rust/tantivy
1
0
Fork 0
mirror of https://github.com/quickwit-oss/tantivy.git synced 2026-01-05 00:32:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: rust:revert-1222-issue/1195
Branches Tags
rust:main rust:stuhood.lazy-scorers-blocks-upstream rust:paul.masurel/seek_into_the_danger_zone_result rust:paul.masurel/skip-first-doc-on-intersection rust:paul.masurel/search_into_danger_zone_fastfield rust:stuhood.lazy-scorers-blocks 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: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:fix_open_bytes_index
Branches Tags
rust:stuhood.lazy-scorers-blocks-upstream rust:paul.masurel/seek_into_the_danger_zone_result rust:paul.masurel/skip-first-doc-on-intersection rust:main rust:paul.masurel/search_into_danger_zone_fastfield rust:stuhood.lazy-scorers-blocks 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: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

33 Commits
revert-122 ... fix_open_b

Author SHA1 Message Date
Pascal Seitz
9c12860b01 extend proptest to cover bytes field codec bug 2022-02-18 10:50:46 +01:00
Pascal Seitz
886245ad21 Fix opening bytes index with dynamic codec 
Fix #1278
 2022-02-18 07:08:43 +01:00
Shikhar Bhushan
505e6a440c Remove test assertion sensitive to background segment merging (#1274) 2022-02-17 10:59:46 +09:00
Koichi Akabe
fcd651f6a9 Add Vaporetto tokenizer to README (#1271) 
* Add Vaporetto tokenizer to README

* Update README.md
 2022-02-14 18:19:57 +09:00
Paul Masurel
e6653228a9 Renamed github workflows (#1269) 2022-02-04 15:10:24 +09:00
Paul Masurel
bdedefe07d Adding an IndexingContext object (#1268) 2022-02-04 15:08:01 +09:00
Paul Masurel
13a4473faa Removing obsolete clippy allow thingy. 2022-02-01 11:54:01 +09:00
Paul Masurel
2069e3e52b Fixing clippy comments 2022-02-01 10:24:05 +09:00
Paul Masurel
0d8263cba1 Using nightly to format 2022-01-31 16:10:11 +09:00
Paul Masurel
65b365b81c Fixing all-features build. 2022-01-31 14:41:14 +09:00
dependabot[bot]
4c1366da87 Update fastdivide requirement from 0.3 to 0.4 (#1265) 
Updates the requirements on fastdivide to permit the latest version.

---
updated-dependencies:
- dependency-name: fastdivide
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2022-01-31 11:26:50 +09:00
Paul Masurel
eca6628b3c Minor refactoring (#1266) 2022-01-28 15:55:55 +09:00
Paul Masurel
9679c5f306 Rename quickwit-inc -> quickwit-oss 2022-01-27 15:37:09 +09:00
Shikhar Bhushan
5a2497b6fd Avoid exposing TrackedObject from Warmer API (#1264) 2022-01-25 10:04:08 +09:00
Shikhar Bhushan
99d4b1a177 Searcher Warming API (#1261) 
Adds an API to register Warmers in the IndexReader.

Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-01-20 23:40:25 +09:00
Paul Masurel
732f6847c0 Field type with codes (#1255) 
* Term are now typed.

This change is backward compatible:
While the Term has a byte representation that is modified, a Term itself
is a transient object that is not serialized as is in the index.

Its .field() and .value_bytes() on the other hand are unchanged.
This change offers better Debug information for terms.

While not necessary it also will help in the support for JSON types.

* Renamed Hierarchical Facet -> Facet
 2022-01-07 20:49:00 +09:00
Paul Masurel
1c6d9bdc6a Comparison of Value based on serialization. (#1250) 2022-01-07 20:31:26 +09:00
Paul Masurel
3ea6800ac5 Pleasing clippy (#1253) 2022-01-06 16:41:24 +09:00
Antoine G
395303b644 Collector + directory doc fixes (#1247) 
* doc(collector)

* doc(directory)

* doc(misc)

* wording
 2022-01-04 09:22:58 +09:00
Daniel Müller
2c200b46cb Use test-log instead of test-env-log (#1248) 
The test-env-log crate has been renamed to test-log to better reflect
its intent of not only catering to env_logger specific initialization
but also tracing (and potentially others in the future).
This change updates the crate to use test-log instead of the now
deprecated test-env-log.
 2022-01-04 09:20:30 +09:00
Liam Warfield
17e00df112 Change Snippet.fragments -> Snippet.fragment (#1243) 
* Change Snippet.fragments -> Snippet.fragment
* Apply suggestions from code review

Co-authored-by: Liam Warfield <lwarfield@arista.com>
 2022-01-03 22:23:51 +09:00
Antoine G
3129d86743 doc(termdict) expose structs (#1242) 
* doc(termdict) expose structs
also add merger doc + lint
refs #1232
 2022-01-03 22:20:31 +09:00
Shikhar Bhushan
e5e252cbc0 LogMergePolicy knob del_docs_percentage_before_merge (#1238) 
Add a knob to LogMergePolicy to always merge segments that exceed a threshold of deleted docs

Closes #115
 2021-12-20 13:14:56 +09:00
Paul Masurel
b2da82f151 Making MergeCandidate public in order to allow the usage of custom merge (#1237) 
policies.

Closes #1235
 2021-12-13 09:54:21 +09:00
Paul Masurel
c81b3030fa Issue/922b (#1233) 
* Add a NORMED options on field

Make fieldnorm indexation optional:

* for all types except text => added a NORMED options
* for text field
** if STRING, field has not fieldnorm retained
** if TEXT, field has fieldnorm computed

* Finalize making fieldnorm optional for all field types.

- Using Option for fieldnorm readers.
 2021-12-10 21:12:29 +09:00
Paul Masurel
9e66c75fc6 Using stable in CI as rustc nightly seems broken 2021-12-10 18:45:23 +09:00
Paul Masurel
ebdbb6bd2e Fixing compilation warnings & clippy comments. 2021-12-10 16:47:59 +09:00
Antoine G
c980b19dd9 canonicalize path when opening MmapDirectory (#1231) 
* canonicalize path when opening `MmapDirectory`
fixes #1229
 2021-12-09 10:19:52 +09:00
Paul Masurel
098eea843a Reducing the number of call to fsync on the directory. (#1228) 
This work by introducing a new API method in the Directory
trait. The user needs to explicitely call this method.
(In particular, once before a commmit)

Closes #1225
 2021-12-03 03:10:52 +00:00
Paul Masurel
466dc8233c Cargo fmt 2021-12-02 18:46:28 +09:00
Paul Masurel
03c2f6ece2 We are missing 4 bytes in the LZ4 compression buffer. (#1226) 
Closes #831
 2021-12-02 16:00:29 +09:00
Paul Masurel
1d4e9a29db Cargo fmt 2021-12-02 15:51:44 +09:00
Paul Masurel
f378d9a57b Pleasing clippy 2021-12-02 14:48:33 +09:00
226 changed files with 4748 additions and 3576 deletions
Expand all files Collapse all files

2
.github/workflows/long_running.yml vendored
View File

@@ -1,4 +1,4 @@
name: Rust
name: Long running tests
on:
push:

18
.github/workflows/test.yml vendored
View File

@@ -1,4 +1,4 @@
name: Rust
name: Unit tests
on:
push:
@@ -24,7 +24,19 @@ jobs:
toolchain: nightly
override: true
components: rustfmt
- name: Install latest nightly to test also against unstable feature flag
uses: actions-rs/toolchain@v1
with:
toolchain: stable
override: true
components: rustfmt, clippy
- name: Run tests
run: cargo test --all-features --verbose --workspace
run: cargo +stable test --features mmap,brotli-compression,lz4-compression,snappy-compression,failpoints --verbose --workspace
- name: Check Formatting
run: cargo fmt --all -- --check
run: cargo +nightly fmt --all -- --check
- uses: actions-rs/clippy-check@v1
with:
toolchain: stable
token: ${{ secrets.GITHUB_TOKEN }}
args: --tests

8
CHANGELOG.md
View File

@@ -1,6 +1,12 @@
Tantivy 0.17
================================
- LogMergePolicy now triggers merges if the ratio of deleted documents reaches a threshold (@shikhar @fulmicoton) [#115](https://github.com/quickwit-oss/tantivy/issues/115)
- Adds a searcher Warmer API (@shikhar @fulmicoton)
- Change to non-strict schema. Ignore fields in data which are not defined in schema. Previously this returned an error. #1211
- Facets are necessarily indexed. Existing index with indexed facets should work out of the box. Index without facets that are marked with index: false should be broken (but they were already broken in a sense). (@fulmicoton) #1195 .
- Bugfix that could in theory impact durability in theory on some filesystems [#1224](https://github.com/quickwit-oss/tantivy/issues/1224)
- Schema now offers not indexing fieldnorms (@lpouget) [#922](https://github.com/quickwit-oss/tantivy/issues/922)
- Reduce the number of fsync calls [#1225](https://github.com/quickwit-oss/tantivy/issues/1225)
Tantivy 0.16.2
================================
@@ -122,7 +128,7 @@ Tantivy 0.12.0
## How to update?
Crates relying on custom tokenizer, or registering tokenizer in the manager will require some
minor changes. Check https://github.com/quickwit-inc/tantivy/blob/main/examples/custom_tokenizer.rs
minor changes. Check https://github.com/quickwit-oss/tantivy/blob/main/examples/custom_tokenizer.rs
to check for some code sample.
Tantivy 0.11.3

12
Cargo.toml
View File

@@ -6,8 +6,8 @@ license = "MIT"
categories = ["database-implementations", "data-structures"]
description = """Search engine library"""
documentation = "https://docs.rs/tantivy/"
homepage = "https://github.com/quickwit-inc/tantivy"
repository = "https://github.com/quickwit-inc/tantivy"
homepage = "https://github.com/quickwit-oss/tantivy"
repository = "https://github.com/quickwit-oss/tantivy"
readme = "README.md"
keywords = ["search", "information", "retrieval"]
edition = "2018"
@@ -52,7 +52,7 @@ chrono = "0.4.19"
smallvec = "1.6.1"
rayon = "1.5"
lru = "0.7.0"
fastdivide = "0.3"
fastdivide = "0.4"
itertools = "0.10.0"
measure_time = "0.8.0"
@@ -65,7 +65,7 @@ maplit = "1.0.2"
matches = "0.1.8"
proptest = "1.0"
criterion = "0.3.5"
test-env-log = "0.2.7"
test-log = "0.2.8"
env_logger = "0.9.0"
[dev-dependencies.fail]
@@ -91,14 +91,10 @@ snappy-compression = ["snap"]
failpoints = ["fail/failpoints"]
unstable = [] # useful for benches.
wasm-bindgen = ["uuid/wasm-bindgen"]
[workspace]
members = ["query-grammar", "bitpacker", "common", "fastfield_codecs", "ownedbytes"]
[badges]
travis-ci = { repository = "tantivy-search/tantivy" }
# Following the "fail" crate best practises, we isolate
# tests that define specific behavior in fail check points
# in a different binary.

3
Makefile
View File

@@ -1,3 +1,6 @@
test:
echo "Run test only... No examples."
cargo test --tests --lib
fmt:
cargo +nightly fmt --all

25
README.md
View File

@@ -1,22 +1,13 @@
[![Docs](https://docs.rs/tantivy/badge.svg)](https://docs.rs/crate/tantivy/)
[![Build Status](https://github.com/quickwit-inc/tantivy/actions/workflows/test.yml/badge.svg)](https://github.com/quickwit-inc/tantivy/actions/workflows/test.yml)
[![codecov](https://codecov.io/gh/quickwit-inc/tantivy/branch/main/graph/badge.svg)](https://codecov.io/gh/quickwit-inc/tantivy)
[![Build Status](https://github.com/quickwit-oss/tantivy/actions/workflows/test.yml/badge.svg)](https://github.com/quickwit-oss/tantivy/actions/workflows/test.yml)
[![codecov](https://codecov.io/gh/quickwit-oss/tantivy/branch/main/graph/badge.svg)](https://codecov.io/gh/quickwit-oss/tantivy)
[![Join the chat at https://discord.gg/MT27AG5EVE](https://shields.io/discord/908281611840282624?label=chat%20on%20discord)](https://discord.gg/MT27AG5EVE)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
[![Crates.io](https://img.shields.io/crates/v/tantivy.svg)](https://crates.io/crates/tantivy)
![Tantivy](https://tantivy-search.github.io/logo/tantivy-logo.png)
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/0)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/0)
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/1)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/1)
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/2)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/2)
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/3)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/3)
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/4)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/4)
[![](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/images/5)](https://sourcerer.io/fame/fulmicoton/tantivy-search/tantivy/links/5)
[![](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)
**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) or [Apache Solr](https://lucene.apache.org/solr/) in the sense it is not
@@ -27,7 +18,7 @@ Tantivy is, in fact, strongly inspired by Lucene's design.
# Benchmark
The following [benchmark](https://tantivy-search.github.io/bench/) break downs
The following [benchmark](https://tantivy-search.github.io/bench/) break downs
performance for different type of queries / collection.
Your mileage WILL vary depending on the nature of queries and their load.
@@ -35,7 +26,7 @@ Your mileage WILL vary depending on the nature of queries and their load.
# Features
- Full-text search
- Configurable tokenizer (stemming available for 17 Latin languages with third party support for Chinese ([tantivy-jieba](https://crates.io/crates/tantivy-jieba) and [cang-jie](https://crates.io/crates/cang-jie)), Japanese ([lindera](https://github.com/lindera-morphology/lindera-tantivy) and [tantivy-tokenizer-tiny-segmenter](https://crates.io/crates/tantivy-tokenizer-tiny-segmenter)) and Korean ([lindera](https://github.com/lindera-morphology/lindera-tantivy) + [lindera-ko-dic-builder](https://github.com/lindera-morphology/lindera-ko-dic-builder))
- Configurable tokenizer (stemming available for 17 Latin languages with third party support for Chinese ([tantivy-jieba](https://crates.io/crates/tantivy-jieba) and [cang-jie](https://crates.io/crates/cang-jie)), Japanese ([lindera](https://github.com/lindera-morphology/lindera-tantivy), [Vaporetto](https://crates.io/crates/vaporetto_tantivy), and [tantivy-tokenizer-tiny-segmenter](https://crates.io/crates/tantivy-tokenizer-tiny-segmenter)) and Korean ([lindera](https://github.com/lindera-morphology/lindera-tantivy) + [lindera-ko-dic-builder](https://github.com/lindera-morphology/lindera-ko-dic-builder))
- 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)
@@ -57,7 +48,7 @@ Your mileage WILL vary depending on the nature of queries and their load.
## Non-features
- Distributed search is out of the scope of Tantivy. That being said, Tantivy is a
library upon which one could build a distributed search. Serializable/mergeable collector state for instance,
library upon which one could build a distributed search. Serializable/mergeable collector state for instance,
are within the scope of Tantivy.
@@ -66,14 +57,14 @@ are within the scope of Tantivy.
Tantivy works on stable Rust (>= 1.27) and supports Linux, MacOS, and Windows.
- [Tantivy's simple search example](https://tantivy-search.github.io/examples/basic_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,
- [tantivy-cli and its tutorial](https://github.com/quickwit-oss/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 walks 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/)
# How can I support this project?
There are many ways to support this project.
There are many ways to support this project.
- Use Tantivy and tell us about your experience on [Discord](https://discord.gg/MT27AG5EVE) or by email (paul.masurel@gmail.com)
- Report bugs
@@ -92,7 +83,7 @@ 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/quickwit-inc/tantivy.git
git clone https://github.com/quickwit-oss/tantivy.git
cd tantivy
cargo build
```

2
bitpacker/Cargo.toml
View File

@@ -6,7 +6,7 @@ authors = ["Paul Masurel <paul.masurel@gmail.com>"]
license = "MIT"
categories = []
description = """Tantivy-sub crate: bitpacking"""
repository = "https://github.com/quickwit-inc/tantivy"
repository = "https://github.com/quickwit-oss/tantivy"
keywords = []

3
bitpacker/src/bitpacker.rs
View File

@@ -1,4 +1,5 @@
use std::{convert::TryInto, io};
use std::convert::TryInto;
use std::io;
pub struct BitPacker {
mini_buffer: u64,

5
bitpacker/src/blocked_bitpacker.rs
View File

@@ -1,12 +1,11 @@
use super::bitpacker::BitPacker;
use super::compute_num_bits;
use crate::{minmax, BitUnpacker};
use super::{bitpacker::BitPacker, compute_num_bits};
const BLOCK_SIZE: usize = 128;
/// `BlockedBitpacker` compresses data in blocks of
/// 128 elements, while keeping an index on it
///
#[derive(Debug, Clone)]
pub struct BlockedBitpacker {
// bitpacked blocks

3
bitpacker/src/lib.rs
View File

@@ -1,8 +1,7 @@
mod bitpacker;
mod blocked_bitpacker;
pub use crate::bitpacker::BitPacker;
pub use crate::bitpacker::BitUnpacker;
pub use crate::bitpacker::{BitPacker, BitUnpacker};
pub use crate::blocked_bitpacker::BlockedBitpacker;
/// Computes the number of bits that will be used for bitpacking.

35
common/src/bitset.rs
View File

@@ -1,8 +1,8 @@
use ownedbytes::OwnedBytes;
use std::convert::TryInto;
use std::io::Write;
use std::u64;
use std::{fmt, io};
use std::{fmt, io, u64};
use ownedbytes::OwnedBytes;
#[derive(Clone, Copy, Eq, PartialEq)]
pub struct TinySet(u64);
@@ -62,29 +62,30 @@ impl TinySet {
self.0 = 0u64;
}
#[inline]
/// Returns the complement of the set in `[0, 64[`.
///
/// Careful on making this function public, as it will break the padding handling in the last
/// bucket.
#[inline]
fn complement(self) -> TinySet {
TinySet(!self.0)
}
#[inline]
/// Returns true iff the `TinySet` contains the element `el`.
#[inline]
pub fn contains(self, el: u32) -> bool {
!self.intersect(TinySet::singleton(el)).is_empty()
}
#[inline]
/// Returns the number of elements in the TinySet.
#[inline]
pub fn len(self) -> u32 {
self.0.count_ones()
}
#[inline]
/// Returns the intersection of `self` and `other`
#[inline]
#[must_use]
pub fn intersect(self, other: TinySet) -> TinySet {
TinySet(self.0 & other.0)
}
@@ -98,12 +99,14 @@ impl TinySet {
/// Insert a new element within [0..64)
#[inline]
#[must_use]
pub fn insert(self, el: u32) -> TinySet {
self.union(TinySet::singleton(el))
}
/// Removes an element within [0..64)
#[inline]
#[must_use]
pub fn remove(self, el: u32) -> TinySet {
self.intersect(TinySet::singleton(el).complement())
}
@@ -130,6 +133,7 @@ impl TinySet {
/// Returns the union of two tinysets
#[inline]
#[must_use]
pub fn union(self, other: TinySet) -> TinySet {
TinySet(self.0 | other.0)
}
@@ -183,7 +187,6 @@ fn num_buckets(max_val: u32) -> u32 {
impl BitSet {
/// serialize a `BitSet`.
///
pub fn serialize<T: Write>(&self, writer: &mut T) -> io::Result<()> {
writer.write_all(self.max_value.to_le_bytes().as_ref())?;
for tinyset in self.tinysets.iter().cloned() {
@@ -349,7 +352,6 @@ impl ReadOnlyBitSet {
}
/// Iterate the tinyset on the fly from serialized data.
///
#[inline]
fn iter_tinysets(&self) -> impl Iterator<Item = TinySet> + '_ {
self.data.chunks_exact(8).map(move |chunk| {
@@ -359,7 +361,6 @@ impl ReadOnlyBitSet {
}
/// Iterate over the positions of the elements.
///
#[inline]
pub fn iter(&self) -> impl Iterator<Item = u32> + '_ {
self.iter_tinysets()
@@ -411,14 +412,14 @@ impl<'a> From<&'a BitSet> for ReadOnlyBitSet {
#[cfg(test)]
mod tests {
use super::BitSet;
use super::ReadOnlyBitSet;
use super::TinySet;
use std::collections::HashSet;
use ownedbytes::OwnedBytes;
use rand::distributions::Bernoulli;
use rand::rngs::StdRng;
use rand::{Rng, SeedableRng};
use std::collections::HashSet;
use super::{BitSet, ReadOnlyBitSet, TinySet};
#[test]
fn test_read_serialized_bitset_full_multi() {
@@ -439,7 +440,7 @@ mod tests {
bitset.serialize(&mut out).unwrap();
let bitset = ReadOnlyBitSet::open(OwnedBytes::new(out));
assert_eq!(bitset.len() as usize, 64 as usize);
assert_eq!(bitset.len() as usize, 64);
}
#[test]
@@ -706,10 +707,10 @@ mod tests {
#[cfg(all(test, feature = "unstable"))]
mod bench {
use super::BitSet;
use super::TinySet;
use test;
use super::{BitSet, TinySet};
#[bench]
fn bench_tinyset_pop(b: &mut test::Bencher) {
b.iter(|| {

15
common/src/lib.rs
View File

@@ -104,11 +104,12 @@ pub fn u64_to_f64(val: u64) -> f64 {
#[cfg(test)]
pub mod test {
use super::{f64_to_u64, i64_to_u64, u64_to_f64, u64_to_i64};
use super::{BinarySerializable, FixedSize};
use proptest::prelude::*;
use std::f64;
use proptest::prelude::*;
use super::{f64_to_u64, i64_to_u64, u64_to_f64, u64_to_i64, BinarySerializable, FixedSize};
fn test_i64_converter_helper(val: i64) {
assert_eq!(u64_to_i64(i64_to_u64(val)), val);
}
@@ -157,10 +158,10 @@ pub mod test {
#[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
.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));

14
common/src/serialize.rs
View File

@@ -1,10 +1,9 @@
use crate::Endianness;
use crate::VInt;
use std::io::{Read, Write};
use std::{fmt, io};
use byteorder::{ReadBytesExt, WriteBytesExt};
use std::fmt;
use std::io;
use std::io::Read;
use std::io::Write;
use crate::{Endianness, VInt};
/// Trait for a simple binary serialization.
pub trait BinarySerializable: fmt::Debug + Sized {
@@ -202,8 +201,7 @@ impl BinarySerializable for String {
#[cfg(test)]
pub mod test {
use super::VInt;
use super::*;
use super::{VInt, *};
use crate::serialize::BinarySerializable;
pub fn fixed_size_test<O: BinarySerializable + FixedSize + Default>() {
let mut buffer = Vec::new();

13
common/src/vint.rs
View File

@@ -1,8 +1,9 @@
use super::BinarySerializable;
use byteorder::{ByteOrder, LittleEndian};
use std::io;
use std::io::Read;
use std::io::Write;
use std::io::{Read, Write};
use byteorder::{ByteOrder, LittleEndian};
use super::BinarySerializable;
/// Wrapper over a `u64` that serializes as a variable int.
#[derive(Clone, Copy, Debug, Eq, PartialEq)]
@@ -174,9 +175,7 @@ impl BinarySerializable for VInt {
#[cfg(test)]
mod tests {
use super::serialize_vint_u32;
use super::BinarySerializable;
use super::VInt;
use super::{serialize_vint_u32, BinarySerializable, VInt};
fn aux_test_vint(val: u64) {
let mut v = [14u8; 10];

10
common/src/writer.rs
View File

@@ -54,7 +54,8 @@ impl<W: TerminatingWrite> TerminatingWrite for CountingWriter<W> {
}
}
/// Struct used to prevent from calling [`terminate_ref`](trait.TerminatingWrite#method.terminate_ref) directly
/// Struct used to prevent from calling
/// [`terminate_ref`](trait.TerminatingWrite.html#tymethod.terminate_ref) directly
///
/// The point is that while the type is public, it cannot be built by anyone
/// outside of this module.
@@ -64,9 +65,7 @@ pub struct AntiCallToken(());
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,
{
where Self: Sized {
self.terminate_ref(AntiCallToken(()))
}
@@ -97,9 +96,10 @@ impl<'a> TerminatingWrite for &'a mut Vec<u8> {
#[cfg(test)]
mod test {
use super::CountingWriter;
use std::io::Write;
use super::CountingWriter;
#[test]
fn test_counting_writer() {
let buffer: Vec<u8> = vec![];

4
doc/src/index_sorting.md
View File

@@ -38,7 +38,7 @@ Note: Tantivy 0.16 does not do this optimization yet.
In principle there are many algorithms possible that exploit the monotonically increasing nature. (aggregations maybe?)
## Usage
The index sorting can be configured setting [`sort_by_field`](https://github.com/quickwit-inc/tantivy/blob/000d76b11a139a84b16b9b95060a1c93e8b9851c/src/core/index_meta.rs#L238) on `IndexSettings` and passing it to a `IndexBuilder`. As of tantvy 0.16 only fast fields are allowed to be used.
The index sorting can be configured setting [`sort_by_field`](https://github.com/quickwit-oss/tantivy/blob/000d76b11a139a84b16b9b95060a1c93e8b9851c/src/core/index_meta.rs#L238) on `IndexSettings` and passing it to a `IndexBuilder`. As of tantvy 0.16 only fast fields are allowed to be used.
```
let settings = IndexSettings {
@@ -55,7 +55,7 @@ let index = index_builder.create_in_ram().unwrap();
## Implementation details
Sorting an index is applied in the serialization step. In general there are two serialization steps: [Finishing a single segment](https://github.com/quickwit-inc/tantivy/blob/000d76b11a139a84b16b9b95060a1c93e8b9851c/src/indexer/segment_writer.rs#L338) and [merging multiple segments](https://github.com/quickwit-inc/tantivy/blob/000d76b11a139a84b16b9b95060a1c93e8b9851c/src/indexer/merger.rs#L1073).
Sorting an index is applied in the serialization step. In general there are two serialization steps: [Finishing a single segment](https://github.com/quickwit-oss/tantivy/blob/000d76b11a139a84b16b9b95060a1c93e8b9851c/src/indexer/segment_writer.rs#L338) and [merging multiple segments](https://github.com/quickwit-oss/tantivy/blob/000d76b11a139a84b16b9b95060a1c93e8b9851c/src/indexer/merger.rs#L1073).
In both cases we generate a docid mapping reflecting the sort. This mapping is used when serializing the different components (doc store, fastfields, posting list, normfield, facets).

6
examples/basic_search.rs
View File

@@ -73,7 +73,7 @@ fn main() -> tantivy::Result<()> {
// multithreaded.
//
// Here we give tantivy a budget of `50MB`.
// Using a bigger heap for the indexer may increase
// Using a bigger memory_arena for the indexer may increase
// throughput, but 50 MB is already plenty.
let mut index_writer = index.writer(50_000_000)?;
@@ -91,8 +91,8 @@ fn main() -> tantivy::Result<()> {
old_man_doc.add_text(title, "The Old Man and the Sea");
old_man_doc.add_text(
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.",
"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.",
);
// ... and add it to the `IndexWriter`.

3
examples/custom_collector.rs
View File

@@ -12,8 +12,7 @@
use tantivy::collector::{Collector, SegmentCollector};
use tantivy::fastfield::{DynamicFastFieldReader, FastFieldReader};
use tantivy::query::QueryParser;
use tantivy::schema::Field;
use tantivy::schema::{Schema, FAST, INDEXED, TEXT};
use tantivy::schema::{Field, Schema, FAST, INDEXED, TEXT};
use tantivy::{doc, Index, Score, SegmentReader};
#[derive(Default)]

2
examples/custom_tokenizer.rs
View File

@@ -62,7 +62,7 @@ fn main() -> tantivy::Result<()> {
// multithreaded.
//
// Here we use a buffer of 50MB per thread. Using a bigger
// heap for the indexer can increase its throughput.
// memory arena 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",

5
examples/deleting_updating_documents.rs
View File

@@ -56,8 +56,9 @@ fn main() -> tantivy::Result<()> {
// If it is `text`, let's make sure to keep it `raw` and let's avoid
// running any text processing on it.
// This is done by associating this field to the tokenizer named `raw`.
// Rather than building our [`TextOptions`](//docs.rs/tantivy/~0/tantivy/schema/struct.TextOptions.html) manually,
// We use the `STRING` shortcut. `STRING` stands for indexed (without term frequency or positions)
// Rather than building our
// [`TextOptions`](//docs.rs/tantivy/~0/tantivy/schema/struct.TextOptions.html) manually, We
// use the `STRING` shortcut. `STRING` stands for indexed (without term frequency or positions)
// and untokenized.
//
// Because we also want to be able to see this `id` in our returned documents,

2
examples/faceted_search.rs
View File

@@ -23,7 +23,7 @@ fn main() -> tantivy::Result<()> {
let name = schema_builder.add_text_field("felin_name", TEXT | STORED);
// this is our faceted field: its scientific classification
let classification = schema_builder.add_facet_field("classification", INDEXED);
let classification = schema_builder.add_facet_field("classification", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);

8
examples/faceted_search_with_tweaked_score.rs
View File

@@ -1,15 +1,15 @@
use std::collections::HashSet;
use tantivy::collector::TopDocs;
use tantivy::doc;
use tantivy::query::BooleanQuery;
use tantivy::schema::*;
use tantivy::{DocId, Index, Score, SegmentReader};
use tantivy::{doc, DocId, Index, Score, SegmentReader};
fn main() -> tantivy::Result<()> {
let mut schema_builder = Schema::builder();
let title = schema_builder.add_text_field("title", STORED);
let ingredient = schema_builder.add_facet_field("ingredient", INDEXED);
let ingredient = schema_builder.add_facet_field("ingredient", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -87,7 +87,7 @@ fn main() -> tantivy::Result<()> {
.unwrap()
.get_first(title)
.unwrap()
.text()
.as_text()
.unwrap()
.to_owned()
})

16
examples/iterating_docs_and_positions.rs
View File

@@ -52,11 +52,11 @@ fn main() -> tantivy::Result<()> {
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.
// 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 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)?
{
@@ -109,11 +109,11 @@ fn main() -> tantivy::Result<()> {
let inverted_index = segment_reader.inverted_index(title)?;
// 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.
// 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 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)?
{

4
examples/multiple_producer.rs
View File

@@ -28,6 +28,7 @@
use std::sync::{Arc, RwLock};
use std::thread;
use std::time::Duration;
use tantivy::schema::{Schema, STORED, TEXT};
use tantivy::{doc, Index, IndexWriter, Opstamp, TantivyError};
@@ -90,7 +91,8 @@ fn main() -> tantivy::Result<()> {
// # 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.
// 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()?
};

11
examples/snippet.rs
View File

@@ -57,7 +57,10 @@ fn main() -> tantivy::Result<()> {
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!(
"title: {}",
doc.get_first(title).unwrap().as_text().unwrap()
);
println!("snippet: {}", snippet.to_html());
println!("custom highlighting: {}", highlight(snippet));
}
@@ -70,13 +73,13 @@ fn highlight(snippet: Snippet) -> String {
let mut start_from = 0;
for fragment_range in snippet.highlighted() {
result.push_str(&snippet.fragments()[start_from..fragment_range.start]);
result.push_str(&snippet.fragment()[start_from..fragment_range.start]);
result.push_str(" --> ");
result.push_str(&snippet.fragments()[fragment_range.clone()]);
result.push_str(&snippet.fragment()[fragment_range.clone()]);
result.push_str(" <-- ");
start_from = fragment_range.end;
}
result.push_str(&snippet.fragments()[start_from..]);
result.push_str(&snippet.fragment()[start_from..]);
result
}

224
examples/warmer.rs Normal file
View File

@@ -0,0 +1,224 @@
use std::cmp::Reverse;
use std::collections::{HashMap, HashSet};
use std::sync::{Arc, RwLock, Weak};
use tantivy::collector::TopDocs;
use tantivy::fastfield::FastFieldReader;
use tantivy::query::QueryParser;
use tantivy::schema::{Field, Schema, FAST, TEXT};
use tantivy::{
doc, DocAddress, DocId, Index, IndexReader, Opstamp, Searcher, SearcherGeneration, SegmentId,
SegmentReader, Warmer,
};
// This example shows how warmers can be used to
// load a values from an external sources using the Warmer API.
//
// In this example, we assume an e-commerce search engine.
type ProductId = u64;
/// Price
type Price = u32;
pub trait PriceFetcher: Send + Sync + 'static {
fn fetch_prices(&self, product_ids: &[ProductId]) -> Vec<Price>;
}
struct DynamicPriceColumn {
field: Field,
price_cache: RwLock<HashMap<(SegmentId, Option<Opstamp>), Arc<Vec<Price>>>>,
price_fetcher: Box<dyn PriceFetcher>,
}
impl DynamicPriceColumn {
pub fn with_product_id_field<T: PriceFetcher>(field: Field, price_fetcher: T) -> Self {
DynamicPriceColumn {
field,
price_cache: Default::default(),
price_fetcher: Box::new(price_fetcher),
}
}
pub fn price_for_segment(&self, segment_reader: &SegmentReader) -> Option<Arc<Vec<Price>>> {
let segment_key = (segment_reader.segment_id(), segment_reader.delete_opstamp());
self.price_cache.read().unwrap().get(&segment_key).cloned()
}
}
impl Warmer for DynamicPriceColumn {
fn warm(&self, searcher: &Searcher) -> tantivy::Result<()> {
for segment in searcher.segment_readers() {
let key = (segment.segment_id(), segment.delete_opstamp());
let product_id_reader = segment.fast_fields().u64(self.field)?;
let product_ids: Vec<ProductId> = segment
.doc_ids_alive()
.map(|doc| product_id_reader.get(doc))
.collect();
let mut prices_it = self.price_fetcher.fetch_prices(&product_ids).into_iter();
let mut price_vals: Vec<Price> = Vec::new();
for doc in 0..segment.max_doc() {
if segment.is_deleted(doc) {
price_vals.push(0);
} else {
price_vals.push(prices_it.next().unwrap())
}
}
self.price_cache
.write()
.unwrap()
.insert(key, Arc::new(price_vals));
}
Ok(())
}
fn garbage_collect(&self, live_generations: &[&SearcherGeneration]) {
let live_segment_id_and_delete_ops: HashSet<(SegmentId, Option<Opstamp>)> =
live_generations
.iter()
.flat_map(|gen| gen.segments())
.map(|(&segment_id, &opstamp)| (segment_id, opstamp))
.collect();
let mut price_cache_wrt = self.price_cache.write().unwrap();
// let price_cache = std::mem::take(&mut *price_cache_wrt);
// Drain would be nicer here.
*price_cache_wrt = std::mem::take(&mut *price_cache_wrt)
.into_iter()
.filter(|(seg_id_and_op, _)| !live_segment_id_and_delete_ops.contains(seg_id_and_op))
.collect();
}
}
/// For the sake of this example, the table is just an editable HashMap behind a RwLock.
/// This map represents a map (ProductId -> Price)
///
/// In practise, it could be fetching things from an external service, like a SQL table.
#[derive(Default, Clone)]
pub struct ExternalPriceTable {
prices: Arc<RwLock<HashMap<ProductId, Price>>>,
}
impl ExternalPriceTable {
pub fn update_price(&self, product_id: ProductId, price: Price) {
let mut prices_wrt = self.prices.write().unwrap();
prices_wrt.insert(product_id, price);
}
}
impl PriceFetcher for ExternalPriceTable {
fn fetch_prices(&self, product_ids: &[ProductId]) -> Vec<Price> {
let prices_read = self.prices.read().unwrap();
product_ids
.iter()
.map(|product_id| prices_read.get(product_id).cloned().unwrap_or(0))
.collect()
}
}
fn main() -> tantivy::Result<()> {
// Declaring our schema.
let mut schema_builder = Schema::builder();
// The product id is assumed to be a primary id for our external price source.
let product_id = schema_builder.add_u64_field("product_id", FAST);
let text = schema_builder.add_text_field("text", TEXT);
let schema: Schema = schema_builder.build();
let price_table = ExternalPriceTable::default();
let price_dynamic_column = Arc::new(DynamicPriceColumn::with_product_id_field(
product_id,
price_table.clone(),
));
price_table.update_price(OLIVE_OIL, 12);
price_table.update_price(GLOVES, 13);
price_table.update_price(SNEAKERS, 80);
const OLIVE_OIL: ProductId = 323423;
const GLOVES: ProductId = 3966623;
const SNEAKERS: ProductId = 23222;
let index = Index::create_in_ram(schema);
let mut writer = index.writer_with_num_threads(1, 10_000_000)?;
writer.add_document(doc!(product_id=>OLIVE_OIL, text=>"cooking olive oil from greece"))?;
writer.add_document(doc!(product_id=>GLOVES, text=>"kitchen gloves, perfect for cooking"))?;
writer.add_document(doc!(product_id=>SNEAKERS, text=>"uber sweet sneakers"))?;
writer.commit()?;
let warmers: Vec<Weak<dyn Warmer>> = vec![Arc::downgrade(
&(price_dynamic_column.clone() as Arc<dyn Warmer>),
)];
let reader: IndexReader = index
.reader_builder()
.warmers(warmers)
.num_searchers(1)
.try_into()?;
reader.reload()?;
let query_parser = QueryParser::for_index(&index, vec![text]);
let query = query_parser.parse_query("cooking")?;
let searcher = reader.searcher();
let score_by_price = move |segment_reader: &SegmentReader| {
let price = price_dynamic_column
.price_for_segment(segment_reader)
.unwrap();
move |doc_id: DocId| Reverse(price[doc_id as usize])
};
let most_expensive_first = TopDocs::with_limit(10).custom_score(score_by_price);
let hits = searcher.search(&query, &most_expensive_first)?;
assert_eq!(
&hits,
&[
(
Reverse(12u32),
DocAddress {
segment_ord: 0,
doc_id: 0u32
}
),
(
Reverse(13u32),
DocAddress {
segment_ord: 0,
doc_id: 1u32
}
),
]
);
// Olive oil just got more expensive!
price_table.update_price(OLIVE_OIL, 15);
// The price update are directly reflected on `reload`.
//
// Be careful here though!...
// You may have spotted that we are still using the same `Searcher`.
//
// It is up to the `Warmer` implementer to decide how
// to control this behavior.
reader.reload()?;
let hits_with_new_prices = searcher.search(&query, &most_expensive_first)?;
assert_eq!(
&hits_with_new_prices,
&[
(
Reverse(13u32),
DocAddress {
segment_ord: 0,
doc_id: 1u32
}
),
(
Reverse(15u32),
DocAddress {
segment_ord: 0,
doc_id: 0u32
}
),
]
);
Ok(())
}

14
fastfield_codecs/benches/bench.rs
View File

@@ -4,14 +4,14 @@ extern crate test;
#[cfg(test)]
mod tests {
use fastfield_codecs::{
bitpacked::{BitpackedFastFieldReader, BitpackedFastFieldSerializer},
linearinterpol::{LinearInterpolFastFieldReader, LinearInterpolFastFieldSerializer},
multilinearinterpol::{
MultiLinearInterpolFastFieldReader, MultiLinearInterpolFastFieldSerializer,
},
*,
use fastfield_codecs::bitpacked::{BitpackedFastFieldReader, BitpackedFastFieldSerializer};
use fastfield_codecs::linearinterpol::{
LinearInterpolFastFieldReader, LinearInterpolFastFieldSerializer,
};
use fastfield_codecs::multilinearinterpol::{
MultiLinearInterpolFastFieldReader, MultiLinearInterpolFastFieldSerializer,
};
use fastfield_codecs::*;
fn get_data() -> Vec<u64> {
let mut data: Vec<_> = (100..55000_u64)

12
fastfield_codecs/src/bitpacked.rs
View File

@@ -1,13 +1,9 @@
use crate::FastFieldCodecReader;
use crate::FastFieldCodecSerializer;
use crate::FastFieldDataAccess;
use crate::FastFieldStats;
use common::BinarySerializable;
use std::io::{self, Write};
use tantivy_bitpacker::compute_num_bits;
use tantivy_bitpacker::BitPacker;
use tantivy_bitpacker::BitUnpacker;
use common::BinarySerializable;
use tantivy_bitpacker::{compute_num_bits, BitPacker, BitUnpacker};
use crate::{FastFieldCodecReader, FastFieldCodecSerializer, FastFieldDataAccess, FastFieldStats};
/// Depending on the field type, a different
/// fast field is required.

13
fastfield_codecs/src/lib.rs
View File

@@ -53,7 +53,8 @@ pub trait FastFieldCodecSerializer {
pub trait FastFieldDataAccess {
/// Return the value associated to the given position.
///
/// Whenever possible use the Iterator passed to the fastfield creation instead, for performance reasons.
/// Whenever possible use the Iterator passed to the fastfield creation instead, for performance
/// reasons.
///
/// # Panics
///
@@ -82,12 +83,10 @@ impl FastFieldDataAccess for Vec<u64> {
#[cfg(test)]
mod tests {
use crate::{
bitpacked::{BitpackedFastFieldReader, BitpackedFastFieldSerializer},
linearinterpol::{LinearInterpolFastFieldReader, LinearInterpolFastFieldSerializer},
multilinearinterpol::{
MultiLinearInterpolFastFieldReader, MultiLinearInterpolFastFieldSerializer,
},
use crate::bitpacked::{BitpackedFastFieldReader, BitpackedFastFieldSerializer};
use crate::linearinterpol::{LinearInterpolFastFieldReader, LinearInterpolFastFieldSerializer};
use crate::multilinearinterpol::{
MultiLinearInterpolFastFieldReader, MultiLinearInterpolFastFieldSerializer,
};
pub fn create_and_validate<S: FastFieldCodecSerializer, R: FastFieldCodecReader>(

21
fastfield_codecs/src/linearinterpol.rs
View File

@@ -1,15 +1,10 @@
use crate::FastFieldCodecReader;
use crate::FastFieldCodecSerializer;
use crate::FastFieldDataAccess;
use crate::FastFieldStats;
use std::io::{self, Read, Write};
use std::ops::Sub;
use tantivy_bitpacker::compute_num_bits;
use tantivy_bitpacker::BitPacker;
use common::BinarySerializable;
use common::FixedSize;
use tantivy_bitpacker::BitUnpacker;
use common::{BinarySerializable, FixedSize};
use tantivy_bitpacker::{compute_num_bits, BitPacker, BitUnpacker};
use crate::{FastFieldCodecReader, FastFieldCodecSerializer, FastFieldDataAccess, FastFieldStats};
/// Depending on the field type, a different
/// fast field is required.
@@ -137,7 +132,7 @@ impl FastFieldCodecSerializer for LinearInterpolFastFieldSerializer {
// will be offset to 0
offset = offset.max(calculated_value - actual_value);
} else {
//positive value no offset reuqired
// positive value no offset reuqired
rel_positive_max = rel_positive_max.max(actual_value - calculated_value);
}
}
@@ -171,7 +166,7 @@ impl FastFieldCodecSerializer for LinearInterpolFastFieldSerializer {
stats: FastFieldStats,
) -> bool {
if stats.num_vals < 3 {
return false; //disable compressor for this case
return false; // disable compressor for this case
}
// On serialisation the offset is added to the actual value.
// We need to make sure this won't run into overflow calculation issues.
@@ -211,8 +206,8 @@ impl FastFieldCodecSerializer for LinearInterpolFastFieldSerializer {
.max()
.unwrap_or(0);
// the theory would be that we don't have the actual max_distance, but we are close within 50%
// threshold.
// the theory would be that we don't have the actual max_distance, but we are close within
// 50% threshold.
// It is multiplied by 2 because in a log case scenario the line would be as much above as
// below. So the offset would = max_distance
//

15
fastfield_codecs/src/main.rs
View File

@@ -1,10 +1,8 @@
#[macro_use]
extern crate prettytable;
use fastfield_codecs::{
linearinterpol::LinearInterpolFastFieldSerializer,
multilinearinterpol::MultiLinearInterpolFastFieldSerializer, FastFieldCodecSerializer,
FastFieldStats,
};
use fastfield_codecs::linearinterpol::LinearInterpolFastFieldSerializer;
use fastfield_codecs::multilinearinterpol::MultiLinearInterpolFastFieldSerializer;
use fastfield_codecs::{FastFieldCodecSerializer, FastFieldStats};
use prettytable::{Cell, Row, Table};
fn main() {
@@ -24,7 +22,7 @@ fn main() {
);
results.push(res);
//let best_estimation_codec = results
// let best_estimation_codec = results
//.iter()
//.min_by(|res1, res2| res1.partial_cmp(&res2).unwrap())
//.unwrap();
@@ -41,7 +39,6 @@ fn main() {
} else {
(est.to_string(), comp.to_string())
};
#[allow(clippy::all)]
let style = if comp == best_compression_ratio_codec.1 {
"Fb"
} else {
@@ -49,7 +46,7 @@ fn main() {
};
table.add_row(Row::new(vec![
Cell::new(&name.to_string()).style_spec("bFg"),
Cell::new(name).style_spec("bFg"),
Cell::new(&ratio_cell).style_spec(style),
Cell::new(&est_cell).style_spec(""),
]));
@@ -73,7 +70,7 @@ pub fn get_codec_test_data_sets() -> Vec<(Vec<u64>, &'static str)> {
current_cumulative
})
.collect::<Vec<_>>();
//let data = (1..=200000_u64).map(|num| num + num).collect::<Vec<_>>();
// let data = (1..=200000_u64).map(|num| num + num).collect::<Vec<_>>();
data_and_names.push((data, "Monotonically increasing concave"));
let mut current_cumulative = 0;

48
fastfield_codecs/src/multilinearinterpol.rs
View File

@@ -1,30 +1,22 @@
/*!
//! MultiLinearInterpol compressor uses linear interpolation to guess a values and stores the
//! offset, but in blocks of 512.
//!
//! With a CHUNK_SIZE of 512 and 29 byte metadata per block, we get a overhead for metadata of 232 /
//! 512 = 0,45 bits per element. The additional space required per element in a block is the the
//! maximum deviation of the linear interpolation estimation function.
//!
//! E.g. if the maximum deviation of an element is 12, all elements cost 4bits.
//!
//! Size per block:
//! Num Elements * Maximum Deviation from Interpolation + 29 Byte Metadata
MultiLinearInterpol compressor uses linear interpolation to guess a values and stores the offset, but in blocks of 512.
With a CHUNK_SIZE of 512 and 29 byte metadata per block, we get a overhead for metadata of 232 / 512 = 0,45 bits per element.
The additional space required per element in a block is the the maximum deviation of the linear interpolation estimation function.
E.g. if the maximum deviation of an element is 12, all elements cost 4bits.
Size per block:
Num Elements * Maximum Deviation from Interpolation + 29 Byte Metadata
*/
use crate::FastFieldCodecReader;
use crate::FastFieldCodecSerializer;
use crate::FastFieldDataAccess;
use crate::FastFieldStats;
use common::CountingWriter;
use std::io::{self, Read, Write};
use std::ops::Sub;
use tantivy_bitpacker::compute_num_bits;
use tantivy_bitpacker::BitPacker;
use common::BinarySerializable;
use common::DeserializeFrom;
use tantivy_bitpacker::BitUnpacker;
use common::{BinarySerializable, CountingWriter, DeserializeFrom};
use tantivy_bitpacker::{compute_num_bits, BitPacker, BitUnpacker};
use crate::{FastFieldCodecReader, FastFieldCodecSerializer, FastFieldDataAccess, FastFieldStats};
const CHUNK_SIZE: u64 = 512;
@@ -252,11 +244,11 @@ impl FastFieldCodecSerializer for MultiLinearInterpolFastFieldSerializer {
);
if calculated_value > actual_value {
// negative value we need to apply an offset
// we ignore negative values in the max value calculation, because negative values
// will be offset to 0
// we ignore negative values in the max value calculation, because negative
// values will be offset to 0
offset = offset.max(calculated_value - actual_value);
} else {
//positive value no offset reuqired
// positive value no offset reuqired
rel_positive_max = rel_positive_max.max(actual_value - calculated_value);
}
}
@@ -350,8 +342,8 @@ impl FastFieldCodecSerializer for MultiLinearInterpolFastFieldSerializer {
.unwrap();
// Estimate one block and extrapolate the cost to all blocks.
// the theory would be that we don't have the actual max_distance, but we are close within 50%
// threshold.
// the theory would be that we don't have the actual max_distance, but we are close within
// 50% threshold.
// It is multiplied by 2 because in a log case scenario the line would be as much above as
// below. So the offset would = max_distance
//

15
ownedbytes/src/lib.rs
View File

@@ -1,9 +1,9 @@
use stable_deref_trait::StableDeref;
use std::convert::TryInto;
use std::mem;
use std::ops::{Deref, Range};
use std::sync::Arc;
use std::{fmt, io};
use std::{fmt, io, mem};
use stable_deref_trait::StableDeref;
/// An OwnedBytes simply wraps an object that owns a slice of data and exposes
/// this data as a static slice.
@@ -35,6 +35,8 @@ impl OwnedBytes {
}
/// creates a fileslice that is just a view over a slice of the data.
#[must_use]
#[inline]
pub fn slice(&self, range: Range<usize>) -> Self {
OwnedBytes {
data: &self.data[range],
@@ -63,6 +65,8 @@ impl OwnedBytes {
/// On the other hand, both `left` and `right` retain a handle over
/// the entire slice of memory. In other words, the memory will only
/// be released when both left and right are dropped.
#[inline]
#[must_use]
pub fn split(self, split_len: usize) -> (OwnedBytes, OwnedBytes) {
let right_box_stable_deref = self.box_stable_deref.clone();
let left = OwnedBytes {
@@ -96,8 +100,6 @@ impl OwnedBytes {
}
/// Drops the left most `advance_len` bytes.
///
/// See also [.clip(clip_len: usize))](#method.clip).
#[inline]
pub fn advance(&mut self, advance_len: usize) {
self.data = &self.data[advance_len..]
@@ -158,8 +160,7 @@ impl PartialEq<str> for OwnedBytes {
}
impl<'a, T: ?Sized> PartialEq<&'a T> for OwnedBytes
where
OwnedBytes: PartialEq<T>,
where OwnedBytes: PartialEq<T>
{
fn eq(&self, other: &&'a T) -> bool {
*self == **other

5
query-grammar/Cargo.toml
View File

@@ -5,9 +5,8 @@ authors = ["Paul Masurel <paul.masurel@gmail.com>"]
license = "MIT"
categories = ["database-implementations", "data-structures"]
description = """Search engine library"""
documentation = "https://quickwit-inc.github.io/tantivy/tantivy/index.html"
homepage = "https://github.com/quickwit-inc/tantivy"
repository = "https://github.com/quickwit-inc/tantivy"
homepage = "https://github.com/quickwit-oss/tantivy"
repository = "https://github.com/quickwit-oss/tantivy"
readme = "README.md"
keywords = ["search", "information", "retrieval"]
edition = "2018"

14
query-grammar/src/query_grammar.rs
View File

@@ -1,17 +1,20 @@
use super::user_input_ast::{UserInputAst, UserInputBound, UserInputLeaf, UserInputLiteral};
use crate::Occur;
use combine::error::StringStreamError;
use combine::parser::char::{char, digit, space, spaces, string};
use combine::parser::combinator::recognize;
use combine::parser::range::{take_while, take_while1};
use combine::parser::repeat::escaped;
use combine::parser::Parser;
use combine::{
attempt, choice, eof, many, many1, one_of, optional, parser, satisfy, skip_many1, value,
};
use combine::{error::StringStreamError, parser::combinator::recognize};
use once_cell::sync::Lazy;
use regex::Regex;
// Note: '-' char is only forbidden at the beginning of a field name, would be clearer to add it to special characters.
use super::user_input_ast::{UserInputAst, UserInputBound, UserInputLeaf, UserInputLiteral};
use crate::Occur;
// Note: '-' char is only forbidden at the beginning of a field name, would be clearer to add it to
// special characters.
const SPECIAL_CHARS: &[char] = &[
'+', '^', '`', ':', '{', '}', '"', '[', ']', '(', ')', '~', '!', '\\', '*', ' ',
];
@@ -363,9 +366,10 @@ mod test {
type TestParseResult = Result<(), StringStreamError>;
use super::*;
use combine::parser::Parser;
use super::*;
pub fn nearly_equals(a: f64, b: f64) -> bool {
(a - b).abs() < 0.0005 * (a + b).abs()
}

1
query-grammar/src/user_input_ast.rs
View File

@@ -91,6 +91,7 @@ pub enum UserInputAst {
}
impl UserInputAst {
#[must_use]
pub fn unary(self, occur: Occur) -> UserInputAst {
UserInputAst::Clause(vec![(Some(occur), self)])
}

8
rustfmt.toml
View File

@@ -1 +1,7 @@
use_try_shorthand = true
comment_width = 120
format_strings = true
group_imports = "StdExternalCrate"
imports_granularity = "Module"
normalize_comments = true
where_single_line = true
wrap_comments = true

8
src/collector/count_collector.rs
View File

@@ -1,9 +1,6 @@
use super::Collector;
use crate::collector::SegmentCollector;
use crate::DocId;
use crate::Score;
use crate::SegmentOrdinal;
use crate::SegmentReader;
use crate::{DocId, Score, SegmentOrdinal, SegmentReader};
/// `CountCollector` collector only counts how many
/// documents match the query.
@@ -80,8 +77,7 @@ impl SegmentCollector for SegmentCountCollector {
#[cfg(test)]
mod tests {
use super::{Count, SegmentCountCollector};
use crate::collector::Collector;
use crate::collector::SegmentCollector;
use crate::collector::{Collector, SegmentCollector};
#[test]
fn test_count_collect_does_not_requires_scoring() {

6
src/collector/custom_score_top_collector.rs
View File

@@ -8,8 +8,7 @@ pub(crate) struct CustomScoreTopCollector<TCustomScorer, TScore = Score> {
}
impl<TCustomScorer, TScore> CustomScoreTopCollector<TCustomScorer, TScore>
where
TScore: Clone + PartialOrd,
where TScore: Clone + PartialOrd
{
pub(crate) fn new(
custom_scorer: TCustomScorer,
@@ -114,8 +113,7 @@ where
}
impl<F, TScore> CustomSegmentScorer<TScore> for F
where
F: 'static + FnMut(DocId) -> TScore,
where F: 'static + FnMut(DocId) -> TScore
{
fn score(&mut self, doc: DocId) -> TScore {
(self)(doc)

3
src/collector/docset_collector.rs
View File

@@ -1,8 +1,7 @@
use std::collections::HashSet;
use crate::{DocAddress, DocId, Score};
use super::{Collector, SegmentCollector};
use crate::{DocAddress, DocId, Score};
/// Collectors that returns the set of DocAddress that matches the query.
///

75
src/collector/facet_collector.rs
View File

@@ -1,21 +1,14 @@
use crate::collector::Collector;
use crate::collector::SegmentCollector;
use crate::fastfield::FacetReader;
use crate::schema::Facet;
use crate::schema::Field;
use crate::DocId;
use crate::Score;
use crate::SegmentOrdinal;
use crate::SegmentReader;
use std::cmp::Ordering;
use std::collections::btree_map;
use std::collections::BTreeMap;
use std::collections::BTreeSet;
use std::collections::BinaryHeap;
use std::collections::{btree_map, BTreeMap, BTreeSet, BinaryHeap};
use std::iter::Peekable;
use std::ops::Bound;
use std::{u64, usize};
use crate::collector::{Collector, SegmentCollector};
use crate::fastfield::FacetReader;
use crate::schema::{Facet, Field};
use crate::{DocId, Score, SegmentOrdinal, SegmentReader};
struct Hit<'a> {
count: u64,
facet: &'a Facet,
@@ -83,7 +76,7 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
/// ```rust
/// use tantivy::collector::FacetCollector;
/// use tantivy::query::AllQuery;
/// use tantivy::schema::{Facet, Schema, INDEXED, TEXT};
/// use tantivy::schema::{Facet, Schema, FacetOptions, TEXT};
/// use tantivy::{doc, Index};
///
/// fn example() -> tantivy::Result<()> {
@@ -92,7 +85,7 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
/// // Facet have their own specific type.
/// // It is not a bad practise to put all of your
/// // facet information in the same field.
/// let facet = schema_builder.add_facet_field("facet", INDEXED);
/// let facet = schema_builder.add_facet_field("facet", FacetOptions::default());
/// let title = schema_builder.add_text_field("title", TEXT);
/// let schema = schema_builder.build();
/// let index = Index::create_in_ram(schema);
@@ -240,9 +233,7 @@ impl FacetCollector {
/// If you need the correct number of unique documents for two such facets,
/// just add them in separate `FacetCollector`.
pub fn add_facet<T>(&mut self, facet_from: T)
where
Facet: From<T>,
{
where Facet: From<T> {
let facet = Facet::from(facet_from);
for old_facet in &self.facets {
assert!(
@@ -400,11 +391,9 @@ impl<'a> Iterator for FacetChildIterator<'a> {
impl FacetCounts {
/// Returns an iterator over all of the facet count pairs inside this result.
/// See the documentation for `FacetCollector` for a usage example.
/// See the documentation for [FacetCollector] for a usage example.
pub fn get<T>(&self, facet_from: T) -> FacetChildIterator<'_>
where
Facet: From<T>,
{
where Facet: From<T> {
let facet = Facet::from(facet_from);
let left_bound = Bound::Excluded(facet.clone());
let right_bound = if facet.is_root() {
@@ -421,11 +410,9 @@ impl FacetCounts {
}
/// Returns a vector of top `k` facets with their counts, sorted highest-to-lowest by counts.
/// See the documentation for `FacetCollector` for a usage example.
/// See the documentation for [FacetCollector] for a usage example.
pub fn top_k<T>(&self, facet: T, k: usize) -> Vec<(&Facet, u64)>
where
Facet: From<T>,
{
where Facet: From<T> {
let mut heap = BinaryHeap::with_capacity(k);
let mut it = self.get(facet);
@@ -458,21 +445,23 @@ impl FacetCounts {
#[cfg(test)]
mod tests {
use std::iter;
use rand::distributions::Uniform;
use rand::prelude::SliceRandom;
use rand::{thread_rng, Rng};
use super::{FacetCollector, FacetCounts};
use crate::collector::Count;
use crate::core::Index;
use crate::query::{AllQuery, QueryParser, TermQuery};
use crate::schema::{Document, Facet, Field, IndexRecordOption, Schema, INDEXED};
use crate::schema::{Document, Facet, FacetOptions, Field, IndexRecordOption, Schema};
use crate::Term;
use rand::distributions::Uniform;
use rand::prelude::SliceRandom;
use rand::{thread_rng, Rng};
use std::iter;
#[test]
fn test_facet_collector_drilldown() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -522,8 +511,9 @@ mod tests {
}
#[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::from_field_id(0));
facet_collector.add_facet(Facet::from("/country"));
@@ -533,7 +523,7 @@ mod tests {
#[test]
fn test_doc_unsorted_multifacet() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let facet_field = schema_builder.add_facet_field("facets", INDEXED);
let facet_field = schema_builder.add_facet_field("facets", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
@@ -558,7 +548,7 @@ mod tests {
#[test]
fn test_doc_search_by_facet() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
@@ -615,7 +605,7 @@ mod tests {
#[test]
fn test_facet_collector_topk() {
let mut schema_builder = Schema::builder();
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -664,7 +654,7 @@ mod tests {
#[test]
fn test_facet_collector_topk_tie_break() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
@@ -700,13 +690,14 @@ mod tests {
#[cfg(all(test, feature = "unstable"))]
mod bench {
use rand::seq::SliceRandom;
use rand::thread_rng;
use test::Bencher;
use crate::collector::FacetCollector;
use crate::query::AllQuery;
use crate::schema::{Facet, Schema, INDEXED};
use crate::Index;
use rand::seq::SliceRandom;
use rand::thread_rng;
use test::Bencher;
#[bench]
fn bench_facet_collector(b: &mut Bencher) {
@@ -727,7 +718,7 @@ mod bench {
let mut index_writer = index.writer_for_tests().unwrap();
for doc in docs {
index_writer.add_document(doc);
index_writer.add_document(doc).unwrap();
}
index_writer.commit().unwrap();
let reader = index.reader().unwrap();

8
src/collector/filter_collector_wrapper.rs
View File

@@ -16,8 +16,9 @@ use crate::fastfield::{DynamicFastFieldReader, FastFieldReader, FastValue};
use crate::schema::Field;
use crate::{Score, SegmentReader, TantivyError};
/// The `FilterCollector` collector filters docs using a fast field value and a predicate.
/// Only the documents for which the predicate returned "true" will be passed on to the next collector.
/// The `FilterCollector` filters docs using a fast field value and a predicate.
/// Only the documents for which the predicate returned "true" will be passed on to the next
/// collector.
///
/// ```rust
/// use tantivy::collector::{TopDocs, FilterCollector};
@@ -58,8 +59,7 @@ use crate::{Score, SegmentReader, TantivyError};
/// # }
/// ```
pub struct FilterCollector<TCollector, TPredicate, TPredicateValue: FastValue>
where
TPredicate: 'static + Clone,
where TPredicate: 'static + Clone
{
field: Field,
collector: TCollector,

12
src/collector/histogram_collector.rs
View File

@@ -1,8 +1,9 @@
use fastdivide::DividerU64;
use crate::collector::{Collector, SegmentCollector};
use crate::fastfield::{DynamicFastFieldReader, FastFieldReader, FastValue};
use crate::schema::{Field, Type};
use crate::{DocId, Score};
use fastdivide::DividerU64;
/// Histogram builds an histogram of the values of a fastfield for the
/// collected DocSet.
@@ -36,8 +37,8 @@ impl HistogramCollector {
/// - `bucket_width`: the length of the interval that is associated to each buckets.
/// - `num_buckets`: The overall number of buckets.
///
/// Together, this parameters define a partition of `[min_value, min_value + num_buckets * bucket_width)`
/// into `num_buckets` intervals of width bucket that we call `bucket`.
/// Together, this parameters define a partition of `[min_value, min_value + num_buckets *
/// bucket_width)` into `num_buckets` intervals of width bucket that we call `bucket`.
///
/// # Disclaimer
/// This function panics if the field given is of type f64.
@@ -147,12 +148,13 @@ fn add_vecs(mut vals_list: Vec<Vec<u64>>, len: usize) -> Vec<u64> {
#[cfg(test)]
mod tests {
use fastdivide::DividerU64;
use query::AllQuery;
use super::{add_vecs, HistogramCollector, HistogramComputer};
use crate::chrono::{TimeZone, Utc};
use crate::schema::{Schema, FAST};
use crate::{doc, query, Index};
use fastdivide::DividerU64;
use query::AllQuery;
#[test]
fn test_add_histograms_simple() {

176
src/collector/mod.rs
View File

@@ -1,95 +1,90 @@
/*!
//! # 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.
# 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.
*/
use crate::DocId;
use crate::Score;
use crate::SegmentOrdinal;
use crate::SegmentReader;
use downcast_rs::impl_downcast;
use crate::{DocId, Score, SegmentOrdinal, SegmentReader};
mod count_collector;
pub use self::count_collector::Count;
@@ -111,8 +106,7 @@ mod tweak_score_top_collector;
pub use self::tweak_score_top_collector::{ScoreSegmentTweaker, ScoreTweaker};
mod facet_collector;
pub use self::facet_collector::FacetCollector;
pub use self::facet_collector::FacetCounts;
pub use self::facet_collector::{FacetCollector, FacetCounts};
use crate::query::Weight;
mod docset_collector;

21
src/collector/multi_collector.rs
View File

@@ -1,14 +1,10 @@
use super::Collector;
use super::SegmentCollector;
use crate::collector::Fruit;
use crate::DocId;
use crate::Score;
use crate::SegmentOrdinal;
use crate::SegmentReader;
use crate::TantivyError;
use std::marker::PhantomData;
use std::ops::Deref;
use super::{Collector, SegmentCollector};
use crate::collector::Fruit;
use crate::{DocId, Score, SegmentOrdinal, SegmentReader, TantivyError};
pub struct MultiFruit {
sub_fruits: Vec<Option<Box<dyn Fruit>>>,
}
@@ -104,7 +100,8 @@ impl<TFruit: Fruit> FruitHandle<TFruit> {
///
/// 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).
/// [Combining several collectors section of the collector
/// documentation](./index.html#combining-several-collectors).
///
/// ```rust
/// use tantivy::collector::{Count, TopDocs, MultiCollector};
@@ -248,10 +245,8 @@ 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 crate::schema::{IndexRecordOption, Schema, TEXT};
use crate::{Index, Term};
#[test]
fn test_multi_collector() -> crate::Result<()> {

24
src/collector/tests.rs
View File

@@ -1,21 +1,13 @@
use super::*;
use crate::core::SegmentReader;
use crate::fastfield::BytesFastFieldReader;
use crate::fastfield::DynamicFastFieldReader;
use crate::fastfield::FastFieldReader;
use crate::schema::Field;
use crate::DocId;
use crate::Score;
use crate::SegmentOrdinal;
use crate::{DocAddress, Document, Searcher};
use crate::collector::{Count, FilterCollector, TopDocs};
use crate::query::{AllQuery, QueryParser};
use crate::schema::{Schema, FAST, TEXT};
use crate::DateTime;
use crate::{doc, Index};
use std::str::FromStr;
use super::*;
use crate::collector::{Count, FilterCollector, TopDocs};
use crate::core::SegmentReader;
use crate::fastfield::{BytesFastFieldReader, DynamicFastFieldReader, FastFieldReader};
use crate::query::{AllQuery, QueryParser};
use crate::schema::{Field, Schema, FAST, TEXT};
use crate::{doc, DateTime, DocAddress, DocId, Document, Index, Score, Searcher, SegmentOrdinal};
pub const TEST_COLLECTOR_WITH_SCORE: TestCollector = TestCollector {
compute_score: true,
};

14
src/collector/top_collector.rs
View File

@@ -1,11 +1,9 @@
use crate::DocAddress;
use crate::DocId;
use crate::SegmentOrdinal;
use crate::SegmentReader;
use std::cmp::Ordering;
use std::collections::BinaryHeap;
use std::marker::PhantomData;
use crate::{DocAddress, DocId, SegmentOrdinal, 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
@@ -62,8 +60,7 @@ pub(crate) struct TopCollector<T> {
}
impl<T> TopCollector<T>
where
T: PartialOrd + Clone,
where T: PartialOrd + Clone
{
/// Creates a top collector, with a number of documents equal to "limit".
///
@@ -253,7 +250,7 @@ mod tests {
// when harvesting we have to guarantee stable sorting in case of a tie
// on the score
let doc_ids_collection = [4, 5, 6];
let score = 3.14;
let score = 3.3f32;
let mut top_collector_limit_2 = TopSegmentCollector::new(0, 2);
for id in &doc_ids_collection {
@@ -322,9 +319,10 @@ mod tests {
#[cfg(all(test, feature = "unstable"))]
mod bench {
use super::TopSegmentCollector;
use test::Bencher;
use super::TopSegmentCollector;
#[bench]
fn bench_top_segment_collector_collect_not_at_capacity(b: &mut Bencher) {
let mut top_collector = TopSegmentCollector::new(0, 400);

43
src/collector/top_score_collector.rs
View File

@@ -1,21 +1,18 @@
use std::collections::BinaryHeap;
use std::fmt;
use std::marker::PhantomData;
use super::Collector;
use crate::collector::top_collector::{ComparableDoc, TopCollector};
use crate::collector::custom_score_top_collector::CustomScoreTopCollector;
use crate::collector::top_collector::{ComparableDoc, TopCollector, TopSegmentCollector};
use crate::collector::tweak_score_top_collector::TweakedScoreTopCollector;
use crate::collector::{
CustomScorer, CustomSegmentScorer, ScoreSegmentTweaker, ScoreTweaker, SegmentCollector,
};
use crate::fastfield::{DynamicFastFieldReader, FastFieldReader};
use crate::fastfield::{DynamicFastFieldReader, FastFieldReader, FastValue};
use crate::query::Weight;
use crate::schema::Field;
use crate::DocAddress;
use crate::DocId;
use crate::Score;
use crate::SegmentOrdinal;
use crate::SegmentReader;
use crate::{collector::custom_score_top_collector::CustomScoreTopCollector, fastfield::FastValue};
use crate::{collector::top_collector::TopSegmentCollector, TantivyError};
use std::fmt;
use std::{collections::BinaryHeap, marker::PhantomData};
use crate::{DocAddress, DocId, Score, SegmentOrdinal, SegmentReader, TantivyError};
struct FastFieldConvertCollector<
TCollector: Collector<Fruit = Vec<(u64, DocAddress)>>,
@@ -210,17 +207,19 @@ impl TopDocs {
/// Ok(())
/// # }
/// ```
#[must_use]
pub fn and_offset(self, offset: usize) -> TopDocs {
TopDocs(self.0.and_offset(offset))
}
/// Set top-K to rank documents by a given fast field.
///
/// If the field is not a fast or does not exist, this method returns successfully (it is not aware of any schema).
/// An error will be returned at the moment of search.
/// If the field is not a fast or does not exist, this method returns successfully (it is not
/// aware of any schema). An error will be returned at the moment of search.
///
/// If the field is a FAST field but not a u64 field, search will return successfully but it will return
/// returns a monotonic u64-representation (ie. the order is still correct) of the requested field type.
/// If the field is a FAST field but not a u64 field, search will return successfully but it
/// will return returns a monotonic u64-representation (ie. the order is still correct) of
/// the requested field type.
///
/// # Example
///
@@ -295,14 +294,15 @@ impl TopDocs {
/// Set top-K to rank documents by a given fast field.
///
/// If the field is not a fast field, or its field type does not match the generic type, this method does not panic,
/// but an explicit error will be returned at the moment of collection.
/// If the field is not a fast field, or its field type does not match the generic type, this
/// method does not panic, but an explicit error will be returned at the moment of
/// collection.
///
/// Note that this method is a generic. The requested fast field type will be often
/// inferred in your code by the rust compiler.
///
/// Implementation-wise, for performance reason, tantivy will manipulate the u64 representation of your fast
/// field until the last moment.
/// Implementation-wise, for performance reason, tantivy will manipulate the u64 representation
/// of your fast field until the last moment.
///
/// # Example
///
@@ -714,10 +714,7 @@ mod tests {
use crate::collector::Collector;
use crate::query::{AllQuery, Query, QueryParser};
use crate::schema::{Field, Schema, FAST, STORED, TEXT};
use crate::Index;
use crate::IndexWriter;
use crate::Score;
use crate::{DocAddress, DocId, SegmentReader};
use crate::{DocAddress, DocId, Index, IndexWriter, Score, SegmentReader};
fn make_index() -> crate::Result<Index> {
let mut schema_builder = Schema::builder();

9
src/collector/tweak_score_top_collector.rs
View File

@@ -1,7 +1,6 @@
use crate::collector::top_collector::{TopCollector, TopSegmentCollector};
use crate::collector::{Collector, SegmentCollector};
use crate::DocAddress;
use crate::{DocId, Result, Score, SegmentReader};
use crate::{DocAddress, DocId, Result, Score, SegmentReader};
pub(crate) struct TweakedScoreTopCollector<TScoreTweaker, TScore = Score> {
score_tweaker: TScoreTweaker,
@@ -9,8 +8,7 @@ pub(crate) struct TweakedScoreTopCollector<TScoreTweaker, TScore = Score> {
}
impl<TScoreTweaker, TScore> TweakedScoreTopCollector<TScoreTweaker, TScore>
where
TScore: Clone + PartialOrd,
where TScore: Clone + PartialOrd
{
pub fn new(
score_tweaker: TScoreTweaker,
@@ -118,8 +116,7 @@ where
}
impl<F, TScore> ScoreSegmentTweaker<TScore> for F
where
F: 'static + FnMut(DocId, Score) -> TScore,
where F: 'static + FnMut(DocId, Score) -> TScore
{
fn score(&mut self, doc: DocId, score: Score) -> TScore {
(self)(doc, score)

6
src/core/executor.rs
View File

@@ -57,7 +57,11 @@ impl Executor {
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);
error!(
"Failed to send search task. It probably means all search \
threads have panicked. {:?}",
err
);
}
});
}

111
src/core/index.rs
View File

@@ -1,35 +1,27 @@
use super::{segment::Segment, IndexSettings};
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::error::OpenReadError;
use crate::directory::ManagedDirectory;
#[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, MAX_NUM_THREAD};
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::{TextAnalyzer, TokenizerManager};
use crate::IndexWriter;
use std::collections::HashSet;
use std::fmt;
#[cfg(feature = "mmap")]
use std::path::Path;
use std::path::PathBuf;
use std::sync::Arc;
use super::segment::Segment;
use super::IndexSettings;
use crate::core::{
Executor, IndexMeta, SegmentId, SegmentMeta, SegmentMetaInventory, META_FILEPATH,
};
use crate::directory::error::OpenReadError;
#[cfg(feature = "mmap")]
use crate::directory::MmapDirectory;
use crate::directory::{Directory, ManagedDirectory, RamDirectory, INDEX_WRITER_LOCK};
use crate::error::{DataCorruption, TantivyError};
use crate::indexer::index_writer::{MAX_NUM_THREAD, MEMORY_ARENA_NUM_BYTES_MIN};
use crate::indexer::segment_updater::save_new_metas;
use crate::reader::{IndexReader, IndexReaderBuilder};
use crate::schema::{Field, FieldType, Schema};
use crate::tokenizer::{TextAnalyzer, TokenizerManager};
use crate::IndexWriter;
fn load_metas(
directory: &dyn Directory,
inventory: &SegmentMetaInventory,
@@ -78,7 +70,6 @@ fn load_metas(
/// let schema = schema_builder.build();
/// let settings = IndexSettings{sort_by_field: Some(IndexSortByField{field:"number".to_string(), order:Order::Asc}), ..Default::default()};
/// let index = Index::builder().schema(schema).settings(settings).create_in_ram();
///
/// ```
pub struct IndexBuilder {
schema: Option<Schema>,
@@ -97,16 +88,21 @@ impl IndexBuilder {
index_settings: IndexSettings::default(),
}
}
/// Set the settings
#[must_use]
pub fn settings(mut self, settings: IndexSettings) -> Self {
self.index_settings = settings;
self
}
/// Set the schema
#[must_use]
pub fn schema(mut self, schema: Schema) -> Self {
self.schema = Some(schema);
self
}
/// Creates a new index using the `RAMDirectory`.
///
/// The index will be allocated in anonymous memory.
@@ -117,6 +113,7 @@ impl IndexBuilder {
.create(ram_directory)
.expect("Creating a RAMDirectory should never fail"))
}
/// Creates a new index in a given filepath.
/// The index will use the `MMapDirectory`.
///
@@ -129,6 +126,7 @@ impl IndexBuilder {
}
self.create(mmap_directory)
}
/// Creates a new index in a temp directory.
///
/// The index will use the `MMapDirectory` in a newly created directory.
@@ -142,12 +140,14 @@ impl IndexBuilder {
let mmap_directory: Box<dyn Directory> = Box::new(MmapDirectory::create_from_tempdir()?);
self.create(mmap_directory)
}
fn get_expect_schema(&self) -> crate::Result<Schema> {
self.schema
.as_ref()
.cloned()
.ok_or(TantivyError::IndexBuilderMissingArgument("schema"))
}
/// Opens or creates a new index in the provided directory
pub fn open_or_create<T: Into<Box<dyn Directory>>>(self, dir: T) -> crate::Result<Index> {
let dir = dir.into();
@@ -217,7 +217,7 @@ impl Index {
/// 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) -> crate::Result<()> {
self.executor = Arc::new(Executor::multi_thread(num_threads, "thrd-tantivy-search-")?);
self.executor = Arc::new(Executor::multi_thread(num_threads, "tantivy-search-")?);
Ok(())
}
@@ -397,17 +397,18 @@ impl Index {
/// - `num_threads` defines the number of indexing workers that
/// should work at the same time.
///
/// - `overall_heap_size_in_bytes` sets the amount of memory
/// - `overall_memory_arena_in_bytes` sets the amount of memory
/// allocated for all indexing thread.
/// Each thread will receive a budget of `overall_heap_size_in_bytes / num_threads`.
/// Each thread will receive a budget of `overall_memory_arena_in_bytes / num_threads`.
///
/// # Errors
/// If the lockfile already exists, returns `Error::DirectoryLockBusy` or an `Error::IoError`.
/// If the heap size per thread is too small or too big, returns `TantivyError::InvalidArgument`
/// If the memory arena per thread is too small or too big, returns
/// `TantivyError::InvalidArgument`
pub fn writer_with_num_threads(
&self,
num_threads: usize,
overall_heap_size_in_bytes: usize,
overall_memory_arena_in_bytes: usize,
) -> crate::Result<IndexWriter> {
let directory_lock = self
.directory
@@ -416,26 +417,25 @@ impl Index {
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."
"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 heap_size_in_bytes_per_thread = overall_heap_size_in_bytes / num_threads;
let memory_arena_in_bytes_per_thread = overall_memory_arena_in_bytes / num_threads;
IndexWriter::new(
self,
num_threads,
heap_size_in_bytes_per_thread,
memory_arena_in_bytes_per_thread,
directory_lock,
)
}
/// Helper to create an index writer for tests.
///
/// That index writer only simply has a single thread and a heap of 10 MB.
/// That index writer only simply has a single thread and a memory arena of 10 MB.
/// Using a single thread gives us a deterministic allocation of DocId.
#[cfg(test)]
pub fn writer_for_tests(&self) -> crate::Result<IndexWriter> {
@@ -446,29 +446,28 @@ impl Index {
///
/// Tantivy will automatically define the number of threads to use, but
/// no more than 8 threads.
/// `overall_heap_size_in_bytes` is the total target memory usage that will be split
/// `overall_memory_arena_in_bytes` is the total target memory usage that will be split
/// between a given number of threads.
///
/// # Errors
/// If the lockfile already exists, returns `Error::FileAlreadyExists`.
/// If the heap size per thread is too small or too big, returns `TantivyError::InvalidArgument`
pub fn writer(&self, overall_heap_size_in_bytes: usize) -> crate::Result<IndexWriter> {
/// If the memory arena per thread is too small or too big, returns
/// `TantivyError::InvalidArgument`
pub fn writer(&self, memory_arena_num_bytes: usize) -> crate::Result<IndexWriter> {
let mut num_threads = std::cmp::min(num_cpus::get(), MAX_NUM_THREAD);
let heap_size_in_bytes_per_thread = overall_heap_size_in_bytes / num_threads;
if heap_size_in_bytes_per_thread < HEAP_SIZE_MIN {
num_threads = (overall_heap_size_in_bytes / HEAP_SIZE_MIN).max(1);
let memory_arena_num_bytes_per_thread = memory_arena_num_bytes / num_threads;
if memory_arena_num_bytes_per_thread < MEMORY_ARENA_NUM_BYTES_MIN {
num_threads = (memory_arena_num_bytes / MEMORY_ARENA_NUM_BYTES_MIN).max(1);
}
self.writer_with_num_threads(num_threads, overall_heap_size_in_bytes)
self.writer_with_num_threads(num_threads, memory_arena_num_bytes)
}
/// Accessor to the index settings
///
pub fn settings(&self) -> &IndexSettings {
&self.settings
}
/// Accessor to the index settings
///
pub fn settings_mut(&mut self) -> &mut IndexSettings {
&mut self.settings
}
@@ -556,15 +555,9 @@ impl fmt::Debug for Index {
#[cfg(test)]
mod tests {
use crate::schema::Field;
use crate::schema::{Schema, INDEXED, TEXT};
use crate::IndexReader;
use crate::ReloadPolicy;
use crate::{
directory::{RamDirectory, WatchCallback},
IndexSettings,
};
use crate::{Directory, Index};
use crate::directory::{RamDirectory, WatchCallback};
use crate::schema::{Field, Schema, INDEXED, TEXT};
use crate::{Directory, Index, IndexReader, IndexSettings, ReloadPolicy};
#[test]
fn test_indexer_for_field() {
@@ -673,10 +666,12 @@ mod tests {
#[cfg(feature = "mmap")]
mod mmap_specific {
use std::path::PathBuf;
use tempfile::TempDir;
use super::*;
use crate::Directory;
use std::path::PathBuf;
use tempfile::TempDir;
#[test]
fn test_index_on_commit_reload_policy_mmap() -> crate::Result<()> {

34
src/core/index_meta.rs
View File

@@ -1,12 +1,16 @@
use super::SegmentComponent;
use crate::schema::Schema;
use crate::Opstamp;
use crate::{core::SegmentId, store::Compressor};
use census::{Inventory, TrackedObject};
use serde::{Deserialize, Serialize};
use std::collections::HashSet;
use std::fmt;
use std::path::PathBuf;
use std::{collections::HashSet, sync::atomic::AtomicBool};
use std::{fmt, sync::Arc};
use std::sync::atomic::AtomicBool;
use std::sync::Arc;
use serde::{Deserialize, Serialize};
use super::SegmentComponent;
use crate::core::SegmentId;
use crate::schema::Schema;
use crate::store::Compressor;
use crate::{Inventory, Opstamp, TrackedObject};
#[derive(Clone, Debug, Serialize, Deserialize)]
struct DeleteMeta {
@@ -188,7 +192,12 @@ impl SegmentMeta {
}
#[doc(hidden)]
#[must_use]
pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> SegmentMeta {
assert!(
num_deleted_docs <= self.max_doc(),
"There cannot be more deleted docs than there are docs."
);
let delete_meta = DeleteMeta {
num_deleted_docs,
opstamp,
@@ -278,7 +287,6 @@ impl Order {
/// * the searchable segments,
/// * the index `docstamp`
/// * the schema
///
#[derive(Clone, Serialize)]
pub struct IndexMeta {
/// `IndexSettings` to configure index options.
@@ -366,10 +374,8 @@ impl fmt::Debug for IndexMeta {
mod tests {
use super::IndexMeta;
use crate::{
schema::{Schema, TEXT},
IndexSettings, IndexSortByField, Order,
};
use crate::schema::{Schema, TEXT};
use crate::{IndexSettings, IndexSortByField, Order};
#[test]
fn test_serialize_metas() {
@@ -394,7 +400,7 @@ mod tests {
let json = serde_json::ser::to_string(&index_metas).expect("serialization failed");
assert_eq!(
json,
r#"{"index_settings":{"sort_by_field":{"field":"text","order":"Asc"},"docstore_compression":"lz4"},"segments":[],"schema":[{"name":"text","type":"text","options":{"indexing":{"record":"position","tokenizer":"default"},"stored":false}}],"opstamp":0}"#
r#"{"index_settings":{"sort_by_field":{"field":"text","order":"Asc"},"docstore_compression":"lz4"},"segments":[],"schema":[{"name":"text","type":"text","options":{"indexing":{"record":"position","fieldnorms":true,"tokenizer":"default"},"stored":false}}],"opstamp":0}"#
);
}
}

9
src/core/inverted_index_reader.rs
View File

@@ -1,13 +1,12 @@
use std::io;
use common::BinarySerializable;
use crate::directory::FileSlice;
use crate::positions::PositionReader;
use crate::postings::TermInfo;
use crate::postings::{BlockSegmentPostings, SegmentPostings};
use crate::schema::IndexRecordOption;
use crate::schema::Term;
use crate::postings::{BlockSegmentPostings, SegmentPostings, TermInfo};
use crate::schema::{IndexRecordOption, Term};
use crate::termdict::TermDictionary;
use common::BinarySerializable;
/// The inverted index reader is in charge of accessing
/// the inverted index associated to a specific field.

9
src/core/mod.rs
View File

@@ -8,21 +8,22 @@ mod segment_component;
mod segment_id;
mod segment_reader;
use std::path::Path;
use once_cell::sync::Lazy;
pub use self::executor::Executor;
pub use self::index::{Index, IndexBuilder};
pub use self::index_meta::{
IndexMeta, IndexSettings, IndexSortByField, Order, SegmentMeta, SegmentMetaInventory,
};
pub use self::inverted_index_reader::InvertedIndexReader;
pub use self::searcher::Searcher;
pub use self::searcher::{Searcher, SearcherGeneration};
pub use self::segment::Segment;
pub use self::segment_component::SegmentComponent;
pub use self::segment_id::SegmentId;
pub use self::segment_reader::SegmentReader;
use once_cell::sync::Lazy;
use std::path::Path;
/// 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"));

72
src/core/searcher.rs
View File

@@ -1,28 +1,73 @@
use crate::collector::Collector;
use crate::core::Executor;
use std::collections::BTreeMap;
use std::{fmt, io};
use crate::core::SegmentReader;
use crate::collector::Collector;
use crate::core::{Executor, SegmentReader};
use crate::query::Query;
use crate::schema::Document;
use crate::schema::Schema;
use crate::schema::Term;
use crate::schema::{Document, Schema, Term};
use crate::space_usage::SearcherSpaceUsage;
use crate::store::StoreReader;
use crate::DocAddress;
use crate::Index;
use crate::{DocAddress, Index, Opstamp, SegmentId, TrackedObject};
use std::{fmt, io};
/// Identifies the searcher generation accessed by a [Searcher].
///
/// While this might seem redundant, a [SearcherGeneration] contains
/// both a `generation_id` AND a list of `(SegmentId, DeleteOpstamp)`.
///
/// This is on purpose. This object is used by the `Warmer` API.
/// Having both information makes it possible to identify which
/// artifact should be refreshed or garbage collected.
///
/// Depending on the use case, `Warmer`'s implementers can decide to
/// produce artifacts per:
/// - `generation_id` (e.g. some searcher level aggregates)
/// - `(segment_id, delete_opstamp)` (e.g. segment level aggregates)
/// - `segment_id` (e.g. for immutable document level information)
/// - `(generation_id, segment_id)` (e.g. for consistent dynamic column)
/// - ...
#[derive(Debug, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
pub struct SearcherGeneration {
segments: BTreeMap<SegmentId, Option<Opstamp>>,
generation_id: u64,
}
impl SearcherGeneration {
pub(crate) fn from_segment_readers(
segment_readers: &[SegmentReader],
generation_id: u64,
) -> Self {
let mut segment_id_to_del_opstamp = BTreeMap::new();
for segment_reader in segment_readers {
segment_id_to_del_opstamp
.insert(segment_reader.segment_id(), segment_reader.delete_opstamp());
}
Self {
segments: segment_id_to_del_opstamp,
generation_id,
}
}
/// Returns the searcher generation id.
pub fn generation_id(&self) -> u64 {
self.generation_id
}
/// Return a `(SegmentId -> DeleteOpstamp)` mapping.
pub fn segments(&self) -> &BTreeMap<SegmentId, Option<Opstamp>> {
&self.segments
}
}
/// Holds a list of `SegmentReader`s ready for search.
///
/// It guarantees that the `Segment` will not be removed before
/// the destruction of the `Searcher`.
///
pub struct Searcher {
schema: Schema,
index: Index,
segment_readers: Vec<SegmentReader>,
store_readers: Vec<StoreReader>,
generation: TrackedObject<SearcherGeneration>,
}
impl Searcher {
@@ -31,6 +76,7 @@ impl Searcher {
schema: Schema,
index: Index,
segment_readers: Vec<SegmentReader>,
generation: TrackedObject<SearcherGeneration>,
) -> io::Result<Searcher> {
let store_readers: Vec<StoreReader> = segment_readers
.iter()
@@ -41,6 +87,7 @@ impl Searcher {
index,
segment_readers,
store_readers,
generation,
})
}
@@ -49,6 +96,11 @@ impl Searcher {
&self.index
}
/// [SearcherGeneration] which identifies the version of the snapshot held by this `Searcher`.
pub fn generation(&self) -> &SearcherGeneration {
self.generation.as_ref()
}
/// Fetches a document from tantivy's store given a `DocAddress`.
///
/// The searcher uses the segment ordinal to route the

17
src/core/segment.rs
View File

@@ -1,15 +1,13 @@
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::{FileSlice, WritePtr};
use crate::schema::Schema;
use crate::Opstamp;
use std::fmt;
use std::path::PathBuf;
use super::SegmentComponent;
use crate::core::{Index, SegmentId, SegmentMeta};
use crate::directory::error::{OpenReadError, OpenWriteError};
use crate::directory::{Directory, FileSlice, WritePtr};
use crate::schema::Schema;
use crate::Opstamp;
/// A segment is a piece of the index.
#[derive(Clone)]
pub struct Segment {
@@ -56,6 +54,7 @@ impl Segment {
}
#[doc(hidden)]
#[must_use]
pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> Segment {
Segment {
index: self.index,

10
src/core/segment_id.rs
View File

@@ -1,14 +1,14 @@
use std::cmp::{Ord, Ordering};
use std::error::Error;
use std::fmt;
use uuid::Uuid;
use std::str::FromStr;
#[cfg(test)]
use std::sync::atomic;
#[cfg(test)]
use once_cell::sync::Lazy;
use serde::{Deserialize, Serialize};
use std::error::Error;
use std::str::FromStr;
#[cfg(test)]
use std::sync::atomic;
use uuid::Uuid;
/// Uuid identifying a segment.
///

74
src/core/segment_reader.rs
View File

@@ -1,27 +1,19 @@
use crate::core::InvertedIndexReader;
use crate::core::Segment;
use crate::core::SegmentComponent;
use crate::core::SegmentId;
use crate::directory::CompositeFile;
use crate::directory::FileSlice;
use std::collections::HashMap;
use std::sync::{Arc, RwLock};
use std::{fmt, io};
use fail::fail_point;
use crate::core::{InvertedIndexReader, Segment, SegmentComponent, SegmentId};
use crate::directory::{CompositeFile, FileSlice};
use crate::error::DataCorruption;
use crate::fastfield::intersect_alive_bitsets;
use crate::fastfield::AliveBitSet;
use crate::fastfield::FacetReader;
use crate::fastfield::FastFieldReaders;
use crate::fastfield::{intersect_alive_bitsets, AliveBitSet, FacetReader, FastFieldReaders};
use crate::fieldnorm::{FieldNormReader, FieldNormReaders};
use crate::schema::FieldType;
use crate::schema::Schema;
use crate::schema::{Field, IndexRecordOption};
use crate::schema::{Field, FieldType, IndexRecordOption, Schema};
use crate::space_usage::SegmentSpaceUsage;
use crate::store::StoreReader;
use crate::termdict::TermDictionary;
use crate::DocId;
use fail::fail_point;
use std::fmt;
use std::sync::Arc;
use std::sync::RwLock;
use std::{collections::HashMap, io};
use crate::{DocId, Opstamp};
/// Entry point to access all of the datastructures of the `Segment`
///
@@ -38,6 +30,8 @@ pub struct SegmentReader {
inv_idx_reader_cache: Arc<RwLock<HashMap<Field, Arc<InvertedIndexReader>>>>,
segment_id: SegmentId,
delete_opstamp: Option<Opstamp>,
max_doc: DocId,
num_docs: DocId,
@@ -100,7 +94,7 @@ impl SegmentReader {
let field_entry = self.schema.get_field_entry(field);
match field_entry.field_type() {
FieldType::HierarchicalFacet(_) => {
FieldType::Facet(_) => {
let term_ords_reader = self.fast_fields().u64s(field)?;
let termdict = self
.termdict_composite
@@ -127,13 +121,18 @@ impl SegmentReader {
self.fieldnorm_readers.get_field(field)?.ok_or_else(|| {
let field_name = self.schema.get_field_name(field);
let err_msg = format!(
"Field norm not found for field {:?}. Was it marked as indexed during indexing?",
"Field norm not found for field {:?}. Was the field set to record norm during \
indexing?",
field_name
);
crate::TantivyError::SchemaError(err_msg)
})
}
pub(crate) fn fieldnorms_readers(&self) -> &FieldNormReaders {
&self.fieldnorm_readers
}
/// Accessor to the segment's `StoreReader`.
pub fn get_store_reader(&self) -> io::Result<StoreReader> {
StoreReader::open(self.store_file.clone())
@@ -201,6 +200,7 @@ impl SegmentReader {
fast_fields_readers: fast_field_readers,
fieldnorm_readers,
segment_id: segment.id(),
delete_opstamp: segment.meta().delete_opstamp(),
store_file,
alive_bitset_opt,
positions_composite,
@@ -251,19 +251,24 @@ impl SegmentReader {
let record_option = record_option_opt.unwrap();
let postings_file = postings_file_opt.unwrap();
let termdict_file: FileSlice = self.termdict_composite.open_read(field)
.ok_or_else(||
DataCorruption::comment_only(format!("Failed to open field {:?}'s term dictionary in the composite file. Has the schema been modified?", field_entry.name()))
)?;
let positions_file = self
.positions_composite
.open_read(field)
.ok_or_else(|| {
let error_msg = format!("Failed to open field {:?}'s positions in the composite file. Has the schema been modified?", field_entry.name());
DataCorruption::comment_only(error_msg)
let termdict_file: FileSlice =
self.termdict_composite.open_read(field).ok_or_else(|| {
DataCorruption::comment_only(format!(
"Failed to open field {:?}'s term dictionary in the composite file. Has the \
schema been modified?",
field_entry.name()
))
})?;
let positions_file = self.positions_composite.open_read(field).ok_or_else(|| {
let error_msg = format!(
"Failed to open field {:?}'s positions in the composite file. Has the schema been \
modified?",
field_entry.name()
);
DataCorruption::comment_only(error_msg)
})?;
let inv_idx_reader = Arc::new(InvertedIndexReader::new(
TermDictionary::open(termdict_file)?,
postings_file,
@@ -286,6 +291,11 @@ impl SegmentReader {
self.segment_id
}
/// Returns the delete opstamp
pub fn delete_opstamp(&self) -> Option<Opstamp> {
self.delete_opstamp
}
/// Returns the bitset representing
/// the documents that have been deleted.
pub fn alive_bitset(&self) -> Option<&AliveBitSet> {

24
src/directory/composite_file.rs
View File

@@ -1,17 +1,14 @@
use crate::directory::FileSlice;
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::HasLen;
use common::VInt;
use std::collections::HashMap;
use std::io::{self, Read, Write};
use std::iter::ExactSizeIterator;
use std::ops::Range;
use common::{BinarySerializable, CountingWriter, HasLen, VInt};
use crate::directory::{FileSlice, TerminatingWrite, WritePtr};
use crate::schema::Field;
use crate::space_usage::{FieldUsage, PerFieldSpaceUsage};
#[derive(Eq, PartialEq, Hash, Copy, Ord, PartialOrd, Clone, Debug)]
pub struct FileAddr {
field: Field,
@@ -186,13 +183,14 @@ impl CompositeFile {
#[cfg(test)]
mod test {
use std::io::Write;
use std::path::Path;
use common::{BinarySerializable, VInt};
use super::{CompositeFile, CompositeWrite};
use crate::directory::{Directory, RamDirectory};
use crate::schema::Field;
use common::BinarySerializable;
use common::VInt;
use std::io::Write;
use std::path::Path;
#[test]
fn test_composite_file() -> crate::Result<()> {

43
src/directory/directory.rs
View File

@@ -1,18 +1,12 @@
use crate::directory::directory_lock::Lock;
use crate::directory::error::LockError;
use crate::directory::error::{DeleteError, OpenReadError, OpenWriteError};
use crate::directory::WatchHandle;
use crate::directory::{FileHandle, WatchCallback};
use crate::directory::{FileSlice, 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::thread;
use std::marker::{Send, Sync};
use std::path::{Path, PathBuf};
use std::time::Duration;
use std::{fmt, io, thread};
use crate::directory::directory_lock::Lock;
use crate::directory::error::{DeleteError, LockError, OpenReadError, OpenWriteError};
use crate::directory::{FileHandle, FileSlice, WatchCallback, WatchHandle, WritePtr};
/// Retry the logic of acquiring locks is pretty simple.
/// We just retry `n` times after a given `duratio`, both
@@ -43,10 +37,8 @@ impl RetryPolicy {
}
/// 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`.
/// It is associated to a lock file, that gets deleted on `Drop.`
pub struct DirectoryLock(Box<dyn Send + Sync + 'static>);
struct DirectoryLockGuard {
@@ -142,10 +134,16 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
/// Opens a writer for the *virtual file* associated with
/// a Path.
///
/// Right after this call, the file should be created
/// and any subsequent call to `open_read` for the
/// Right after this call, for the span of the execution of the program
/// the file should be created and any subsequent call to `open_read` for the
/// same path should return a `FileSlice`.
///
/// However, depending on the directory implementation,
/// it might be required to call `sync_directory` to ensure
/// that the file is durably created.
/// (The semantics here are the same when dealing with
/// a posix filesystem.)
///
/// Write operations may be aggressively buffered.
/// The client of this trait is responsible for calling flush
/// to ensure that subsequent `read` operations
@@ -176,6 +174,12 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
/// The file may or may not previously exist.
fn atomic_write(&self, path: &Path, data: &[u8]) -> io::Result<()>;
/// Sync the directory.
///
/// This call is required to ensure that newly created files are
/// effectively stored durably.
fn sync_directory(&self) -> io::Result<()>;
/// Acquire a lock in the given directory.
///
/// The method is blocking or not depending on the `Lock` object.
@@ -223,8 +227,7 @@ pub trait DirectoryClone {
}
impl<T> DirectoryClone for T
where
T: 'static + Directory + Clone,
where T: 'static + Directory + Clone
{
fn box_clone(&self) -> Box<dyn Directory> {
Box::new(self.clone())

8
src/directory/directory_lock.rs
View File

@@ -1,17 +1,17 @@
use once_cell::sync::Lazy;
use std::path::PathBuf;
use once_cell::sync::Lazy;
/// 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)
/// - [INDEX_WRITER_LOCK]
/// - [META_LOCK]
///
/// Check out these locks documentation for more information.
///
#[derive(Debug)]
pub struct Lock {
/// The lock needs to be associated with its own file `path`.

22
src/directory/error.rs
View File

@@ -1,15 +1,17 @@
use crate::Version;
use std::fmt;
use std::io;
use std::path::PathBuf;
use std::{fmt, io};
use crate::Version;
/// Error while trying to acquire a directory lock.
#[derive(Debug, Error)]
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.
/// - 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.
#[error("Could not acquire lock as it is already held, possibly by a different process.")]
LockBusy,
/// Trying to acquire a lock failed with an `IoError`
@@ -39,6 +41,16 @@ pub enum OpenDirectoryError {
},
}
impl OpenDirectoryError {
/// Wraps an io error.
pub fn wrap_io_error(io_error: io::Error, directory_path: PathBuf) -> Self {
Self::IoError {
io_error,
directory_path,
}
}
}
/// Error that may occur when starting to write in a file
#[derive(Debug, Error)]
pub enum OpenWriteError {

25
src/directory/file_slice.rs
View File

@@ -1,11 +1,11 @@
use std::ops::{Deref, Range};
use std::sync::{Arc, Weak};
use std::{fmt, io};
use common::HasLen;
use stable_deref_trait::StableDeref;
use crate::directory::OwnedBytes;
use common::HasLen;
use std::fmt;
use std::ops::Range;
use std::sync::{Arc, Weak};
use std::{io, ops::Deref};
pub type ArcBytes = Arc<dyn Deref<Target = [u8]> + Send + Sync + 'static>;
pub type WeakArcBytes = Weak<dyn Deref<Target = [u8]> + Send + Sync + 'static>;
@@ -33,8 +33,7 @@ impl FileHandle for &'static [u8] {
}
impl<B> From<B> for FileSlice
where
B: StableDeref + Deref<Target = [u8]> + 'static + Send + Sync,
where B: StableDeref + Deref<Target = [u8]> + 'static + Send + Sync
{
fn from(bytes: B) -> FileSlice {
FileSlice::new(Box::new(OwnedBytes::new(bytes)))
@@ -44,7 +43,6 @@ where
/// Logical slice of read only file in tantivy.
///
/// It can be cloned and sliced cheaply.
///
#[derive(Clone)]
pub struct FileSlice {
data: Arc<dyn FileHandle>,
@@ -66,6 +64,7 @@ impl FileSlice {
/// Wraps a FileHandle.
#[doc(hidden)]
#[must_use]
pub fn new_with_num_bytes(file_handle: Box<dyn FileHandle>, num_bytes: usize) -> Self {
FileSlice {
data: Arc::from(file_handle),
@@ -78,6 +77,7 @@ impl FileSlice {
/// # Panics
///
/// Panics if `byte_range.end` exceeds the filesize.
#[must_use]
pub fn slice(&self, byte_range: Range<usize>) -> FileSlice {
assert!(byte_range.end <= self.len());
FileSlice {
@@ -137,6 +137,7 @@ impl FileSlice {
/// boundary.
///
/// Equivalent to `.slice(from_offset, self.len())`
#[must_use]
pub fn slice_from(&self, from_offset: usize) -> FileSlice {
self.slice(from_offset..self.len())
}
@@ -144,6 +145,7 @@ impl FileSlice {
/// Returns a slice from the end.
///
/// Equivalent to `.slice(self.len() - from_offset, self.len())`
#[must_use]
pub fn slice_from_end(&self, from_offset: usize) -> FileSlice {
self.slice(self.len() - from_offset..self.len())
}
@@ -152,6 +154,7 @@ impl FileSlice {
/// boundary.
///
/// Equivalent to `.slice(0, to_offset)`
#[must_use]
pub fn slice_to(&self, to_offset: usize) -> FileSlice {
self.slice(0..to_offset)
}
@@ -171,10 +174,12 @@ impl HasLen for FileSlice {
#[cfg(test)]
mod tests {
use super::{FileHandle, FileSlice};
use common::HasLen;
use std::io;
use common::HasLen;
use super::{FileHandle, FileSlice};
#[test]
fn test_file_slice() -> io::Result<()> {
let file_slice = FileSlice::new(Box::new(b"abcdef".as_ref()));

23
src/directory/file_watcher.rs
View File

@@ -1,13 +1,13 @@
use crate::directory::{WatchCallback, WatchCallbackList, WatchHandle};
use crc32fast::Hasher;
use std::fs;
use std::io;
use std::io::BufRead;
use std::path::Path;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::sync::Arc;
use std::thread;
use std::time::Duration;
use std::{fs, io, thread};
use crc32fast::Hasher;
use crate::directory::{WatchCallback, WatchCallbackList, WatchHandle};
pub const POLLING_INTERVAL: Duration = Duration::from_millis(if cfg!(test) { 1 } else { 500 });
@@ -43,14 +43,16 @@ impl FileWatcher {
thread::Builder::new()
.name("thread-tantivy-meta-file-watcher".to_string())
.spawn(move || {
let mut current_checksum = None;
let mut current_checksum_opt = None;
while state.load(Ordering::SeqCst) == 1 {
if let Ok(checksum) = FileWatcher::compute_checksum(&path) {
// `None.unwrap_or_else(|| !checksum) != checksum` evaluates to `true`
if current_checksum.unwrap_or_else(|| !checksum) != checksum {
let metafile_has_changed = current_checksum_opt
.map(|current_checksum| current_checksum != checksum)
.unwrap_or(true);
if metafile_has_changed {
info!("Meta file {:?} was modified", path);
current_checksum = Some(checksum);
current_checksum_opt = Some(checksum);
futures::executor::block_on(callbacks.broadcast());
}
}
@@ -97,9 +99,8 @@ mod tests {
use std::mem;
use crate::directory::mmap_directory::atomic_write;
use super::*;
use crate::directory::mmap_directory::atomic_write;
#[test]
fn test_file_watcher_drop_watcher() -> crate::Result<()> {

39
src/directory/footer.rs
View File

@@ -1,14 +1,13 @@
use crate::directory::error::Incompatibility;
use crate::directory::FileSlice;
use crate::{
directory::{AntiCallToken, TerminatingWrite},
Version, INDEX_FORMAT_VERSION,
};
use std::io;
use std::io::Write;
use common::{BinarySerializable, CountingWriter, DeserializeFrom, FixedSize, HasLen};
use crc32fast::Hasher;
use serde::{Deserialize, Serialize};
use std::io;
use std::io::Write;
use crate::directory::error::Incompatibility;
use crate::directory::{AntiCallToken, FileSlice, TerminatingWrite};
use crate::{Version, INDEX_FORMAT_VERSION};
const FOOTER_MAX_LEN: u32 = 50_000;
@@ -64,7 +63,9 @@ impl Footer {
if footer_magic_byte != FOOTER_MAGIC_NUMBER {
return Err(io::Error::new(
io::ErrorKind::InvalidData,
"Footer magic byte mismatch. File corrupted or index was created using old an tantivy version which is not supported anymore. Please use tantivy 0.15 or above to recreate the index.",
"Footer magic byte mismatch. File corrupted or index was created using old an \
tantivy version which is not supported anymore. Please use tantivy 0.15 or above \
to recreate the index.",
));
}
@@ -73,7 +74,7 @@ impl Footer {
io::ErrorKind::InvalidData,
format!(
"Footer seems invalid as it suggests a footer len of {}. File is corrupted, \
or the index was created with a different & old version of tantivy.",
or the index was created with a different & old version of tantivy.",
footer_len
),
));
@@ -154,12 +155,13 @@ impl<W: TerminatingWrite> TerminatingWrite for FooterProxy<W> {
#[cfg(test)]
mod tests {
use crate::directory::footer::Footer;
use crate::directory::OwnedBytes;
use crate::directory::{footer::FOOTER_MAGIC_NUMBER, FileSlice};
use common::BinarySerializable;
use std::io;
use common::BinarySerializable;
use crate::directory::footer::{Footer, FOOTER_MAGIC_NUMBER};
use crate::directory::{FileSlice, OwnedBytes};
#[test]
fn test_deserialize_footer() {
let mut buf: Vec<u8> = vec![];
@@ -183,8 +185,9 @@ mod tests {
let err = Footer::extract_footer(fileslice).unwrap_err();
assert_eq!(
err.to_string(),
"Footer magic byte mismatch. File corrupted or index was created using old an tantivy version which \
is not supported anymore. Please use tantivy 0.15 or above to recreate the index."
"Footer magic byte mismatch. File corrupted or index was created using old an tantivy \
version which is not supported anymore. Please use tantivy 0.15 or above to recreate \
the index."
);
}
#[test]
@@ -219,8 +222,8 @@ mod tests {
assert_eq!(err.kind(), io::ErrorKind::InvalidData);
assert_eq!(
err.to_string(),
"Footer seems invalid as it suggests a footer len of 50001. File is corrupted, \
or the index was created with a different & old version of tantivy."
"Footer seems invalid as it suggests a footer len of 50001. File is corrupted, or the \
index was created with a different & old version of tantivy."
);
}
}

54
src/directory/managed_directory.rs
View File

@@ -1,24 +1,21 @@
use std::collections::HashSet;
use std::io::Write;
use std::path::{Path, PathBuf};
use std::sync::{Arc, RwLock, RwLockWriteGuard};
use std::{io, result};
use crc32fast::Hasher;
use crate::core::MANAGED_FILEPATH;
use crate::directory::error::{DeleteError, LockError, OpenReadError, OpenWriteError};
use crate::directory::footer::{Footer, FooterProxy};
use crate::directory::GarbageCollectionResult;
use crate::directory::Lock;
use crate::directory::META_LOCK;
use crate::directory::{DirectoryLock, FileHandle};
use crate::directory::{FileSlice, WritePtr};
use crate::directory::{WatchCallback, WatchHandle};
use crate::directory::{
DirectoryLock, FileHandle, FileSlice, GarbageCollectionResult, Lock, WatchCallback,
WatchHandle, WritePtr, META_LOCK,
};
use crate::error::DataCorruption;
use crate::Directory;
use crc32fast::Hasher;
use std::collections::HashSet;
use std::io;
use std::io::Write;
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.
///
@@ -192,6 +189,7 @@ impl ManagedDirectory {
for delete_file in &deleted_files {
managed_paths_write.remove(delete_file);
}
self.directory.sync_directory()?;
save_managed_paths(self.directory.as_mut(), &meta_informations_wlock)?;
}
@@ -222,9 +220,22 @@ impl ManagedDirectory {
.write()
.expect("Managed file lock poisoned");
let has_changed = meta_wlock.managed_paths.insert(filepath.to_owned());
if has_changed {
save_managed_paths(self.directory.as_ref(), &meta_wlock)?;
if !has_changed {
return Ok(());
}
save_managed_paths(self.directory.as_ref(), &meta_wlock)?;
// This is not the first file we add.
// Therefore, we are sure that `.managed.json` has been already
// properly created and we do not need to sync its parent directory.
//
// (It might seem like a nicer solution to create the managed_json on the
// creation of the ManagedDirectory instance but it would actually
// prevent the use of read-only directories..)
let managed_file_definitely_already_exists = meta_wlock.managed_paths.len() > 1;
if managed_file_definitely_already_exists {
return Ok(());
}
self.directory.sync_directory()?;
Ok(())
}
@@ -310,6 +321,11 @@ impl Directory for ManagedDirectory {
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
self.directory.watch(watch_callback)
}
fn sync_directory(&self) -> io::Result<()> {
self.directory.sync_directory()?;
Ok(())
}
}
impl Clone for ManagedDirectory {
@@ -325,12 +341,14 @@ impl Clone for ManagedDirectory {
#[cfg(test)]
mod tests_mmap_specific {
use crate::directory::{Directory, ManagedDirectory, MmapDirectory, TerminatingWrite};
use std::collections::HashSet;
use std::io::Write;
use std::path::{Path, PathBuf};
use tempfile::TempDir;
use crate::directory::{Directory, ManagedDirectory, MmapDirectory, TerminatingWrite};
#[test]
fn test_managed_directory() {
let tempdir = TempDir::new().unwrap();

165
src/directory/mmap_directory.rs
View File

@@ -1,32 +1,28 @@
use crate::core::META_FILEPATH;
use crate::directory::error::LockError;
use crate::directory::error::{DeleteError, OpenDirectoryError, OpenReadError, OpenWriteError};
use crate::directory::file_watcher::FileWatcher;
use crate::directory::Directory;
use crate::directory::DirectoryLock;
use crate::directory::Lock;
use crate::directory::WatchCallback;
use crate::directory::WatchHandle;
use crate::directory::{AntiCallToken, FileHandle, OwnedBytes};
use crate::directory::{ArcBytes, WeakArcBytes};
use crate::directory::{TerminatingWrite, WritePtr};
use std::collections::HashMap;
use std::convert::From;
use std::fs::{self, File, OpenOptions};
use std::io::{self, BufWriter, Read, Seek, SeekFrom, Write};
use std::ops::Deref;
use std::path::{Path, PathBuf};
use std::sync::{Arc, RwLock};
use std::{fmt, result};
use fs2::FileExt;
use memmap2::Mmap;
use serde::{Deserialize, Serialize};
use stable_deref_trait::StableDeref;
use std::convert::From;
use std::fmt;
use std::fs::OpenOptions;
use std::fs::{self, File};
use std::io::{self, Seek, SeekFrom};
use std::io::{BufWriter, Read, Write};
use std::path::{Path, PathBuf};
use std::result;
use std::sync::Arc;
use std::sync::RwLock;
use std::{collections::HashMap, ops::Deref};
use tempfile::TempDir;
use crate::core::META_FILEPATH;
use crate::directory::error::{
DeleteError, LockError, OpenDirectoryError, OpenReadError, OpenWriteError,
};
use crate::directory::file_watcher::FileWatcher;
use crate::directory::{
AntiCallToken, ArcBytes, Directory, DirectoryLock, FileHandle, Lock, OwnedBytes,
TerminatingWrite, WatchCallback, WatchHandle, WeakArcBytes, WritePtr,
};
/// 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)
@@ -193,16 +189,19 @@ impl MmapDirectory {
pub fn open<P: AsRef<Path>>(directory_path: P) -> Result<MmapDirectory, OpenDirectoryError> {
let directory_path: &Path = directory_path.as_ref();
if !directory_path.exists() {
Err(OpenDirectoryError::DoesNotExist(PathBuf::from(
return Err(OpenDirectoryError::DoesNotExist(PathBuf::from(
directory_path,
)))
} else if !directory_path.is_dir() {
Err(OpenDirectoryError::NotADirectory(PathBuf::from(
directory_path,
)))
} else {
Ok(MmapDirectory::new(PathBuf::from(directory_path), None))
)));
}
let canonical_path: PathBuf = directory_path.canonicalize().map_err(|io_err| {
OpenDirectoryError::wrap_io_error(io_err, PathBuf::from(directory_path))
})?;
if !canonical_path.is_dir() {
return Err(OpenDirectoryError::NotADirectory(PathBuf::from(
directory_path,
)));
}
Ok(MmapDirectory::new(canonical_path, None))
}
/// Joins a relative_path to the directory `root_path`
@@ -211,33 +210,6 @@ impl MmapDirectory {
self.inner.root_path.join(relative_path)
}
/// Sync the root directory.
/// In certain FS, this is required to persistently create
/// a file.
fn sync_directory(&self) -> Result<(), io::Error> {
let mut open_opts = OpenOptions::new();
// Linux needs read to be set, otherwise returns EINVAL
// write must not be set, or it fails with EISDIR
open_opts.read(true);
// On Windows, opening a directory requires FILE_FLAG_BACKUP_SEMANTICS
// and calling sync_all() only works if write access is requested.
#[cfg(windows)]
{
use std::os::windows::fs::OpenOptionsExt;
use winapi::um::winbase;
open_opts
.write(true)
.custom_flags(winbase::FILE_FLAG_BACKUP_SEMANTICS);
}
let fd = open_opts.open(&self.inner.root_path)?;
fd.sync_data()?;
Ok(())
}
/// Returns some statistical information
/// about the Mmap cache.
///
@@ -344,8 +316,7 @@ impl Directory for MmapDirectory {
let mut mmap_cache = self.inner.mmap_cache.write().map_err(|_| {
let msg = format!(
"Failed to acquired write lock \
on mmap cache while reading {:?}",
"Failed to acquired write lock on mmap cache while reading {:?}",
path
);
let io_err = make_io_err(msg);
@@ -367,22 +338,17 @@ impl Directory for MmapDirectory {
/// 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| DeleteError::IoError {
io_error: e,
filepath: path.to_path_buf(),
}),
Err(e) => {
if e.kind() == io::ErrorKind::NotFound {
Err(DeleteError::FileDoesNotExist(path.to_owned()))
} else {
Err(DeleteError::IoError {
io_error: e,
filepath: path.to_path_buf(),
})
fs::remove_file(&full_path).map_err(|e| {
if e.kind() == io::ErrorKind::NotFound {
DeleteError::FileDoesNotExist(path.to_owned())
} else {
DeleteError::IoError {
io_error: e,
filepath: path.to_path_buf(),
}
}
}
})?;
Ok(())
}
fn exists(&self, path: &Path) -> Result<bool, OpenReadError> {
@@ -411,10 +377,13 @@ impl Directory for MmapDirectory {
file.flush()
.map_err(|io_error| OpenWriteError::wrap_io_error(io_error, path.to_path_buf()))?;
// Apparetntly, on some filesystem syncing the parent
// directory is required.
self.sync_directory()
.map_err(|io_err| OpenWriteError::wrap_io_error(io_err, path.to_path_buf()))?;
// Note we actually do not sync the parent directory here.
//
// A newly created file, may, in some case, be created and even flushed to disk.
// and then lost...
//
// The file will only be durably written after we terminate AND
// sync_directory() is called.
let writer = SafeFileWriter::new(file);
Ok(BufWriter::new(Box::new(writer)))
@@ -444,7 +413,7 @@ impl Directory for MmapDirectory {
debug!("Atomic Write {:?}", path);
let full_path = self.resolve_path(path);
atomic_write(&full_path, content)?;
self.sync_directory()
Ok(())
}
fn acquire_lock(&self, lock: &Lock) -> Result<DirectoryLock, LockError> {
@@ -470,6 +439,31 @@ impl Directory for MmapDirectory {
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
Ok(self.inner.watch(watch_callback))
}
fn sync_directory(&self) -> Result<(), io::Error> {
let mut open_opts = OpenOptions::new();
// Linux needs read to be set, otherwise returns EINVAL
// write must not be set, or it fails with EISDIR
open_opts.read(true);
// On Windows, opening a directory requires FILE_FLAG_BACKUP_SEMANTICS
// and calling sync_all() only works if write access is requested.
#[cfg(windows)]
{
use std::os::windows::fs::OpenOptionsExt;
use winapi::um::winbase;
open_opts
.write(true)
.custom_flags(winbase::FILE_FLAG_BACKUP_SEMANTICS);
}
let fd = open_opts.open(&self.inner.root_path)?;
fd.sync_data()?;
Ok(())
}
}
#[cfg(test)]
@@ -478,15 +472,12 @@ mod tests {
// There are more tests in directory/mod.rs
// The following tests are specific to the MmapDirectory
use common::HasLen;
use super::*;
use crate::indexer::LogMergePolicy;
use crate::Index;
use crate::ReloadPolicy;
use crate::{
schema::{Schema, SchemaBuilder, TEXT},
IndexSettings,
};
use common::HasLen;
use crate::schema::{Schema, SchemaBuilder, TEXT};
use crate::{Index, IndexSettings, ReloadPolicy};
#[test]
fn test_open_non_existent_path() {
@@ -523,7 +514,7 @@ mod tests {
{
for path in &paths {
let mut w = mmap_directory.open_write(path).unwrap();
w.write(content).unwrap();
w.write_all(content).unwrap();
w.flush().unwrap();
}
}

21
src/directory/mod.rs
View File

@@ -1,8 +1,4 @@
/*!
WORM directory abstraction.
*/
//! WORM (Write Once Read Many) directory abstraction.
#[cfg(feature = "mmap")]
mod mmap_directory;
@@ -22,19 +18,19 @@ pub mod error;
mod composite_file;
use std::io::BufWriter;
use std::path::PathBuf;
pub use common::{AntiCallToken, TerminatingWrite};
pub(crate) use self::composite_file::{CompositeFile, CompositeWrite};
pub use self::directory::DirectoryLock;
pub use self::directory::{Directory, DirectoryClone};
pub use self::directory::{Directory, DirectoryClone, DirectoryLock};
pub use self::directory_lock::{Lock, INDEX_WRITER_LOCK, META_LOCK};
pub(crate) use self::file_slice::{ArcBytes, WeakArcBytes};
pub use self::file_slice::{FileHandle, FileSlice};
pub use self::owned_bytes::OwnedBytes;
pub use self::ram_directory::RamDirectory;
pub use self::watch_event_router::{WatchCallback, WatchCallbackList, WatchHandle};
pub use common::AntiCallToken;
pub use common::TerminatingWrite;
use std::io::BufWriter;
use std::path::PathBuf;
/// Outcome of the Garbage collection
pub struct GarbageCollectionResult {
@@ -50,11 +46,10 @@ pub struct GarbageCollectionResult {
pub failed_to_delete_files: Vec<PathBuf>,
}
pub use self::managed_directory::ManagedDirectory;
#[cfg(feature = "mmap")]
pub use self::mmap_directory::MmapDirectory;
pub use self::managed_directory::ManagedDirectory;
/// Write object for Directory.
///
/// `WritePtr` are required to implement both Write

3
src/directory/owned_bytes.rs
View File

@@ -1,9 +1,10 @@
use crate::directory::FileHandle;
use std::io;
use std::ops::Range;
pub use ownedbytes::OwnedBytes;
use crate::directory::FileHandle;
impl FileHandle for OwnedBytes {
fn read_bytes(&self, range: Range<usize>) -> io::Result<OwnedBytes> {
Ok(self.slice(range))

34
src/directory/ram_directory.rs
View File

@@ -1,19 +1,19 @@
use crate::core::META_FILEPATH;
use crate::directory::error::{DeleteError, OpenReadError, OpenWriteError};
use crate::directory::AntiCallToken;
use crate::directory::WatchCallbackList;
use crate::directory::{Directory, FileSlice, WatchCallback, WatchHandle};
use crate::directory::{TerminatingWrite, WritePtr};
use common::HasLen;
use fail::fail_point;
use std::collections::HashMap;
use std::fmt;
use std::io::{self, BufWriter, Cursor, Seek, SeekFrom, Write};
use std::path::{Path, PathBuf};
use std::result;
use std::sync::{Arc, RwLock};
use std::{fmt, result};
use common::HasLen;
use fail::fail_point;
use super::FileHandle;
use crate::core::META_FILEPATH;
use crate::directory::error::{DeleteError, OpenReadError, OpenWriteError};
use crate::directory::{
AntiCallToken, Directory, FileSlice, TerminatingWrite, WatchCallback, WatchCallbackList,
WatchHandle, WritePtr,
};
/// Writer associated with the `RamDirectory`
///
@@ -40,7 +40,9 @@ impl Drop for VecWriter {
fn drop(&mut self) {
if !self.is_flushed {
warn!(
"You forgot to flush {:?} before its writter got Drop. Do not rely on drop. This also occurs when the indexer crashed, so you may want to check the logs for the root cause.",
"You forgot to flush {:?} before its writter got Drop. Do not rely on drop. This \
also occurs when the indexer crashed, so you may want to check the logs for the \
root cause.",
self.path
)
}
@@ -123,7 +125,6 @@ impl fmt::Debug for RamDirectory {
///
/// It is mainly meant for unit testing.
/// Writes are only made visible upon flushing.
///
#[derive(Clone, Default)]
pub struct RamDirectory {
fs: Arc<RwLock<InnerDirectory>>,
@@ -225,15 +226,20 @@ impl Directory for RamDirectory {
fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
Ok(self.fs.write().unwrap().watch(watch_callback))
}
fn sync_directory(&self) -> io::Result<()> {
Ok(())
}
}
#[cfg(test)]
mod tests {
use super::RamDirectory;
use crate::Directory;
use std::io::Write;
use std::path::Path;
use super::RamDirectory;
use crate::Directory;
#[test]
fn test_persist() {
let msg_atomic: &'static [u8] = b"atomic is the way";

8
src/directory/tests.rs
View File

@@ -1,6 +1,3 @@
use super::*;
use futures::channel::oneshot;
use futures::executor::block_on;
use std::io::Write;
use std::mem;
use std::path::{Path, PathBuf};
@@ -9,6 +6,11 @@ use std::sync::atomic::{AtomicBool, AtomicUsize};
use std::sync::Arc;
use std::time::Duration;
use futures::channel::oneshot;
use futures::executor::block_on;
use super::*;
#[cfg(feature = "mmap")]
mod mmap_directory_tests {
use crate::directory::MmapDirectory;

11
src/directory/watch_event_router.rs
View File

@@ -1,8 +1,7 @@
use std::sync::{Arc, RwLock, Weak};
use futures::channel::oneshot;
use futures::{Future, TryFutureExt};
use std::sync::Arc;
use std::sync::RwLock;
use std::sync::Weak;
/// Cloneable wrapper for callbacks registered when watching files of a `Directory`.
#[derive(Clone)]
@@ -103,12 +102,14 @@ impl WatchCallbackList {
#[cfg(test)]
mod tests {
use crate::directory::{WatchCallback, WatchCallbackList};
use futures::executor::block_on;
use std::mem;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::sync::Arc;
use futures::executor::block_on;
use crate::directory::{WatchCallback, WatchCallbackList};
#[test]
fn test_watch_event_router_simple() {
let watch_event_router = WatchCallbackList::default();

4
src/docset.rs
View File

@@ -1,7 +1,7 @@
use std::borrow::{Borrow, BorrowMut};
use crate::fastfield::AliveBitSet;
use crate::DocId;
use std::borrow::Borrow;
use std::borrow::BorrowMut;
/// Sentinel value returned when a DocSet has been entirely consumed.
///

17
src/error.rs
View File

@@ -1,17 +1,14 @@
//! Definition of Tantivy's error and result.
use std::io;
use crate::directory::error::{Incompatibility, LockError};
use crate::fastfield::FastFieldNotAvailableError;
use crate::query;
use crate::{
directory::error::{OpenDirectoryError, OpenReadError, OpenWriteError},
schema,
};
use std::fmt;
use std::path::PathBuf;
use std::sync::PoisonError;
use std::{fmt, io};
use crate::directory::error::{
Incompatibility, LockError, OpenDirectoryError, OpenReadError, OpenWriteError,
};
use crate::fastfield::FastFieldNotAvailableError;
use crate::{query, schema};
/// Represents a `DataCorruption` error.
///

15
src/fastfield/alive_bitset.rs
View File

@@ -1,12 +1,12 @@
use crate::space_usage::ByteCount;
use crate::DocId;
use common::intersect_bitsets;
use common::BitSet;
use common::ReadOnlyBitSet;
use ownedbytes::OwnedBytes;
use std::io;
use std::io::Write;
use common::{intersect_bitsets, BitSet, ReadOnlyBitSet};
use ownedbytes::OwnedBytes;
use crate::space_usage::ByteCount;
use crate::DocId;
/// Write a alive `BitSet`
///
/// where `alive_bitset` is the set of alive `DocId`.
@@ -168,11 +168,12 @@ mod tests {
#[cfg(all(test, feature = "unstable"))]
mod bench {
use super::AliveBitSet;
use rand::prelude::IteratorRandom;
use rand::thread_rng;
use test::Bencher;
use super::AliveBitSet;
fn get_alive() -> Vec<u32> {
let mut data = (0..1_000_000_u32).collect::<Vec<u32>>();
for _ in 0..(1_000_000) * 1 / 8 {

9
src/fastfield/bytes/mod.rs
View File

@@ -6,11 +6,12 @@ pub use self::writer::BytesFastFieldWriter;
#[cfg(test)]
mod tests {
use crate::schema::{BytesOptions, IndexRecordOption, Schema, Value};
use crate::{query::TermQuery, schema::FAST, schema::INDEXED, schema::STORED};
use crate::{DocAddress, DocSet, Index, Searcher, Term};
use std::ops::Deref;
use crate::query::TermQuery;
use crate::schema::{BytesOptions, IndexRecordOption, Schema, Value, FAST, INDEXED, STORED};
use crate::{DocAddress, DocSet, Index, Searcher, Term};
#[test]
fn test_bytes() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
@@ -62,7 +63,7 @@ mod tests {
assert_eq!(values.len(), 2);
let values_bytes: Vec<&[u8]> = values
.into_iter()
.flat_map(|value| value.bytes_value())
.flat_map(|value| value.as_bytes())
.collect();
assert_eq!(values_bytes, &[&b"tantivy"[..], &b"lucene"[..]]);
Ok(())

9
src/fastfield/bytes/reader.rs
View File

@@ -1,6 +1,5 @@
use crate::directory::FileSlice;
use crate::directory::OwnedBytes;
use crate::fastfield::{BitpackedFastFieldReader, FastFieldReader, MultiValueLength};
use crate::directory::{FileSlice, OwnedBytes};
use crate::fastfield::{DynamicFastFieldReader, FastFieldReader, MultiValueLength};
use crate::DocId;
/// Reader for byte array fast fields
@@ -15,13 +14,13 @@ use crate::DocId;
/// and the start index for the next document, and keeping the bytes in between.
#[derive(Clone)]
pub struct BytesFastFieldReader {
idx_reader: BitpackedFastFieldReader<u64>,
idx_reader: DynamicFastFieldReader<u64>,
values: OwnedBytes,
}
impl BytesFastFieldReader {
pub(crate) fn open(
idx_reader: BitpackedFastFieldReader<u64>,
idx_reader: DynamicFastFieldReader<u64>,
values_file: FileSlice,
) -> crate::Result<BytesFastFieldReader> {
let values = values_file.read_bytes()?;

5
src/fastfield/bytes/writer.rs
View File

@@ -1,10 +1,9 @@
use std::io;
use crate::fastfield::serializer::CompositeFastFieldSerializer;
use crate::indexer::doc_id_mapping::DocIdMapping;
use crate::schema::{Document, Field, Value};
use crate::DocId;
use crate::{
fastfield::serializer::CompositeFastFieldSerializer, indexer::doc_id_mapping::DocIdMapping,
};
/// Writer for byte array (as in, any number of bytes per document) fast fields
///

3
src/fastfield/error.rs
View File

@@ -1,6 +1,7 @@
use crate::schema::FieldEntry;
use std::result;
use crate::schema::FieldEntry;
/// `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.

113
src/fastfield/facet_reader.rs
View File

@@ -1,10 +1,10 @@
use std::str;
use super::MultiValuedFastFieldReader;
use crate::error::DataCorruption;
use crate::schema::Facet;
use crate::termdict::TermDictionary;
use crate::termdict::TermOrdinal;
use crate::termdict::{TermDictionary, TermOrdinal};
use crate::DocId;
use std::str;
/// The facet reader makes it possible to access the list of
/// facets associated to a given document in a specific
@@ -82,83 +82,11 @@ impl FacetReader {
#[cfg(test)]
mod tests {
use crate::Index;
use crate::{
schema::{Facet, FacetOptions, SchemaBuilder, Value, INDEXED, STORED},
DocAddress, Document,
};
use crate::schema::{Facet, FacetOptions, SchemaBuilder, Value, STORED};
use crate::{DocAddress, Document, Index};
#[test]
fn test_facet_only_indexed() -> crate::Result<()> {
let mut schema_builder = SchemaBuilder::default();
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
index_writer.add_document(doc!(facet_field=>Facet::from_text("/a/b").unwrap()))?;
index_writer.commit()?;
let searcher = index.reader()?.searcher();
let facet_reader = searcher
.segment_reader(0u32)
.facet_reader(facet_field)
.unwrap();
let mut facet_ords = Vec::new();
facet_reader.facet_ords(0u32, &mut facet_ords);
assert_eq!(&facet_ords, &[2u64]);
let doc = searcher.doc(DocAddress::new(0u32, 0u32))?;
let value = doc.get_first(facet_field).and_then(Value::path);
assert_eq!(value, None);
Ok(())
}
#[test]
fn test_facet_only_stored() -> crate::Result<()> {
let mut schema_builder = SchemaBuilder::default();
let facet_field = schema_builder.add_facet_field("facet", STORED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
index_writer.add_document(doc!(facet_field=>Facet::from_text("/a/b").unwrap()))?;
index_writer.commit()?;
let searcher = index.reader()?.searcher();
let facet_reader = searcher
.segment_reader(0u32)
.facet_reader(facet_field)
.unwrap();
let mut facet_ords = Vec::new();
facet_reader.facet_ords(0u32, &mut facet_ords);
assert!(facet_ords.is_empty());
let doc = searcher.doc(DocAddress::new(0u32, 0u32))?;
let value = doc.get_first(facet_field).and_then(Value::path);
assert_eq!(value, Some("/a/b".to_string()));
Ok(())
}
#[test]
fn test_facet_stored_and_indexed() -> crate::Result<()> {
let mut schema_builder = SchemaBuilder::default();
let facet_field = schema_builder.add_facet_field("facet", STORED | INDEXED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
index_writer.add_document(doc!(facet_field=>Facet::from_text("/a/b").unwrap()))?;
index_writer.commit()?;
let searcher = index.reader()?.searcher();
let facet_reader = searcher
.segment_reader(0u32)
.facet_reader(facet_field)
.unwrap();
let mut facet_ords = Vec::new();
facet_reader.facet_ords(0u32, &mut facet_ords);
assert_eq!(&facet_ords, &[2u64]);
let doc = searcher.doc(DocAddress::new(0u32, 0u32))?;
let value = doc.get_first(facet_field).and_then(Value::path);
assert_eq!(value, Some("/a/b".to_string()));
Ok(())
}
#[test]
fn test_facet_neither_stored_and_indexed() -> crate::Result<()> {
let mut schema_builder = SchemaBuilder::default();
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
@@ -173,17 +101,40 @@ mod tests {
.unwrap();
let mut facet_ords = Vec::new();
facet_reader.facet_ords(0u32, &mut facet_ords);
assert!(facet_ords.is_empty());
assert_eq!(&facet_ords, &[2u64]);
let doc = searcher.doc(DocAddress::new(0u32, 0u32))?;
let value = doc.get_first(facet_field).and_then(Value::path);
let value = doc.get_first(facet_field).and_then(Value::as_facet);
assert_eq!(value, None);
Ok(())
}
#[test]
fn test_facet_stored_and_indexed() -> crate::Result<()> {
let mut schema_builder = SchemaBuilder::default();
let facet_field = schema_builder.add_facet_field("facet", STORED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
index_writer.add_document(doc!(facet_field=>Facet::from_text("/a/b").unwrap()))?;
index_writer.commit()?;
let searcher = index.reader()?.searcher();
let facet_reader = searcher
.segment_reader(0u32)
.facet_reader(facet_field)
.unwrap();
let mut facet_ords = Vec::new();
facet_reader.facet_ords(0u32, &mut facet_ords);
assert_eq!(&facet_ords, &[2u64]);
let doc = searcher.doc(DocAddress::new(0u32, 0u32))?;
let value: Option<&Facet> = doc.get_first(facet_field).and_then(Value::as_facet);
assert_eq!(value, Facet::from_text("/a/b").ok().as_ref());
Ok(())
}
#[test]
fn test_facet_not_populated_for_all_docs() -> crate::Result<()> {
let mut schema_builder = SchemaBuilder::default();
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
@@ -206,7 +157,7 @@ mod tests {
#[test]
fn test_facet_not_populated_for_any_docs() -> crate::Result<()> {
let mut schema_builder = SchemaBuilder::default();
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;

103
src/fastfield/mod.rs
View File

@@ -1,51 +1,38 @@
/*!
Column oriented field storage for tantivy.
//! Column oriented field storage for tantivy.
//!
//! It is the equivalent of `Lucene`'s `DocValues`.
//!
//! Fast fields is a column-oriented fashion storage of `tantivy`.
//!
//! It is designed for the fast random access of some document
//! fields given a document id.
//!
//! `FastField` are useful when a field is required for all or most of
//! the `DocSet` : for instance for scoring, grouping, filtering, or faceting.
//!
//!
//! Fields have to be declared as `FAST` in the schema.
//! Currently only 64-bits integers (signed or unsigned) are
//! supported.
//!
//! They are stored in a bit-packed fashion so that their
//! memory usage is directly linear with the amplitude of the
//! values stored.
//!
//! Read access performance is comparable to that of an array lookup.
It is the equivalent of `Lucene`'s `DocValues`.
Fast fields is a column-oriented fashion storage of `tantivy`.
It is designed for the fast random access of some document
fields given a document id.
`FastField` are useful when a field is required for all or most of
the `DocSet` : for instance for scoring, grouping, filtering, or faceting.
Fields have to be declared as `FAST` in the schema.
Currently only 64-bits integers (signed or unsigned) are
supported.
They are stored in a bit-packed fashion so that their
memory usage is directly linear with the amplitude of the
values stored.
Read access performance is comparable to that of an array lookup.
*/
pub use self::alive_bitset::intersect_alive_bitsets;
pub use self::alive_bitset::write_alive_bitset;
pub use self::alive_bitset::AliveBitSet;
pub use self::alive_bitset::{intersect_alive_bitsets, write_alive_bitset, AliveBitSet};
pub use self::bytes::{BytesFastFieldReader, BytesFastFieldWriter};
pub use self::error::{FastFieldNotAvailableError, Result};
pub use self::facet_reader::FacetReader;
pub use self::multivalued::{MultiValuedFastFieldReader, MultiValuedFastFieldWriter};
pub(crate) use self::reader::BitpackedFastFieldReader;
pub use self::reader::DynamicFastFieldReader;
pub use self::reader::FastFieldReader;
pub use self::reader::{DynamicFastFieldReader, FastFieldReader};
pub use self::readers::FastFieldReaders;
pub use self::serializer::CompositeFastFieldSerializer;
pub use self::serializer::FastFieldDataAccess;
pub use self::serializer::FastFieldStats;
pub use self::serializer::{CompositeFastFieldSerializer, FastFieldDataAccess, FastFieldStats};
pub use self::writer::{FastFieldsWriter, IntFastFieldWriter};
use crate::schema::Cardinality;
use crate::schema::FieldType;
use crate::schema::Value;
use crate::chrono::{NaiveDateTime, Utc};
use crate::schema::{Cardinality, FieldType, Type, Value};
use crate::DocId;
use crate::{
chrono::{NaiveDateTime, Utc},
schema::Type,
};
mod alive_bitset;
mod bytes;
@@ -110,7 +97,7 @@ impl FastValue for u64 {
fn fast_field_cardinality(field_type: &FieldType) -> Option<Cardinality> {
match *field_type {
FieldType::U64(ref integer_options) => integer_options.get_fastfield_cardinality(),
FieldType::HierarchicalFacet(_) => Some(Cardinality::MultiValues),
FieldType::Facet(_) => Some(Cardinality::MultiValues),
_ => None,
}
}
@@ -213,22 +200,20 @@ fn value_to_u64(value: &Value) -> u64 {
#[cfg(test)]
mod tests {
use super::*;
use crate::directory::CompositeFile;
use crate::directory::{Directory, RamDirectory, WritePtr};
use crate::merge_policy::NoMergePolicy;
use crate::schema::Field;
use crate::schema::Schema;
use crate::schema::FAST;
use crate::schema::{Document, IntOptions};
use crate::{Index, SegmentId, SegmentReader};
use std::collections::HashMap;
use std::path::Path;
use common::HasLen;
use once_cell::sync::Lazy;
use rand::prelude::SliceRandom;
use rand::rngs::StdRng;
use rand::SeedableRng;
use std::collections::HashMap;
use std::path::Path;
use super::*;
use crate::directory::{CompositeFile, Directory, RamDirectory, WritePtr};
use crate::merge_policy::NoMergePolicy;
use crate::schema::{Document, Field, IntOptions, Schema, FAST};
use crate::{Index, SegmentId, SegmentReader};
pub static SCHEMA: Lazy<Schema> = Lazy::new(|| {
let mut schema_builder = Schema::builder();
@@ -407,7 +392,7 @@ mod tests {
serializer.close().unwrap();
}
let file = directory.open_read(path).unwrap();
//assert_eq!(file.len(), 17710 as usize); //bitpacked size
// assert_eq!(file.len(), 17710 as usize); //bitpacked size
assert_eq!(file.len(), 10175_usize); // linear interpol size
{
let fast_fields_composite = CompositeFile::open(&file)?;
@@ -587,16 +572,16 @@ mod tests {
#[cfg(all(test, feature = "unstable"))]
mod bench {
use super::tests::FIELD;
use super::tests::{generate_permutation, SCHEMA};
use super::*;
use crate::directory::CompositeFile;
use crate::directory::{Directory, RamDirectory, WritePtr};
use crate::fastfield::FastFieldReader;
use std::collections::HashMap;
use std::path::Path;
use test::{self, Bencher};
use super::tests::{generate_permutation, FIELD, SCHEMA};
use super::*;
use crate::directory::{CompositeFile, Directory, RamDirectory, WritePtr};
use crate::fastfield::FastFieldReader;
#[bench]
fn bench_intfastfield_linear_veclookup(b: &mut Bencher) {
let permutation = generate_permutation();

37
src/fastfield/multivalued/mod.rs
View File

@@ -7,23 +7,17 @@ pub use self::writer::MultiValuedFastFieldWriter;
#[cfg(test)]
mod tests {
use chrono::Duration;
use futures::executor::block_on;
use proptest::strategy::Strategy;
use proptest::{prop_oneof, proptest};
use test_log::test;
use crate::collector::TopDocs;
use crate::indexer::NoMergePolicy;
use crate::query::QueryParser;
use crate::schema::Cardinality;
use crate::schema::Facet;
use crate::schema::IntOptions;
use crate::schema::Schema;
use crate::schema::INDEXED;
use crate::Document;
use crate::Index;
use crate::Term;
use chrono::Duration;
use futures::executor::block_on;
use proptest::prop_oneof;
use proptest::proptest;
use proptest::strategy::Strategy;
use test_env_log::test;
use crate::schema::{Cardinality, Facet, FacetOptions, IntOptions, Schema};
use crate::{Document, Index, Term};
#[test]
fn test_multivalued_u64() -> crate::Result<()> {
@@ -68,6 +62,7 @@ mod tests {
IntOptions::default()
.set_fast(Cardinality::MultiValues)
.set_indexed()
.set_fieldnorm()
.set_stored(),
);
let time_i =
@@ -109,7 +104,7 @@ mod tests {
retrieved_doc
.get_first(date_field)
.expect("cannot find value")
.date_value()
.as_date()
.unwrap()
.timestamp(),
first_time_stamp.timestamp()
@@ -118,7 +113,7 @@ mod tests {
retrieved_doc
.get_first(time_i)
.expect("cannot find value")
.i64_value(),
.as_i64(),
Some(1i64)
);
}
@@ -137,7 +132,7 @@ mod tests {
retrieved_doc
.get_first(date_field)
.expect("cannot find value")
.date_value()
.as_date()
.unwrap()
.timestamp(),
two_secs_ahead.timestamp()
@@ -146,7 +141,7 @@ mod tests {
retrieved_doc
.get_first(time_i)
.expect("cannot find value")
.i64_value(),
.as_i64(),
Some(3i64)
);
}
@@ -179,7 +174,7 @@ mod tests {
retrieved_doc
.get_first(date_field)
.expect("cannot find value")
.date_value()
.as_date()
.expect("value not of Date type")
.timestamp(),
(first_time_stamp + Duration::seconds(offset_sec)).timestamp()
@@ -188,7 +183,7 @@ mod tests {
retrieved_doc
.get_first(time_i)
.expect("cannot find value")
.i64_value(),
.as_i64(),
Some(time_i_val)
);
}
@@ -334,7 +329,7 @@ mod tests {
#[ignore]
fn test_many_facets() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let field = schema_builder.add_facet_field("facetfield", INDEXED);
let field = schema_builder.add_facet_field("facetfield", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;

5
src/fastfield/multivalued/reader.rs
View File

@@ -10,7 +10,6 @@ use crate::DocId;
/// The `vals_reader` will access the concatenated list of all
/// values for all reader.
/// The `idx_reader` associated, for each document, the index of its first value.
///
#[derive(Clone)]
pub struct MultiValuedFastFieldReader<Item: FastValue> {
idx_reader: DynamicFastFieldReader<u64>,
@@ -91,12 +90,12 @@ impl<Item: FastValue> MultiValueLength for MultiValuedFastFieldReader<Item> {
mod tests {
use crate::core::Index;
use crate::schema::{Cardinality, Facet, IntOptions, Schema, INDEXED};
use crate::schema::{Cardinality, Facet, FacetOptions, IntOptions, Schema};
#[test]
fn test_multifastfield_reader() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let facet_field = schema_builder.add_facet_field("facets", INDEXED);
let facet_field = schema_builder.add_facet_field("facets", FacetOptions::default());
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;

18
src/fastfield/multivalued/writer.rs
View File

@@ -1,13 +1,15 @@
use std::io;
use fnv::FnvHashMap;
use tantivy_bitpacker::minmax;
use crate::fastfield::serializer::BitpackedFastFieldSerializerLegacy;
use crate::fastfield::CompositeFastFieldSerializer;
use crate::fastfield::{value_to_u64, CompositeFastFieldSerializer};
use crate::indexer::doc_id_mapping::DocIdMapping;
use crate::postings::UnorderedTermId;
use crate::schema::{Document, Field};
use crate::termdict::TermOrdinal;
use crate::DocId;
use crate::{fastfield::value_to_u64, indexer::doc_id_mapping::DocIdMapping};
use fnv::FnvHashMap;
use std::io;
use tantivy_bitpacker::minmax;
/// Writer for multi-valued (as in, more than one value per document)
/// int fast field.
@@ -20,7 +22,8 @@ use tantivy_bitpacker::minmax;
/// - add your document simply by calling `.add_document(...)`.
///
/// The `MultiValuedFastFieldWriter` can be acquired from the
/// fastfield writer, by calling [`.get_multivalue_writer(...)`](./struct.FastFieldsWriter.html#method.get_multivalue_writer).
/// fastfield writer, by calling
/// [`.get_multivalue_writer(...)`](./struct.FastFieldsWriter.html#method.get_multivalue_writer).
///
/// Once acquired, writing is done by calling calls to
/// `.add_document_vals(&[u64])` once per document.
@@ -76,7 +79,7 @@ impl MultiValuedFastFieldWriter {
// facets are indexed in the `SegmentWriter` as we encode their unordered id.
if !self.is_facet {
for field_value in doc.field_values() {
if field_value.field() == self.field {
if field_value.field == self.field {
self.add_val(value_to_u64(field_value.value()));
}
}
@@ -131,7 +134,6 @@ impl MultiValuedFastFieldWriter {
/// During the serialization of the segment, terms gets sorted and
/// `tantivy` builds a mapping to convert this `UnorderedTermId` into
/// term ordinals.
///
pub fn serialize(
&self,
serializer: &mut CompositeFastFieldSerializer,

43
src/fastfield/reader.rs
View File

@@ -1,25 +1,25 @@
use super::FastValue;
use crate::directory::CompositeFile;
use crate::directory::FileSlice;
use crate::directory::OwnedBytes;
use crate::directory::{Directory, RamDirectory, WritePtr};
use crate::fastfield::{CompositeFastFieldSerializer, FastFieldsWriter};
use crate::schema::Schema;
use crate::schema::FAST;
use crate::DocId;
use common::BinarySerializable;
use fastfield_codecs::bitpacked::BitpackedFastFieldReader as BitpackedReader;
use fastfield_codecs::bitpacked::BitpackedFastFieldSerializer;
use fastfield_codecs::linearinterpol::LinearInterpolFastFieldReader;
use fastfield_codecs::linearinterpol::LinearInterpolFastFieldSerializer;
use fastfield_codecs::multilinearinterpol::MultiLinearInterpolFastFieldReader;
use fastfield_codecs::multilinearinterpol::MultiLinearInterpolFastFieldSerializer;
use fastfield_codecs::FastFieldCodecReader;
use fastfield_codecs::FastFieldCodecSerializer;
use std::collections::HashMap;
use std::marker::PhantomData;
use std::path::Path;
use common::BinarySerializable;
use fastfield_codecs::bitpacked::{
BitpackedFastFieldReader as BitpackedReader, BitpackedFastFieldSerializer,
};
use fastfield_codecs::linearinterpol::{
LinearInterpolFastFieldReader, LinearInterpolFastFieldSerializer,
};
use fastfield_codecs::multilinearinterpol::{
MultiLinearInterpolFastFieldReader, MultiLinearInterpolFastFieldSerializer,
};
use fastfield_codecs::{FastFieldCodecReader, FastFieldCodecSerializer};
use super::FastValue;
use crate::directory::{CompositeFile, Directory, FileSlice, OwnedBytes, RamDirectory, WritePtr};
use crate::fastfield::{CompositeFastFieldSerializer, FastFieldsWriter};
use crate::schema::{Schema, FAST};
use crate::DocId;
/// FastFieldReader is the trait to access fast field data.
pub trait FastFieldReader<Item: FastValue>: Clone {
/// Return the value associated to the given document.
@@ -64,7 +64,6 @@ pub trait FastFieldReader<Item: FastValue>: Clone {
#[derive(Clone)]
/// DynamicFastFieldReader wraps different readers to access
/// the various encoded fastfield data
///
pub enum DynamicFastFieldReader<Item: FastValue> {
/// Bitpacked compressed fastfield data.
Bitpacked(FastFieldReaderCodecWrapper<Item, BitpackedReader>),
@@ -146,7 +145,6 @@ impl<Item: FastValue> FastFieldReader<Item> for DynamicFastFieldReader<Item> {
/// Wrapper for accessing a fastfield.
///
/// Holds the data and the codec to the read the data.
///
#[derive(Clone)]
pub struct FastFieldReaderCodecWrapper<Item: FastValue, CodecReader> {
reader: CodecReader,
@@ -162,7 +160,8 @@ impl<Item: FastValue, C: FastFieldCodecReader> FastFieldReaderCodecWrapper<Item,
assert_eq!(
BitpackedFastFieldSerializer::ID,
id,
"Tried to open fast field as bitpacked encoded (id=1), but got serializer with different id"
"Tried to open fast field as bitpacked encoded (id=1), but got serializer with \
different id"
);
Self::open_from_bytes(bytes)
}
@@ -249,8 +248,6 @@ impl<Item: FastValue, C: FastFieldCodecReader + Clone> FastFieldReader<Item>
}
}
pub(crate) type BitpackedFastFieldReader<Item> = FastFieldReaderCodecWrapper<Item, BitpackedReader>;
impl<Item: FastValue> From<Vec<Item>> for DynamicFastFieldReader<Item> {
fn from(vals: Vec<Item>) -> DynamicFastFieldReader<Item> {
let mut schema_builder = Schema::builder();

27
src/fastfield/readers.rs
View File

@@ -1,14 +1,12 @@
use crate::directory::CompositeFile;
use crate::directory::FileSlice;
use crate::fastfield::MultiValuedFastFieldReader;
use crate::fastfield::{BitpackedFastFieldReader, FastFieldNotAvailableError};
use crate::fastfield::{BytesFastFieldReader, FastValue};
use super::reader::DynamicFastFieldReader;
use crate::directory::{CompositeFile, FileSlice};
use crate::fastfield::{
BytesFastFieldReader, FastFieldNotAvailableError, FastValue, MultiValuedFastFieldReader,
};
use crate::schema::{Cardinality, Field, FieldType, Schema};
use crate::space_usage::PerFieldSpaceUsage;
use crate::TantivyError;
use super::reader::DynamicFastFieldReader;
/// Provides access to all of the BitpackedFastFieldReader.
///
/// Internally, `FastFieldReaders` have preloaded fast field readers,
@@ -40,7 +38,7 @@ fn type_and_cardinality(field_type: &FieldType) -> Option<(FastType, Cardinality
FieldType::Date(options) => options
.get_fastfield_cardinality()
.map(|cardinality| (FastType::Date, cardinality)),
FieldType::HierarchicalFacet(_) => Some((FastType::U64, Cardinality::MultiValues)),
FieldType::Facet(_) => Some((FastType::U64, Cardinality::MultiValues)),
_ => None,
}
}
@@ -131,10 +129,11 @@ impl FastFieldReaders {
self.typed_fast_field_reader(field)
}
/// Returns the `u64` fast field reader reader associated to `field`, regardless of whether the given
/// field is effectively of type `u64` or not.
/// Returns the `u64` fast field reader reader associated to `field`, regardless of whether the
/// given field is effectively of type `u64` or not.
///
/// If not, the fastfield reader will returns the u64-value associated to the original FastValue.
/// If not, the fastfield reader will returns the u64-value associated to the original
/// FastValue.
pub fn u64_lenient(&self, field: Field) -> crate::Result<DynamicFastFieldReader<u64>> {
self.typed_fast_field_reader(field)
}
@@ -171,8 +170,8 @@ impl FastFieldReaders {
self.typed_fast_field_multi_reader(field)
}
/// Returns a `u64s` multi-valued fast field reader reader associated to `field`, regardless of whether the given
/// field is effectively of type `u64` or not.
/// Returns a `u64s` multi-valued fast field reader reader associated to `field`, regardless of
/// whether the given field is effectively of type `u64` or not.
///
/// If `field` is not a u64 multi-valued fast field, this method returns an Error.
pub fn u64s_lenient(&self, field: Field) -> crate::Result<MultiValuedFastFieldReader<u64>> {
@@ -219,7 +218,7 @@ impl FastFieldReaders {
)));
}
let fast_field_idx_file = self.fast_field_data(field, 0)?;
let idx_reader = BitpackedFastFieldReader::open(fast_field_idx_file)?;
let idx_reader = DynamicFastFieldReader::open(fast_field_idx_file)?;
let data = self.fast_field_data(field, 1)?;
BytesFastFieldReader::open(idx_reader, data)
} else {

30
src/fastfield/serializer/mod.rs
View File

@@ -1,16 +1,15 @@
use crate::directory::CompositeWrite;
use crate::directory::WritePtr;
use crate::schema::Field;
use common::BinarySerializable;
use common::CountingWriter;
pub use fastfield_codecs::bitpacked::BitpackedFastFieldSerializer;
pub use fastfield_codecs::bitpacked::BitpackedFastFieldSerializerLegacy;
use std::io::{self, Write};
use common::{BinarySerializable, CountingWriter};
pub use fastfield_codecs::bitpacked::{
BitpackedFastFieldSerializer, BitpackedFastFieldSerializerLegacy,
};
use fastfield_codecs::linearinterpol::LinearInterpolFastFieldSerializer;
use fastfield_codecs::multilinearinterpol::MultiLinearInterpolFastFieldSerializer;
pub use fastfield_codecs::FastFieldCodecSerializer;
pub use fastfield_codecs::FastFieldDataAccess;
pub use fastfield_codecs::FastFieldStats;
use std::io::{self, Write};
pub use fastfield_codecs::{FastFieldCodecSerializer, FastFieldDataAccess, FastFieldStats};
use crate::directory::{CompositeWrite, WritePtr};
use crate::schema::Field;
/// `CompositeFastFieldSerializer` is in charge of serializing
/// fastfields on disk.
@@ -58,7 +57,8 @@ impl CompositeFastFieldSerializer {
Ok(CompositeFastFieldSerializer { composite_write })
}
/// Serialize data into a new u64 fast field. The best compression codec will be chosen automatically.
/// Serialize data into a new u64 fast field. The best compression codec will be chosen
/// automatically.
pub fn create_auto_detect_u64_fast_field(
&mut self,
field: Field,
@@ -76,7 +76,8 @@ impl CompositeFastFieldSerializer {
0,
)
}
/// Serialize data into a new u64 fast field. The best compression codec will be chosen automatically.
/// Serialize data into a new u64 fast field. The best compression codec will be chosen
/// automatically.
pub fn create_auto_detect_u64_fast_field_with_idx(
&mut self,
field: Field,
@@ -112,7 +113,8 @@ impl CompositeFastFieldSerializer {
broken_estimation.1
);
}
// removing nan values for codecs with broken calculations, and max values which disables codecs
// removing nan values for codecs with broken calculations, and max values which disables
// codecs
estimations.retain(|estimation| !estimation.0.is_nan() && estimation.0 != f32::MAX);
estimations.sort_by(|a, b| a.0.partial_cmp(&b.0).unwrap());
let (_ratio, name, id) = estimations[0];

21
src/fastfield/writer.rs
View File

@@ -1,3 +1,10 @@
use std::collections::HashMap;
use std::io;
use common;
use fnv::FnvHashMap;
use tantivy_bitpacker::BlockedBitpacker;
use super::multivalued::MultiValuedFastFieldWriter;
use super::serializer::FastFieldStats;
use super::FastFieldDataAccess;
@@ -6,11 +13,6 @@ use crate::indexer::doc_id_mapping::DocIdMapping;
use crate::postings::UnorderedTermId;
use crate::schema::{Cardinality, Document, Field, FieldEntry, FieldType, Schema};
use crate::termdict::TermOrdinal;
use common;
use fnv::FnvHashMap;
use std::collections::HashMap;
use std::io;
use tantivy_bitpacker::BlockedBitpacker;
/// The fastfieldswriter regroup all of the fast field writers.
pub struct FastFieldsWriter {
@@ -54,7 +56,7 @@ impl FastFieldsWriter {
None => {}
}
}
FieldType::HierarchicalFacet(_) => {
FieldType::Facet(_) => {
let fast_field_writer = MultiValuedFastFieldWriter::new(field, true);
multi_values_writers.push(fast_field_writer);
}
@@ -324,7 +326,8 @@ struct WriterFastFieldAccessProvider<'map, 'bitp> {
impl<'map, 'bitp> FastFieldDataAccess for WriterFastFieldAccessProvider<'map, 'bitp> {
/// Return the value associated to the given doc.
///
/// Whenever possible use the Iterator passed to the fastfield creation instead, for performance reasons.
/// Whenever possible use the Iterator passed to the fastfield creation instead, for performance
/// reasons.
///
/// # Panics
///
@@ -332,7 +335,9 @@ impl<'map, 'bitp> FastFieldDataAccess for WriterFastFieldAccessProvider<'map, 'b
fn get_val(&self, doc: u64) -> u64 {
if let Some(doc_id_map) = self.doc_id_map {
self.vals
.get(doc_id_map.get_old_doc_id(doc as u32) as usize) // consider extra FastFieldReader wrapper for non doc_id_map
.get(doc_id_map.get_old_doc_id(doc as u32) as usize) // consider extra
// FastFieldReader wrapper for
// non doc_id_map
} else {
self.vals.get(doc as usize)
}

128
src/fieldnorm/mod.rs
View File

@@ -21,8 +21,134 @@ mod reader;
mod serializer;
mod writer;
use self::code::{fieldnorm_to_id, id_to_fieldnorm};
pub use self::reader::{FieldNormReader, FieldNormReaders};
pub use self::serializer::FieldNormsSerializer;
pub use self::writer::FieldNormsWriter;
use self::code::{fieldnorm_to_id, id_to_fieldnorm};
#[cfg(test)]
mod tests {
use std::path::Path;
use once_cell::sync::Lazy;
use crate::directory::{CompositeFile, Directory, RamDirectory, WritePtr};
use crate::fieldnorm::{FieldNormReader, FieldNormsSerializer, FieldNormsWriter};
use crate::query::{Query, TermQuery};
use crate::schema::{
Field, IndexRecordOption, Schema, TextFieldIndexing, TextOptions, STORED, TEXT,
};
use crate::{Index, Term, TERMINATED};
pub static SCHEMA: Lazy<Schema> = Lazy::new(|| {
let mut schema_builder = Schema::builder();
schema_builder.add_text_field("field", STORED);
schema_builder.add_text_field("txt_field", TEXT);
schema_builder.add_text_field(
"str_field",
TextOptions::default().set_indexing_options(
TextFieldIndexing::default()
.set_index_option(IndexRecordOption::Basic)
.set_fieldnorms(false),
),
);
schema_builder.build()
});
pub static FIELD: Lazy<Field> = Lazy::new(|| SCHEMA.get_field("field").unwrap());
pub static TXT_FIELD: Lazy<Field> = Lazy::new(|| SCHEMA.get_field("txt_field").unwrap());
pub static STR_FIELD: Lazy<Field> = Lazy::new(|| SCHEMA.get_field("str_field").unwrap());
#[test]
#[should_panic(expected = "Cannot register a given fieldnorm twice")]
pub fn test_should_panic_when_recording_fieldnorm_twice_for_same_doc() {
let mut fieldnorm_writers = FieldNormsWriter::for_schema(&SCHEMA);
fieldnorm_writers.record(0u32, *TXT_FIELD, 5);
fieldnorm_writers.record(0u32, *TXT_FIELD, 3);
}
#[test]
pub fn test_fieldnorm() -> crate::Result<()> {
let path = Path::new("test");
let directory: RamDirectory = RamDirectory::create();
{
let write: WritePtr = directory.open_write(Path::new("test"))?;
let serializer = FieldNormsSerializer::from_write(write)?;
let mut fieldnorm_writers = FieldNormsWriter::for_schema(&SCHEMA);
fieldnorm_writers.record(2u32, *TXT_FIELD, 5);
fieldnorm_writers.record(3u32, *TXT_FIELD, 3);
fieldnorm_writers.serialize(serializer, None)?;
}
let file = directory.open_read(path)?;
{
let fields_composite = CompositeFile::open(&file)?;
assert!(fields_composite.open_read(*FIELD).is_none());
assert!(fields_composite.open_read(*STR_FIELD).is_none());
let data = fields_composite.open_read(*TXT_FIELD).unwrap();
let fieldnorm_reader = FieldNormReader::open(data)?;
assert_eq!(fieldnorm_reader.fieldnorm(0u32), 0u32);
assert_eq!(fieldnorm_reader.fieldnorm(1u32), 0u32);
assert_eq!(fieldnorm_reader.fieldnorm(2u32), 5u32);
assert_eq!(fieldnorm_reader.fieldnorm(3u32), 3u32);
}
Ok(())
}
#[test]
fn test_fieldnorm_disabled() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let text_options = TextOptions::default()
.set_indexing_options(TextFieldIndexing::default().set_fieldnorms(false));
let text = schema_builder.add_text_field("text", text_options);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut writer = index.writer_for_tests()?;
writer.add_document(doc!(text=>"hello"))?;
writer.add_document(doc!(text=>"hello hello hello"))?;
writer.commit()?;
let reader = index.reader()?;
let searcher = reader.searcher();
let query = TermQuery::new(
Term::from_field_text(text, "hello"),
IndexRecordOption::WithFreqs,
);
let weight = query.weight(&*searcher, true)?;
let mut scorer = weight.scorer(searcher.segment_reader(0), 1.0f32)?;
assert_eq!(scorer.doc(), 0);
assert!((scorer.score() - 0.22920431).abs() < 0.001f32);
assert_eq!(scorer.advance(), 1);
assert_eq!(scorer.doc(), 1);
assert!((scorer.score() - 0.22920431).abs() < 0.001f32);
assert_eq!(scorer.advance(), TERMINATED);
Ok(())
}
#[test]
fn test_fieldnorm_enabled() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let text_options = TextOptions::default()
.set_indexing_options(TextFieldIndexing::default().set_fieldnorms(true));
let text = schema_builder.add_text_field("text", text_options);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut writer = index.writer_for_tests()?;
writer.add_document(doc!(text=>"hello"))?;
writer.add_document(doc!(text=>"hello hello hello"))?;
writer.commit()?;
let reader = index.reader()?;
let searcher = reader.searcher();
let query = TermQuery::new(
Term::from_field_text(text, "hello"),
IndexRecordOption::WithFreqs,
);
let weight = query.weight(&*searcher, true)?;
let mut scorer = weight.scorer(searcher.segment_reader(0), 1.0f32)?;
assert_eq!(scorer.doc(), 0);
assert!((scorer.score() - 0.22920431).abs() < 0.001f32);
assert_eq!(scorer.advance(), 1);
assert_eq!(scorer.doc(), 1);
assert!((scorer.score() - 0.15136132).abs() < 0.001f32);
assert_eq!(scorer.advance(), TERMINATED);
Ok(())
}
}

7
src/fieldnorm/reader.rs
View File

@@ -1,11 +1,10 @@
use std::sync::Arc;
use super::{fieldnorm_to_id, id_to_fieldnorm};
use crate::directory::CompositeFile;
use crate::directory::FileSlice;
use crate::directory::OwnedBytes;
use crate::directory::{CompositeFile, FileSlice, OwnedBytes};
use crate::schema::Field;
use crate::space_usage::PerFieldSpaceUsage;
use crate::DocId;
use std::sync::Arc;
/// Reader for the fieldnorm (for each document, the number of tokens indexed in the
/// field) of all indexed fields in the index.

6
src/fieldnorm/serializer.rs
View File

@@ -1,9 +1,9 @@
use crate::directory::CompositeWrite;
use crate::directory::WritePtr;
use crate::schema::Field;
use std::io;
use std::io::Write;
use crate::directory::{CompositeWrite, WritePtr};
use crate::schema::Field;
/// The fieldnorms serializer is in charge of
/// the serialization of field norms for all fields.
pub struct FieldNormsSerializer {

88
src/fieldnorm/writer.rs
View File

@@ -1,19 +1,18 @@
use crate::{indexer::doc_id_mapping::DocIdMapping, DocId};
use super::fieldnorm_to_id;
use super::FieldNormsSerializer;
use crate::schema::Field;
use crate::schema::Schema;
use std::cmp::Ordering;
use std::{io, iter};
use super::{fieldnorm_to_id, FieldNormsSerializer};
use crate::indexer::doc_id_mapping::DocIdMapping;
use crate::schema::{Field, Schema};
use crate::DocId;
/// The `FieldNormsWriter` is in charge of tracking the fieldnorm byte
/// of each document for each field with field norms.
///
/// `FieldNormsWriter` stores a Vec<u8> for each tracked field, using a
/// byte per document per field.
pub struct FieldNormsWriter {
fields: Vec<Field>,
fieldnorms_buffer: Vec<Vec<u8>>,
fieldnorms_buffers: Vec<Option<Vec<u8>>>,
}
impl FieldNormsWriter {
@@ -23,7 +22,7 @@ impl FieldNormsWriter {
schema
.fields()
.filter_map(|(field, field_entry)| {
if field_entry.is_indexed() {
if field_entry.is_indexed() && field_entry.has_fieldnorms() {
Some(field)
} else {
None
@@ -35,25 +34,20 @@ impl FieldNormsWriter {
/// Initialize with state for tracking the field norm fields
/// specified in the schema.
pub fn for_schema(schema: &Schema) -> FieldNormsWriter {
let fields = FieldNormsWriter::fields_with_fieldnorm(schema);
let max_field = fields
.iter()
.map(Field::field_id)
.max()
.map(|max_field_id| max_field_id as usize + 1)
.unwrap_or(0);
FieldNormsWriter {
fields,
fieldnorms_buffer: iter::repeat_with(Vec::new)
.take(max_field)
.collect::<Vec<_>>(),
let mut fieldnorms_buffers: Vec<Option<Vec<u8>>> = iter::repeat_with(|| None)
.take(schema.num_fields())
.collect();
for field in FieldNormsWriter::fields_with_fieldnorm(schema) {
fieldnorms_buffers[field.field_id() as usize] = Some(Vec::with_capacity(1_000));
}
FieldNormsWriter { fieldnorms_buffers }
}
/// The memory used inclusive childs
pub fn mem_usage(&self) -> usize {
self.fieldnorms_buffer
self.fieldnorms_buffers
.iter()
.flatten()
.map(|buf| buf.capacity())
.sum()
}
@@ -62,8 +56,10 @@ impl FieldNormsWriter {
///
/// Will extend with 0-bytes for documents that have not been seen.
pub fn fill_up_to_max_doc(&mut self, max_doc: DocId) {
for field in self.fields.iter() {
self.fieldnorms_buffer[field.field_id() as usize].resize(max_doc as usize, 0u8);
for fieldnorms_buffer_opt in self.fieldnorms_buffers.iter_mut() {
if let Some(fieldnorms_buffer) = fieldnorms_buffer_opt.as_mut() {
fieldnorms_buffer.resize(max_doc as usize, 0u8);
}
}
}
@@ -76,14 +72,23 @@ impl FieldNormsWriter {
/// * field - the field being set
/// * fieldnorm - the number of terms present in document `doc` in field `field`
pub fn record(&mut self, doc: DocId, field: Field, fieldnorm: u32) {
let fieldnorm_buffer: &mut Vec<u8> = &mut self.fieldnorms_buffer[field.field_id() as usize];
assert!(
fieldnorm_buffer.len() <= doc as usize,
"Cannot register a given fieldnorm twice"
);
// we fill intermediary `DocId` as having a fieldnorm of 0.
fieldnorm_buffer.resize(doc as usize + 1, 0u8);
fieldnorm_buffer[doc as usize] = fieldnorm_to_id(fieldnorm);
if let Some(fieldnorm_buffer) = self
.fieldnorms_buffers
.get_mut(field.field_id() as usize)
.and_then(Option::as_mut)
{
match fieldnorm_buffer.len().cmp(&(doc as usize)) {
Ordering::Less => {
// we fill intermediary `DocId` as having a fieldnorm of 0.
fieldnorm_buffer.resize(doc as usize, 0u8);
}
Ordering::Equal => {}
Ordering::Greater => {
panic!("Cannot register a given fieldnorm twice")
}
}
fieldnorm_buffer.push(fieldnorm_to_id(fieldnorm));
}
}
/// Serialize the seen fieldnorm values to the serializer for all fields.
@@ -92,17 +97,18 @@ impl FieldNormsWriter {
mut fieldnorms_serializer: FieldNormsSerializer,
doc_id_map: Option<&DocIdMapping>,
) -> io::Result<()> {
for &field in self.fields.iter() {
let fieldnorm_values: &[u8] = &self.fieldnorms_buffer[field.field_id() as usize][..];
for (field, fieldnorms_buffer) in self.fieldnorms_buffers.iter().enumerate().filter_map(
|(field_id, fieldnorms_buffer_opt)| {
fieldnorms_buffer_opt.as_ref().map(|fieldnorms_buffer| {
(Field::from_field_id(field_id as u32), fieldnorms_buffer)
})
},
) {
if let Some(doc_id_map) = doc_id_map {
let mut mapped_fieldnorm_values = vec![];
mapped_fieldnorm_values.resize(fieldnorm_values.len(), 0u8);
for (new_doc_id, old_doc_id) in doc_id_map.iter_old_doc_ids().enumerate() {
mapped_fieldnorm_values[new_doc_id] = fieldnorm_values[old_doc_id as usize];
}
fieldnorms_serializer.serialize_field(field, &mapped_fieldnorm_values)?;
let remapped_fieldnorm_buffer = doc_id_map.remap(fieldnorms_buffer);
fieldnorms_serializer.serialize_field(field, &remapped_fieldnorm_buffer)?;
} else {
fieldnorms_serializer.serialize_field(field, fieldnorm_values)?;
fieldnorms_serializer.serialize_field(field, fieldnorms_buffer)?;
}
}
fieldnorms_serializer.close()?;

28
src/functional_test.rs
View File

@@ -1,14 +1,10 @@
use crate::schema;
use crate::Index;
use crate::IndexSettings;
use crate::IndexSortByField;
use crate::Order;
use crate::Searcher;
use crate::{doc, schema::*};
use rand::thread_rng;
use rand::Rng;
use std::collections::HashSet;
use rand::{thread_rng, Rng};
use crate::schema::*;
use crate::{doc, schema, Index, IndexSettings, IndexSortByField, Order, Searcher};
fn check_index_content(searcher: &Searcher, vals: &[u64]) -> crate::Result<()> {
assert!(searcher.segment_readers().len() < 20);
assert_eq!(searcher.num_docs() as usize, vals.len());
@@ -130,14 +126,12 @@ fn test_functional_indexing_sorted() -> crate::Result<()> {
Ok(())
}
const LOREM: &str = "Doc Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed \
do eiusmod tempor incididunt ut labore et dolore magna aliqua. \
Ut enim ad minim veniam, quis nostrud exercitation ullamco \
laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure \
dolor in reprehenderit in voluptate velit esse cillum dolore eu \
fugiat nulla pariatur. Excepteur sint occaecat cupidatat non \
proident, sunt in culpa qui officia deserunt mollit anim id est \
laborum.";
const LOREM: &str = "Doc Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod \
tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, \
quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo \
consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse \
cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat \
non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.";
fn get_text() -> String {
use rand::seq::SliceRandom;
let mut rng = thread_rng();

20
src/indexer/delete_queue.rs
View File

@@ -1,9 +1,9 @@
use super::operation::DeleteOperation;
use crate::Opstamp;
use std::ops::DerefMut;
use std::sync::{Arc, RwLock, Weak};
use super::operation::DeleteOperation;
use crate::Opstamp;
// The DeleteQueue is similar in conceptually to a multiple
// consumer single producer broadcast channel.
//
@@ -13,12 +13,10 @@ use std::sync::{Arc, RwLock, Weak};
// which points to a specific place of the `DeleteQueue`.
//
// New consumer can be created in two ways
// - calling `delete_queue.cursor()` returns a cursor, that
// will include all future delete operation (and some or none
// of the past operations... The client is in charge of checking the opstamps.).
// - cloning an existing cursor returns a new cursor, that
// is at the exact same position, and can now advance independently
// from the original cursor.
// - calling `delete_queue.cursor()` returns a cursor, that will include all future delete operation
// (and some or none of the past operations... The client is in charge of checking the opstamps.).
// - cloning an existing cursor returns a new cursor, that is at the exact same position, and can
// now advance independently from the original cursor.
#[derive(Default)]
struct InnerDeleteQueue {
writer: Vec<DeleteOperation>,
@@ -179,8 +177,8 @@ pub struct DeleteCursor {
impl DeleteCursor {
/// Skips operations and position it so that
/// - either all of the delete operation currently in the
/// queue are consume and the next get will return None.
/// - either all of the delete operation currently in the 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) {

32
src/indexer/demuxer.rs
View File

@@ -5,8 +5,8 @@ use crate::fastfield::AliveBitSet;
use crate::{merge_filtered_segments, Directory, Index, IndexSettings, Segment, SegmentOrdinal};
/// DemuxMapping can be used to reorganize data from multiple segments.
///
/// DemuxMapping is useful in a multitenant settings, in which each document might actually belong to a different tenant.
/// It allows to reorganize documents as follows:
/// DemuxMapping is useful in a multitenant settings, in which each document might actually belong
/// to a different tenant. It allows to reorganize documents as follows:
///
/// e.g. if you have two tenant ids TENANT_A and TENANT_B and two segments with
/// the documents (simplified)
@@ -18,7 +18,8 @@ use crate::{merge_filtered_segments, Directory, Index, IndexSettings, Segment, S
/// Seg 2 [TENANT_B, TENANT_B]
///
/// Demuxing is the tool for that.
/// Semantically you can define a mapping from [old segment ordinal, old doc_id] -> [new segment ordinal].
/// Semantically you can define a mapping from [old segment ordinal, old doc_id] -> [new segment
/// ordinal].
#[derive(Debug, Default)]
pub struct DemuxMapping {
/// [index old segment ordinal] -> [index doc_id] = new segment ordinal
@@ -132,27 +133,24 @@ pub fn demux(
#[cfg(test)]
mod tests {
use crate::{
collector::TopDocs,
directory::RamDirectory,
query::QueryParser,
schema::{Schema, TEXT},
DocAddress, Term,
};
use super::*;
use crate::collector::TopDocs;
use crate::directory::RamDirectory;
use crate::query::QueryParser;
use crate::schema::{Schema, TEXT};
use crate::{DocAddress, Term};
#[test]
fn test_demux_map_to_deletebitset() {
let max_value = 2;
let mut demux_mapping = DemuxMapping::default();
//segment ordinal 0 mapping
// segment ordinal 0 mapping
let mut doc_id_to_segment = DocIdToSegmentOrdinal::with_max_doc(max_value);
doc_id_to_segment.set(0, 1);
doc_id_to_segment.set(1, 0);
demux_mapping.add(doc_id_to_segment);
//segment ordinal 1 mapping
// segment ordinal 1 mapping
let mut doc_id_to_segment = DocIdToSegmentOrdinal::with_max_doc(max_value);
doc_id_to_segment.set(0, 1);
doc_id_to_segment.set(1, 1);
@@ -235,13 +233,13 @@ mod tests {
let mut demux_mapping = DemuxMapping::default();
{
let max_value = 2;
//segment ordinal 0 mapping
// segment ordinal 0 mapping
let mut doc_id_to_segment = DocIdToSegmentOrdinal::with_max_doc(max_value);
doc_id_to_segment.set(0, 1);
doc_id_to_segment.set(1, 0);
demux_mapping.add(doc_id_to_segment);
//segment ordinal 1 mapping
// segment ordinal 1 mapping
let mut doc_id_to_segment = DocIdToSegmentOrdinal::with_max_doc(max_value);
doc_id_to_segment.set(0, 1);
doc_id_to_segment.set(1, 1);
@@ -274,7 +272,7 @@ mod tests {
let text_field = index.schema().get_field("text").unwrap();
let do_search = |term: &str| {
let query = QueryParser::for_index(&index, vec![text_field])
let query = QueryParser::for_index(index, vec![text_field])
.parse_query(term)
.unwrap();
let top_docs: Vec<(f32, DocAddress)> =
@@ -303,7 +301,7 @@ mod tests {
let text_field = index.schema().get_field("text").unwrap();
let do_search = |term: &str| {
let query = QueryParser::for_index(&index, vec![text_field])
let query = QueryParser::for_index(index, vec![text_field])
.parse_query(term)
.unwrap();
let top_docs: Vec<(f32, DocAddress)> =

88
src/indexer/doc_id_mapping.rs
View File

@@ -1,13 +1,12 @@
//! This module is used when sorting the index by a property, e.g.
//! to get mappings from old doc_id to new doc_id and vice versa, after sorting
//!
use std::cmp::Reverse;
use std::ops::Index;
use super::SegmentWriter;
use crate::{
schema::{Field, Schema},
DocId, IndexSortByField, Order, SegmentOrdinal, TantivyError,
};
use std::{cmp::Reverse, ops::Index};
use crate::schema::{Field, Schema};
use crate::{DocId, IndexSortByField, Order, SegmentOrdinal, TantivyError};
/// Struct to provide mapping from new doc_id to old doc_id and segment.
#[derive(Clone)]
@@ -63,6 +62,24 @@ pub struct DocIdMapping {
}
impl DocIdMapping {
pub fn from_new_id_to_old_id(new_doc_id_to_old: Vec<DocId>) -> Self {
let max_doc = new_doc_id_to_old.len();
let old_max_doc = new_doc_id_to_old
.iter()
.cloned()
.max()
.map(|n| n + 1)
.unwrap_or(0);
let mut old_doc_id_to_new = vec![0; old_max_doc as usize];
for i in 0..max_doc {
old_doc_id_to_new[new_doc_id_to_old[i] as usize] = i as DocId;
}
DocIdMapping {
new_doc_id_to_old,
old_doc_id_to_new,
}
}
/// returns the new doc_id for the old doc_id
pub fn get_new_doc_id(&self, doc_id: DocId) -> DocId {
self.old_doc_id_to_new[doc_id as usize]
@@ -75,6 +92,13 @@ impl DocIdMapping {
pub fn iter_old_doc_ids(&self) -> impl Iterator<Item = DocId> + Clone + '_ {
self.new_doc_id_to_old.iter().cloned()
}
/// Remaps a given array to the new doc ids.
pub fn remap<T: Copy>(&self, els: &[T]) -> Vec<T> {
self.new_doc_id_to_old
.iter()
.map(|old_doc| els[*old_doc as usize])
.collect()
}
}
pub(crate) fn expect_field_id_for_sort_field(
@@ -122,26 +146,17 @@ pub(crate) fn get_doc_id_mapping_from_field(
.into_iter()
.map(|el| el.0)
.collect::<Vec<_>>();
// create old doc_id to new doc_id index (used in posting recorder)
let max_doc = new_doc_id_to_old.len();
let mut old_doc_id_to_new = vec![0; max_doc];
for i in 0..max_doc {
old_doc_id_to_new[new_doc_id_to_old[i] as usize] = i as DocId;
}
let doc_id_map = DocIdMapping {
new_doc_id_to_old,
old_doc_id_to_new,
};
Ok(doc_id_map)
Ok(DocIdMapping::from_new_id_to_old_id(new_doc_id_to_old))
}
#[cfg(test)]
mod tests_indexsorting {
use crate::collector::TopDocs;
use crate::fastfield::FastFieldReader;
use crate::{collector::TopDocs, query::QueryParser, schema::*};
use crate::{schema::Schema, DocAddress};
use crate::{Index, IndexSettings, IndexSortByField, Order};
use crate::indexer::doc_id_mapping::DocIdMapping;
use crate::query::QueryParser;
use crate::schema::{Schema, *};
use crate::{DocAddress, Index, IndexSettings, IndexSortByField, Order};
fn create_test_index(
index_settings: Option<IndexSettings>,
@@ -202,7 +217,7 @@ mod tests_indexsorting {
];
for option in options {
//let options = get_text_options();
// let options = get_text_options();
// no index_sort
let index = create_test_index(None, option.clone())?;
let my_text_field = index.schema().get_field("text_field").unwrap();
@@ -303,7 +318,7 @@ mod tests_indexsorting {
.doc(DocAddress::new(0, 3))?
.get_first(my_string_field)
.unwrap()
.text(),
.as_text(),
Some("blublub")
);
}
@@ -326,7 +341,7 @@ mod tests_indexsorting {
.doc(DocAddress::new(0, 0))?
.get_first(my_string_field)
.unwrap()
.text(),
.as_text(),
Some("blublub")
);
let doc = searcher.doc(DocAddress::new(0, 4))?;
@@ -348,7 +363,7 @@ mod tests_indexsorting {
{
let doc = searcher.doc(DocAddress::new(0, 4))?;
assert_eq!(
doc.get_first(my_string_field).unwrap().text(),
doc.get_first(my_string_field).unwrap().as_text(),
Some("blublub")
);
}
@@ -475,4 +490,27 @@ mod tests_indexsorting {
assert_eq!(vals, &[3]);
Ok(())
}
#[test]
fn test_doc_mapping() {
let doc_mapping = DocIdMapping::from_new_id_to_old_id(vec![3, 2, 5]);
assert_eq!(doc_mapping.get_old_doc_id(0), 3);
assert_eq!(doc_mapping.get_old_doc_id(1), 2);
assert_eq!(doc_mapping.get_old_doc_id(2), 5);
assert_eq!(doc_mapping.get_new_doc_id(0), 0);
assert_eq!(doc_mapping.get_new_doc_id(1), 0);
assert_eq!(doc_mapping.get_new_doc_id(2), 1);
assert_eq!(doc_mapping.get_new_doc_id(3), 0);
assert_eq!(doc_mapping.get_new_doc_id(4), 0);
assert_eq!(doc_mapping.get_new_doc_id(5), 2);
}
#[test]
fn test_doc_mapping_remap() {
let doc_mapping = DocIdMapping::from_new_id_to_old_id(vec![2, 8, 3]);
assert_eq!(
&doc_mapping.remap(&[0, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000]),
&[2000, 8000, 3000]
);
}
}

3
src/indexer/doc_opstamp_mapping.rs
View File

@@ -1,5 +1,4 @@
use crate::DocId;
use crate::Opstamp;
use crate::{DocId, Opstamp};
// Doc to opstamp is used to identify which
// document should be deleted.

162
src/indexer/index_writer.rs
View File

@@ -1,14 +1,19 @@
use std::ops::Range;
use std::sync::Arc;
use std::thread;
use std::thread::JoinHandle;
use common::BitSet;
use crossbeam::channel;
use futures::executor::block_on;
use futures::future::Future;
use smallvec::smallvec;
use super::operation::{AddOperation, UserOperation};
use super::segment_updater::SegmentUpdater;
use super::PreparedCommit;
use crate::core::Index;
use crate::core::Segment;
use crate::core::SegmentComponent;
use crate::core::SegmentId;
use crate::core::SegmentMeta;
use crate::core::SegmentReader;
use crate::directory::TerminatingWrite;
use crate::directory::{DirectoryLock, GarbageCollectionResult};
use super::{AddBatch, AddBatchReceiver, AddBatchSender, PreparedCommit};
use crate::core::{Index, Segment, SegmentComponent, SegmentId, SegmentMeta, SegmentReader};
use crate::directory::{DirectoryLock, GarbageCollectionResult, TerminatingWrite};
use crate::docset::{DocSet, TERMINATED};
use crate::error::TantivyError;
use crate::fastfield::write_alive_bitset;
@@ -17,32 +22,17 @@ use crate::indexer::doc_opstamp_mapping::DocToOpstampMapping;
use crate::indexer::index_writer_status::IndexWriterStatus;
use crate::indexer::operation::DeleteOperation;
use crate::indexer::stamper::Stamper;
use crate::indexer::MergePolicy;
use crate::indexer::SegmentEntry;
use crate::indexer::SegmentWriter;
use crate::schema::Document;
use crate::schema::IndexRecordOption;
use crate::schema::Term;
use crate::indexer::{MergePolicy, SegmentEntry, SegmentWriter};
use crate::schema::{Document, IndexRecordOption, Term};
use crate::Opstamp;
use common::BitSet;
use crossbeam::channel;
use futures::executor::block_on;
use futures::future::Future;
use smallvec::smallvec;
use std::ops::Range;
use std::sync::Arc;
use std::thread;
use std::thread::JoinHandle;
use super::{AddBatch, AddBatchReceiver, AddBatchSender};
// Size of the margin for the heap. A segment is closed when the remaining memory
// in the heap goes below MARGIN_IN_BYTES.
// Size of the margin for the `memory_arena`. A segment is closed when the remaining memory
// in the `memory_arena` goes below MARGIN_IN_BYTES.
pub const MARGIN_IN_BYTES: usize = 1_000_000;
// We impose the memory per thread to be at least 3 MB.
pub const HEAP_SIZE_MIN: usize = ((MARGIN_IN_BYTES as u32) * 3u32) as usize;
pub const HEAP_SIZE_MAX: usize = u32::max_value() as usize - MARGIN_IN_BYTES;
pub const MEMORY_ARENA_NUM_BYTES_MIN: usize = ((MARGIN_IN_BYTES as u32) * 3u32) as usize;
pub const MEMORY_ARENA_NUM_BYTES_MAX: usize = u32::max_value() as usize - MARGIN_IN_BYTES;
// We impose the number of index writter thread to be at most this.
pub const MAX_NUM_THREAD: usize = 8;
@@ -71,7 +61,7 @@ pub struct IndexWriter {
index: Index,
heap_size_in_bytes_per_thread: usize,
memory_arena_in_bytes_per_thread: usize,
workers_join_handle: Vec<JoinHandle<crate::Result<()>>>,
@@ -189,10 +179,10 @@ fn index_documents(
) -> crate::Result<()> {
let schema = segment.schema();
let mut segment_writer = SegmentWriter::for_segment(memory_budget, segment.clone(), &schema)?;
let mut segment_writer = SegmentWriter::for_segment(memory_budget, segment.clone(), schema)?;
for document_group in grouped_document_iterator {
for doc in document_group {
segment_writer.add_document(doc, &schema)?;
segment_writer.add_document(doc)?;
}
let mem_usage = segment_writer.mem_usage();
if mem_usage >= memory_budget - MARGIN_IN_BYTES {
@@ -278,22 +268,26 @@ impl IndexWriter {
/// should work at the same time.
/// # Errors
/// If the lockfile already exists, returns `Error::FileAlreadyExists`.
/// If the heap size per thread is too small or too big, returns `TantivyError::InvalidArgument`
/// If the memory arena per thread is too small or too big, returns
/// `TantivyError::InvalidArgument`
pub(crate) fn new(
index: &Index,
num_threads: usize,
heap_size_in_bytes_per_thread: usize,
memory_arena_in_bytes_per_thread: usize,
directory_lock: DirectoryLock,
) -> crate::Result<IndexWriter> {
if heap_size_in_bytes_per_thread < HEAP_SIZE_MIN {
if memory_arena_in_bytes_per_thread < MEMORY_ARENA_NUM_BYTES_MIN {
let err_msg = format!(
"The heap size per thread needs to be at least {}.",
HEAP_SIZE_MIN
"The memory arena in bytes per thread needs to be at least {}.",
MEMORY_ARENA_NUM_BYTES_MIN
);
return Err(TantivyError::InvalidArgument(err_msg));
}
if heap_size_in_bytes_per_thread >= HEAP_SIZE_MAX {
let err_msg = format!("The heap size per thread cannot exceed {}", HEAP_SIZE_MAX);
if memory_arena_in_bytes_per_thread >= MEMORY_ARENA_NUM_BYTES_MAX {
let err_msg = format!(
"The memory arena in bytes per thread cannot exceed {}",
MEMORY_ARENA_NUM_BYTES_MAX
);
return Err(TantivyError::InvalidArgument(err_msg));
}
let (document_sender, document_receiver): (AddBatchSender, AddBatchReceiver) =
@@ -311,7 +305,7 @@ impl IndexWriter {
let mut index_writer = IndexWriter {
_directory_lock: Some(directory_lock),
heap_size_in_bytes_per_thread,
memory_arena_in_bytes_per_thread,
index: index.clone(),
index_writer_status: IndexWriterStatus::from(document_receiver),
@@ -392,7 +386,13 @@ impl IndexWriter {
fn operation_receiver(&self) -> crate::Result<AddBatchReceiver> {
self.index_writer_status
.operation_receiver()
.ok_or_else(|| crate::TantivyError::ErrorInThread("The index writer was killed. It can happen if an indexing worker encounterred an Io error for instance.".to_string()))
.ok_or_else(|| {
crate::TantivyError::ErrorInThread(
"The index writer was killed. It can happen if an indexing worker \
encounterred an Io error for instance."
.to_string(),
)
})
}
/// Spawns a new worker thread for indexing.
@@ -405,7 +405,7 @@ impl IndexWriter {
let mut delete_cursor = self.delete_queue.cursor();
let mem_budget = self.heap_size_in_bytes_per_thread;
let mem_budget = self.memory_arena_in_bytes_per_thread;
let index = self.index.clone();
let join_handle: JoinHandle<crate::Result<()>> = thread::Builder::new()
.name(format!("thrd-tantivy-index{}", self.worker_id))
@@ -553,7 +553,7 @@ impl IndexWriter {
// marks the segment updater as killed. From now on, all
// segment updates will be ignored.
self.segment_updater.kill();
let document_receiver = self.operation_receiver();
let document_receiver_res = self.operation_receiver();
// take the directory lock to create a new index_writer.
let directory_lock = self
@@ -564,7 +564,7 @@ impl IndexWriter {
let new_index_writer: IndexWriter = IndexWriter::new(
&self.index,
self.num_threads,
self.heap_size_in_bytes_per_thread,
self.memory_arena_in_bytes_per_thread,
directory_lock,
)?;
@@ -579,7 +579,9 @@ impl IndexWriter {
//
// This will reach an end as the only document_sender
// was dropped with the index_writer.
for _ in document_receiver {}
if let Ok(document_receiver) = document_receiver_res {
for _ in document_receiver {}
}
Ok(self.committed_opstamp)
}
@@ -651,7 +653,6 @@ impl IndexWriter {
///
/// Commit returns the `opstamp` of the last document
/// that made it in the commit.
///
pub fn commit(&mut self) -> crate::Result<Opstamp> {
self.prepare_commit()?.commit()
}
@@ -778,8 +779,7 @@ impl Drop for IndexWriter {
#[cfg(test)]
mod tests {
use std::collections::HashMap;
use std::collections::HashSet;
use std::collections::{HashMap, HashSet};
use futures::executor::block_on;
use proptest::prelude::*;
@@ -792,30 +792,20 @@ mod tests {
use crate::error::*;
use crate::fastfield::FastFieldReader;
use crate::indexer::NoMergePolicy;
use crate::query::QueryParser;
use crate::query::TermQuery;
use crate::schema::Cardinality;
use crate::schema::Facet;
use crate::schema::IntOptions;
use crate::schema::TextFieldIndexing;
use crate::schema::TextOptions;
use crate::schema::STORED;
use crate::schema::TEXT;
use crate::schema::{self, IndexRecordOption, FAST, INDEXED, STRING};
use crate::DocAddress;
use crate::Index;
use crate::ReloadPolicy;
use crate::Term;
use crate::{IndexSettings, IndexSortByField, Order};
use crate::query::{QueryParser, TermQuery};
use crate::schema::{
self, Cardinality, Facet, FacetOptions, IndexRecordOption, IntOptions, TextFieldIndexing,
TextOptions, FAST, INDEXED, STORED, STRING, TEXT,
};
use crate::{DocAddress, Index, IndexSettings, IndexSortByField, Order, ReloadPolicy, Term};
const LOREM: &str = "Doc Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed \
do eiusmod tempor incididunt ut labore et dolore magna aliqua. \
Ut enim ad minim veniam, quis nostrud exercitation ullamco \
laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure \
dolor in reprehenderit in voluptate velit esse cillum dolore eu \
fugiat nulla pariatur. Excepteur sint occaecat cupidatat non \
proident, sunt in culpa qui officia deserunt mollit anim id est \
laborum.";
const LOREM: &str = "Doc Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do \
eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad \
minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip \
ex ea commodo consequat. Duis aute irure dolor in reprehenderit in \
voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur \
sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt \
mollit anim id est laborum.";
#[test]
fn test_operations_group() {
@@ -970,8 +960,8 @@ mod tests {
let index_writer = index.writer(3_000_000).unwrap();
assert_eq!(
format!("{:?}", index_writer.get_merge_policy()),
"LogMergePolicy { min_num_segments: 8, max_docs_before_merge: 10000000, min_layer_size: 10000, \
level_log_size: 0.75 }"
"LogMergePolicy { min_num_segments: 8, max_docs_before_merge: 10000000, \
min_layer_size: 10000, level_log_size: 0.75, del_docs_ratio_before_merge: 1.0 }"
);
let merge_policy = Box::new(NoMergePolicy::default());
index_writer.set_merge_policy(merge_policy);
@@ -1399,6 +1389,7 @@ mod tests {
) -> crate::Result<()> {
let mut schema_builder = schema::Schema::builder();
let id_field = schema_builder.add_u64_field("id", FAST | INDEXED | STORED);
let bytes_field = schema_builder.add_bytes_field("bytes", FAST | INDEXED | STORED);
let text_field = schema_builder.add_text_field(
"text_field",
TextOptions::default()
@@ -1417,7 +1408,7 @@ mod tests {
.set_fast(Cardinality::MultiValues)
.set_stored(),
);
let facet_field = schema_builder.add_facet_field("facet", INDEXED);
let facet_field = schema_builder.add_facet_field("facet", FacetOptions::default());
let schema = schema_builder.build();
let settings = if sort_index {
IndexSettings {
@@ -1445,8 +1436,14 @@ mod tests {
match op {
IndexingOp::AddDoc { id } => {
let facet = Facet::from(&("/cola/".to_string() + &id.to_string()));
index_writer
.add_document(doc!(id_field=>id, multi_numbers=> id, multi_numbers => id, text_field => id.to_string(), facet_field => facet, large_text_field=> LOREM))?;
index_writer.add_document(doc!(id_field=>id,
bytes_field => id.to_le_bytes().as_slice(),
multi_numbers=> id,
multi_numbers => id,
text_field => id.to_string(),
facet_field => facet,
large_text_field=> LOREM
))?;
}
IndexingOp::DeleteDoc { id } => {
index_writer.delete_term(Term::from_field_u64(id_field, id));
@@ -1544,12 +1541,7 @@ mod tests {
let store_reader = segment_reader.get_store_reader().unwrap();
// test store iterator
for doc in store_reader.iter(segment_reader.alive_bitset()) {
let id = doc
.unwrap()
.get_first(id_field)
.unwrap()
.u64_value()
.unwrap();
let id = doc.unwrap().get_first(id_field).unwrap().as_u64().unwrap();
assert!(expected_ids_and_num_occurences.contains_key(&id));
}
// test store random access
@@ -1559,7 +1551,7 @@ mod tests {
.unwrap()
.get_first(id_field)
.unwrap()
.u64_value()
.as_u64()
.unwrap();
assert!(expected_ids_and_num_occurences.contains_key(&id));
let id2 = store_reader
@@ -1567,7 +1559,7 @@ mod tests {
.unwrap()
.get_first(multi_numbers)
.unwrap()
.u64_value()
.as_u64()
.unwrap();
assert_eq!(id, id2);
}

8
src/indexer/index_writer_status.rs
View File

@@ -22,7 +22,7 @@ impl IndexWriterStatus {
.receive_channel
.read()
.expect("This lock should never be poisoned");
rlock.as_ref().map(|receiver| receiver.clone())
rlock.as_ref().cloned()
}
/// Create an index writer bomb.
@@ -90,10 +90,12 @@ impl Drop for IndexWriterBomb {
#[cfg(test)]
mod tests {
use super::IndexWriterStatus;
use crossbeam::channel;
use std::mem;
use crossbeam::channel;
use super::IndexWriterStatus;
#[test]
fn test_bomb_goes_boom() {
let (_tx, rx) = channel::bounded(10);

113
src/indexer/log_merge_policy.rs
View File

@@ -1,13 +1,18 @@
use std::cmp;
use itertools::Itertools;
use super::merge_policy::{MergeCandidate, MergePolicy};
use crate::core::SegmentMeta;
use itertools::Itertools;
use std::cmp;
use std::f64;
const DEFAULT_LEVEL_LOG_SIZE: f64 = 0.75;
const DEFAULT_MIN_LAYER_SIZE: u32 = 10_000;
const DEFAULT_MIN_NUM_SEGMENTS_IN_MERGE: usize = 8;
const DEFAULT_MAX_DOCS_BEFORE_MERGE: usize = 10_000_000;
// The default value of 1 means that deletes are not taken in account when
// identifying merge candidates. This is not a very sensible default: it was
// set like that for backward compatibility and might change in the near future.
const DEFAULT_DEL_DOCS_RATIO_BEFORE_MERGE: f32 = 1.0f32;
/// `LogMergePolicy` tries to merge segments that have a similar number of
/// documents.
@@ -17,6 +22,7 @@ pub struct LogMergePolicy {
max_docs_before_merge: usize,
min_layer_size: u32,
level_log_size: f64,
del_docs_ratio_before_merge: f32,
}
impl LogMergePolicy {
@@ -52,19 +58,49 @@ impl LogMergePolicy {
pub fn set_level_log_size(&mut self, level_log_size: f64) {
self.level_log_size = level_log_size;
}
/// Set the ratio of deleted documents in a segment to tolerate.
///
/// If it is exceeded by any segment at a log level, a merge
/// will be triggered for that level.
///
/// If there is a single segment at a level, we effectively end up expunging
/// deleted documents from it.
///
/// # Panics
///
/// Panics if del_docs_ratio_before_merge is not within (0..1].
pub fn set_del_docs_ratio_before_merge(&mut self, del_docs_ratio_before_merge: f32) {
assert!(del_docs_ratio_before_merge <= 1.0f32);
assert!(del_docs_ratio_before_merge > 0f32);
self.del_docs_ratio_before_merge = del_docs_ratio_before_merge;
}
fn has_segment_above_deletes_threshold(&self, level: &[&SegmentMeta]) -> bool {
level
.iter()
.any(|segment| deletes_ratio(segment) > self.del_docs_ratio_before_merge)
}
}
fn deletes_ratio(segment: &SegmentMeta) -> f32 {
if segment.max_doc() == 0 {
return 0f32;
}
segment.num_deleted_docs() as f32 / segment.max_doc() as f32
}
impl MergePolicy for LogMergePolicy {
fn compute_merge_candidates(&self, segments: &[SegmentMeta]) -> Vec<MergeCandidate> {
let mut size_sorted_segments = segments
let size_sorted_segments = segments
.iter()
.filter(|segment_meta| segment_meta.num_docs() <= (self.max_docs_before_merge as u32))
.filter(|seg| seg.num_docs() <= (self.max_docs_before_merge as u32))
.sorted_by_key(|seg| std::cmp::Reverse(seg.max_doc()))
.collect::<Vec<&SegmentMeta>>();
if size_sorted_segments.len() <= 1 {
if size_sorted_segments.is_empty() {
return vec![];
}
size_sorted_segments.sort_by_key(|seg| std::cmp::Reverse(seg.num_docs()));
let mut current_max_log_size = f64::MAX;
let mut levels = vec![];
@@ -82,7 +118,10 @@ impl MergePolicy for LogMergePolicy {
levels
.iter()
.filter(|level| level.len() >= self.min_num_segments)
.filter(|level| {
level.len() >= self.min_num_segments
|| self.has_segment_above_deletes_threshold(level)
})
.map(|segments| MergeCandidate(segments.iter().map(|&seg| seg.id()).collect()))
.collect()
}
@@ -95,20 +134,21 @@ impl Default for LogMergePolicy {
max_docs_before_merge: DEFAULT_MAX_DOCS_BEFORE_MERGE,
min_layer_size: DEFAULT_MIN_LAYER_SIZE,
level_log_size: DEFAULT_LEVEL_LOG_SIZE,
del_docs_ratio_before_merge: DEFAULT_DEL_DOCS_RATIO_BEFORE_MERGE,
}
}
}
#[cfg(test)]
mod tests {
use super::*;
use crate::{
core::{SegmentId, SegmentMeta, SegmentMetaInventory},
schema,
};
use crate::{indexer::merge_policy::MergePolicy, schema::INDEXED};
use once_cell::sync::Lazy;
use super::*;
use crate::core::{SegmentId, SegmentMeta, SegmentMetaInventory};
use crate::indexer::merge_policy::MergePolicy;
use crate::schema;
use crate::schema::INDEXED;
static INVENTORY: Lazy<SegmentMetaInventory> = Lazy::new(SegmentMetaInventory::default);
use crate::Index;
@@ -288,4 +328,49 @@ mod tests {
assert_eq!(result_list[0].0[1], test_input[4].id());
assert_eq!(result_list[0].0[2], test_input[5].id());
}
#[test]
fn test_merge_single_segment_with_deletes_below_threshold() {
let mut test_merge_policy = test_merge_policy();
test_merge_policy.set_del_docs_ratio_before_merge(0.25f32);
let test_input = vec![create_random_segment_meta(40_000).with_delete_meta(10_000, 1)];
let merge_candidates = test_merge_policy.compute_merge_candidates(&test_input);
assert!(merge_candidates.is_empty());
}
#[test]
fn test_merge_single_segment_with_deletes_above_threshold() {
let mut test_merge_policy = test_merge_policy();
test_merge_policy.set_del_docs_ratio_before_merge(0.25f32);
let test_input = vec![create_random_segment_meta(40_000).with_delete_meta(10_001, 1)];
let merge_candidates = test_merge_policy.compute_merge_candidates(&test_input);
assert_eq!(merge_candidates.len(), 1);
}
#[test]
fn test_merge_segments_with_deletes_above_threshold_all_in_level() {
let mut test_merge_policy = test_merge_policy();
test_merge_policy.set_del_docs_ratio_before_merge(0.25f32);
let test_input = vec![
create_random_segment_meta(40_000).with_delete_meta(10_001, 1),
create_random_segment_meta(40_000),
];
let merge_candidates = test_merge_policy.compute_merge_candidates(&test_input);
assert_eq!(merge_candidates.len(), 1);
assert_eq!(merge_candidates[0].0.len(), 2);
}
#[test]
fn test_merge_segments_with_deletes_above_threshold_different_level_not_involved() {
let mut test_merge_policy = test_merge_policy();
test_merge_policy.set_del_docs_ratio_before_merge(0.25f32);
let test_input = vec![
create_random_segment_meta(100),
create_random_segment_meta(40_000).with_delete_meta(10_001, 1),
];
let merge_candidates = test_merge_policy.compute_merge_candidates(&test_input);
assert_eq!(merge_candidates.len(), 1);
assert_eq!(merge_candidates[0].0.len(), 1);
assert_eq!(merge_candidates[0].0[0], test_input[1].id());
}
}

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