
postings_composite: CompositeFile,
positions_composite: CompositeFile,
positions_idx_composite: CompositeFile,
fast_fields_readers: Arc<FastFieldReaders>,
fast_fields_composite: CompositeFile,
fieldnorms_composite: CompositeFile,
store_source: ReadOnlySource,
store_reader: StoreReader,
delete_bitset_opt: Option<DeleteBitSet>,
schema: Schema,
}
@@ -62,7 +64,7 @@ impl SegmentReader {
/// Today, `tantivy` does not handle deletes, so it happens
/// to also be the number of documents in the index.
pub fn max_doc(&self) -> DocId {
self.max_doc
self.segment_meta.max_doc()
}
/// Returns the number of documents.
@@ -71,7 +73,7 @@ impl SegmentReader {
/// Today, `tantivy` does not handle deletes so max doc and
/// num_docs are the same.
pub fn num_docs(&self) -> DocId {
self.num_docs
self.segment_meta.num_docs()
}
/// Returns the schema of the index this segment belongs to.
@@ -102,27 +104,98 @@ impl SegmentReader {
///
/// # Panics
/// May panic if the index is corrupted.
pub fn fast_fields(&self) -> &FastFieldReaders {
&self.fast_fields_readers
pub fn fast_field_reader<Item: FastValue>(
&self,
field: Field,
) -> fastfield::Result<FastFieldReader<Item>> {
let field_entry = self.schema.get_field_entry(field);
if Item::fast_field_cardinality(field_entry.field_type()) == Some(Cardinality::SingleValue)
{
self.fast_fields_composite
.open_read(field)
.ok_or_else(|| FastFieldNotAvailableError::new(field_entry))
.map(FastFieldReader::open)
} else {
Err(FastFieldNotAvailableError::new(field_entry))
}
}
pub(crate) fn fast_field_reader_with_idx<Item: FastValue>(
&self,
field: Field,
idx: usize,
) -> fastfield::Result<FastFieldReader<Item>> {
if let Some(ff_source) = self.fast_fields_composite.open_read_with_idx(field, idx) {
Ok(FastFieldReader::open(ff_source))
} else {
let field_entry = self.schema.get_field_entry(field);
Err(FastFieldNotAvailableError::new(field_entry))
}
}
/// Accessor to the `MultiValueIntFastFieldReader` associated to a given `Field`.
/// May panick if the field is not a multivalued fastfield of the type `Item`.
pub fn multi_fast_field_reader<Item: FastValue>(
&self,
field: Field,
) -> fastfield::Result<MultiValueIntFastFieldReader<Item>> {
let field_entry = self.schema.get_field_entry(field);
if Item::fast_field_cardinality(field_entry.field_type()) == Some(Cardinality::MultiValues)
{
let idx_reader = self.fast_field_reader_with_idx(field, 0)?;
let vals_reader = self.fast_field_reader_with_idx(field, 1)?;
Ok(MultiValueIntFastFieldReader::open(idx_reader, vals_reader))
} else {
Err(FastFieldNotAvailableError::new(field_entry))
}
}
/// Accessor to the `BytesFastFieldReader` associated to a given `Field`.
pub fn bytes_fast_field_reader(&self, field: Field) -> fastfield::Result<BytesFastFieldReader> {
let field_entry = self.schema.get_field_entry(field);
match field_entry.field_type() {
&FieldType::Bytes => {}
_ => return Err(FastFieldNotAvailableError::new(field_entry)),
}
let idx_reader = self.fast_fields_composite
.open_read_with_idx(field, 0)
.ok_or_else(|| FastFieldNotAvailableError::new(field_entry))
.map(FastFieldReader::open)?;
let values = self.fast_fields_composite
.open_read_with_idx(field, 1)
.ok_or_else(|| FastFieldNotAvailableError::new(field_entry))?;
Ok(BytesFastFieldReader::open(idx_reader, values))
}
/// Accessor to the `FacetReader` associated to a given `Field`.
pub fn facet_reader(&self, field: Field) -> Option<FacetReader> {
pub fn facet_reader(&self, field: Field) -> Result<FacetReader> {
let field_entry = self.schema.get_field_entry(field);
if field_entry.field_type() != &FieldType::HierarchicalFacet {
return None;
return Err(TantivyError::InvalidArgument(format!(
"The field {:?} is not a \
hierarchical facet.",
field_entry
)).into());
}
let term_ords_reader = self.fast_fields().u64s(field)?;
let termdict_source = self.termdict_composite.open_read(field)?;
let termdict = TermDictionary::from_source(&termdict_source);
let term_ords_reader = self.multi_fast_field_reader(field)?;
let termdict_source = self.termdict_composite.open_read(field).ok_or_else(|| {
TantivyError::InvalidArgument(format!(
"The field \"{}\" is a hierarchical \
but this segment does not seem to have the field term \
dictionary.",
field_entry.name()
))
})?;
let termdict = TermDictionary::from_source(termdict_source);
let facet_reader = FacetReader::new(term_ords_reader, termdict);
Some(facet_reader)
Ok(facet_reader)
}
/// Accessor to the segment's `Field norms`'s reader.
///
/// Field norms are the length (in tokens) of the fields.
/// It is used in the computation of the [TfIdf](https://fulmicoton.gitbooks.io/tantivy-doc/content/tfidf.html).
/// It is used in the computation of the [TfIdf]
/// (https://fulmicoton.gitbooks.io/tantivy-doc/content/tfidf.html).
///
/// They are simply stored as a fast field, serialized in
/// the `.fieldnorm` file of the segment.
@@ -140,8 +213,8 @@ impl SegmentReader {
}
/// Accessor to the segment's `StoreReader`.
pub fn get_store_reader(&self) -> StoreReader {
StoreReader::from_source(self.store_source.clone())
pub fn get_store_reader(&self) -> &StoreReader {
&self.store_reader
}
/// Open a new segment for reading.
@@ -150,8 +223,7 @@ impl SegmentReader {
let termdict_composite = CompositeFile::open(&termdict_source)?;
let store_source = segment.open_read(SegmentComponent::STORE)?;
fail_point!("SegmentReader::open#middle");
let store_reader = StoreReader::from_source(store_source);
let postings_source = segment.open_read(SegmentComponent::POSTINGS)?;
let postings_composite = CompositeFile::open(&postings_source)?;
@@ -172,12 +244,8 @@ impl SegmentReader {
}
};
let schema = segment.schema();
let fast_fields_data = segment.open_read(SegmentComponent::FASTFIELDS)?;
let fast_fields_composite = CompositeFile::open(&fast_fields_data)?;
let fast_field_readers =
Arc::new(FastFieldReaders::load_all(&schema, &fast_fields_composite)?);
let fieldnorms_data = segment.open_read(SegmentComponent::FIELDNORMS)?;
let fieldnorms_composite = CompositeFile::open(&fieldnorms_data)?;
@@ -189,16 +257,16 @@ impl SegmentReader {
None
};
let schema = segment.schema();
Ok(SegmentReader {
inv_idx_reader_cache: Arc::new(RwLock::new(HashMap::new())),
max_doc: segment.meta().max_doc(),
num_docs: segment.meta().num_docs(),
segment_meta: segment.meta().clone(),
termdict_composite,
postings_composite,
fast_fields_readers: fast_field_readers,
fast_fields_composite,
fieldnorms_composite,
segment_id: segment.id(),
store_source,
store_reader,
delete_bitset_opt,
positions_composite,
positions_idx_composite,
@@ -214,8 +282,7 @@ impl SegmentReader {
/// term dictionary associated to a specific field,
/// and opening the posting list associated to any term.
pub fn inverted_index(&self, field: Field) -> Arc<InvertedIndexReader> {
if let Some(inv_idx_reader) = self
.inv_idx_reader_cache
if let Some(inv_idx_reader) = self.inv_idx_reader_cache
.read()
.expect("Lock poisoned. This should never happen")
.get(&field)
@@ -239,27 +306,25 @@ impl SegmentReader {
// As a result, no data is associated to the inverted index.
//
// Returns an empty inverted index.
return Arc::new(InvertedIndexReader::empty(field_type));
return Arc::new(InvertedIndexReader::empty(field_type.clone()));
}
let postings_source = postings_source_opt.unwrap();
let termdict_source = self.termdict_composite.open_read(field).expect(
"Failed to open field term dictionary in composite file. Is the field indexed?",
);
let termdict_source = self.termdict_composite
.open_read(field)
.expect("Failed to open field term dictionary in composite file. Is the field indexed");
let positions_source = self
.positions_composite
let positions_source = self.positions_composite
.open_read(field)
.expect("Index corrupted. Failed to open field positions in composite file.");
let positions_idx_source = self
.positions_idx_composite
let positions_idx_source = self.positions_idx_composite
.open_read(field)
.expect("Index corrupted. Failed to open field positions in composite file.");
let inv_idx_reader = Arc::new(InvertedIndexReader::new(
TermDictionary::from_source(&termdict_source),
TermDictionary::from_source(termdict_source),
postings_source,
positions_source,
positions_idx_source,
@@ -276,6 +341,14 @@ impl SegmentReader {
inv_idx_reader
}
/// Returns the document (or to be accurate, its stored field)
/// bearing the given doc id.
/// This method is slow and should seldom be called from
/// within a collector.
pub fn doc(&self, doc_id: DocId) -> Result<Document> {
self.store_reader.get(doc_id)
}
/// Returns the segment id
pub fn segment_id(&self) -> SegmentId {
self.segment_id
@@ -296,31 +369,13 @@ impl SegmentReader {
}
/// Returns an iterator that will iterate over the alive document ids
pub fn doc_ids_alive(&self) -> SegmentReaderAliveDocsIterator<'_> {
pub fn doc_ids_alive(&self) -> SegmentReaderAliveDocsIterator {
SegmentReaderAliveDocsIterator::new(&self)
}
/// Summarize total space usage of this segment.
pub fn space_usage(&self) -> SegmentSpaceUsage {
SegmentSpaceUsage::new(
self.num_docs(),
self.termdict_composite.space_usage(),
self.postings_composite.space_usage(),
self.positions_composite.space_usage(),
self.positions_idx_composite.space_usage(),
self.fast_fields_readers.space_usage(),
self.fieldnorms_composite.space_usage(),
self.get_store_reader().space_usage(),
self.delete_bitset_opt
.as_ref()
.map(DeleteBitSet::space_usage)
.unwrap_or(0),
)
}
}
impl fmt::Debug for SegmentReader {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "SegmentReader({:?})", self.segment_id)
}
}
@@ -336,7 +391,7 @@ pub struct SegmentReaderAliveDocsIterator<'a> {
impl<'a> SegmentReaderAliveDocsIterator<'a> {
pub fn new(reader: &'a SegmentReader) -> SegmentReaderAliveDocsIterator<'a> {
SegmentReaderAliveDocsIterator {
reader,
reader: reader,
max_doc: reader.max_doc(),
current: 0,
}
@@ -373,13 +428,13 @@ impl<'a> Iterator for SegmentReaderAliveDocsIterator<'a> {
#[cfg(test)]
mod test {
use crate::core::Index;
use crate::schema::{Schema, Term, STORED, TEXT};
use crate::DocId;
use core::Index;
use schema::{SchemaBuilder, Term, STORED, TEXT};
use DocId;
#[test]
fn test_alive_docs_iterator() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
schema_builder.add_text_field("name", TEXT | STORED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema.clone());
@@ -404,7 +459,9 @@ mod test {
// ok, now we should have a deleted doc
index_writer2.commit().unwrap();
}
let searcher = index.reader().unwrap().searcher();
index.load_searchers().unwrap();
let searcher = index.searcher();
let docs: Vec<DocId> = searcher.segment_reader(0).doc_ids_alive().collect();
assert_eq!(vec![0u32, 2u32], docs);
}

152
src/directory/directory.rs
View File

@@ -1,104 +1,11 @@
use crate::directory::directory_lock::Lock;
use crate::directory::error::LockError;
use crate::directory::error::{DeleteError, OpenReadError, OpenWriteError};
use crate::directory::WatchCallback;
use crate::directory::WatchHandle;
use crate::directory::{ReadOnlySource, WritePtr};
use directory::error::{DeleteError, OpenReadError, OpenWriteError};
use directory::{ReadOnlySource, WritePtr};
use std::fmt;
use std::io;
use std::io::Write;
use std::marker::Send;
use std::marker::Sync;
use std::path::Path;
use std::path::PathBuf;
use std::result;
use std::thread;
use std::time::Duration;
/// Retry the logic of acquiring locks is pretty simple.
/// We just retry `n` times after a given `duratio`, both
/// depending on the type of lock.
struct RetryPolicy {
num_retries: usize,
wait_in_ms: u64,
}
impl RetryPolicy {
fn no_retry() -> RetryPolicy {
RetryPolicy {
num_retries: 0,
wait_in_ms: 0,
}
}
fn wait_and_retry(&mut self) -> bool {
if self.num_retries == 0 {
false
} else {
self.num_retries -= 1;
let wait_duration = Duration::from_millis(self.wait_in_ms);
thread::sleep(wait_duration);
true
}
}
}
/// The `DirectoryLock` is an object that represents a file lock.
/// See [`LockType`](struct.LockType.html)
///
/// It is transparently associated to a lock file, that gets deleted
/// on `Drop.` The lock is released automatically on `Drop`.
pub struct DirectoryLock(Box<dyn Send + Sync + 'static>);
struct DirectoryLockGuard {
directory: Box<dyn Directory>,
path: PathBuf,
}
impl<T: Send + Sync + 'static> From<Box<T>> for DirectoryLock {
fn from(underlying: Box<T>) -> Self {
DirectoryLock(underlying)
}
}
impl Drop for DirectoryLockGuard {
fn drop(&mut self) {
if let Err(e) = self.directory.delete(&*self.path) {
error!("Failed to remove the lock file. {:?}", e);
}
}
}
enum TryAcquireLockError {
FileExists,
IOError(io::Error),
}
fn try_acquire_lock(
filepath: &Path,
directory: &mut dyn Directory,
) -> Result<DirectoryLock, TryAcquireLockError> {
let mut write = directory.open_write(filepath).map_err(|e| match e {
OpenWriteError::FileAlreadyExists(_) => TryAcquireLockError::FileExists,
OpenWriteError::IOError(io_error) => TryAcquireLockError::IOError(io_error.into()),
})?;
write.flush().map_err(TryAcquireLockError::IOError)?;
Ok(DirectoryLock::from(Box::new(DirectoryLockGuard {
directory: directory.box_clone(),
path: filepath.to_owned(),
})))
}
fn retry_policy(is_blocking: bool) -> RetryPolicy {
if is_blocking {
RetryPolicy {
num_retries: 100,
wait_in_ms: 100,
}
} else {
RetryPolicy::no_retry()
}
}
/// Write-once read many (WORM) abstraction for where
/// tantivy's data should be stored.
@@ -118,8 +25,6 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
///
/// Specifically, subsequent writes or flushes should
/// have no effect on the returned `ReadOnlySource` object.
///
/// You should only use this to read files create with [`open_write`]
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError>;
/// Removes a file
@@ -159,8 +64,6 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
/// atomic_write.
///
/// This should only be used for small files.
///
/// You should only use this to read files create with [`atomic_write`]
fn atomic_read(&self, path: &Path) -> Result<Vec<u8>, OpenReadError>;
/// Atomically replace the content of a file with data.
@@ -170,58 +73,19 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
///
/// The file may or may not previously exist.
fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()>;
/// Acquire a lock in the given directory.
///
/// The method is blocking or not depending on the `Lock` object.
fn acquire_lock(&self, lock: &Lock) -> Result<DirectoryLock, LockError> {
let mut box_directory = self.box_clone();
let mut retry_policy = retry_policy(lock.is_blocking);
loop {
match try_acquire_lock(&lock.filepath, &mut *box_directory) {
Ok(result) => {
return Ok(result);
}
Err(TryAcquireLockError::FileExists) => {
if !retry_policy.wait_and_retry() {
return Err(LockError::LockBusy);
}
}
Err(TryAcquireLockError::IOError(io_error)) => {
return Err(LockError::IOError(io_error));
}
}
}
}
/// Registers a callback that will be called whenever a change on the `meta.json`
/// using the `atomic_write` API is detected.
///
/// The behavior when using `.watch()` on a file using `.open_write(...)` is, on the other
/// hand, undefined.
///
/// The file will be watched for the lifetime of the returned `WatchHandle`. The caller is
/// required to keep it.
/// It does not override previous callbacks. When the file is modified, all callback that are
/// registered (and whose `WatchHandle` is still alive) are triggered.
///
/// Internally, tantivy only uses this API to detect new commits to implement the
/// `OnCommit` `ReloadPolicy`. Not implementing watch in a `Directory` only prevents the
/// `OnCommit` `ReloadPolicy` to work properly.
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle>;
}
/// DirectoryClone
pub trait DirectoryClone {
/// Clones the directory and boxes the clone
fn box_clone(&self) -> Box<dyn Directory>;
/// Clones the directory and boxes the clone
fn box_clone(&self) -> Box<Directory>;
}
impl<T> DirectoryClone for T
where
T: 'static + Directory + Clone,
T: 'static + Directory + Clone,
{
fn box_clone(&self) -> Box<dyn Directory> {
Box::new(self.clone())
}
fn box_clone(&self) -> Box<Directory> {
Box::new(self.clone())
}
}

55
src/directory/directory_lock.rs
View File

@@ -1,55 +0,0 @@
use once_cell::sync::Lazy;
use std::path::PathBuf;
/// A directory lock.
///
/// A lock is associated to a specific path and some
/// [`LockParams`](./enum.LockParams.html).
/// Tantivy itself uses only two locks but client application
/// can use the directory facility to define their own locks.
/// - [INDEX_WRITER_LOCK](./struct.INDEX_WRITER_LOCK.html)
/// - [META_LOCK](./struct.META_LOCK.html)
///
/// Check out these locks documentation for more information.
///
#[derive(Debug)]
pub struct Lock {
/// The lock needs to be associated with its own file `path`.
/// Depending on the platform, the lock might rely on the creation
/// and deletion of this filepath.
pub filepath: PathBuf,
/// `lock_params` describes whether acquiring the lock is meant
/// to be a blocking operation or a non-blocking.
///
/// Acquiring a blocking lock blocks until the lock is
/// available.
/// Acquiring a blocking lock returns rapidly, either successfully
/// or with an error signifying that someone is already holding
/// the lock.
pub is_blocking: bool,
}
/// Only one process should be able to write tantivy's index at a time.
/// This lock file, when present, is in charge of preventing other processes to open an IndexWriter.
///
/// If the process is killed and this file remains, it is safe to remove it manually.
///
/// Failing to acquire this lock usually means a misuse of tantivy's API,
/// (creating more than one instance of the `IndexWriter`), are a spurious
/// lock file remaining after a crash. In the latter case, removing the file after
/// checking no process running tantivy is running is safe.
pub static INDEX_WRITER_LOCK: Lazy<Lock> = Lazy::new(|| Lock {
filepath: PathBuf::from(".tantivy-writer.lock"),
is_blocking: false,
});
/// The meta lock file is here to protect the segment files being opened by
/// `IndexReader::reload()` from being garbage collected.
/// It makes it possible for another process to safely consume
/// our index in-writing. Ideally, we may have prefered `RWLock` semantics
/// here, but it is difficult to achieve on Windows.
///
/// Opening segment readers is a very fast process.
pub static META_LOCK: Lazy<Lock> = Lazy::new(|| Lock {
filepath: PathBuf::from(".tantivy-meta.lock"),
is_blocking: true,
});

55
src/directory/error.rs
View File

@@ -3,22 +3,6 @@ use std::fmt;
use std::io;
use std::path::PathBuf;
/// Error while trying to acquire a directory lock.
#[derive(Debug, Fail)]
pub enum LockError {
/// Failed to acquired a lock as it is already held by another
/// client.
/// - In the context of a blocking lock, this means the lock was not released within some `timeout` period.
/// - In the context of a non-blocking lock, this means the lock was busy at the moment of the call.
#[fail(
display = "Could not acquire lock as it is already held, possibly by a different process."
)]
LockBusy,
/// Trying to acquire a lock failed with an `IOError`
#[fail(display = "Failed to acquire the lock due to an io:Error.")]
IOError(io::Error),
}
/// General IO error with an optional path to the offending file.
#[derive(Debug)]
pub struct IOError {
@@ -26,14 +10,8 @@ pub struct IOError {
err: io::Error,
}
impl Into<io::Error> for IOError {
fn into(self) -> io::Error {
self.err
}
}
impl fmt::Display for IOError {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match self.path {
Some(ref path) => write!(f, "io error occurred on path '{:?}': '{}'", path, self.err),
None => write!(f, "io error occurred: '{}'", self.err),
@@ -46,7 +24,7 @@ impl StdError for IOError {
"io error occurred"
}
fn cause(&self) -> Option<&dyn StdError> {
fn cause(&self) -> Option<&StdError> {
Some(&self.err)
}
}
@@ -73,18 +51,10 @@ pub enum OpenDirectoryError {
DoesNotExist(PathBuf),
/// The path exists but is not a directory.
NotADirectory(PathBuf),
/// IoError
IoError(io::Error),
}
impl From<io::Error> for OpenDirectoryError {
fn from(io_err: io::Error) -> Self {
OpenDirectoryError::IoError(io_err)
}
}
impl fmt::Display for OpenDirectoryError {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match *self {
OpenDirectoryError::DoesNotExist(ref path) => {
write!(f, "the underlying directory '{:?}' does not exist", path)
@@ -92,11 +62,6 @@ impl fmt::Display for OpenDirectoryError {
OpenDirectoryError::NotADirectory(ref path) => {
write!(f, "the path '{:?}' exists but is not a directory", path)
}
OpenDirectoryError::IoError(ref err) => write!(
f,
"IOError while trying to open/create the directory. {:?}",
err
),
}
}
}
@@ -106,7 +71,7 @@ impl StdError for OpenDirectoryError {
"error occurred while opening a directory"
}
fn cause(&self) -> Option<&dyn StdError> {
fn cause(&self) -> Option<&StdError> {
None
}
}
@@ -129,7 +94,7 @@ impl From<IOError> for OpenWriteError {
}
impl fmt::Display for OpenWriteError {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match *self {
OpenWriteError::FileAlreadyExists(ref path) => {
write!(f, "the file '{:?}' already exists", path)
@@ -148,7 +113,7 @@ impl StdError for OpenWriteError {
"error occurred while opening a file for writing"
}
fn cause(&self) -> Option<&dyn StdError> {
fn cause(&self) -> Option<&StdError> {
match *self {
OpenWriteError::FileAlreadyExists(_) => None,
OpenWriteError::IOError(ref err) => Some(err),
@@ -173,7 +138,7 @@ impl From<IOError> for OpenReadError {
}
impl fmt::Display for OpenReadError {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match *self {
OpenReadError::FileDoesNotExist(ref path) => {
write!(f, "the file '{:?}' does not exist", path)
@@ -192,7 +157,7 @@ impl StdError for OpenReadError {
"error occurred while opening a file for reading"
}
fn cause(&self) -> Option<&dyn StdError> {
fn cause(&self) -> Option<&StdError> {
match *self {
OpenReadError::FileDoesNotExist(_) => None,
OpenReadError::IOError(ref err) => Some(err),
@@ -217,7 +182,7 @@ impl From<IOError> for DeleteError {
}
impl fmt::Display for DeleteError {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match *self {
DeleteError::FileDoesNotExist(ref path) => {
write!(f, "the file '{:?}' does not exist", path)
@@ -234,7 +199,7 @@ impl StdError for DeleteError {
"error occurred while deleting a file"
}
fn cause(&self) -> Option<&dyn StdError> {
fn cause(&self) -> Option<&StdError> {
match *self {
DeleteError::FileDoesNotExist(_) => None,
DeleteError::IOError(ref err) => Some(err),

213
src/directory/footer.rs
View File

@@ -1,213 +0,0 @@
use crate::directory::read_only_source::ReadOnlySource;
use crate::directory::{AntiCallToken, TerminatingWrite};
use byteorder::{ByteOrder, LittleEndian};
use crc32fast::Hasher;
use std::io;
use std::io::Write;
const COMMON_FOOTER_SIZE: usize = 4 * 5;
#[derive(Debug, Clone, PartialEq)]
pub struct Footer {
pub tantivy_version: (u32, u32, u32),
pub meta: String,
pub versioned_footer: VersionedFooter,
}
impl Footer {
pub fn new(versioned_footer: VersionedFooter) -> Self {
let tantivy_version = (
env!("CARGO_PKG_VERSION_MAJOR").parse().unwrap(),
env!("CARGO_PKG_VERSION_MINOR").parse().unwrap(),
env!("CARGO_PKG_VERSION_PATCH").parse().unwrap(),
);
Footer {
tantivy_version,
meta: format!(
"tantivy {}.{}.{}, index v{}",
tantivy_version.0,
tantivy_version.1,
tantivy_version.2,
versioned_footer.version()
),
versioned_footer,
}
}
pub fn to_bytes(&self) -> Vec<u8> {
let mut res = self.versioned_footer.to_bytes();
res.extend_from_slice(self.meta.as_bytes());
let len = res.len();
res.resize(len + COMMON_FOOTER_SIZE, 0);
let mut common_footer = &mut res[len..];
LittleEndian::write_u32(&mut common_footer, self.meta.len() as u32);
LittleEndian::write_u32(&mut common_footer[4..], self.tantivy_version.0);
LittleEndian::write_u32(&mut common_footer[8..], self.tantivy_version.1);
LittleEndian::write_u32(&mut common_footer[12..], self.tantivy_version.2);
LittleEndian::write_u32(&mut common_footer[16..], (len + COMMON_FOOTER_SIZE) as u32);
res
}
pub fn from_bytes(data: &[u8]) -> Result<Self, io::Error> {
let len = data.len();
if len < COMMON_FOOTER_SIZE + 4 {
// 4 bytes for index version, stored in versioned footer
return Err(io::Error::new(
io::ErrorKind::UnexpectedEof,
format!("File corrupted. The footer len must be over 24, while the entire file len is {}", len)
)
);
}
let size = LittleEndian::read_u32(&data[len - 4..]) as usize;
if len < size as usize {
return Err(io::Error::new(
io::ErrorKind::UnexpectedEof,
format!(
"File corrupted. The footer len is {}, while the entire file len is {}",
size, len
),
));
}
let footer = &data[len - size as usize..];
let meta_len = LittleEndian::read_u32(&footer[size - 20..]) as usize;
let tantivy_major = LittleEndian::read_u32(&footer[size - 16..]);
let tantivy_minor = LittleEndian::read_u32(&footer[size - 12..]);
let tantivy_patch = LittleEndian::read_u32(&footer[size - 8..]);
Ok(Footer {
tantivy_version: (tantivy_major, tantivy_minor, tantivy_patch),
meta: String::from_utf8_lossy(&footer[size - meta_len - 20..size - 20]).into_owned(),
versioned_footer: VersionedFooter::from_bytes(&footer[..size - meta_len - 20])?,
})
}
pub fn extract_footer(source: ReadOnlySource) -> Result<(Footer, ReadOnlySource), io::Error> {
let footer = Footer::from_bytes(source.as_slice())?;
let reader = source.slice_to(source.as_slice().len() - footer.size());
Ok((footer, reader))
}
pub fn size(&self) -> usize {
self.versioned_footer.size() as usize + self.meta.len() + 20
}
}
#[derive(Debug, Clone, PartialEq)]
pub enum VersionedFooter {
UnknownVersion { version: u32, size: u32 },
V0(u32), // crc
}
impl VersionedFooter {
pub fn to_bytes(&self) -> Vec<u8> {
match self {
VersionedFooter::V0(crc) => {
let mut res = vec![0; 8];
LittleEndian::write_u32(&mut res, 0);
LittleEndian::write_u32(&mut res[4..], *crc);
res
}
VersionedFooter::UnknownVersion { .. } => {
panic!("Unsupported index should never get serialized");
}
}
}
pub fn from_bytes(footer: &[u8]) -> Result<Self, io::Error> {
assert!(footer.len() >= 4);
let version = LittleEndian::read_u32(footer);
match version {
0 => {
if footer.len() == 8 {
Ok(VersionedFooter::V0(LittleEndian::read_u32(&footer[4..])))
} else {
Err(io::Error::new(
io::ErrorKind::UnexpectedEof,
format!(
"File corrupted. The versioned footer len is {}, while it should be 8",
footer.len()
),
))
}
}
version => Ok(VersionedFooter::UnknownVersion {
version,
size: footer.len() as u32,
}),
}
}
pub fn size(&self) -> u32 {
match self {
VersionedFooter::V0(_) => 8,
VersionedFooter::UnknownVersion { size, .. } => *size,
}
}
pub fn version(&self) -> u32 {
match self {
VersionedFooter::V0(_) => 0,
VersionedFooter::UnknownVersion { version, .. } => *version,
}
}
pub fn crc(&self) -> Option<u32> {
match self {
VersionedFooter::V0(crc) => Some(*crc),
VersionedFooter::UnknownVersion { .. } => None,
}
}
}
pub(crate) struct FooterProxy<W: TerminatingWrite> {
/// always Some except after terminate call
hasher: Option<Hasher>,
/// always Some except after terminate call
writer: Option<W>,
}
impl<W: TerminatingWrite> FooterProxy<W> {
pub fn new(writer: W) -> Self {
FooterProxy {
hasher: Some(Hasher::new()),
writer: Some(writer),
}
}
}
impl<W: TerminatingWrite> Write for FooterProxy<W> {
fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
let count = self.writer.as_mut().unwrap().write(buf)?;
self.hasher.as_mut().unwrap().update(&buf[..count]);
Ok(count)
}
fn flush(&mut self) -> io::Result<()> {
self.writer.as_mut().unwrap().flush()
}
}
impl<W: TerminatingWrite> TerminatingWrite for FooterProxy<W> {
fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()> {
let crc = self.hasher.take().unwrap().finalize();
let footer = Footer::new(VersionedFooter::V0(crc)).to_bytes();
let mut writer = self.writer.take().unwrap();
writer.write_all(&footer)?;
writer.terminate()
}
}
#[cfg(test)]
mod tests {
use crate::directory::footer::{Footer, VersionedFooter};
#[test]
fn test_serialize_deserialize_footer() {
let crc = 123456;
let footer = Footer::new(VersionedFooter::V0(crc));
let footer_bytes = footer.to_bytes();
assert_eq!(Footer::from_bytes(&footer_bytes).unwrap(), footer);
}
}

338
src/directory/managed_directory.rs
View File

@@ -1,15 +1,7 @@
use crate::core::MANAGED_FILEPATH;
use crate::directory::error::{DeleteError, IOError, LockError, OpenReadError, OpenWriteError};
use crate::directory::footer::{Footer, FooterProxy};
use crate::directory::DirectoryLock;
use crate::directory::Lock;
use crate::directory::META_LOCK;
use crate::directory::{ReadOnlySource, WritePtr};
use crate::directory::{WatchCallback, WatchHandle};
use crate::error::DataCorruption;
use crate::Directory;
use crate::Result;
use crc32fast::Hasher;
use core::MANAGED_FILEPATH;
use directory::error::{DeleteError, IOError, OpenReadError, OpenWriteError};
use directory::{ReadOnlySource, WritePtr};
use error::TantivyError;
use serde_json;
use std::collections::HashSet;
use std::io;
@@ -18,17 +10,8 @@ use std::path::{Path, PathBuf};
use std::result;
use std::sync::RwLockWriteGuard;
use std::sync::{Arc, RwLock};
/// Returns true iff the file is "managed".
/// Non-managed file are not subject to garbage collection.
///
/// Filenames that starts by a "." -typically locks-
/// are not managed.
fn is_managed(path: &Path) -> bool {
path.to_str()
.map(|p_str| !p_str.starts_with('.'))
.unwrap_or(true)
}
use Directory;
use Result;
/// Wrapper of directories that keeps track of files created by Tantivy.
///
@@ -41,7 +24,7 @@ fn is_managed(path: &Path) -> bool {
/// useful anymore.
#[derive(Debug)]
pub struct ManagedDirectory {
directory: Box<dyn Directory>,
directory: Box<Directory>,
meta_informations: Arc<RwLock<MetaInformation>>,
}
@@ -53,28 +36,23 @@ struct MetaInformation {
/// Saves the file containing the list of existing files
/// that were created by tantivy.
fn save_managed_paths(
directory: &mut dyn Directory,
wlock: &RwLockWriteGuard<'_, MetaInformation>,
directory: &mut Directory,
wlock: &RwLockWriteGuard<MetaInformation>,
) -> io::Result<()> {
let mut w = serde_json::to_vec(&wlock.managed_paths)?;
writeln!(&mut w)?;
write!(&mut w, "\n")?;
directory.atomic_write(&MANAGED_FILEPATH, &w[..])?;
Ok(())
}
impl ManagedDirectory {
/// Wraps a directory as managed directory.
pub fn wrap<Dir: Directory>(directory: Dir) -> Result<ManagedDirectory> {
pub fn new<Dir: Directory>(directory: Dir) -> Result<ManagedDirectory> {
match directory.atomic_read(&MANAGED_FILEPATH) {
Ok(data) => {
let managed_files_json = String::from_utf8_lossy(&data);
let managed_files: HashSet<PathBuf> = serde_json::from_str(&managed_files_json)
.map_err(|e| {
DataCorruption::new(
MANAGED_FILEPATH.to_path_buf(),
format!("Managed file cannot be deserialized: {:?}. ", e),
)
})?;
.map_err(|_| TantivyError::CorruptedFile(MANAGED_FILEPATH.clone()))?;
Ok(ManagedDirectory {
directory: Box::new(directory),
meta_informations: Arc::new(RwLock::new(MetaInformation {
@@ -97,9 +75,6 @@ impl ManagedDirectory {
///
/// * `living_files` - List of files that are still used by the index.
///
/// The use a callback ensures that the list of living_files is computed
/// while we hold the lock on meta.
///
/// This method does not panick nor returns errors.
/// If a file cannot be deleted (for permission reasons for instance)
/// an error is simply logged, and the file remains in the list of managed
@@ -107,58 +82,48 @@ impl ManagedDirectory {
pub fn garbage_collect<L: FnOnce() -> HashSet<PathBuf>>(&mut self, get_living_files: L) {
info!("Garbage collect");
let mut files_to_delete = vec![];
// It is crucial to get the living files after acquiring the
// read lock of meta informations. That way, we
// avoid the following scenario.
//
// 1) we get the list of living files.
// 2) someone creates a new file.
// 3) we start garbage collection and remove this file
// even though it is a living file.
//
// releasing the lock as .delete() will use it too.
{
let meta_informations_rlock = self
.meta_informations
// releasing the lock as .delete() will use it too.
let meta_informations_rlock = self.meta_informations
.read()
.expect("Managed directory rlock poisoned in garbage collect.");
// The point of this second "file" lock is to enforce the following scenario
// 1) process B tries to load a new set of searcher.
// The list of segments is loaded
// 2) writer change meta.json (for instance after a merge or a commit)
// 3) gc kicks in.
// 4) gc removes a file that was useful for process B, before process B opened it.
if let Ok(_meta_lock) = self.acquire_lock(&META_LOCK) {
let living_files = get_living_files();
for managed_path in &meta_informations_rlock.managed_paths {
if !living_files.contains(managed_path) {
files_to_delete.push(managed_path.clone());
}
// It is crucial to get the living files after acquiring the
// read lock of meta informations. That way, we
// avoid the following scenario.
//
// 1) we get the list of living files.
// 2) someone creates a new file.
// 3) we start garbage collection and remove this file
// even though it is a living file.
let living_files = get_living_files();
for managed_path in &meta_informations_rlock.managed_paths {
if !living_files.contains(managed_path) {
files_to_delete.push(managed_path.clone());
}
} else {
error!("Failed to acquire lock for GC");
}
}
let mut deleted_files = vec![];
for file_to_delete in files_to_delete {
match self.delete(&file_to_delete) {
Ok(_) => {
info!("Deleted {:?}", file_to_delete);
deleted_files.push(file_to_delete);
}
Err(file_error) => {
match file_error {
DeleteError::FileDoesNotExist(_) => {
deleted_files.push(file_to_delete);
}
DeleteError::IOError(_) => {
if !cfg!(target_os = "windows") {
// On windows, delete is expected to fail if the file
// is mmapped.
error!("Failed to delete {:?}", file_to_delete);
{
for file_to_delete in files_to_delete {
match self.delete(&file_to_delete) {
Ok(_) => {
info!("Deleted {:?}", file_to_delete);
deleted_files.push(file_to_delete);
}
Err(file_error) => {
match file_error {
DeleteError::FileDoesNotExist(_) => {
deleted_files.push(file_to_delete);
}
DeleteError::IOError(_) => {
if !cfg!(target_os = "windows") {
// On windows, delete is expected to fail if the file
// is mmapped.
error!("Failed to delete {:?}", file_to_delete);
}
}
}
}
@@ -169,13 +134,14 @@ impl ManagedDirectory {
if !deleted_files.is_empty() {
// update the list of managed files by removing
// the file that were removed.
let mut meta_informations_wlock = self
.meta_informations
let mut meta_informations_wlock = self.meta_informations
.write()
.expect("Managed directory wlock poisoned (2).");
let managed_paths_write = &mut meta_informations_wlock.managed_paths;
for delete_file in &deleted_files {
managed_paths_write.remove(delete_file);
{
let managed_paths_write = &mut meta_informations_wlock.managed_paths;
for delete_file in &deleted_files {
managed_paths_write.remove(delete_file);
}
}
if save_managed_paths(self.directory.as_mut(), &meta_informations_wlock).is_err() {
error!("Failed to save the list of managed files.");
@@ -190,17 +156,8 @@ impl ManagedDirectory {
/// registering the filepath and creating the file
/// will not lead to garbage files that will
/// never get removed.
///
/// File starting by "." are reserved to locks.
/// They are not managed and cannot be subjected
/// to garbage collection.
fn register_file_as_managed(&mut self, filepath: &Path) -> io::Result<()> {
// Files starting by "." (e.g. lock files) are not managed.
if !is_managed(filepath) {
return Ok(());
}
let mut meta_wlock = self
.meta_informations
let mut meta_wlock = self.meta_informations
.write()
.expect("Managed file lock poisoned");
let has_changed = meta_wlock.managed_paths.insert(filepath.to_owned());
@@ -209,59 +166,17 @@ impl ManagedDirectory {
}
Ok(())
}
/// Verify checksum of a managed file
pub fn validate_checksum(&self, path: &Path) -> result::Result<bool, OpenReadError> {
let reader = self.directory.open_read(path)?;
let (footer, data) = Footer::extract_footer(reader)
.map_err(|err| IOError::with_path(path.to_path_buf(), err))?;
let mut hasher = Hasher::new();
hasher.update(data.as_slice());
let crc = hasher.finalize();
Ok(footer
.versioned_footer
.crc()
.map(|v| v == crc)
.unwrap_or(false))
}
/// List files for which checksum does not match content
pub fn list_damaged(&self) -> result::Result<HashSet<PathBuf>, OpenReadError> {
let mut hashset = HashSet::new();
let managed_paths = self
.meta_informations
.read()
.expect("Managed directory rlock poisoned in list damaged.")
.managed_paths
.clone();
for path in managed_paths.into_iter() {
if !self.validate_checksum(&path)? {
hashset.insert(path);
}
}
Ok(hashset)
}
}
impl Directory for ManagedDirectory {
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
let read_only_source = self.directory.open_read(path)?;
let (_footer, reader) = Footer::extract_footer(read_only_source)
.map_err(|err| IOError::with_path(path.to_path_buf(), err))?;
Ok(reader)
self.directory.open_read(path)
}
fn open_write(&mut self, path: &Path) -> result::Result<WritePtr, OpenWriteError> {
self.register_file_as_managed(path)
.map_err(|e| IOError::with_path(path.to_owned(), e))?;
Ok(io::BufWriter::new(Box::new(FooterProxy::new(
self.directory
.open_write(path)?
.into_inner()
.map_err(|_| ())
.expect("buffer should be empty"),
))))
self.directory.open_write(path)
}
fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()> {
@@ -280,14 +195,6 @@ impl Directory for ManagedDirectory {
fn exists(&self, path: &Path) -> bool {
self.directory.exists(path)
}
fn acquire_lock(&self, lock: &Lock) -> result::Result<DirectoryLock, LockError> {
self.directory.acquire_lock(lock)
}
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
self.directory.watch(watch_callback)
}
}
impl Clone for ManagedDirectory {
@@ -299,115 +206,98 @@ impl Clone for ManagedDirectory {
}
}
#[cfg(feature = "mmap")]
#[cfg(test)]
mod tests_mmap_specific {
mod tests {
use crate::directory::{Directory, ManagedDirectory, MmapDirectory, TerminatingWrite};
use std::collections::HashSet;
use std::fs::OpenOptions;
use super::*;
#[cfg(feature = "mmap")]
use directory::MmapDirectory;
use std::io::Write;
use std::path::{Path, PathBuf};
use tempfile::TempDir;
use std::path::Path;
use tempdir::TempDir;
lazy_static! {
static ref TEST_PATH1: &'static Path = Path::new("some_path_for_test");
static ref TEST_PATH2: &'static Path = Path::new("some_path_for_test2");
}
#[test]
#[cfg(feature = "mmap")]
fn test_managed_directory() {
let tempdir = TempDir::new().unwrap();
let tempdir = TempDir::new("index").unwrap();
let tempdir_path = PathBuf::from(tempdir.path());
let test_path1: &'static Path = Path::new("some_path_for_test");
let test_path2: &'static Path = Path::new("some_path_for_test_2");
{
let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
let mut managed_directory = ManagedDirectory::wrap(mmap_directory).unwrap();
let write_file = managed_directory.open_write(test_path1).unwrap();
write_file.terminate().unwrap();
managed_directory
.atomic_write(test_path2, &[0u8, 1u8])
.unwrap();
assert!(managed_directory.exists(test_path1));
assert!(managed_directory.exists(test_path2));
let living_files: HashSet<PathBuf> =
[test_path1.to_owned()].into_iter().cloned().collect();
managed_directory.garbage_collect(|| living_files);
assert!(managed_directory.exists(test_path1));
assert!(!managed_directory.exists(test_path2));
let mut managed_directory = ManagedDirectory::new(mmap_directory).unwrap();
{
let mut write_file = managed_directory.open_write(*TEST_PATH1).unwrap();
write_file.flush().unwrap();
}
{
managed_directory
.atomic_write(*TEST_PATH2, &vec![0u8, 1u8])
.unwrap();
}
{
assert!(managed_directory.exists(*TEST_PATH1));
assert!(managed_directory.exists(*TEST_PATH2));
}
{
let living_files: HashSet<PathBuf> =
[TEST_PATH1.to_owned()].into_iter().cloned().collect();
managed_directory.garbage_collect(|| living_files);
}
{
assert!(managed_directory.exists(*TEST_PATH1));
assert!(!managed_directory.exists(*TEST_PATH2));
}
}
{
let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
let mut managed_directory = ManagedDirectory::wrap(mmap_directory).unwrap();
assert!(managed_directory.exists(test_path1));
assert!(!managed_directory.exists(test_path2));
let living_files: HashSet<PathBuf> = HashSet::new();
managed_directory.garbage_collect(|| living_files);
assert!(!managed_directory.exists(test_path1));
assert!(!managed_directory.exists(test_path2));
let mut managed_directory = ManagedDirectory::new(mmap_directory).unwrap();
{
assert!(managed_directory.exists(*TEST_PATH1));
assert!(!managed_directory.exists(*TEST_PATH2));
}
{
let living_files: HashSet<PathBuf> = HashSet::new();
managed_directory.garbage_collect(|| living_files);
}
{
assert!(!managed_directory.exists(*TEST_PATH1));
assert!(!managed_directory.exists(*TEST_PATH2));
}
}
}
#[test]
#[cfg(feature = "mmap ")]
fn test_managed_directory_gc_while_mmapped() {
let test_path1: &'static Path = Path::new("some_path_for_test");
let tempdir = TempDir::new().unwrap();
let tempdir = TempDir::new("index").unwrap();
let tempdir_path = PathBuf::from(tempdir.path());
let living_files = HashSet::new();
let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
let mut managed_directory = ManagedDirectory::wrap(mmap_directory).unwrap();
let mut write = managed_directory.open_write(test_path1).unwrap();
write.write_all(&[0u8, 1u8]).unwrap();
write.terminate().unwrap();
assert!(managed_directory.exists(test_path1));
let mut managed_directory = ManagedDirectory::new(mmap_directory).unwrap();
managed_directory
.atomic_write(*TEST_PATH1, &vec![0u8, 1u8])
.unwrap();
assert!(managed_directory.exists(*TEST_PATH1));
let _mmap_read = managed_directory.open_read(test_path1).unwrap();
let _mmap_read = managed_directory.open_read(*TEST_PATH1).unwrap();
managed_directory.garbage_collect(|| living_files.clone());
if cfg!(target_os = "windows") {
// On Windows, gc should try and fail the file as it is mmapped.
assert!(managed_directory.exists(test_path1));
assert!(managed_directory.exists(*TEST_PATH1));
// unmap should happen here.
drop(_mmap_read);
// The file should still be in the list of managed file and
// eventually be deleted once mmap is released.
managed_directory.garbage_collect(|| living_files);
assert!(!managed_directory.exists(test_path1));
assert!(!managed_directory.exists(*TEST_PATH1));
} else {
assert!(!managed_directory.exists(test_path1));
assert!(!managed_directory.exists(*TEST_PATH1));
}
}
#[test]
fn test_checksum() {
let test_path1: &'static Path = Path::new("some_path_for_test");
let test_path2: &'static Path = Path::new("other_test_path");
let tempdir = TempDir::new().unwrap();
let tempdir_path = PathBuf::from(tempdir.path());
let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
let mut managed_directory = ManagedDirectory::wrap(mmap_directory).unwrap();
let mut write = managed_directory.open_write(test_path1).unwrap();
write.write_all(&[0u8, 1u8]).unwrap();
write.terminate().unwrap();
let mut write = managed_directory.open_write(test_path2).unwrap();
write.write_all(&[3u8, 4u8, 5u8]).unwrap();
write.terminate().unwrap();
assert!(managed_directory.list_damaged().unwrap().is_empty());
let mut corrupted_path = tempdir_path.clone();
corrupted_path.push(test_path2);
let mut file = OpenOptions::new()
.write(true)
.open(&corrupted_path)
.unwrap();
file.write_all(&[255u8]).unwrap();
file.flush().unwrap();
drop(file);
let damaged = managed_directory.list_damaged().unwrap();
assert_eq!(damaged.len(), 1);
assert!(damaged.contains(test_path2));
}
}

495
src/directory/mmap_directory.rs
View File

@@ -1,27 +1,12 @@
use fs2;
use notify;
use self::fs2::FileExt;
use self::notify::RawEvent;
use self::notify::RecursiveMode;
use self::notify::Watcher;
use crate::core::META_FILEPATH;
use crate::directory::error::LockError;
use crate::directory::error::{
DeleteError, IOError, OpenDirectoryError, OpenReadError, OpenWriteError,
};
use crate::directory::read_only_source::BoxedData;
use crate::directory::AntiCallToken;
use crate::directory::Directory;
use crate::directory::DirectoryLock;
use crate::directory::Lock;
use crate::directory::ReadOnlySource;
use crate::directory::WatchCallback;
use crate::directory::WatchCallbackList;
use crate::directory::WatchHandle;
use crate::directory::{TerminatingWrite, WritePtr};
use atomicwrites;
use memmap::Mmap;
use common::make_io_err;
use directory::error::{DeleteError, IOError, OpenDirectoryError, OpenReadError, OpenWriteError};
use directory::shared_vec_slice::SharedVecSlice;
use directory::Directory;
use directory::ReadOnlySource;
use directory::WritePtr;
use fst::raw::MmapReadOnly;
use std::collections::hash_map::Entry as HashMapEntry;
use std::collections::HashMap;
use std::convert::From;
use std::fmt;
@@ -31,22 +16,14 @@ use std::io::{self, Seek, SeekFrom};
use std::io::{BufWriter, Read, Write};
use std::path::{Path, PathBuf};
use std::result;
use std::sync::mpsc::{channel, Receiver, Sender};
use std::sync::Arc;
use std::sync::Mutex;
use std::sync::RwLock;
use std::sync::Weak;
use std::thread;
use tempfile::TempDir;
/// Create a default io error given a string.
pub(crate) fn make_io_err(msg: String) -> io::Error {
io::Error::new(io::ErrorKind::Other, msg)
}
use tempdir::TempDir;
/// Returns None iff the file exists, can be read, but is empty (and hence
/// cannot be mmapped)
fn open_mmap(full_path: &Path) -> result::Result<Option<Mmap>, OpenReadError> {
/// cannot be mmapped).
///
fn open_mmap(full_path: &Path) -> result::Result<Option<MmapReadOnly>, OpenReadError> {
let file = File::open(full_path).map_err(|e| {
if e.kind() == io::ErrorKind::NotFound {
OpenReadError::FileDoesNotExist(full_path.to_owned())
@@ -55,8 +32,7 @@ fn open_mmap(full_path: &Path) -> result::Result<Option<Mmap>, OpenReadError> {
}
})?;
let meta_data = file
.metadata()
let meta_data = file.metadata()
.map_err(|e| IOError::with_path(full_path.to_owned(), e))?;
if meta_data.len() == 0 {
// if the file size is 0, it will not be possible
@@ -65,7 +41,7 @@ fn open_mmap(full_path: &Path) -> result::Result<Option<Mmap>, OpenReadError> {
return Ok(None);
}
unsafe {
memmap::Mmap::map(&file)
MmapReadOnly::open(&file)
.map(Some)
.map_err(|e| From::from(IOError::with_path(full_path.to_owned(), e)))
}
@@ -88,7 +64,7 @@ pub struct CacheInfo {
struct MmapCache {
counters: CacheCounters,
cache: HashMap<PathBuf, Weak<BoxedData>>,
cache: HashMap<PathBuf, MmapReadOnly>,
}
impl Default for MmapCache {
@@ -101,7 +77,12 @@ impl Default for MmapCache {
}
impl MmapCache {
fn get_info(&self) -> CacheInfo {
/// Removes a `MmapReadOnly` entry from the mmap cache.
fn discard_from_cache(&mut self, full_path: &Path) -> bool {
self.cache.remove(full_path).is_some()
}
fn get_info(&mut self) -> CacheInfo {
let paths: Vec<PathBuf> = self.cache.keys().cloned().collect();
CacheInfo {
counters: self.counters.clone(),
@@ -109,183 +90,57 @@ impl MmapCache {
}
}
fn remove_weak_ref(&mut self) {
let keys_to_remove: Vec<PathBuf> = self
.cache
.iter()
.filter(|(_, mmap_weakref)| mmap_weakref.upgrade().is_none())
.map(|(key, _)| key.clone())
.collect();
for key in keys_to_remove {
self.cache.remove(&key);
}
}
// Returns None if the file exists but as a len of 0 (and hence is not mmappable).
fn get_mmap(&mut self, full_path: &Path) -> Result<Option<Arc<BoxedData>>, OpenReadError> {
if let Some(mmap_weak) = self.cache.get(full_path) {
if let Some(mmap_arc) = mmap_weak.upgrade() {
fn get_mmap(&mut self, full_path: &Path) -> Result<Option<MmapReadOnly>, OpenReadError> {
Ok(match self.cache.entry(full_path.to_owned()) {
HashMapEntry::Occupied(occupied_entry) => {
let mmap = occupied_entry.get();
self.counters.hit += 1;
return Ok(Some(mmap_arc));
Some(mmap.clone())
}
}
self.cache.remove(full_path);
self.counters.miss += 1;
Ok(if let Some(mmap) = open_mmap(full_path)? {
let mmap_arc: Arc<BoxedData> = Arc::new(Box::new(mmap));
let mmap_weak = Arc::downgrade(&mmap_arc);
self.cache.insert(full_path.to_owned(), mmap_weak);
Some(mmap_arc)
} else {
None
})
}
}
struct WatcherWrapper {
_watcher: Mutex<notify::RecommendedWatcher>,
watcher_router: Arc<WatchCallbackList>,
}
impl WatcherWrapper {
pub fn new(path: &Path) -> Result<Self, OpenDirectoryError> {
let (tx, watcher_recv): (Sender<RawEvent>, Receiver<RawEvent>) = channel();
// We need to initialize the
let watcher = notify::raw_watcher(tx)
.and_then(|mut watcher| {
watcher.watch(path, RecursiveMode::Recursive)?;
Ok(watcher)
})
.map_err(|err| match err {
notify::Error::PathNotFound => OpenDirectoryError::DoesNotExist(path.to_owned()),
_ => {
panic!("Unknown error while starting watching directory {:?}", path);
HashMapEntry::Vacant(vacant_entry) => {
self.counters.miss += 1;
if let Some(mmap) = open_mmap(full_path)? {
vacant_entry.insert(mmap.clone());
Some(mmap)
} else {
None
}
})?;
let watcher_router: Arc<WatchCallbackList> = Default::default();
let watcher_router_clone = watcher_router.clone();
thread::Builder::new()
.name("meta-file-watch-thread".to_string())
.spawn(move || {
loop {
match watcher_recv.recv().map(|evt| evt.path) {
Ok(Some(changed_path)) => {
// ... Actually subject to false positive.
// We might want to be more accurate than this at one point.
if let Some(filename) = changed_path.file_name() {
if filename == *META_FILEPATH {
watcher_router_clone.broadcast();
}
}
}
Ok(None) => {
// not an event we are interested in.
}
Err(_e) => {
// the watch send channel was dropped
break;
}
}
}
})?;
Ok(WatcherWrapper {
_watcher: Mutex::new(watcher),
watcher_router,
}
})
}
pub fn watch(&mut self, watch_callback: WatchCallback) -> WatchHandle {
self.watcher_router.subscribe(watch_callback)
}
}
/// Directory storing data in files, read via mmap.
///
/// The Mmap object are cached to limit the
/// system calls.
///
/// In the `MmapDirectory`, locks are implemented using the `fs2` crate definition of locks.
///
/// On MacOS & linux, it relies on `flock` (aka `BSD Lock`). These locks solve most of the
/// problems related to POSIX Locks, but may their contract may not be respected on `NFS`
/// depending on the implementation.
///
/// On Windows the semantics are again different.
#[derive(Clone)]
pub struct MmapDirectory {
inner: Arc<MmapDirectoryInner>,
}
struct MmapDirectoryInner {
root_path: PathBuf,
mmap_cache: RwLock<MmapCache>,
_temp_directory: Option<TempDir>,
watcher: RwLock<Option<WatcherWrapper>>,
}
impl MmapDirectoryInner {
fn new(
root_path: PathBuf,
temp_directory: Option<TempDir>,
) -> Result<MmapDirectoryInner, OpenDirectoryError> {
let mmap_directory_inner = MmapDirectoryInner {
root_path,
mmap_cache: Default::default(),
_temp_directory: temp_directory,
watcher: RwLock::new(None),
};
Ok(mmap_directory_inner)
}
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
// a lot of juggling here, to ensure we don't do anything that panics
// while the rwlock is held. That way we ensure that the rwlock cannot
// be poisoned.
//
// The downside is that we might create a watch wrapper that is not useful.
let need_initialization = self.watcher.read().unwrap().is_none();
if need_initialization {
let watch_wrapper = WatcherWrapper::new(&self.root_path)?;
let mut watch_wlock = self.watcher.write().unwrap();
// the watcher could have been initialized when we released the lock, and
// we do not want to lose the watched files that were set.
if watch_wlock.is_none() {
*watch_wlock = Some(watch_wrapper);
}
}
if let Some(watch_wrapper) = self.watcher.write().unwrap().as_mut() {
Ok(watch_wrapper.watch(watch_callback))
} else {
unreachable!("At this point, watch wrapper is supposed to be initialized");
}
}
mmap_cache: Arc<RwLock<MmapCache>>,
_temp_directory: Arc<Option<TempDir>>,
}
impl fmt::Debug for MmapDirectory {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(f, "MmapDirectory({:?})", self.inner.root_path)
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "MmapDirectory({:?})", self.root_path)
}
}
impl MmapDirectory {
fn new(
root_path: PathBuf,
temp_directory: Option<TempDir>,
) -> Result<MmapDirectory, OpenDirectoryError> {
let inner = MmapDirectoryInner::new(root_path, temp_directory)?;
Ok(MmapDirectory {
inner: Arc::new(inner),
})
}
/// Creates a new MmapDirectory in a temporary directory.
///
/// This is mostly useful to test the MmapDirectory itself.
/// For your unit tests, prefer the RAMDirectory.
pub fn create_from_tempdir() -> Result<MmapDirectory, OpenDirectoryError> {
let tempdir = TempDir::new().map_err(OpenDirectoryError::IoError)?;
pub fn create_from_tempdir() -> io::Result<MmapDirectory> {
let tempdir = TempDir::new("index")?;
let tempdir_path = PathBuf::from(tempdir.path());
MmapDirectory::new(tempdir_path, Some(tempdir))
let directory = MmapDirectory {
root_path: tempdir_path,
mmap_cache: Arc::new(RwLock::new(MmapCache::default())),
_temp_directory: Arc::new(Some(tempdir)),
};
Ok(directory)
}
/// Opens a MmapDirectory in a directory.
@@ -303,14 +158,18 @@ impl MmapDirectory {
directory_path,
)))
} else {
Ok(MmapDirectory::new(PathBuf::from(directory_path), None)?)
Ok(MmapDirectory {
root_path: PathBuf::from(directory_path),
mmap_cache: Arc::new(RwLock::new(MmapCache::default())),
_temp_directory: Arc::new(None),
})
}
}
/// Joins a relative_path to the directory `root_path`
/// to create a proper complete `filepath`.
fn resolve_path(&self, relative_path: &Path) -> PathBuf {
self.inner.root_path.join(relative_path)
self.root_path.join(relative_path)
}
/// Sync the root directory.
@@ -328,14 +187,14 @@ impl MmapDirectory {
#[cfg(windows)]
{
use std::os::windows::fs::OpenOptionsExt;
use winapi::um::winbase;
use winapi::winbase;
open_opts
.write(true)
.custom_flags(winbase::FILE_FLAG_BACKUP_SEMANTICS);
}
let fd = open_opts.open(&self.inner.root_path)?;
let fd = open_opts.open(&self.root_path)?;
fd.sync_all()?;
Ok(())
}
@@ -345,35 +204,14 @@ impl MmapDirectory {
///
/// The `MmapDirectory` embeds a `MmapDirectory`
/// to avoid multiplying the `mmap` system calls.
pub fn get_cache_info(&self) -> CacheInfo {
self.inner
.mmap_cache
pub fn get_cache_info(&mut self) -> CacheInfo {
self.mmap_cache
.write()
.expect("mmap cache lock is poisoned")
.remove_weak_ref();
self.inner
.mmap_cache
.read()
.expect("Mmap cache lock is poisoned.")
.get_info()
}
}
/// We rely on fs2 for file locking. On Windows & MacOS this
/// uses BSD locks (`flock`). The lock is actually released when
/// the `File` object is dropped and its associated file descriptor
/// is closed.
struct ReleaseLockFile {
_file: File,
path: PathBuf,
}
impl Drop for ReleaseLockFile {
fn drop(&mut self) {
debug!("Releasing lock {:?}", self.path);
}
}
/// This Write wraps a File, but has the specificity of
/// call `sync_all` on flush.
struct SafeFileWriter(File);
@@ -401,18 +239,12 @@ impl Seek for SafeFileWriter {
}
}
impl TerminatingWrite for SafeFileWriter {
fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()> {
self.flush()
}
}
impl Directory for MmapDirectory {
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
debug!("Open Read {:?}", path);
let full_path = self.resolve_path(path);
let mut mmap_cache = self.inner.mmap_cache.write().map_err(|_| {
let mut mmap_cache = self.mmap_cache.write().map_err(|_| {
let msg = format!(
"Failed to acquired write lock \
on mmap cache while reading {:?}",
@@ -420,33 +252,11 @@ impl Directory for MmapDirectory {
);
IOError::with_path(path.to_owned(), make_io_err(msg))
})?;
Ok(mmap_cache
.get_mmap(&full_path)?
.map(ReadOnlySource::from)
.unwrap_or_else(ReadOnlySource::empty))
}
/// Any entry associated to the path in the mmap will be
/// removed before the file is deleted.
fn delete(&self, path: &Path) -> result::Result<(), DeleteError> {
let full_path = self.resolve_path(path);
match fs::remove_file(&full_path) {
Ok(_) => self
.sync_directory()
.map_err(|e| IOError::with_path(path.to_owned(), e).into()),
Err(e) => {
if e.kind() == io::ErrorKind::NotFound {
Err(DeleteError::FileDoesNotExist(path.to_owned()))
} else {
Err(IOError::with_path(path.to_owned(), e).into())
}
}
}
}
fn exists(&self, path: &Path) -> bool {
let full_path = self.resolve_path(path);
full_path.exists()
.map(ReadOnlySource::Mmap)
.unwrap_or_else(|| ReadOnlySource::Anonymous(SharedVecSlice::empty())))
}
fn open_write(&mut self, path: &Path) -> Result<WritePtr, OpenWriteError> {
@@ -479,6 +289,43 @@ impl Directory for MmapDirectory {
Ok(BufWriter::new(Box::new(writer)))
}
/// Any entry associated to the path in the mmap will be
/// removed before the file is deleted.
fn delete(&self, path: &Path) -> result::Result<(), DeleteError> {
debug!("Deleting file {:?}", path);
let full_path = self.resolve_path(path);
let mut mmap_cache = self.mmap_cache.write().map_err(|_| {
let msg = format!(
"Failed to acquired write lock \
on mmap cache while deleting {:?}",
path
);
IOError::with_path(path.to_owned(), make_io_err(msg))
})?;
mmap_cache.discard_from_cache(path);
// Removing the entry in the MMap cache.
// The munmap will appear on Drop,
// when the last reference is gone.
mmap_cache.cache.remove(&full_path);
match fs::remove_file(&full_path) {
Ok(_) => self.sync_directory()
.map_err(|e| IOError::with_path(path.to_owned(), e).into()),
Err(e) => {
if e.kind() == io::ErrorKind::NotFound {
Err(DeleteError::FileDoesNotExist(path.to_owned()))
} else {
Err(IOError::with_path(path.to_owned(), e).into())
}
}
}
}
fn exists(&self, path: &Path) -> bool {
let full_path = self.resolve_path(path);
full_path.exists()
}
fn atomic_read(&self, path: &Path) -> Result<Vec<u8>, OpenReadError> {
let full_path = self.resolve_path(path);
let mut buffer = Vec::new();
@@ -505,30 +352,6 @@ impl Directory for MmapDirectory {
meta_file.write(|f| f.write_all(data))?;
Ok(())
}
fn acquire_lock(&self, lock: &Lock) -> Result<DirectoryLock, LockError> {
let full_path = self.resolve_path(&lock.filepath);
// We make sure that the file exists.
let file: File = OpenOptions::new()
.write(true)
.create(true) //< if the file does not exist yet, create it.
.open(&full_path)
.map_err(LockError::IOError)?;
if lock.is_blocking {
file.lock_exclusive().map_err(LockError::IOError)?;
} else {
file.try_lock_exclusive().map_err(|_| LockError::LockBusy)?
}
// dropping the file handle will release the lock.
Ok(DirectoryLock::from(Box::new(ReleaseLockFile {
path: lock.filepath.clone(),
_file: file,
})))
}
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
self.inner.watch(watch_callback)
}
}
#[cfg(test)]
@@ -538,18 +361,6 @@ mod tests {
// The following tests are specific to the MmapDirectory
use super::*;
use crate::schema::{Schema, SchemaBuilder, TEXT};
use crate::Index;
use crate::ReloadPolicy;
use std::fs;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::thread;
use std::time::Duration;
#[test]
fn test_open_non_existant_path() {
assert!(MmapDirectory::open(PathBuf::from("./nowhere")).is_err());
}
#[test]
fn test_open_empty() {
@@ -569,7 +380,7 @@ mod tests {
#[test]
fn test_cache() {
let content = b"abc";
let content = "abc".as_bytes();
// here we test if the cache releases
// mmaps correctly.
@@ -585,104 +396,26 @@ mod tests {
w.flush().unwrap();
}
}
let mut keep = vec![];
for (i, path) in paths.iter().enumerate() {
keep.push(mmap_directory.open_read(path).unwrap());
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), i + 1);
}
assert_eq!(mmap_directory.get_cache_info().counters.hit, 0);
assert_eq!(mmap_directory.get_cache_info().counters.miss, 10);
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 10);
for path in paths.iter() {
let _r = mmap_directory.open_read(path).unwrap();
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), num_paths);
{
for (i, path) in paths.iter().enumerate() {
let _r = mmap_directory.open_read(path).unwrap();
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), i + 1);
}
for path in paths.iter() {
let _r = mmap_directory.open_read(path).unwrap();
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), num_paths);
}
for (i, path) in paths.iter().enumerate() {
mmap_directory.delete(path).unwrap();
assert_eq!(
mmap_directory.get_cache_info().mmapped.len(),
num_paths - i - 1
);
}
}
assert_eq!(mmap_directory.get_cache_info().counters.hit, 10);
assert_eq!(mmap_directory.get_cache_info().counters.miss, 10);
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 10);
for path in paths.iter() {
let _r = mmap_directory.open_read(path).unwrap();
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 10);
}
assert_eq!(mmap_directory.get_cache_info().counters.hit, 20);
assert_eq!(mmap_directory.get_cache_info().counters.miss, 10);
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 10);
drop(keep);
for path in paths.iter() {
let _r = mmap_directory.open_read(path).unwrap();
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 1);
}
assert_eq!(mmap_directory.get_cache_info().counters.hit, 20);
assert_eq!(mmap_directory.get_cache_info().counters.miss, 20);
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 0);
for path in &paths {
mmap_directory.delete(path).unwrap();
}
assert_eq!(mmap_directory.get_cache_info().counters.hit, 20);
assert_eq!(mmap_directory.get_cache_info().counters.miss, 20);
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 0);
for path in paths.iter() {
assert!(mmap_directory.open_read(path).is_err());
}
assert_eq!(mmap_directory.get_cache_info().counters.hit, 20);
assert_eq!(mmap_directory.get_cache_info().counters.miss, 30);
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 0);
}
#[test]
fn test_watch_wrapper() {
let counter: Arc<AtomicUsize> = Default::default();
let counter_clone = counter.clone();
let tmp_dir = tempfile::TempDir::new().unwrap();
let tmp_dirpath = tmp_dir.path().to_owned();
let mut watch_wrapper = WatcherWrapper::new(&tmp_dirpath).unwrap();
let tmp_file = tmp_dirpath.join("coucou");
let _handle = watch_wrapper.watch(Box::new(move || {
counter_clone.fetch_add(1, Ordering::SeqCst);
}));
assert_eq!(counter.load(Ordering::SeqCst), 0);
fs::write(&tmp_file, b"whateverwilldo").unwrap();
thread::sleep(Duration::new(0, 1_000u32));
}
#[test]
fn test_mmap_released() {
let mmap_directory = MmapDirectory::create_from_tempdir().unwrap();
let mut schema_builder: SchemaBuilder = Schema::builder();
let text_field = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
{
let index = Index::create(mmap_directory.clone(), schema).unwrap();
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
for _num_commits in 0..16 {
for _ in 0..10 {
index_writer.add_document(doc!(text_field=>"abc"));
}
index_writer.commit().unwrap();
}
let reader = index
.reader_builder()
.reload_policy(ReloadPolicy::Manual)
.try_into()
.unwrap();
for _ in 0..30 {
index_writer.add_document(doc!(text_field=>"abc"));
index_writer.commit().unwrap();
reader.reload().unwrap();
}
index_writer.wait_merging_threads().unwrap();
reader.reload().unwrap();
let num_segments = reader.searcher().segment_readers().len();
assert_eq!(num_segments, 4);
assert_eq!(
num_segments * 7,
mmap_directory.get_cache_info().mmapped.len()
);
}
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 0);
}
}

174
src/directory/mod.rs
View File

@@ -8,66 +8,158 @@ WORM directory abstraction.
mod mmap_directory;
mod directory;
mod directory_lock;
mod footer;
mod managed_directory;
mod ram_directory;
mod read_only_source;
mod watch_event_router;
mod shared_vec_slice;
/// Errors specific to the directory module.
pub mod error;
pub use self::directory::DirectoryLock;
use std::io::{BufWriter, Seek, Write};
pub use self::directory::{Directory, DirectoryClone};
pub use self::directory_lock::{Lock, INDEX_WRITER_LOCK, META_LOCK};
pub use self::ram_directory::RAMDirectory;
pub use self::read_only_source::ReadOnlySource;
pub(crate) use self::watch_event_router::WatchCallbackList;
pub use self::watch_event_router::{WatchCallback, WatchHandle};
use std::io::{self, BufWriter, Write};
#[cfg(feature = "mmap")]
pub use self::mmap_directory::MmapDirectory;
pub use self::managed_directory::ManagedDirectory;
pub(crate) use self::managed_directory::ManagedDirectory;
/// Struct used to prevent from calling [`terminate_ref`](trait.TerminatingWrite#method.terminate_ref) directly
pub struct AntiCallToken(());
/// Trait used to indicate when no more write need to be done on a writer
pub trait TerminatingWrite: Write {
/// Indicate that the writer will no longer be used. Internally call terminate_ref.
fn terminate(mut self) -> io::Result<()>
where
Self: Sized,
{
self.terminate_ref(AntiCallToken(()))
}
/// You should implement this function to define custom behavior.
/// This function should flush any buffer it may hold.
fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()>;
}
impl<W: TerminatingWrite + ?Sized> TerminatingWrite for Box<W> {
fn terminate_ref(&mut self, token: AntiCallToken) -> io::Result<()> {
self.as_mut().terminate_ref(token)
}
}
impl<W: TerminatingWrite> TerminatingWrite for BufWriter<W> {
fn terminate_ref(&mut self, a: AntiCallToken) -> io::Result<()> {
self.flush()?;
self.get_mut().terminate_ref(a)
}
}
/// Synonym of Seek + Write
pub trait SeekableWrite: Seek + Write {}
impl<T: Seek + Write> SeekableWrite for T {}
/// Write object for Directory.
///
/// `WritePtr` are required to implement both Write
/// and Seek.
pub type WritePtr = BufWriter<Box<dyn TerminatingWrite>>;
pub type WritePtr = BufWriter<Box<SeekableWrite>>;
#[cfg(test)]
mod tests;
mod tests {
use super::*;
use std::io::{Seek, SeekFrom, Write};
use std::path::Path;
lazy_static! {
static ref TEST_PATH: &'static Path = Path::new("some_path_for_test");
}
#[test]
fn test_ram_directory() {
let mut ram_directory = RAMDirectory::create();
test_directory(&mut ram_directory);
}
#[test]
#[cfg(feature = "mmap")]
fn test_mmap_directory() {
let mut mmap_directory = MmapDirectory::create_from_tempdir().unwrap();
test_directory(&mut mmap_directory);
}
#[test]
#[should_panic]
fn ram_directory_panics_if_flush_forgotten() {
let mut ram_directory = RAMDirectory::create();
let mut write_file = ram_directory.open_write(*TEST_PATH).unwrap();
assert!(write_file.write_all(&[4]).is_ok());
}
fn test_simple(directory: &mut Directory) {
{
{
let mut write_file = directory.open_write(*TEST_PATH).unwrap();
assert!(directory.exists(*TEST_PATH));
write_file.write_all(&[4]).unwrap();
write_file.write_all(&[3]).unwrap();
write_file.write_all(&[7, 3, 5]).unwrap();
write_file.flush().unwrap();
}
let read_file = directory.open_read(*TEST_PATH).unwrap();
let data: &[u8] = &*read_file;
assert_eq!(data, &[4u8, 3u8, 7u8, 3u8, 5u8]);
}
assert!(directory.delete(*TEST_PATH).is_ok());
assert!(!directory.exists(*TEST_PATH));
}
fn test_seek(directory: &mut Directory) {
{
{
let mut write_file = directory.open_write(*TEST_PATH).unwrap();
write_file.write_all(&[4, 3, 7, 3, 5]).unwrap();
write_file.seek(SeekFrom::Start(0)).unwrap();
write_file.write_all(&[3, 1]).unwrap();
write_file.flush().unwrap();
}
let read_file = directory.open_read(*TEST_PATH).unwrap();
let data: &[u8] = &*read_file;
assert_eq!(data, &[3u8, 1u8, 7u8, 3u8, 5u8]);
}
assert!(directory.delete(*TEST_PATH).is_ok());
}
fn test_rewrite_forbidden(directory: &mut Directory) {
{
directory.open_write(*TEST_PATH).unwrap();
assert!(directory.exists(*TEST_PATH));
}
{
assert!(directory.open_write(*TEST_PATH).is_err());
}
assert!(directory.delete(*TEST_PATH).is_ok());
}
fn test_write_create_the_file(directory: &mut Directory) {
{
assert!(directory.open_read(*TEST_PATH).is_err());
let _w = directory.open_write(*TEST_PATH).unwrap();
assert!(directory.exists(*TEST_PATH));
assert!(directory.open_read(*TEST_PATH).is_ok());
assert!(directory.delete(*TEST_PATH).is_ok());
}
}
fn test_directory_delete(directory: &mut Directory) {
assert!(directory.open_read(*TEST_PATH).is_err());
let mut write_file = directory.open_write(*TEST_PATH).unwrap();
write_file.write_all(&[1, 2, 3, 4]).unwrap();
write_file.flush().unwrap();
{
let read_handle = directory.open_read(*TEST_PATH).unwrap();
{
assert_eq!(&*read_handle, &[1u8, 2u8, 3u8, 4u8]);
// Mapped files can't be deleted on Windows
if !cfg!(windows) {
assert!(directory.delete(*TEST_PATH).is_ok());
assert_eq!(&*read_handle, &[1u8, 2u8, 3u8, 4u8]);
}
assert!(directory.delete(Path::new("SomeOtherPath")).is_err());
}
}
if cfg!(windows) {
assert!(directory.delete(*TEST_PATH).is_ok());
}
assert!(directory.open_read(*TEST_PATH).is_err());
assert!(directory.delete(*TEST_PATH).is_err());
}
fn test_directory(directory: &mut Directory) {
test_simple(directory);
test_seek(directory);
test_rewrite_forbidden(directory);
test_write_create_the_file(directory);
test_directory_delete(directory);
}
}

174
src/directory/ram_directory.rs
View File

@@ -1,10 +1,8 @@
use crate::core::META_FILEPATH;
use crate::directory::error::{DeleteError, OpenReadError, OpenWriteError};
use crate::directory::AntiCallToken;
use crate::directory::WatchCallbackList;
use crate::directory::{Directory, ReadOnlySource, WatchCallback, WatchHandle};
use crate::directory::{TerminatingWrite, WritePtr};
use fail::fail_point;
use super::shared_vec_slice::SharedVecSlice;
use common::make_io_err;
use directory::error::{DeleteError, IOError, OpenReadError, OpenWriteError};
use directory::WritePtr;
use directory::{Directory, ReadOnlySource};
use std::collections::HashMap;
use std::fmt;
use std::io::{self, BufWriter, Cursor, Seek, SeekFrom, Write};
@@ -24,13 +22,13 @@ use std::sync::{Arc, RwLock};
///
struct VecWriter {
path: PathBuf,
shared_directory: RAMDirectory,
shared_directory: InnerDirectory,
data: Cursor<Vec<u8>>,
is_flushed: bool,
}
impl VecWriter {
fn new(path_buf: PathBuf, shared_directory: RAMDirectory) -> VecWriter {
fn new(path_buf: PathBuf, shared_directory: InnerDirectory) -> VecWriter {
VecWriter {
path: path_buf,
data: Cursor::new(Vec::new()),
@@ -66,59 +64,80 @@ impl Write for VecWriter {
fn flush(&mut self) -> io::Result<()> {
self.is_flushed = true;
let mut fs = self.shared_directory.fs.write().unwrap();
fs.write(self.path.clone(), self.data.get_ref());
self.shared_directory
.write(self.path.clone(), self.data.get_ref())?;
Ok(())
}
}
impl TerminatingWrite for VecWriter {
fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()> {
self.flush()
}
}
#[derive(Default)]
struct InnerDirectory {
fs: HashMap<PathBuf, ReadOnlySource>,
watch_router: WatchCallbackList,
}
#[derive(Clone)]
struct InnerDirectory(Arc<RwLock<HashMap<PathBuf, Arc<Vec<u8>>>>>);
impl InnerDirectory {
fn write(&mut self, path: PathBuf, data: &[u8]) -> bool {
let data = ReadOnlySource::new(Vec::from(data));
self.fs.insert(path, data).is_some()
fn new() -> InnerDirectory {
InnerDirectory(Arc::new(RwLock::new(HashMap::new())))
}
fn write(&self, path: PathBuf, data: &[u8]) -> io::Result<bool> {
let mut map = self.0.write().map_err(|_| {
make_io_err(format!(
"Failed to lock the directory, when trying to write {:?}",
path
))
})?;
let prev_value = map.insert(path, Arc::new(Vec::from(data)));
Ok(prev_value.is_some())
}
fn open_read(&self, path: &Path) -> Result<ReadOnlySource, OpenReadError> {
self.fs
.get(path)
.ok_or_else(|| OpenReadError::FileDoesNotExist(PathBuf::from(path)))
.map(Clone::clone)
self.0
.read()
.map_err(|_| {
let msg = format!(
"Failed to acquire read lock for the \
directory when trying to read {:?}",
path
);
let io_err = make_io_err(msg);
OpenReadError::IOError(IOError::with_path(path.to_owned(), io_err))
})
.and_then(|readable_map| {
readable_map
.get(path)
.ok_or_else(|| OpenReadError::FileDoesNotExist(PathBuf::from(path)))
.map(Arc::clone)
.map(|data| ReadOnlySource::Anonymous(SharedVecSlice::new(data)))
})
}
fn delete(&mut self, path: &Path) -> result::Result<(), DeleteError> {
match self.fs.remove(path) {
Some(_) => Ok(()),
None => Err(DeleteError::FileDoesNotExist(PathBuf::from(path))),
}
fn delete(&self, path: &Path) -> result::Result<(), DeleteError> {
self.0
.write()
.map_err(|_| {
let msg = format!(
"Failed to acquire write lock for the \
directory when trying to delete {:?}",
path
);
let io_err = make_io_err(msg);
DeleteError::IOError(IOError::with_path(path.to_owned(), io_err))
})
.and_then(|mut writable_map| match writable_map.remove(path) {
Some(_) => Ok(()),
None => Err(DeleteError::FileDoesNotExist(PathBuf::from(path))),
})
}
fn exists(&self, path: &Path) -> bool {
self.fs.contains_key(path)
}
fn watch(&mut self, watch_handle: WatchCallback) -> WatchHandle {
self.watch_router.subscribe(watch_handle)
}
fn total_mem_usage(&self) -> usize {
self.fs.values().map(|f| f.len()).sum()
self.0
.read()
.expect("Failed to get read lock directory.")
.contains_key(path)
}
}
impl fmt::Debug for RAMDirectory {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "RAMDirectory")
}
}
@@ -128,47 +147,33 @@ impl fmt::Debug for RAMDirectory {
/// It is mainly meant for unit testing.
/// Writes are only made visible upon flushing.
///
#[derive(Clone, Default)]
#[derive(Clone)]
pub struct RAMDirectory {
fs: Arc<RwLock<InnerDirectory>>,
fs: InnerDirectory,
}
impl RAMDirectory {
/// Constructor
pub fn create() -> RAMDirectory {
Self::default()
}
/// Returns the sum of the size of the different files
/// in the RAMDirectory.
pub fn total_mem_usage(&self) -> usize {
self.fs.read().unwrap().total_mem_usage()
RAMDirectory {
fs: InnerDirectory::new(),
}
}
}
impl Directory for RAMDirectory {
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
self.fs.read().unwrap().open_read(path)
}
fn delete(&self, path: &Path) -> result::Result<(), DeleteError> {
fail_point!("RAMDirectory::delete", |_| {
use crate::directory::error::IOError;
let io_error = IOError::from(io::Error::from(io::ErrorKind::Other));
Err(DeleteError::from(io_error))
});
self.fs.write().unwrap().delete(path)
}
fn exists(&self, path: &Path) -> bool {
self.fs.read().unwrap().exists(path)
self.fs.open_read(path)
}
fn open_write(&mut self, path: &Path) -> Result<WritePtr, OpenWriteError> {
let mut fs = self.fs.write().unwrap();
let path_buf = PathBuf::from(path);
let vec_writer = VecWriter::new(path_buf.clone(), self.clone());
let exists = fs.write(path_buf.clone(), &[]);
let vec_writer = VecWriter::new(path_buf.clone(), self.fs.clone());
let exists = self.fs
.write(path_buf.clone(), &Vec::new())
.map_err(|err| IOError::with_path(path.to_owned(), err))?;
// force the creation of the file to mimic the MMap directory.
if exists {
Err(OpenWriteError::FileAlreadyExists(path_buf))
@@ -177,30 +182,25 @@ impl Directory for RAMDirectory {
}
}
fn delete(&self, path: &Path) -> result::Result<(), DeleteError> {
self.fs.delete(path)
}
fn exists(&self, path: &Path) -> bool {
self.fs.exists(path)
}
fn atomic_read(&self, path: &Path) -> Result<Vec<u8>, OpenReadError> {
Ok(self.open_read(path)?.as_slice().to_owned())
let read = self.open_read(path)?;
Ok(read.as_slice().to_owned())
}
fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()> {
fail_point!("RAMDirectory::atomic_write", |msg| Err(io::Error::new(
io::ErrorKind::Other,
msg.unwrap_or_else(|| "Undefined".to_string())
)));
let path_buf = PathBuf::from(path);
// Reserve the path to prevent calls to .write() to succeed.
self.fs.write().unwrap().write(path_buf.clone(), &[]);
let mut vec_writer = VecWriter::new(path_buf.clone(), self.clone());
let mut vec_writer = VecWriter::new(path_buf.clone(), self.fs.clone());
self.fs.write(path_buf, &Vec::new())?;
vec_writer.write_all(data)?;
vec_writer.flush()?;
if path == Path::new(&*META_FILEPATH) {
self.fs.write().unwrap().watch_router.broadcast();
}
Ok(())
}
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
Ok(self.fs.write().unwrap().watch(watch_callback))
}
}

80
src/directory/read_only_source.rs
View File

@@ -1,9 +1,10 @@
use crate::common::HasLen;
use super::shared_vec_slice::SharedVecSlice;
use common::HasLen;
#[cfg(feature = "mmap")]
use fst::raw::MmapReadOnly;
use stable_deref_trait::{CloneStableDeref, StableDeref};
use std::ops::Deref;
use std::sync::Arc;
pub type BoxedData = Box<dyn Deref<Target = [u8]> + Send + Sync + 'static>;
/// Read object that represents files in tantivy.
///
@@ -11,10 +12,12 @@ pub type BoxedData = Box<dyn Deref<Target = [u8]> + Send + Sync + 'static>;
/// the data in the form of a constant read-only `&[u8]`.
/// Whatever happens to the directory file, the data
/// hold by this object should never be altered or destroyed.
pub struct ReadOnlySource {
data: Arc<BoxedData>,
start: usize,
stop: usize,
pub enum ReadOnlySource {
/// Mmap source of data
#[cfg(feature = "mmap")]
Mmap(MmapReadOnly),
/// Wrapping a `Vec<u8>`
Anonymous(SharedVecSlice),
}
unsafe impl StableDeref for ReadOnlySource {}
@@ -28,38 +31,19 @@ impl Deref for ReadOnlySource {
}
}
impl From<Arc<BoxedData>> for ReadOnlySource {
fn from(data: Arc<BoxedData>) -> Self {
let len = data.len();
ReadOnlySource {
data,
start: 0,
stop: len,
}
}
}
impl ReadOnlySource {
pub(crate) fn new<D>(data: D) -> ReadOnlySource
where
D: Deref<Target = [u8]> + Send + Sync + 'static,
{
let len = data.len();
ReadOnlySource {
data: Arc::new(Box::new(data)),
start: 0,
stop: len,
}
}
/// Creates an empty ReadOnlySource
pub fn empty() -> ReadOnlySource {
ReadOnlySource::new(&[][..])
ReadOnlySource::Anonymous(SharedVecSlice::empty())
}
/// Returns the data underlying the ReadOnlySource object.
pub fn as_slice(&self) -> &[u8] {
&self.data[self.start..self.stop]
match *self {
#[cfg(feature = "mmap")]
ReadOnlySource::Mmap(ref mmap_read_only) => mmap_read_only.as_slice(),
ReadOnlySource::Anonymous(ref shared_vec) => shared_vec.as_slice(),
}
}
/// Splits into 2 `ReadOnlySource`, at the offset given
@@ -80,18 +64,22 @@ impl ReadOnlySource {
/// worth of data in anonymous memory, and only a
/// 1KB slice is remaining, the whole `500MBs`
/// are retained in memory.
pub fn slice(&self, start: usize, stop: usize) -> ReadOnlySource {
pub fn slice(&self, from_offset: usize, to_offset: usize) -> ReadOnlySource {
assert!(
start <= stop,
from_offset <= to_offset,
"Requested negative slice [{}..{}]",
start,
stop
from_offset,
to_offset
);
assert!(stop <= self.len());
ReadOnlySource {
data: self.data.clone(),
start: self.start + start,
stop: self.start + stop,
match *self {
#[cfg(feature = "mmap")]
ReadOnlySource::Mmap(ref mmap_read_only) => {
let sliced_mmap = mmap_read_only.range(from_offset, to_offset - from_offset);
ReadOnlySource::Mmap(sliced_mmap)
}
ReadOnlySource::Anonymous(ref shared_vec) => {
ReadOnlySource::Anonymous(shared_vec.slice(from_offset, to_offset))
}
}
}
@@ -100,7 +88,8 @@ impl ReadOnlySource {
///
/// Equivalent to `.slice(from_offset, self.len())`
pub fn slice_from(&self, from_offset: usize) -> ReadOnlySource {
self.slice(from_offset, self.len())
let len = self.len();
self.slice(from_offset, len)
}
/// Like `.slice(...)` but enforcing only the `to`
@@ -114,18 +103,19 @@ impl ReadOnlySource {
impl HasLen for ReadOnlySource {
fn len(&self) -> usize {
self.stop - self.start
self.as_slice().len()
}
}
impl Clone for ReadOnlySource {
fn clone(&self) -> Self {
self.slice_from(0)
self.slice(0, self.len())
}
}
impl From<Vec<u8>> for ReadOnlySource {
fn from(data: Vec<u8>) -> ReadOnlySource {
ReadOnlySource::new(data)
let shared_data = SharedVecSlice::from(data);
ReadOnlySource::Anonymous(shared_data)
}
}

41
src/directory/shared_vec_slice.rs Normal file
View File

@@ -0,0 +1,41 @@
use std::sync::Arc;
#[derive(Clone)]
pub struct SharedVecSlice {
pub data: Arc<Vec<u8>>,
pub start: usize,
pub len: usize,
}
impl SharedVecSlice {
pub fn empty() -> SharedVecSlice {
SharedVecSlice::new(Arc::new(Vec::new()))
}
pub fn new(data: Arc<Vec<u8>>) -> SharedVecSlice {
let data_len = data.len();
SharedVecSlice {
data,
start: 0,
len: data_len,
}
}
pub fn as_slice(&self) -> &[u8] {
&self.data[self.start..self.start + self.len]
}
pub fn slice(&self, from_offset: usize, to_offset: usize) -> SharedVecSlice {
SharedVecSlice {
data: Arc::clone(&self.data),
start: self.start + from_offset,
len: to_offset - from_offset,
}
}
}
impl From<Vec<u8>> for SharedVecSlice {
fn from(data: Vec<u8>) -> SharedVecSlice {
SharedVecSlice::new(Arc::new(data))
}
}

202
src/directory/tests.rs
View File

@@ -1,202 +0,0 @@
use super::*;
use std::io::Write;
use std::mem;
use std::path::{Path, PathBuf};
use std::sync::atomic::AtomicUsize;
use std::sync::atomic::Ordering;
use std::sync::Arc;
use std::thread;
use std::time;
use std::time::Duration;
#[test]
fn test_ram_directory() {
let mut ram_directory = RAMDirectory::create();
test_directory(&mut ram_directory);
}
#[test]
#[cfg(feature = "mmap")]
fn test_mmap_directory() {
let mut mmap_directory = MmapDirectory::create_from_tempdir().unwrap();
test_directory(&mut mmap_directory);
}
#[test]
#[should_panic]
fn ram_directory_panics_if_flush_forgotten() {
let test_path: &'static Path = Path::new("some_path_for_test");
let mut ram_directory = RAMDirectory::create();
let mut write_file = ram_directory.open_write(test_path).unwrap();
assert!(write_file.write_all(&[4]).is_ok());
}
fn test_simple(directory: &mut dyn Directory) {
let test_path: &'static Path = Path::new("some_path_for_test");
{
let mut write_file = directory.open_write(test_path).unwrap();
assert!(directory.exists(test_path));
write_file.write_all(&[4]).unwrap();
write_file.write_all(&[3]).unwrap();
write_file.write_all(&[7, 3, 5]).unwrap();
write_file.flush().unwrap();
}
{
let read_file = directory.open_read(test_path).unwrap();
let data: &[u8] = &*read_file;
assert_eq!(data, &[4u8, 3u8, 7u8, 3u8, 5u8]);
}
assert!(directory.delete(test_path).is_ok());
assert!(!directory.exists(test_path));
}
fn test_rewrite_forbidden(directory: &mut dyn Directory) {
let test_path: &'static Path = Path::new("some_path_for_test");
{
directory.open_write(test_path).unwrap();
assert!(directory.exists(test_path));
}
{
assert!(directory.open_write(test_path).is_err());
}
assert!(directory.delete(test_path).is_ok());
}
fn test_write_create_the_file(directory: &mut dyn Directory) {
let test_path: &'static Path = Path::new("some_path_for_test");
{
assert!(directory.open_read(test_path).is_err());
let _w = directory.open_write(test_path).unwrap();
assert!(directory.exists(test_path));
assert!(directory.open_read(test_path).is_ok());
assert!(directory.delete(test_path).is_ok());
}
}
fn test_directory_delete(directory: &mut dyn Directory) {
let test_path: &'static Path = Path::new("some_path_for_test");
assert!(directory.open_read(test_path).is_err());
let mut write_file = directory.open_write(&test_path).unwrap();
write_file.write_all(&[1, 2, 3, 4]).unwrap();
write_file.flush().unwrap();
{
let read_handle = directory.open_read(&test_path).unwrap();
assert_eq!(&*read_handle, &[1u8, 2u8, 3u8, 4u8]);
// Mapped files can't be deleted on Windows
if !cfg!(windows) {
assert!(directory.delete(&test_path).is_ok());
assert_eq!(&*read_handle, &[1u8, 2u8, 3u8, 4u8]);
}
assert!(directory.delete(Path::new("SomeOtherPath")).is_err());
}
if cfg!(windows) {
assert!(directory.delete(&test_path).is_ok());
}
assert!(directory.open_read(&test_path).is_err());
assert!(directory.delete(&test_path).is_err());
}
fn test_directory(directory: &mut dyn Directory) {
test_simple(directory);
test_rewrite_forbidden(directory);
test_write_create_the_file(directory);
test_directory_delete(directory);
test_lock_non_blocking(directory);
test_lock_blocking(directory);
test_watch(directory);
}
fn test_watch(directory: &mut dyn Directory) {
let counter: Arc<AtomicUsize> = Default::default();
let counter_clone = counter.clone();
let watch_callback = Box::new(move || {
counter_clone.fetch_add(1, Ordering::SeqCst);
});
assert!(directory
.atomic_write(Path::new("meta.json"), b"random_test_data")
.is_ok());
thread::sleep(Duration::new(0, 10_000));
assert_eq!(0, counter.load(Ordering::SeqCst));
let watch_handle = directory.watch(watch_callback).unwrap();
for i in 0..10 {
assert_eq!(i, counter.load(Ordering::SeqCst));
assert!(directory
.atomic_write(Path::new("meta.json"), b"random_test_data_2")
.is_ok());
for _ in 0..1_000 {
if counter.load(Ordering::SeqCst) > i {
break;
}
thread::sleep(Duration::from_millis(10));
}
assert_eq!(i + 1, counter.load(Ordering::SeqCst));
}
mem::drop(watch_handle);
assert!(directory
.atomic_write(Path::new("meta.json"), b"random_test_data")
.is_ok());
thread::sleep(Duration::from_millis(200));
assert_eq!(10, counter.load(Ordering::SeqCst));
}
fn test_lock_non_blocking(directory: &mut dyn Directory) {
{
let lock_a_res = directory.acquire_lock(&Lock {
filepath: PathBuf::from("a.lock"),
is_blocking: false,
});
assert!(lock_a_res.is_ok());
let lock_b_res = directory.acquire_lock(&Lock {
filepath: PathBuf::from("b.lock"),
is_blocking: false,
});
assert!(lock_b_res.is_ok());
let lock_a_res2 = directory.acquire_lock(&Lock {
filepath: PathBuf::from("a.lock"),
is_blocking: false,
});
assert!(lock_a_res2.is_err());
}
let lock_a_res = directory.acquire_lock(&Lock {
filepath: PathBuf::from("a.lock"),
is_blocking: false,
});
assert!(lock_a_res.is_ok());
}
fn test_lock_blocking(directory: &mut dyn Directory) {
let lock_a_res = directory.acquire_lock(&Lock {
filepath: PathBuf::from("a.lock"),
is_blocking: true,
});
assert!(lock_a_res.is_ok());
std::thread::spawn(move || {
//< lock_a_res is sent to the thread.
std::thread::sleep(time::Duration::from_millis(10));
// explicitely droping lock_a_res. It would have been sufficient to just force it
// to be part of the move, but the intent seems clearer that way.
drop(lock_a_res);
});
{
// A non-blocking call should fail, as the thread is running and holding the lock.
let lock_a_res = directory.acquire_lock(&Lock {
filepath: PathBuf::from("a.lock"),
is_blocking: false,
});
assert!(lock_a_res.is_err());
}
{
// the blocking call should wait for at least 10ms.
let start = time::Instant::now();
let lock_a_res = directory.acquire_lock(&Lock {
filepath: PathBuf::from("a.lock"),
is_blocking: true,
});
assert!(lock_a_res.is_ok());
assert!(start.elapsed().subsec_millis() >= 10);
}
}

155
src/directory/watch_event_router.rs
View File

@@ -1,155 +0,0 @@
use std::sync::Arc;
use std::sync::RwLock;
use std::sync::Weak;
/// Type alias for callbacks registered when watching files of a `Directory`.
pub type WatchCallback = Box<dyn Fn() -> () + Sync + Send>;
/// Helper struct to implement the watch method in `Directory` implementations.
///
/// It registers callbacks (See `.subscribe(...)`) and
/// calls them upon calls to `.broadcast(...)`.
#[derive(Default)]
pub struct WatchCallbackList {
router: RwLock<Vec<Weak<WatchCallback>>>,
}
/// Controls how long a directory should watch for a file change.
///
/// After all the clones of `WatchHandle` are dropped, the associated will not be called when a
/// file change is detected.
#[must_use = "This `WatchHandle` controls the lifetime of the watch and should therefore be used."]
#[derive(Clone)]
pub struct WatchHandle(Arc<WatchCallback>);
impl WatchCallbackList {
/// Suscribes a new callback and returns a handle that controls the lifetime of the callback.
pub fn subscribe(&self, watch_callback: WatchCallback) -> WatchHandle {
let watch_callback_arc = Arc::new(watch_callback);
let watch_callback_weak = Arc::downgrade(&watch_callback_arc);
self.router.write().unwrap().push(watch_callback_weak);
WatchHandle(watch_callback_arc)
}
fn list_callback(&self) -> Vec<Arc<WatchCallback>> {
let mut callbacks = vec![];
let mut router_wlock = self.router.write().unwrap();
let mut i = 0;
while i < router_wlock.len() {
if let Some(watch) = router_wlock[i].upgrade() {
callbacks.push(watch);
i += 1;
} else {
router_wlock.swap_remove(i);
}
}
callbacks
}
/// Triggers all callbacks
pub fn broadcast(&self) {
let callbacks = self.list_callback();
let spawn_res = std::thread::Builder::new()
.name("watch-callbacks".to_string())
.spawn(move || {
for callback in callbacks {
callback();
}
});
if let Err(err) = spawn_res {
error!(
"Failed to spawn thread to call watch callbacks. Cause: {:?}",
err
);
}
}
}
#[cfg(test)]
mod tests {
use crate::directory::WatchCallbackList;
use std::mem;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::sync::Arc;
use std::thread;
use std::time::Duration;
const WAIT_TIME: u64 = 20;
#[test]
fn test_watch_event_router_simple() {
let watch_event_router = WatchCallbackList::default();
let counter: Arc<AtomicUsize> = Default::default();
let counter_clone = counter.clone();
let inc_callback = Box::new(move || {
counter_clone.fetch_add(1, Ordering::SeqCst);
});
watch_event_router.broadcast();
assert_eq!(0, counter.load(Ordering::SeqCst));
let handle_a = watch_event_router.subscribe(inc_callback);
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(0, counter.load(Ordering::SeqCst));
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(1, counter.load(Ordering::SeqCst));
watch_event_router.broadcast();
watch_event_router.broadcast();
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(4, counter.load(Ordering::SeqCst));
mem::drop(handle_a);
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(4, counter.load(Ordering::SeqCst));
}
#[test]
fn test_watch_event_router_multiple_callback_same_key() {
let watch_event_router = WatchCallbackList::default();
let counter: Arc<AtomicUsize> = Default::default();
let inc_callback = |inc: usize| {
let counter_clone = counter.clone();
Box::new(move || {
counter_clone.fetch_add(inc, Ordering::SeqCst);
})
};
let handle_a = watch_event_router.subscribe(inc_callback(1));
let handle_a2 = watch_event_router.subscribe(inc_callback(10));
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(0, counter.load(Ordering::SeqCst));
watch_event_router.broadcast();
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(22, counter.load(Ordering::SeqCst));
mem::drop(handle_a);
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(32, counter.load(Ordering::SeqCst));
mem::drop(handle_a2);
watch_event_router.broadcast();
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(32, counter.load(Ordering::SeqCst));
}
#[test]
fn test_watch_event_router_multiple_callback_different_key() {
let watch_event_router = WatchCallbackList::default();
let counter: Arc<AtomicUsize> = Default::default();
let counter_clone = counter.clone();
let inc_callback = Box::new(move || {
counter_clone.fetch_add(1, Ordering::SeqCst);
});
let handle_a = watch_event_router.subscribe(inc_callback);
assert_eq!(0, counter.load(Ordering::SeqCst));
watch_event_router.broadcast();
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(2, counter.load(Ordering::SeqCst));
thread::sleep(Duration::from_millis(WAIT_TIME));
mem::drop(handle_a);
watch_event_router.broadcast();
thread::sleep(Duration::from_millis(WAIT_TIME));
assert_eq!(2, counter.load(Ordering::SeqCst));
}
}

32
src/docset.rs
View File

@@ -1,9 +1,8 @@
use crate::common::BitSet;
use crate::fastfield::DeleteBitSet;
use crate::DocId;
use common::BitSet;
use std::borrow::Borrow;
use std::borrow::BorrowMut;
use std::cmp::Ordering;
use DocId;
/// Expresses the outcome of a call to `DocSet`'s `.skip_next(...)`.
#[derive(PartialEq, Eq, Debug)]
@@ -96,23 +95,9 @@ pub trait DocSet {
}
/// Returns the number documents matching.
/// Calling this method consumes the `DocSet`.
fn count(&mut self, delete_bitset: &DeleteBitSet) -> u32 {
let mut count = 0u32;
while self.advance() {
if !delete_bitset.is_deleted(self.doc()) {
count += 1u32;
}
}
count
}
/// Returns the count of documents, deleted or not.
/// Calling this method consumes the `DocSet`.
///
/// Of course, the result is an upper bound of the result
/// given by `count()`.
fn count_including_deleted(&mut self) -> u32 {
/// Calling this method consumes the `DocSet`.
fn count(&mut self) -> u32 {
let mut count = 0u32;
while self.advance() {
count += 1u32;
@@ -142,14 +127,9 @@ impl<TDocSet: DocSet + ?Sized> DocSet for Box<TDocSet> {
unboxed.size_hint()
}
fn count(&mut self, delete_bitset: &DeleteBitSet) -> u32 {
fn count(&mut self) -> u32 {
let unboxed: &mut TDocSet = self.borrow_mut();
unboxed.count(delete_bitset)
}
fn count_including_deleted(&mut self) -> u32 {
let unboxed: &mut TDocSet = self.borrow_mut();
unboxed.count_including_deleted()
unboxed.count()
}
fn append_to_bitset(&mut self, bitset: &mut BitSet) {

117
src/error.rs
View File

@@ -2,140 +2,92 @@
use std::io;
use crate::directory::error::LockError;
use crate::directory::error::{IOError, OpenDirectoryError, OpenReadError, OpenWriteError};
use crate::fastfield::FastFieldNotAvailableError;
use crate::query;
use crate::schema;
use directory::error::{IOError, OpenDirectoryError, OpenReadError, OpenWriteError};
use fastfield::FastFieldNotAvailableError;
use query;
use schema;
use serde_json;
use std::fmt;
use std::path::PathBuf;
use std::sync::PoisonError;
pub struct DataCorruption {
filepath: Option<PathBuf>,
comment: String,
}
impl DataCorruption {
pub fn new(filepath: PathBuf, comment: String) -> DataCorruption {
DataCorruption {
filepath: Some(filepath),
comment,
}
}
pub fn comment_only(comment: String) -> DataCorruption {
DataCorruption {
filepath: None,
comment,
}
}
}
impl fmt::Debug for DataCorruption {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
write!(f, "Data corruption: ")?;
if let Some(ref filepath) = &self.filepath {
write!(f, "(in file `{:?}`)", filepath)?;
}
write!(f, ": {}.", self.comment)?;
Ok(())
}
}
/// The library's failure based error enum
#[derive(Debug, Fail)]
pub enum TantivyError {
/// Path does not exist.
#[fail(display = "Path does not exist: '{:?}'", _0)]
#[fail(display = "path does not exist: '{:?}'", _0)]
PathDoesNotExist(PathBuf),
/// File already exists, this is a problem when we try to write into a new file.
#[fail(display = "File already exists: '{:?}'", _0)]
#[fail(display = "file already exists: '{:?}'", _0)]
FileAlreadyExists(PathBuf),
/// Index already exists in this directory
#[fail(display = "Index already exists")]
IndexAlreadyExists,
/// Failed to acquire file lock
#[fail(display = "Failed to acquire Lockfile: {:?}. {:?}", _0, _1)]
LockFailure(LockError, Option<String>),
/// IO Error.
#[fail(display = "An IO error occurred: '{}'", _0)]
#[fail(display = "an IO error occurred: '{}'", _0)]
IOError(#[cause] IOError),
/// Data corruption.
#[fail(display = "{:?}", _0)]
DataCorruption(DataCorruption),
/// The data within is corrupted.
///
/// For instance, it contains invalid JSON.
#[fail(display = "file contains corrupted data: '{:?}'", _0)]
CorruptedFile(PathBuf),
/// A thread holding the locked panicked and poisoned the lock.
#[fail(display = "A thread holding the locked panicked and poisoned the lock")]
#[fail(display = "a thread holding the locked panicked and poisoned the lock")]
Poisoned,
/// Invalid argument was passed by the user.
#[fail(display = "An invalid argument was passed: '{}'", _0)]
#[fail(display = "an invalid argument was passed: '{}'", _0)]
InvalidArgument(String),
/// An Error happened in one of the thread.
#[fail(display = "An error occurred in a thread: '{}'", _0)]
#[fail(display = "an error occurred in a thread: '{}'", _0)]
ErrorInThread(String),
/// An Error appeared related to the schema.
#[fail(display = "Schema error: '{}'", _0)]
SchemaError(String),
/// System error. (e.g.: We failed spawning a new thread)
#[fail(display = "System error.'{}'", _0)]
SystemError(String),
}
impl From<DataCorruption> for TantivyError {
fn from(data_corruption: DataCorruption) -> TantivyError {
TantivyError::DataCorruption(data_corruption)
}
/// Tried to access a fastfield reader for a field not configured accordingly.
#[fail(display = "fast field not available: '{:?}'", _0)]
FastFieldError(#[cause] FastFieldNotAvailableError),
}
impl From<FastFieldNotAvailableError> for TantivyError {
fn from(fastfield_error: FastFieldNotAvailableError) -> TantivyError {
TantivyError::SchemaError(format!("{}", fastfield_error))
}
}
impl From<LockError> for TantivyError {
fn from(lock_error: LockError) -> TantivyError {
TantivyError::LockFailure(lock_error, None)
TantivyError::FastFieldError(fastfield_error).into()
}
}
impl From<IOError> for TantivyError {
fn from(io_error: IOError) -> TantivyError {
TantivyError::IOError(io_error)
TantivyError::IOError(io_error).into()
}
}
impl From<io::Error> for TantivyError {
fn from(io_error: io::Error) -> TantivyError {
TantivyError::IOError(io_error.into())
TantivyError::IOError(io_error.into()).into()
}
}
impl From<query::QueryParserError> for TantivyError {
fn from(parsing_error: query::QueryParserError) -> TantivyError {
TantivyError::InvalidArgument(format!("Query is invalid. {:?}", parsing_error))
TantivyError::InvalidArgument(format!("Query is invalid. {:?}", parsing_error)).into()
}
}
impl<Guard> From<PoisonError<Guard>> for TantivyError {
fn from(_: PoisonError<Guard>) -> TantivyError {
TantivyError::Poisoned
TantivyError::Poisoned.into()
}
}
impl From<OpenReadError> for TantivyError {
fn from(error: OpenReadError) -> TantivyError {
match error {
OpenReadError::FileDoesNotExist(filepath) => TantivyError::PathDoesNotExist(filepath),
OpenReadError::IOError(io_error) => TantivyError::IOError(io_error),
OpenReadError::FileDoesNotExist(filepath) => {
TantivyError::PathDoesNotExist(filepath).into()
}
OpenReadError::IOError(io_error) => TantivyError::IOError(io_error).into(),
}
}
}
impl From<schema::DocParsingError> for TantivyError {
fn from(error: schema::DocParsingError) -> TantivyError {
TantivyError::InvalidArgument(format!("Failed to parse document {:?}", error))
TantivyError::InvalidArgument(format!("Failed to parse document {:?}", error)).into()
}
}
@@ -146,7 +98,7 @@ impl From<OpenWriteError> for TantivyError {
TantivyError::FileAlreadyExists(filepath)
}
OpenWriteError::IOError(io_error) => TantivyError::IOError(io_error),
}
}.into()
}
}
@@ -154,12 +106,11 @@ impl From<OpenDirectoryError> for TantivyError {
fn from(error: OpenDirectoryError) -> TantivyError {
match error {
OpenDirectoryError::DoesNotExist(directory_path) => {
TantivyError::PathDoesNotExist(directory_path)
TantivyError::PathDoesNotExist(directory_path).into()
}
OpenDirectoryError::NotADirectory(directory_path) => {
TantivyError::InvalidArgument(format!("{:?} is not a directory", directory_path))
}
OpenDirectoryError::IoError(err) => TantivyError::IOError(IOError::from(err)),
OpenDirectoryError::NotADirectory(directory_path) => TantivyError::InvalidArgument(
format!("{:?} is not a directory", directory_path),
).into(),
}
}
}
@@ -167,6 +118,6 @@ impl From<OpenDirectoryError> for TantivyError {
impl From<serde_json::Error> for TantivyError {
fn from(error: serde_json::Error) -> TantivyError {
let io_err = io::Error::from(error);
TantivyError::IOError(io_err.into())
TantivyError::IOError(io_err.into()).into()
}
}

24
src/fastfield/bytes/mod.rs
View File

@@ -6,12 +6,12 @@ pub use self::writer::BytesFastFieldWriter;
#[cfg(test)]
mod tests {
use crate::schema::Schema;
use crate::Index;
use schema::SchemaBuilder;
use Index;
#[test]
fn test_bytes() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
let field = schema_builder.add_bytes_field("bytesfield");
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -22,15 +22,17 @@ mod tests {
index_writer.add_document(doc!(field=>vec![1u8, 3, 5, 7, 9]));
index_writer.add_document(doc!(field=>vec![0u8; 1000]));
assert!(index_writer.commit().is_ok());
let searcher = index.reader().unwrap().searcher();
let segment_reader = searcher.segment_reader(0);
let bytes_reader = segment_reader.fast_fields().bytes(field).unwrap();
assert_eq!(bytes_reader.get_bytes(0), &[0u8, 1, 2, 3]);
assert!(bytes_reader.get_bytes(1).is_empty());
assert_eq!(bytes_reader.get_bytes(2), &[255u8]);
assert_eq!(bytes_reader.get_bytes(3), &[1u8, 3, 5, 7, 9]);
index.load_searchers().unwrap();
let searcher = index.searcher();
let reader = searcher.segment_reader(0);
let bytes_reader = reader.bytes_fast_field_reader(field).unwrap();
assert_eq!(bytes_reader.get_val(0), &[0u8, 1, 2, 3]);
assert!(bytes_reader.get_val(1).is_empty());
assert_eq!(bytes_reader.get_val(2), &[255u8]);
assert_eq!(bytes_reader.get_val(3), &[1u8, 3, 5, 7, 9]);
let long = vec![0u8; 1000];
assert_eq!(bytes_reader.get_bytes(4), long.as_slice());
assert_eq!(bytes_reader.get_val(4), long.as_slice());
}
}

21
src/fastfield/bytes/reader.rs
View File

@@ -1,8 +1,8 @@
use owning_ref::OwningRef;
use crate::directory::ReadOnlySource;
use crate::fastfield::FastFieldReader;
use crate::DocId;
use directory::ReadOnlySource;
use fastfield::FastFieldReader;
use DocId;
/// Reader for byte array fast fields
///
@@ -14,7 +14,6 @@ use crate::DocId;
///
/// Reading the value for a document is done by reading the start index for it,
/// and the start index for the next document, and keeping the bytes in between.
#[derive(Clone)]
pub struct BytesFastFieldReader {
idx_reader: FastFieldReader<u64>,
values: OwningRef<ReadOnlySource, [u8]>,
@@ -29,20 +28,10 @@ impl BytesFastFieldReader {
BytesFastFieldReader { idx_reader, values }
}
fn range(&self, doc: DocId) -> (usize, usize) {
/// Returns the bytes associated to the given `doc`
pub fn get_val(&self, doc: DocId) -> &[u8] {
let start = self.idx_reader.get(doc) as usize;
let stop = self.idx_reader.get(doc + 1) as usize;
(start, stop)
}
/// Returns the bytes associated to the given `doc`
pub fn get_bytes(&self, doc: DocId) -> &[u8] {
let (start, stop) = self.range(doc);
&self.values[start..stop]
}
/// Returns the overall number of bytes in this bytes fast field.
pub fn total_num_bytes(&self) -> usize {
self.values.len()
}
}

8
src/fastfield/bytes/writer.rs
View File

@@ -1,8 +1,8 @@
use std::io;
use crate::fastfield::serializer::FastFieldSerializer;
use crate::schema::{Document, Field, Value};
use crate::DocId;
use fastfield::serializer::FastFieldSerializer;
use schema::{Document, Field, Value};
use DocId;
/// Writer for byte array (as in, any number of bytes per document) fast fields
///
@@ -51,7 +51,7 @@ impl BytesFastFieldWriter {
self.next_doc();
for field_value in doc.field_values() {
if field_value.field() == self.field {
if let Value::Bytes(ref bytes) = *field_value.value() {
if let &Value::Bytes(ref bytes) = field_value.value() {
self.vals.extend_from_slice(bytes);
} else {
panic!(

39
src/fastfield/delete.rs
View File

@@ -1,11 +1,10 @@
use crate::common::HasLen;
use crate::directory::ReadOnlySource;
use crate::directory::WritePtr;
use crate::space_usage::ByteCount;
use crate::DocId;
use bit_set::BitSet;
use common::HasLen;
use directory::ReadOnlySource;
use directory::WritePtr;
use std::io;
use std::io::Write;
use DocId;
/// Write a delete `BitSet`
///
@@ -42,8 +41,7 @@ pub struct DeleteBitSet {
impl DeleteBitSet {
/// Opens a delete bitset given its data source.
pub fn open(data: ReadOnlySource) -> DeleteBitSet {
let num_deleted: usize = data
.as_slice()
let num_deleted: usize = data.as_slice()
.iter()
.map(|b| b.count_ones() as usize)
.sum();
@@ -53,23 +51,16 @@ impl DeleteBitSet {
}
}
/// Returns true iff the document is still "alive". In other words, if it has not been deleted.
pub fn is_alive(&self, doc: DocId) -> bool {
!self.is_deleted(doc)
}
/// Returns true iff the document has been marked as deleted.
#[inline(always)]
/// Returns whether the document has been marked as deleted.
pub fn is_deleted(&self, doc: DocId) -> bool {
let byte_offset = doc / 8u32;
let b: u8 = (*self.data)[byte_offset as usize];
let shift = (doc & 7u32) as u8;
b & (1u8 << shift) != 0
}
/// Summarize total space usage of this bitset.
pub fn space_usage(&self) -> ByteCount {
self.data.len()
if self.len == 0 {
false
} else {
let byte_offset = doc / 8u32;
let b: u8 = (*self.data)[byte_offset as usize];
let shift = (doc & 7u32) as u8;
b & (1u8 << shift) != 0
}
}
}
@@ -82,8 +73,8 @@ impl HasLen for DeleteBitSet {
#[cfg(test)]
mod tests {
use super::*;
use crate::directory::*;
use bit_set::BitSet;
use directory::*;
use std::path::PathBuf;
fn test_delete_bitset_helper(bitset: &BitSet) {

4
src/fastfield/error.rs
View File

@@ -1,11 +1,11 @@
use crate::schema::FieldEntry;
use schema::FieldEntry;
use std::result;
/// `FastFieldNotAvailableError` is returned when the
/// user requested for a fast field reader, and the field was not
/// defined in the schema as a fast field.
#[derive(Debug, Fail)]
#[fail(display = "Fast field not available: '{:?}'", field_name)]
#[fail(display = "field not available: '{:?}'", field_name)]
pub struct FastFieldNotAvailableError {
field_name: String,
}

25
src/fastfield/facet_reader.rs
View File

@@ -1,9 +1,8 @@
use super::MultiValueIntFastFieldReader;
use crate::schema::Facet;
use crate::termdict::TermDictionary;
use crate::termdict::TermOrdinal;
use crate::DocId;
use std::str;
use schema::Facet;
use termdict::TermDictionary;
use termdict::TermOrdinal;
use DocId;
/// The facet reader makes it possible to access the list of
/// facets associated to a given document in a specific
@@ -21,7 +20,6 @@ use std::str;
pub struct FacetReader {
term_ords: MultiValueIntFastFieldReader<u64>,
term_dict: TermDictionary,
buffer: Vec<u8>,
}
impl FacetReader {
@@ -39,7 +37,6 @@ impl FacetReader {
FacetReader {
term_ords,
term_dict,
buffer: vec![],
}
}
@@ -58,18 +55,10 @@ impl FacetReader {
}
/// Given a term ordinal returns the term associated to it.
pub fn facet_from_ord(
&mut self,
facet_ord: TermOrdinal,
output: &mut Facet,
) -> Result<(), str::Utf8Error> {
let found_term = self
.term_dict
.ord_to_term(facet_ord as u64, &mut self.buffer);
pub fn facet_from_ord(&self, facet_ord: TermOrdinal, output: &mut Facet) {
let found_term = self.term_dict
.ord_to_term(facet_ord as u64, output.inner_buffer_mut());
assert!(found_term, "Term ordinal {} no found.", facet_ord);
let facet_str = str::from_utf8(&self.buffer[..])?;
output.set_facet_str(facet_str);
Ok(())
}
/// Return the list of facet ordinals associated to a document.

99
src/fastfield/mod.rs
View File

@@ -30,13 +30,12 @@ pub use self::error::{FastFieldNotAvailableError, Result};
pub use self::facet_reader::FacetReader;
pub use self::multivalued::{MultiValueIntFastFieldReader, MultiValueIntFastFieldWriter};
pub use self::reader::FastFieldReader;
pub use self::readers::FastFieldReaders;
pub use self::serializer::FastFieldSerializer;
pub use self::writer::{FastFieldsWriter, IntFastFieldWriter};
use crate::common;
use crate::schema::Cardinality;
use crate::schema::FieldType;
use crate::schema::Value;
use common;
use schema::Cardinality;
use schema::FieldType;
use schema::Value;
mod bytes;
mod delete;
@@ -44,12 +43,11 @@ mod error;
mod facet_reader;
mod multivalued;
mod reader;
mod readers;
mod serializer;
mod writer;
/// Trait for types that are allowed for fast fields: (u64, i64 and f64).
pub trait FastValue: Default + Clone + Copy + Send + Sync + PartialOrd {
/// Trait for types that are allowed for fast fields: (u64 or i64).
pub trait FastValue: Default + Clone + Copy {
/// Converts a value from u64
///
/// Internally all fast field values are encoded as u64.
@@ -80,6 +78,10 @@ impl FastValue for u64 {
*self
}
fn as_u64(&self) -> u64 {
*self
}
fn fast_field_cardinality(field_type: &FieldType) -> Option<Cardinality> {
match *field_type {
FieldType::U64(ref integer_options) => integer_options.get_fastfield_cardinality(),
@@ -87,10 +89,6 @@ impl FastValue for u64 {
_ => None,
}
}
fn as_u64(&self) -> u64 {
*self
}
}
impl FastValue for i64 {
@@ -114,33 +112,11 @@ impl FastValue for i64 {
}
}
impl FastValue for f64 {
fn from_u64(val: u64) -> Self {
common::u64_to_f64(val)
}
fn to_u64(&self) -> u64 {
common::f64_to_u64(*self)
}
fn fast_field_cardinality(field_type: &FieldType) -> Option<Cardinality> {
match *field_type {
FieldType::F64(ref integer_options) => integer_options.get_fastfield_cardinality(),
_ => None,
}
}
fn as_u64(&self) -> u64 {
self.to_bits()
}
}
fn value_to_u64(value: &Value) -> u64 {
match *value {
Value::U64(ref val) => *val,
Value::I64(ref val) => common::i64_to_u64(*val),
Value::F64(ref val) => common::f64_to_u64(*val),
_ => panic!("Expected a u64/i64/f64 field, got {:?} ", value),
_ => panic!("Expected a u64/i64 field, got {:?} ", value),
}
}
@@ -148,27 +124,27 @@ fn value_to_u64(value: &Value) -> u64 {
mod tests {
use super::*;
use crate::common::CompositeFile;
use crate::directory::{Directory, RAMDirectory, WritePtr};
use crate::fastfield::FastFieldReader;
use crate::schema::Document;
use crate::schema::Field;
use crate::schema::Schema;
use crate::schema::FAST;
use once_cell::sync::Lazy;
use rand::prelude::SliceRandom;
use rand::rngs::StdRng;
use common::CompositeFile;
use directory::{Directory, RAMDirectory, WritePtr};
use fastfield::FastFieldReader;
use rand::Rng;
use rand::SeedableRng;
use rand::XorShiftRng;
use schema::Document;
use schema::Field;
use schema::FAST;
use schema::{Schema, SchemaBuilder};
use std::collections::HashMap;
use std::path::Path;
pub static SCHEMA: Lazy<Schema> = Lazy::new(|| {
let mut schema_builder = Schema::builder();
schema_builder.add_u64_field("field", FAST);
schema_builder.build()
});
pub static FIELD: Lazy<Field> = Lazy::new(|| SCHEMA.get_field("field").unwrap());
lazy_static! {
pub static ref SCHEMA: Schema = {
let mut schema_builder = SchemaBuilder::default();
schema_builder.add_u64_field("field", FAST);
schema_builder.build()
};
pub static ref FIELD: Field = { SCHEMA.get_field("field").unwrap() };
}
#[test]
pub fn test_fastfield() {
@@ -322,7 +298,7 @@ mod tests {
fn test_signed_intfastfield() {
let path = Path::new("test");
let mut directory: RAMDirectory = RAMDirectory::create();
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
let i64_field = schema_builder.add_i64_field("field", FAST);
let schema = schema_builder.build();
@@ -366,7 +342,7 @@ mod tests {
fn test_signed_intfastfield_default_val() {
let path = Path::new("test");
let mut directory: RAMDirectory = RAMDirectory::create();
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
let i64_field = schema_builder.add_i64_field("field", FAST);
let schema = schema_builder.build();
@@ -391,10 +367,11 @@ mod tests {
}
}
// Warning: this generates the same permutation at each call
pub fn generate_permutation() -> Vec<u64> {
let mut permutation: Vec<u64> = (0u64..100_000u64).collect();
permutation.shuffle(&mut StdRng::from_seed([1u8; 32]));
let seed: [u8; 16] = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
let mut rng = XorShiftRng::from_seed(seed);
let mut permutation: Vec<u64> = (0u64..1_000_000u64).collect();
rng.shuffle(&mut permutation);
permutation
}
@@ -429,6 +406,7 @@ mod tests {
}
}
}
}
#[cfg(all(test, feature = "unstable"))]
@@ -436,9 +414,9 @@ mod bench {
use super::tests::FIELD;
use super::tests::{generate_permutation, SCHEMA};
use super::*;
use crate::common::CompositeFile;
use crate::directory::{Directory, RAMDirectory, WritePtr};
use crate::fastfield::FastFieldReader;
use common::CompositeFile;
use directory::{Directory, RAMDirectory, WritePtr};
use fastfield::FastFieldReader;
use std::collections::HashMap;
use std::path::Path;
use test::{self, Bencher};
@@ -536,4 +514,5 @@ mod bench {
});
}
}
}

172
src/fastfield/multivalued/mod.rs
View File

@@ -7,20 +7,14 @@ pub use self::writer::MultiValueIntFastFieldWriter;
#[cfg(test)]
mod tests {
use time;
use self::time::Duration;
use crate::collector::TopDocs;
use crate::query::QueryParser;
use crate::schema::Cardinality;
use crate::schema::Facet;
use crate::schema::IntOptions;
use crate::schema::Schema;
use crate::Index;
use schema::Cardinality;
use schema::IntOptions;
use schema::SchemaBuilder;
use Index;
#[test]
fn test_multivalued_u64() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
let field = schema_builder.add_u64_field(
"multifield",
IntOptions::default().set_fast(Cardinality::MultiValues),
@@ -34,10 +28,11 @@ mod tests {
index_writer.add_document(doc!(field=>5u64, field=>20u64,field=>1u64));
assert!(index_writer.commit().is_ok());
let searcher = index.reader().unwrap().searcher();
let segment_reader = searcher.segment_reader(0);
index.load_searchers().unwrap();
let searcher = index.searcher();
let reader = searcher.segment_reader(0);
let mut vals = Vec::new();
let multi_value_reader = segment_reader.fast_fields().u64s(field).unwrap();
let multi_value_reader = reader.multi_fast_field_reader::<u64>(field).unwrap();
{
multi_value_reader.get_vals(2, &mut vals);
assert_eq!(&vals, &[4u64]);
@@ -52,136 +47,9 @@ mod tests {
}
}
#[test]
fn test_multivalued_date() {
let mut schema_builder = Schema::builder();
let date_field = schema_builder.add_date_field(
"multi_date_field",
IntOptions::default()
.set_fast(Cardinality::MultiValues)
.set_indexed()
.set_stored(),
);
let time_i =
schema_builder.add_i64_field("time_stamp_i", IntOptions::default().set_stored());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
let first_time_stamp = chrono::Utc::now();
index_writer.add_document(
doc!(date_field=>first_time_stamp, date_field=>first_time_stamp, time_i=>1i64),
);
index_writer.add_document(doc!(time_i=>0i64));
// add one second
index_writer
.add_document(doc!(date_field=>first_time_stamp + Duration::seconds(1), time_i=>2i64));
// add another second
let two_secs_ahead = first_time_stamp + Duration::seconds(2);
index_writer.add_document(doc!(date_field=>two_secs_ahead, date_field=>two_secs_ahead,date_field=>two_secs_ahead, time_i=>3i64));
assert!(index_writer.commit().is_ok());
let reader = index.reader().unwrap();
let searcher = reader.searcher();
let reader = searcher.segment_reader(0);
assert_eq!(reader.num_docs(), 4);
{
let parser = QueryParser::for_index(&index, vec![date_field]);
let query = parser
.parse_query(&format!("\"{}\"", first_time_stamp.to_rfc3339()).to_string())
.expect("could not parse query");
let results = searcher
.search(&query, &TopDocs::with_limit(5))
.expect("could not query index");
assert_eq!(results.len(), 1);
for (_score, doc_address) in results {
let retrieved_doc = searcher.doc(doc_address).expect("cannot fetch doc");
assert_eq!(
retrieved_doc
.get_first(date_field)
.expect("cannot find value")
.date_value()
.timestamp(),
first_time_stamp.timestamp()
);
assert_eq!(
retrieved_doc
.get_first(time_i)
.expect("cannot find value")
.i64_value(),
1i64
);
}
}
{
let parser = QueryParser::for_index(&index, vec![date_field]);
let query = parser
.parse_query(&format!("\"{}\"", two_secs_ahead.to_rfc3339()).to_string())
.expect("could not parse query");
let results = searcher
.search(&query, &TopDocs::with_limit(5))
.expect("could not query index");
assert_eq!(results.len(), 1);
for (_score, doc_address) in results {
let retrieved_doc = searcher.doc(doc_address).expect("cannot fetch doc");
assert_eq!(
retrieved_doc
.get_first(date_field)
.expect("cannot find value")
.date_value()
.timestamp(),
two_secs_ahead.timestamp()
);
assert_eq!(
retrieved_doc
.get_first(time_i)
.expect("cannot find value")
.i64_value(),
3i64
);
}
}
// TODO: support Date range queries
// {
// let parser = QueryParser::for_index(&index, vec![date_field]);
// let range_q = format!("\"{}\"..\"{}\"",
// (first_time_stamp + Duration::seconds(1)).to_rfc3339(),
// (first_time_stamp + Duration::seconds(3)).to_rfc3339()
// );
// let query = parser.parse_query(&range_q)
// .expect("could not parse query");
// let results = searcher.search(&query, &TopDocs::with_limit(5))
// .expect("could not query index");
//
//
// assert_eq!(results.len(), 2);
// for (i, doc_pair) in results.iter().enumerate() {
// let retrieved_doc = searcher.doc(doc_pair.1).expect("cannot fetch doc");
// let offset_sec = match i {
// 0 => 1,
// 1 => 3,
// _ => panic!("should not have more than 2 docs")
// };
// let time_i_val = match i {
// 0 => 2,
// 1 => 3,
// _ => panic!("should not have more than 2 docs")
// };
// assert_eq!(retrieved_doc.get_first(date_field).expect("cannot find value").date_value().timestamp(),
// (first_time_stamp + Duration::seconds(offset_sec)).timestamp());
// assert_eq!(retrieved_doc.get_first(time_i).expect("cannot find value").i64_value(), time_i_val);
// }
// }
}
#[test]
fn test_multivalued_i64() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
let field = schema_builder.add_i64_field(
"multifield",
IntOptions::default().set_fast(Cardinality::MultiValues),
@@ -195,10 +63,11 @@ mod tests {
index_writer.add_document(doc!(field=> -5i64, field => -20i64, field=>1i64));
assert!(index_writer.commit().is_ok());
let searcher = index.reader().unwrap().searcher();
let segment_reader = searcher.segment_reader(0);
index.load_searchers().unwrap();
let searcher = index.searcher();
let reader = searcher.segment_reader(0);
let mut vals = Vec::new();
let multi_value_reader = segment_reader.fast_fields().i64s(field).unwrap();
let multi_value_reader = reader.multi_fast_field_reader::<i64>(field).unwrap();
{
multi_value_reader.get_vals(2, &mut vals);
assert_eq!(&vals, &[-4i64]);
@@ -216,17 +85,4 @@ mod tests {
assert_eq!(&vals, &[-5i64, -20i64, 1i64]);
}
}
#[test]
#[ignore]
fn test_many_facets() {
let mut schema_builder = Schema::builder();
let field = schema_builder.add_facet_field("facetfield");
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
for i in 0..100_000 {
index_writer.add_document(doc!(field=> Facet::from(format!("/lang/{}", i).as_str())));
}
assert!(index_writer.commit().is_ok());
}
}

63
src/fastfield/multivalued/reader.rs
View File

@@ -1,5 +1,5 @@
use crate::fastfield::{FastFieldReader, FastValue};
use crate::DocId;
use fastfield::{FastFieldReader, FastValue};
use DocId;
/// Reader for a multivalued `u64` fast field.
///
@@ -26,13 +26,6 @@ impl<Item: FastValue> MultiValueIntFastFieldReader<Item> {
}
}
pub(crate) fn into_u64s_reader(self) -> MultiValueIntFastFieldReader<u64> {
MultiValueIntFastFieldReader {
idx_reader: self.idx_reader,
vals_reader: self.vals_reader.into_u64_reader(),
}
}
/// Returns `(start, stop)`, such that the values associated
/// to the given document are `start..stop`.
fn range(&self, doc: DocId) -> (u64, u64) {
@@ -46,63 +39,63 @@ impl<Item: FastValue> MultiValueIntFastFieldReader<Item> {
let (start, stop) = self.range(doc);
let len = (stop - start) as usize;
vals.resize(len, Item::default());
self.vals_reader.get_range_u64(start, &mut vals[..]);
}
/// Returns the number of values associated with the document `DocId`.
pub fn num_vals(&self, doc: DocId) -> usize {
let (start, stop) = self.range(doc);
(stop - start) as usize
}
/// Returns the overall number of values in this field .
pub fn total_num_vals(&self) -> u64 {
self.idx_reader.max_value()
self.vals_reader.get_range(start as u32, &mut vals[..]);
}
}
#[cfg(test)]
mod tests {
use crate::core::Index;
use crate::schema::{Facet, Schema};
use core::Index;
use schema::{Document, Facet, SchemaBuilder};
#[test]
fn test_multifastfield_reader() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::new();
let facet_field = schema_builder.add_facet_field("facets");
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index
.writer_with_num_threads(1, 30_000_000)
.expect("Failed to create index writer.");
index_writer.add_document(doc!(
facet_field => Facet::from("/category/cat2"),
facet_field => Facet::from("/category/cat1"),
));
index_writer.add_document(doc!(facet_field => Facet::from("/category/cat2")));
index_writer.add_document(doc!(facet_field => Facet::from("/category/cat3")));
{
let mut doc = Document::new();
doc.add_facet(facet_field, "/category/cat2");
doc.add_facet(facet_field, "/category/cat1");
index_writer.add_document(doc);
}
{
let mut doc = Document::new();
doc.add_facet(facet_field, "/category/cat2");
index_writer.add_document(doc);
}
{
let mut doc = Document::new();
doc.add_facet(facet_field, "/category/cat3");
index_writer.add_document(doc);
}
index_writer.commit().expect("Commit failed");
let searcher = index.reader().unwrap().searcher();
index.load_searchers().expect("Reloading searchers");
let searcher = index.searcher();
let segment_reader = searcher.segment_reader(0);
let mut facet_reader = segment_reader.facet_reader(facet_field).unwrap();
let mut facet = Facet::root();
{
facet_reader.facet_from_ord(1, &mut facet).unwrap();
facet_reader.facet_from_ord(1, &mut facet);
assert_eq!(facet, Facet::from("/category"));
}
{
facet_reader.facet_from_ord(2, &mut facet).unwrap();
facet_reader.facet_from_ord(2, &mut facet);
assert_eq!(facet, Facet::from("/category/cat1"));
}
{
facet_reader.facet_from_ord(3, &mut facet).unwrap();
facet_reader.facet_from_ord(3, &mut facet);
assert_eq!(format!("{}", facet), "/category/cat2");
assert_eq!(facet, Facet::from("/category/cat2"));
}
{
facet_reader.facet_from_ord(4, &mut facet).unwrap();
facet_reader.facet_from_ord(4, &mut facet);
assert_eq!(facet, Facet::from("/category/cat3"));
}

26
src/fastfield/multivalued/writer.rs
View File

@@ -1,13 +1,13 @@
use crate::fastfield::serializer::FastSingleFieldSerializer;
use crate::fastfield::value_to_u64;
use crate::fastfield::FastFieldSerializer;
use crate::postings::UnorderedTermId;
use crate::schema::{Document, Field};
use crate::termdict::TermOrdinal;
use crate::DocId;
use fnv::FnvHashMap;
use fastfield::serializer::FastSingleFieldSerializer;
use fastfield::value_to_u64;
use fastfield::FastFieldSerializer;
use itertools::Itertools;
use postings::UnorderedTermId;
use schema::{Document, Field};
use std::collections::HashMap;
use std::io;
use termdict::TermOrdinal;
use DocId;
/// Writer for multi-valued (as in, more than one value per document)
/// int fast field.
@@ -32,7 +32,7 @@ use std::io;
/// term ids when the segment is getting serialized.
pub struct MultiValueIntFastFieldWriter {
field: Field,
vals: Vec<UnorderedTermId>,
vals: Vec<u64>,
doc_index: Vec<u64>,
is_facet: bool,
}
@@ -102,7 +102,7 @@ impl MultiValueIntFastFieldWriter {
pub fn serialize(
&self,
serializer: &mut FastFieldSerializer,
mapping_opt: Option<&FnvHashMap<UnorderedTermId, TermOrdinal>>,
mapping_opt: Option<&HashMap<UnorderedTermId, TermOrdinal>>,
) -> io::Result<()> {
{
// writing the offset index
@@ -116,7 +116,7 @@ impl MultiValueIntFastFieldWriter {
}
{
// writing the values themselves.
let mut value_serializer: FastSingleFieldSerializer<'_, _>;
let mut value_serializer: FastSingleFieldSerializer<_>;
match mapping_opt {
Some(mapping) => {
value_serializer = serializer.new_u64_fast_field_with_idx(
@@ -132,8 +132,7 @@ impl MultiValueIntFastFieldWriter {
);
let mut doc_vals: Vec<u64> = Vec::with_capacity(100);
for (start, stop) in self
.doc_index
for (start, stop) in self.doc_index
.windows(2)
.map(|interval| (interval[0], interval[1]))
.chain(Some(last_interval).into_iter())
@@ -149,6 +148,7 @@ impl MultiValueIntFastFieldWriter {
value_serializer.add_val(val)?;
}
}
}
None => {
let val_min_max = self.vals.iter().cloned().minmax();

67
src/fastfield/reader.rs
View File

@@ -1,18 +1,19 @@
use super::FastValue;
use crate::common::bitpacker::BitUnpacker;
use crate::common::compute_num_bits;
use crate::common::BinarySerializable;
use crate::common::CompositeFile;
use crate::directory::ReadOnlySource;
use crate::directory::{Directory, RAMDirectory, WritePtr};
use crate::fastfield::{FastFieldSerializer, FastFieldsWriter};
use crate::schema::Schema;
use crate::schema::FAST;
use crate::DocId;
use common::bitpacker::BitUnpacker;
use common::compute_num_bits;
use common::BinarySerializable;
use common::CompositeFile;
use directory::ReadOnlySource;
use directory::{Directory, RAMDirectory, WritePtr};
use fastfield::{FastFieldSerializer, FastFieldsWriter};
use owning_ref::OwningRef;
use schema::SchemaBuilder;
use schema::FAST;
use std::collections::HashMap;
use std::marker::PhantomData;
use std::mem;
use std::path::Path;
use DocId;
/// Trait for accessing a fastfield.
///
@@ -50,15 +51,6 @@ impl<Item: FastValue> FastFieldReader<Item> {
}
}
pub(crate) fn into_u64_reader(self) -> FastFieldReader<u64> {
FastFieldReader {
bit_unpacker: self.bit_unpacker,
min_value_u64: self.min_value_u64,
max_value_u64: self.max_value_u64,
_phantom: PhantomData,
}
}
/// Return the value associated to the given document.
///
/// This accessor should return as fast as possible.
@@ -68,29 +60,7 @@ impl<Item: FastValue> FastFieldReader<Item> {
/// May panic if `doc` is greater than the segment
// `maxdoc`.
pub fn get(&self, doc: DocId) -> Item {
self.get_u64(u64::from(doc))
}
pub(crate) fn get_u64(&self, doc: u64) -> Item {
Item::from_u64(self.min_value_u64 + self.bit_unpacker.get(doc))
}
/// Internally `multivalued` also use SingleValue Fast fields.
/// It works as follows... A first column contains the list of start index
/// for each document, a second column contains the actual values.
///
/// The values associated to a given doc, are then
/// `second_column[first_column.get(doc)..first_column.get(doc+1)]`.
///
/// Which means single value fast field reader can be indexed internally with
/// something different from a `DocId`. For this use case, we want to use `u64`
/// values.
///
/// See `get_range` for an actual documentation about this method.
pub(crate) fn get_range_u64(&self, start: u64, output: &mut [Item]) {
for (i, out) in output.iter_mut().enumerate() {
*out = self.get_u64(start + (i as u64));
}
Item::from_u64(self.min_value_u64 + self.bit_unpacker.get(doc as usize))
}
/// Fills an output buffer with the fast field values
@@ -106,8 +76,15 @@ impl<Item: FastValue> FastFieldReader<Item> {
///
/// May panic if `start + output.len()` is greater than
/// the segment's `maxdoc`.
pub fn get_range(&self, start: DocId, output: &mut [Item]) {
self.get_range_u64(u64::from(start), output);
///
// TODO change start to `u64`.
// For multifastfield, start is an index in a second fastfield, not a `DocId`
pub fn get_range(&self, start: u32, output: &mut [Item]) {
let output_u64: &mut [u64] = unsafe { mem::transmute(output) }; // ok: Item is either `u64` or `i64`
self.bit_unpacker.get_range(start, output_u64);
for out in output_u64.iter_mut() {
*out = Item::from_u64(*out + self.min_value_u64).as_u64();
}
}
/// Returns the minimum value for this fast field.
@@ -131,7 +108,7 @@ impl<Item: FastValue> FastFieldReader<Item> {
impl<Item: FastValue> From<Vec<Item>> for FastFieldReader<Item> {
fn from(vals: Vec<Item>) -> FastFieldReader<Item> {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
let field = schema_builder.add_u64_field("field", FAST);
let schema = schema_builder.build();
let path = Path::new("__dummy__");

229
src/fastfield/readers.rs
View File

@@ -1,229 +0,0 @@
use crate::common::CompositeFile;
use crate::fastfield::BytesFastFieldReader;
use crate::fastfield::MultiValueIntFastFieldReader;
use crate::fastfield::{FastFieldNotAvailableError, FastFieldReader};
use crate::schema::{Cardinality, Field, FieldType, Schema};
use crate::space_usage::PerFieldSpaceUsage;
use crate::Result;
use std::collections::HashMap;
/// Provides access to all of the FastFieldReader.
///
/// Internally, `FastFieldReaders` have preloaded fast field readers,
/// and just wraps several `HashMap`.
pub struct FastFieldReaders {
fast_field_i64: HashMap<Field, FastFieldReader<i64>>,
fast_field_u64: HashMap<Field, FastFieldReader<u64>>,
fast_field_f64: HashMap<Field, FastFieldReader<f64>>,
fast_field_i64s: HashMap<Field, MultiValueIntFastFieldReader<i64>>,
fast_field_u64s: HashMap<Field, MultiValueIntFastFieldReader<u64>>,
fast_field_f64s: HashMap<Field, MultiValueIntFastFieldReader<f64>>,
fast_bytes: HashMap<Field, BytesFastFieldReader>,
fast_fields_composite: CompositeFile,
}
enum FastType {
I64,
U64,
F64,
}
fn type_and_cardinality(field_type: &FieldType) -> Option<(FastType, Cardinality)> {
match field_type {
FieldType::U64(options) => options
.get_fastfield_cardinality()
.map(|cardinality| (FastType::U64, cardinality)),
FieldType::I64(options) => options
.get_fastfield_cardinality()
.map(|cardinality| (FastType::I64, cardinality)),
FieldType::F64(options) => options
.get_fastfield_cardinality()
.map(|cardinality| (FastType::F64, cardinality)),
FieldType::HierarchicalFacet => Some((FastType::U64, Cardinality::MultiValues)),
_ => None,
}
}
impl FastFieldReaders {
pub(crate) fn load_all(
schema: &Schema,
fast_fields_composite: &CompositeFile,
) -> Result<FastFieldReaders> {
let mut fast_field_readers = FastFieldReaders {
fast_field_i64: Default::default(),
fast_field_u64: Default::default(),
fast_field_f64: Default::default(),
fast_field_i64s: Default::default(),
fast_field_u64s: Default::default(),
fast_field_f64s: Default::default(),
fast_bytes: Default::default(),
fast_fields_composite: fast_fields_composite.clone(),
};
for (field_id, field_entry) in schema.fields().iter().enumerate() {
let field = Field(field_id as u32);
let field_type = field_entry.field_type();
if field_type == &FieldType::Bytes {
let idx_reader = fast_fields_composite
.open_read_with_idx(field, 0)
.ok_or_else(|| FastFieldNotAvailableError::new(field_entry))
.map(FastFieldReader::open)?;
let data = fast_fields_composite
.open_read_with_idx(field, 1)
.ok_or_else(|| FastFieldNotAvailableError::new(field_entry))?;
fast_field_readers
.fast_bytes
.insert(field, BytesFastFieldReader::open(idx_reader, data));
} else if let Some((fast_type, cardinality)) = type_and_cardinality(field_type) {
match cardinality {
Cardinality::SingleValue => {
if let Some(fast_field_data) = fast_fields_composite.open_read(field) {
match fast_type {
FastType::U64 => {
let fast_field_reader = FastFieldReader::open(fast_field_data);
fast_field_readers
.fast_field_u64
.insert(field, fast_field_reader);
}
FastType::I64 => {
fast_field_readers.fast_field_i64.insert(
field,
FastFieldReader::open(fast_field_data.clone()),
);
}
FastType::F64 => {
fast_field_readers.fast_field_f64.insert(
field,
FastFieldReader::open(fast_field_data.clone()),
);
}
}
} else {
return Err(From::from(FastFieldNotAvailableError::new(field_entry)));
}
}
Cardinality::MultiValues => {
let idx_opt = fast_fields_composite.open_read_with_idx(field, 0);
let data_opt = fast_fields_composite.open_read_with_idx(field, 1);
if let (Some(fast_field_idx), Some(fast_field_data)) = (idx_opt, data_opt) {
let idx_reader = FastFieldReader::open(fast_field_idx);
match fast_type {
FastType::I64 => {
let vals_reader = FastFieldReader::open(fast_field_data);
let multivalued_int_fast_field =
MultiValueIntFastFieldReader::open(idx_reader, vals_reader);
fast_field_readers
.fast_field_i64s
.insert(field, multivalued_int_fast_field);
}
FastType::U64 => {
let vals_reader = FastFieldReader::open(fast_field_data);
let multivalued_int_fast_field =
MultiValueIntFastFieldReader::open(idx_reader, vals_reader);
fast_field_readers
.fast_field_u64s
.insert(field, multivalued_int_fast_field);
}
FastType::F64 => {
let vals_reader = FastFieldReader::open(fast_field_data);
let multivalued_int_fast_field =
MultiValueIntFastFieldReader::open(idx_reader, vals_reader);
fast_field_readers
.fast_field_f64s
.insert(field, multivalued_int_fast_field);
}
}
} else {
return Err(From::from(FastFieldNotAvailableError::new(field_entry)));
}
}
}
}
}
Ok(fast_field_readers)
}
pub(crate) fn space_usage(&self) -> PerFieldSpaceUsage {
self.fast_fields_composite.space_usage()
}
/// Returns the `u64` fast field reader reader associated to `field`.
///
/// If `field` is not a u64 fast field, this method returns `None`.
pub fn u64(&self, field: Field) -> Option<FastFieldReader<u64>> {
self.fast_field_u64.get(&field).cloned()
}
/// If the field is a u64-fast field return the associated reader.
/// If the field is a i64-fast field, return the associated u64 reader. Values are
/// mapped from i64 to u64 using a (well the, it is unique) monotonic mapping. ///
///
///TODO should it also be lenient with f64?
///
/// This method is useful when merging segment reader.
pub(crate) fn u64_lenient(&self, field: Field) -> Option<FastFieldReader<u64>> {
if let Some(u64_ff_reader) = self.u64(field) {
return Some(u64_ff_reader);
}
if let Some(i64_ff_reader) = self.i64(field) {
return Some(i64_ff_reader.into_u64_reader());
}
None
}
/// Returns the `i64` fast field reader reader associated to `field`.
///
/// If `field` is not a i64 fast field, this method returns `None`.
pub fn i64(&self, field: Field) -> Option<FastFieldReader<i64>> {
self.fast_field_i64.get(&field).cloned()
}
/// Returns the `f64` fast field reader reader associated to `field`.
///
/// If `field` is not a f64 fast field, this method returns `None`.
pub fn f64(&self, field: Field) -> Option<FastFieldReader<f64>> {
self.fast_field_f64.get(&field).cloned()
}
/// Returns a `u64s` multi-valued fast field reader reader associated to `field`.
///
/// If `field` is not a u64 multi-valued fast field, this method returns `None`.
pub fn u64s(&self, field: Field) -> Option<MultiValueIntFastFieldReader<u64>> {
self.fast_field_u64s.get(&field).cloned()
}
/// If the field is a u64s-fast field return the associated reader.
/// If the field is a i64s-fast field, return the associated u64s reader. Values are
/// mapped from i64 to u64 using a (well the, it is unique) monotonic mapping.
///
/// This method is useful when merging segment reader.
pub(crate) fn u64s_lenient(&self, field: Field) -> Option<MultiValueIntFastFieldReader<u64>> {
if let Some(u64s_ff_reader) = self.u64s(field) {
return Some(u64s_ff_reader);
}
if let Some(i64s_ff_reader) = self.i64s(field) {
return Some(i64s_ff_reader.into_u64s_reader());
}
None
}
/// Returns a `i64s` multi-valued fast field reader reader associated to `field`.
///
/// If `field` is not a i64 multi-valued fast field, this method returns `None`.
pub fn i64s(&self, field: Field) -> Option<MultiValueIntFastFieldReader<i64>> {
self.fast_field_i64s.get(&field).cloned()
}
/// Returns a `f64s` multi-valued fast field reader reader associated to `field`.
///
/// If `field` is not a f64 multi-valued fast field, this method returns `None`.
pub fn f64s(&self, field: Field) -> Option<MultiValueIntFastFieldReader<f64>> {
self.fast_field_f64s.get(&field).cloned()
}
/// Returns the `bytes` fast field reader associated to `field`.
///
/// If `field` is not a bytes fast field, returns `None`.
pub fn bytes(&self, field: Field) -> Option<BytesFastFieldReader> {
self.fast_bytes.get(&field).cloned()
}
}

24
src/fastfield/serializer.rs
View File

@@ -1,10 +1,10 @@
use crate::common::bitpacker::BitPacker;
use crate::common::compute_num_bits;
use crate::common::BinarySerializable;
use crate::common::CompositeWrite;
use crate::common::CountingWriter;
use crate::directory::WritePtr;
use crate::schema::Field;
use common::bitpacker::BitPacker;
use common::compute_num_bits;
use common::BinarySerializable;
use common::CompositeWrite;
use common::CountingWriter;
use directory::WritePtr;
use schema::Field;
use std::io::{self, Write};
/// `FastFieldSerializer` is in charge of serializing
@@ -45,7 +45,7 @@ impl FastFieldSerializer {
field: Field,
min_value: u64,
max_value: u64,
) -> io::Result<FastSingleFieldSerializer<'_, CountingWriter<WritePtr>>> {
) -> io::Result<FastSingleFieldSerializer<CountingWriter<WritePtr>>> {
self.new_u64_fast_field_with_idx(field, min_value, max_value, 0)
}
@@ -56,7 +56,7 @@ impl FastFieldSerializer {
min_value: u64,
max_value: u64,
idx: usize,
) -> io::Result<FastSingleFieldSerializer<'_, CountingWriter<WritePtr>>> {
) -> io::Result<FastSingleFieldSerializer<CountingWriter<WritePtr>>> {
let field_write = self.composite_write.for_field_with_idx(field, idx);
FastSingleFieldSerializer::open(field_write, min_value, max_value)
}
@@ -66,7 +66,7 @@ impl FastFieldSerializer {
&mut self,
field: Field,
idx: usize,
) -> io::Result<FastBytesFieldSerializer<'_, CountingWriter<WritePtr>>> {
) -> io::Result<FastBytesFieldSerializer<CountingWriter<WritePtr>>> {
let field_write = self.composite_write.for_field_with_idx(field, idx);
FastBytesFieldSerializer::open(field_write)
}
@@ -79,7 +79,7 @@ impl FastFieldSerializer {
}
}
pub struct FastSingleFieldSerializer<'a, W: Write> {
pub struct FastSingleFieldSerializer<'a, W: Write + 'a> {
bit_packer: BitPacker,
write: &'a mut W,
min_value: u64,
@@ -127,7 +127,7 @@ impl<'a, W: Write> FastSingleFieldSerializer<'a, W> {
}
}
pub struct FastBytesFieldSerializer<'a, W: Write> {
pub struct FastBytesFieldSerializer<'a, W: Write + 'a> {
write: &'a mut W,
}

39
src/fastfield/writer.rs
View File

@@ -1,14 +1,13 @@
use super::multivalued::MultiValueIntFastFieldWriter;
use crate::common;
use crate::common::BinarySerializable;
use crate::common::VInt;
use crate::fastfield::{BytesFastFieldWriter, FastFieldSerializer};
use crate::postings::UnorderedTermId;
use crate::schema::{Cardinality, Document, Field, FieldType, Schema};
use crate::termdict::TermOrdinal;
use fnv::FnvHashMap;
use common;
use common::BinarySerializable;
use common::VInt;
use fastfield::{BytesFastFieldWriter, FastFieldSerializer};
use postings::UnorderedTermId;
use schema::{Cardinality, Document, Field, FieldType, Schema};
use std::collections::HashMap;
use std::io;
use termdict::TermOrdinal;
/// The fastfieldswriter regroup all of the fast field writers.
pub struct FastFieldsWriter {
@@ -26,15 +25,13 @@ impl FastFieldsWriter {
for (field_id, field_entry) in schema.fields().iter().enumerate() {
let field = Field(field_id as u32);
let default_value = match *field_entry.field_type() {
FieldType::I64(_) => common::i64_to_u64(0i64),
FieldType::F64(_) => common::f64_to_u64(0.0f64),
_ => 0u64,
let default_value = if let FieldType::I64(_) = *field_entry.field_type() {
common::i64_to_u64(0i64)
} else {
0u64
};
match *field_entry.field_type() {
FieldType::I64(ref int_options)
| FieldType::U64(ref int_options)
| FieldType::F64(ref int_options) => {
FieldType::I64(ref int_options) | FieldType::U64(ref int_options) => {
match int_options.get_fastfield_cardinality() {
Some(Cardinality::SingleValue) => {
let mut fast_field_writer = IntFastFieldWriter::new(field);
@@ -117,7 +114,7 @@ impl FastFieldsWriter {
pub fn serialize(
&self,
serializer: &mut FastFieldSerializer,
mapping: &HashMap<Field, FnvHashMap<UnorderedTermId, TermOrdinal>>,
mapping: &HashMap<Field, HashMap<UnorderedTermId, TermOrdinal>>,
) -> io::Result<()> {
for field_writer in &self.single_value_writers {
field_writer.serialize(serializer)?;
@@ -145,9 +142,9 @@ impl FastFieldsWriter {
/// bitpacked and the number of bits required for bitpacking
/// can only been known once we have seen all of the values.
///
/// Both u64, i64 and f64 use the same writer.
/// i64 and f64 are just remapped to the `0..2^64 - 1`
/// using `common::i64_to_u64` and `common::f64_to_u64`.
/// Both u64, and i64 use the same writer.
/// i64 are just remapped to the `0..2^64 - 1`
/// using `common::i64_to_u64`.
pub struct IntFastFieldWriter {
field: Field,
vals: Vec<u8>,
@@ -206,8 +203,8 @@ impl IntFastFieldWriter {
/// Extract the value associated to the fast field for
/// this document.
///
/// i64 and f64 are remapped to u64 using the logic
/// in `common::i64_to_u64` and `common::f64_to_u64`.
/// i64 are remapped to u64 using the logic
/// in `common::i64_to_u64`.
///
/// If the value is missing, then the default value is used
/// instead.

276
src/fieldnorm/code.rs
View File

@@ -11,262 +11,26 @@ pub fn fieldnorm_to_id(fieldnorm: u32) -> u8 {
}
pub const FIELD_NORMS_TABLE: [u32; 256] = [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28,
29,
30,
31,
32,
33,
34,
35,
36,
37,
38,
39,
40,
42,
44,
46,
48,
50,
52,
54,
56,
60,
64,
68,
72,
76,
80,
84,
88,
96,
104,
112,
120,
128,
136,
144,
152,
168,
184,
200,
216,
232,
248,
264,
280,
312,
344,
376,
408,
440,
472,
504,
536,
600,
664,
728,
792,
856,
920,
984,
1_048,
1_176,
1_304,
1_432,
1_560,
1_688,
1_816,
1_944,
2_072,
2_328,
2_584,
2_840,
3_096,
3_352,
3_608,
3_864,
4_120,
4_632,
5_144,
5_656,
6_168,
6_680,
7_192,
7_704,
8_216,
9_240,
10_264,
11_288,
12_312,
13_336,
14_360,
15_384,
16_408,
18_456,
20_504,
22_552,
24_600,
26_648,
28_696,
30_744,
32_792,
36_888,
40_984,
45_080,
49_176,
53_272,
57_368,
61_464,
65_560,
73_752,
81_944,
90_136,
98_328,
106_520,
114_712,
122_904,
131_096,
147_480,
163_864,
180_248,
196_632,
213_016,
229_400,
245_784,
262_168,
294_936,
327_704,
360_472,
393_240,
426_008,
458_776,
491_544,
524_312,
589_848,
655_384,
720_920,
786_456,
851_992,
917_528,
983_064,
1_048_600,
1_179_672,
1_310_744,
1_441_816,
1_572_888,
1_703_960,
1_835_032,
1_966_104,
2_097_176,
2_359_320,
2_621_464,
2_883_608,
3_145_752,
3_407_896,
3_670_040,
3_932_184,
4_194_328,
4_718_616,
5_242_904,
5_767_192,
6_291_480,
6_815_768,
7_340_056,
7_864_344,
8_388_632,
9_437_208,
10_485_784,
11_534_360,
12_582_936,
13_631_512,
14_680_088,
15_728_664,
16_777_240,
18_874_392,
20_971_544,
23_068_696,
25_165_848,
27_263_000,
29_360_152,
31_457_304,
33_554_456,
37_748_760,
41_943_064,
46_137_368,
50_331_672,
54_525_976,
58_720_280,
62_914_584,
67_108_888,
75_497_496,
83_886_104,
92_274_712,
100_663_320,
109_051_928,
117_440_536,
125_829_144,
134_217_752,
150_994_968,
167_772_184,
184_549_400,
201_326_616,
218_103_832,
234_881_048,
251_658_264,
268_435_480,
301_989_912,
335_544_344,
369_098_776,
402_653_208,
436_207_640,
469_762_072,
503_316_504,
536_870_936,
603_979_800,
671_088_664,
738_197_528,
805_306_392,
872_415_256,
939_524_120,
1_006_632_984,
1_073_741_848,
1_207_959_576,
1_342_177_304,
1_476_395_032,
1_610_612_760,
1_744_830_488,
1_879_048_216,
2_013_265_944,
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,
26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 42, 44, 46, 48, 50, 52, 54, 56, 60,
64, 68, 72, 76, 80, 84, 88, 96, 104, 112, 120, 128, 136, 144, 152, 168, 184, 200, 216, 232,
248, 264, 280, 312, 344, 376, 408, 440, 472, 504, 536, 600, 664, 728, 792, 856, 920, 984, 1048,
1176, 1304, 1432, 1560, 1688, 1816, 1944, 2072, 2328, 2584, 2840, 3096, 3352, 3608, 3864, 4120,
4632, 5144, 5656, 6168, 6680, 7192, 7704, 8216, 9240, 10264, 11288, 12312, 13336, 14360, 15384,
16408, 18456, 20504, 22552, 24600, 26648, 28696, 30744, 32792, 36888, 40984, 45080, 49176,
53272, 57368, 61464, 65560, 73752, 81944, 90136, 98328, 106520, 114712, 122904, 131096, 147480,
163864, 180248, 196632, 213016, 229400, 245784, 262168, 294936, 327704, 360472, 393240, 426008,
458776, 491544, 524312, 589848, 655384, 720920, 786456, 851992, 917528, 983064, 1048600,
1179672, 1310744, 1441816, 1572888, 1703960, 1835032, 1966104, 2097176, 2359320, 2621464,
2883608, 3145752, 3407896, 3670040, 3932184, 4194328, 4718616, 5242904, 5767192, 6291480,
6815768, 7340056, 7864344, 8388632, 9437208, 10485784, 11534360, 12582936, 13631512, 14680088,
15728664, 16777240, 18874392, 20971544, 23068696, 25165848, 27263000, 29360152, 31457304,
33554456, 37748760, 41943064, 46137368, 50331672, 54525976, 58720280, 62914584, 67108888,
75497496, 83886104, 92274712, 100663320, 109051928, 117440536, 125829144, 134217752, 150994968,
167772184, 184549400, 201326616, 218103832, 234881048, 251658264, 268435480, 301989912,
335544344, 369098776, 402653208, 436207640, 469762072, 503316504, 536870936, 603979800,
671088664, 738197528, 805306392, 872415256, 939524120, 1006632984, 1073741848, 1207959576,
1342177304, 1476395032, 1610612760, 1744830488, 1879048216, 2013265944,
];
#[cfg(test)]

2
src/fieldnorm/mod.rs
View File

@@ -15,7 +15,7 @@
//! precompute computationally expensive functions of the fieldnorm
//! in a very short array.
//!
//! This trick is used by the BM25 similarity.
//! This trick is used by the [BM25 similarity]().
mod code;
mod reader;
mod serializer;

4
src/fieldnorm/reader.rs
View File

@@ -1,6 +1,6 @@
use super::{fieldnorm_to_id, id_to_fieldnorm};
use crate::directory::ReadOnlySource;
use crate::DocId;
use directory::ReadOnlySource;
use DocId;
/// Reads the fieldnorm associated to a document.
/// The fieldnorm represents the length associated to

6
src/fieldnorm/serializer.rs
View File

@@ -1,6 +1,6 @@
use crate::common::CompositeWrite;
use crate::directory::WritePtr;
use crate::schema::Field;
use common::CompositeWrite;
use directory::WritePtr;
use schema::Field;
use std::io;
use std::io::Write;

6
src/fieldnorm/writer.rs
View File

@@ -1,9 +1,9 @@
use crate::DocId;
use DocId;
use super::fieldnorm_to_id;
use super::FieldNormsSerializer;
use crate::schema::Field;
use crate::schema::Schema;
use schema::Field;
use schema::Schema;
use std::io;
/// The `FieldNormsWriter` is in charge of tracking the fieldnorm byte

22
src/functional_test.rs
View File

@@ -1,10 +1,11 @@
use rand::thread_rng;
use std::collections::HashSet;
use crate::schema::*;
use crate::Index;
use crate::Searcher;
use rand::Rng;
use rand::distributions::Range;
use schema::*;
use Index;
use Searcher;
fn check_index_content(searcher: &Searcher, vals: &HashSet<u64>) {
assert!(searcher.segment_readers().len() < 20);
@@ -13,16 +14,17 @@ fn check_index_content(searcher: &Searcher, vals: &HashSet<u64>) {
#[test]
#[ignore]
#[cfg(feature = "mmap")]
fn test_indexing() {
let mut schema_builder = Schema::builder();
let mut schema_builder = SchemaBuilder::default();
let id_field = schema_builder.add_u64_field("id", INDEXED);
let multiples_field = schema_builder.add_u64_field("multiples", INDEXED);
let id_field = schema_builder.add_u64_field("id", INT_INDEXED);
let multiples_field = schema_builder.add_u64_field("multiples", INT_INDEXED);
let schema = schema_builder.build();
let index = Index::create_from_tempdir(schema).unwrap();
let reader = index.reader().unwrap();
let universe = Range::new(0u64, 20u64);
let mut rng = thread_rng();
let mut index_writer = index.writer_with_num_threads(3, 120_000_000).unwrap();
@@ -31,13 +33,13 @@ fn test_indexing() {
let mut uncommitted_docs: HashSet<u64> = HashSet::new();
for _ in 0..200 {
let random_val = rng.gen_range(0, 20);
let random_val = rng.sample(&universe);
if random_val == 0 {
index_writer.commit().expect("Commit failed");
committed_docs.extend(&uncommitted_docs);
uncommitted_docs.clear();
reader.reload().unwrap();
let searcher = reader.searcher();
index.load_searchers().unwrap();
let searcher = index.searcher();
// check that everything is correct.
check_index_content(&searcher, &committed_docs);
} else {

37
src/indexer/delete_queue.rs
View File

@@ -1,5 +1,4 @@
use super::operation::DeleteOperation;
use crate::Opstamp;
use std::mem;
use std::ops::DerefMut;
use std::sync::{Arc, RwLock};
@@ -24,7 +23,7 @@ struct InnerDeleteQueue {
last_block: Option<Arc<Block>>,
}
#[derive(Clone)]
#[derive(Clone, Default)]
pub struct DeleteQueue {
inner: Arc<RwLock<InnerDeleteQueue>>,
}
@@ -37,7 +36,6 @@ impl DeleteQueue {
};
let next_block = NextBlock::from(delete_queue.clone());
{
let mut delete_queue_wlock = delete_queue.inner.write().unwrap();
delete_queue_wlock.last_block = Some(Arc::new(Block {
@@ -54,8 +52,7 @@ impl DeleteQueue {
//
// Past delete operations are not accessible.
pub fn cursor(&self) -> DeleteCursor {
let last_block = self
.inner
let last_block = self.inner
.read()
.expect("Read lock poisoned when opening delete queue cursor")
.last_block
@@ -95,8 +92,7 @@ impl DeleteQueue {
// be some unflushed operations.
//
fn flush(&self) -> Option<Arc<Block>> {
let mut self_wlock = self
.inner
let mut self_wlock = self.inner
.write()
.expect("Failed to acquire write lock on delete queue writer");
@@ -136,8 +132,7 @@ impl From<DeleteQueue> for NextBlock {
impl NextBlock {
fn next_block(&self) -> Option<Arc<Block>> {
{
let next_read_lock = self
.0
let next_read_lock = self.0
.read()
.expect("Failed to acquire write lock in delete queue");
if let InnerNextBlock::Closed(ref block) = *next_read_lock {
@@ -146,8 +141,7 @@ impl NextBlock {
}
let next_block;
{
let mut next_write_lock = self
.0
let mut next_write_lock = self.0
.write()
.expect("Failed to acquire write lock in delete queue");
match *next_write_lock {
@@ -186,20 +180,21 @@ impl DeleteCursor {
/// queue are consume and the next get will return None.
/// - the next get will return the first operation with an
/// `opstamp >= target_opstamp`.
pub fn skip_to(&mut self, target_opstamp: Opstamp) {
pub fn skip_to(&mut self, target_opstamp: u64) {
// TODO Can be optimize as we work with block.
while self.is_behind_opstamp(target_opstamp) {
#[cfg_attr(feature = "cargo-clippy", allow(while_let_loop))]
loop {
if let Some(operation) = self.get() {
if operation.opstamp >= target_opstamp {
break;
}
} else {
break;
}
self.advance();
}
}
#[cfg_attr(feature = "cargo-clippy", allow(clippy::wrong_self_convention))]
fn is_behind_opstamp(&mut self, target_opstamp: Opstamp) -> bool {
self.get()
.map(|operation| operation.opstamp < target_opstamp)
.unwrap_or(false)
}
/// If the current block has been entirely
/// consumed, try to load the next one.
///
@@ -251,7 +246,7 @@ impl DeleteCursor {
mod tests {
use super::{DeleteOperation, DeleteQueue};
use crate::schema::{Field, Term};
use schema::{Field, Term};
#[test]
fn test_deletequeue() {

27
src/indexer/directory_lock.rs Normal file
View File

@@ -0,0 +1,27 @@
use core::LOCKFILE_FILEPATH;
use directory::error::OpenWriteError;
use Directory;
/// The directory lock is a mechanism used to
/// prevent the creation of two [`IndexWriter`](struct.IndexWriter.html)
///
/// Only one lock can exist at a time for a given directory.
/// The lock is release automatically on `Drop`.
pub struct DirectoryLock {
directory: Box<Directory>,
}
impl DirectoryLock {
pub fn lock(mut directory: Box<Directory>) -> Result<DirectoryLock, OpenWriteError> {
directory.open_write(&*LOCKFILE_FILEPATH)?;
Ok(DirectoryLock { directory })
}
}
impl Drop for DirectoryLock {
fn drop(&mut self) {
if let Err(e) = self.directory.delete(&*LOCKFILE_FILEPATH) {
error!("Failed to remove the lock file. {:?}", e);
}
}
}

25
src/indexer/doc_opstamp_mapping.rs
View File

@@ -1,5 +1,5 @@
use crate::DocId;
use crate::Opstamp;
use std::sync::Arc;
use DocId;
// Doc to opstamp is used to identify which
// document should be deleted.
@@ -17,25 +17,25 @@ use crate::Opstamp;
// This mapping is (for the moment) stricly increasing
// because of the way document id are allocated.
#[derive(Clone)]
pub enum DocToOpstampMapping<'a> {
WithMap(&'a [Opstamp]),
pub enum DocToOpstampMapping {
WithMap(Arc<Vec<u64>>),
None,
}
impl<'a> From<&'a [u64]> for DocToOpstampMapping<'a> {
fn from(opstamps: &[Opstamp]) -> DocToOpstampMapping {
DocToOpstampMapping::WithMap(opstamps)
impl From<Vec<u64>> for DocToOpstampMapping {
fn from(opstamps: Vec<u64>) -> DocToOpstampMapping {
DocToOpstampMapping::WithMap(Arc::new(opstamps))
}
}
impl<'a> DocToOpstampMapping<'a> {
impl DocToOpstampMapping {
/// Given an opstamp return the limit doc id L
/// such that all doc id D such that
// D >= L iff opstamp(D) >= than `target_opstamp`.
//
// The edge case opstamp = some doc opstamp is in practise
// never called.
pub fn compute_doc_limit(&self, target_opstamp: Opstamp) -> DocId {
pub fn compute_doc_limit(&self, target_opstamp: u64) -> DocId {
match *self {
DocToOpstampMapping::WithMap(ref doc_opstamps) => {
match doc_opstamps.binary_search(&target_opstamp) {
@@ -64,18 +64,17 @@ mod tests {
#[test]
fn test_doc_to_opstamp_mapping_complex() {
{
let doc_to_opstamp_mapping = DocToOpstampMapping::from(&[][..]);
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec![]);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(0u64), 0);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(2u64), 0);
}
{
let doc_to_opstamp_mapping = DocToOpstampMapping::from(&[1u64][..]);
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec![1u64]);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(0u64), 0);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(2u64), 1);
}
{
let doc_to_opstamp_mapping =
DocToOpstampMapping::from(&[1u64, 12u64, 17u64, 23u64][..]);
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec![1u64, 12u64, 17u64, 23u64]);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(0u64), 0);
for i in 2u64..13u64 {
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(i), 1);

894
src/indexer/index_writer.rs
View File

File diff suppressed because it is too large Load Diff

13
src/indexer/log_merge_policy.rs
View File

@@ -1,5 +1,5 @@
use super::merge_policy::{MergeCandidate, MergePolicy};
use crate::core::SegmentMeta;
use core::SegmentMeta;
use std::cmp;
use std::f64;
@@ -52,7 +52,7 @@ impl MergePolicy for LogMergePolicy {
let mut size_sorted_tuples = segments
.iter()
.map(SegmentMeta::num_docs)
.map(|x| x.num_docs())
.enumerate()
.collect::<Vec<(usize, u32)>>();
@@ -95,11 +95,8 @@ impl Default for LogMergePolicy {
#[cfg(test)]
mod tests {
use super::*;
use crate::core::{SegmentId, SegmentMeta, SegmentMetaInventory};
use crate::indexer::merge_policy::MergePolicy;
use once_cell::sync::Lazy;
static INVENTORY: Lazy<SegmentMetaInventory> = Lazy::new(SegmentMetaInventory::default);
use core::{SegmentId, SegmentMeta};
use indexer::merge_policy::MergePolicy;
fn test_merge_policy() -> LogMergePolicy {
let mut log_merge_policy = LogMergePolicy::default();
@@ -116,7 +113,7 @@ mod tests {
}
fn create_random_segment_meta(num_docs: u32) -> SegmentMeta {
INVENTORY.new_segment_meta(SegmentId::generate_random(), num_docs)
SegmentMeta::new(SegmentId::generate_random(), num_docs)
}
#[test]

65
src/indexer/merge_operation.rs
View File

@@ -1,65 +0,0 @@
use crate::Opstamp;
use crate::SegmentId;
use census::{Inventory, TrackedObject};
use std::collections::HashSet;
#[derive(Default)]
pub struct MergeOperationInventory(Inventory<InnerMergeOperation>);
impl MergeOperationInventory {
pub fn segment_in_merge(&self) -> HashSet<SegmentId> {
let mut segment_in_merge = HashSet::default();
for merge_op in self.0.list() {
for &segment_id in &merge_op.segment_ids {
segment_in_merge.insert(segment_id);
}
}
segment_in_merge
}
}
/// A `MergeOperation` has two roles.
/// It carries all of the information required to describe a merge:
/// - `target_opstamp` is the opstamp up to which we want to consume the
/// delete queue and reflect their deletes.
/// - `segment_ids` is the list of segment to be merged.
///
/// The second role is to ensure keep track of the fact that these
/// segments are in merge and avoid starting a merge operation that
/// may conflict with this one.
///
/// This works by tracking merge operations. When considering computing
/// merge candidates, we simply list tracked merge operations and remove
/// their segments from possible merge candidates.
pub struct MergeOperation {
inner: TrackedObject<InnerMergeOperation>,
}
struct InnerMergeOperation {
target_opstamp: Opstamp,
segment_ids: Vec<SegmentId>,
}
impl MergeOperation {
pub fn new(
inventory: &MergeOperationInventory,
target_opstamp: Opstamp,
segment_ids: Vec<SegmentId>,
) -> MergeOperation {
let inner_merge_operation = InnerMergeOperation {
target_opstamp,
segment_ids,
};
MergeOperation {
inner: inventory.0.track(inner_merge_operation),
}
}
pub fn target_opstamp(&self) -> Opstamp {
self.inner.target_opstamp
}
pub fn segment_ids(&self) -> &[SegmentId] {
&self.inner.segment_ids[..]
}
}

Some files were not shown because too many files have changed in this diff Show More