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

Compare commits

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

180 Commits
sparse_cod ... 0.19

Author SHA1 Message Date
PSeitz
2c50b02eb3 Fix max bucket limit in histogram (#1703) 
* Fix max bucket limit in histogram

The max bucket limit in histogram was broken, since some code introduced temporary filtering of buckets, which then resulted into an incorrect increment on the bucket count.
The provided solution covers more scenarios, but there are still some scenarios unhandled (See #1702).

* Apply suggestions from code review

Co-authored-by: Paul Masurel <paul@quickwit.io>

Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-12-12 04:40:15 +01:00
PSeitz
509adab79d Bump version (#1715) 
* group workspace deps

* update cargo.toml

* revert tant version

* chore: Release
 2022-12-12 04:39:43 +01:00
PSeitz
96c93a6ba3 Merge pull request #1700 from quickwit-oss/PSeitz-patch-1 
Update CHANGELOG.md
 2022-12-02 16:31:11 +01:00
boraarslan
495824361a Move split_full_path to Schema (#1692) 2022-11-29 20:56:13 +09:00
PSeitz
485a8f507e Update CHANGELOG.md 2022-11-28 15:41:31 +01:00
PSeitz
1119e59eae prepare fastfield format for null index (#1691) 
* prepare fastfield format for null index
* add format version for fastfield
* Update fastfield_codecs/src/compact_space/mod.rs
* switch to variable size footer
* serialize delta of end
 2022-11-28 17:15:24 +09:00
PSeitz
ee1f2c1f28 add aggregation support for date type (#1693) 
* add aggregation support for date type
fixes #1332

* serialize key_as_string as rfc3339 in date histogram
* update docs
* enable date for range aggregation
 2022-11-28 09:12:08 +09:00
PSeitz
600548fd26 Merge pull request #1694 from quickwit-oss/dependabot/cargo/zstd-0.12 
Update zstd requirement from 0.11 to 0.12
 2022-11-25 05:48:59 +01:00
PSeitz
9929c0c221 Merge pull request #1696 from quickwit-oss/dependabot/cargo/env_logger-0.10.0 
Update env_logger requirement from 0.9.0 to 0.10.0
 2022-11-25 03:28:10 +01:00
dependabot[bot]
f53e65648b Update env_logger requirement from 0.9.0 to 0.10.0 
Updates the requirements on [env_logger](https://github.com/rust-cli/env_logger) to permit the latest version.
- [Release notes](https://github.com/rust-cli/env_logger/releases)
- [Changelog](https://github.com/rust-cli/env_logger/blob/main/CHANGELOG.md)
- [Commits](https://github.com/rust-cli/env_logger/compare/v0.9.0...v0.10.0)

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

Signed-off-by: dependabot[bot] <support@github.com>
 2022-11-24 20:07:52 +00:00
PSeitz
0281b22b77 update create_in_ram docs (#1695) 2022-11-24 17:30:09 +01:00
dependabot[bot]
a05c184830 Update zstd requirement from 0.11 to 0.12 
Updates the requirements on [zstd](https://github.com/gyscos/zstd-rs) to permit the latest version.
- [Release notes](https://github.com/gyscos/zstd-rs/releases)
- [Commits](https://github.com/gyscos/zstd-rs/commits)

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

Signed-off-by: dependabot[bot] <support@github.com>
 2022-11-23 20:15:32 +00:00
Paul Masurel
0b40a7fe43 Added a expand_dots JsonObjectOptions. (#1687) 
Related with quickwit#2345.
 2022-11-21 23:03:00 +09:00
trinity-1686a
e758080465 add support for TermSetQuery in query parser (#1683) 2022-11-17 16:49:49 +01:00
Paul Masurel
2a39289a1b Handle escaped dot in json path in the QueryParser. (#1682) 2022-11-16 07:18:34 +09:00
Adam Reichold
ca6231170e Make the built-in stop word lists selectable via the Language enum already used by the Stemmer filter. (#1671) 2022-11-15 17:40:25 +09:00
PSeitz
eda6e5a10a Merge pull request #1681 from quickwit-oss/ip_range_query_multi 
remove Column from MultiValuedU128FastFieldReader
 2022-11-15 09:27:46 +08:00
Pascal Seitz
8641155cbb remove column from MultiValuedU128FastFieldReader 2022-11-14 18:49:15 +08:00
PSeitz
9a090ed994 Merge pull request #1659 from quickwit-oss/ip_range_query_multi 
add support for ip range query on multivalue fastfields
 2022-11-14 15:17:41 +08:00
Pascal Seitz
b7d0dd154a fmt 2022-11-14 14:49:15 +08:00
PSeitz
ce10fab20f Apply suggestions from code review 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-11-14 14:21:53 +08:00
Pascal Seitz
e034328a8b Improve position_to_docid, refactor, add tests 2022-11-14 14:21:53 +08:00
Pascal Seitz
f811d1616b add support for ip range query on multivalue fastfields 2022-11-14 14:21:52 +08:00
PSeitz
c665b16ff0 Merge pull request #1672 from quickwit-oss/allow_range_without_indexed 
Allow range query on fastfield without INDEXED
 2022-11-14 12:45:12 +08:00
PSeitz
3b5f810051 Merge pull request #1677 from quickwit-oss/switch_to_u32 
switch total_num_val to u32
 2022-11-14 12:01:40 +08:00
trinity-1686a
5765c261aa allow warming up of the full posting list (#1673) 
* allow warming up of the full posting list

* cargo fmt
 2022-11-14 10:27:56 +09:00
Pascal Seitz
fb9f03118d switch total_num_val to u32 2022-11-11 17:35:52 +08:00
PSeitz
55a9d808d4 Merge pull request #1674 from quickwit-oss/u128_codec_header 
add header with codec type for u128
 2022-11-11 13:47:51 +08:00
Pascal Seitz
32166682b3 add header deser test 2022-11-11 13:28:12 +08:00
Pascal Seitz
e6acf8f76d add header with codec type for u128 2022-11-11 11:52:17 +08:00
Pascal Seitz
9e8a0c2cca Allow range query on fastfield without INDEXED 2022-11-10 15:56:08 +08:00
Paul Masurel
3edf0a2724 Using the manual reload policy in IndexWriter. (#1667) 2022-11-09 11:20:41 +01:00
Paul Masurel
8ca12a5683 Added stop word filter to CHANGELOG.md 2022-11-09 17:00:45 +09:00
Adam Reichold
a4b759d2fe Include stop word lists from Lucene and the Snowball project (#1666) 2022-11-09 16:57:35 +09:00
PSeitz
3e9c806890 Merge pull request #1665 from quickwit-oss/fix_num_vals 
fix num_vals on u128 value index after merge
 2022-11-07 21:46:02 +08:00
Pascal Seitz
c69a873dd3 fix num_vals on value index after merge 2022-11-07 21:05:21 +08:00
PSeitz
666afcf641 Merge pull request #1663 from PSeitz/fix_clippy 
fix clippy
 2022-11-07 18:11:20 +08:00
Pascal Seitz
38ad46e580 fix clippy 2022-11-07 16:09:55 +08:00
PSeitz
e948889f4c Merge pull request #1662 from quickwit-oss/fix_num_vals 
fix num_vals in multivalue index after merge
 2022-11-07 15:57:32 +08:00
Pascal Seitz
6e636c9cea fix num_vals in multivalue index after merge 2022-11-07 15:00:52 +08:00
PSeitz
5a610efbc1 Merge pull request #1661 from quickwit-oss/upgrade_criterion 
update criterion to 0.4
 2022-11-04 14:45:34 +08:00
Pascal Seitz
500a0d5e48 update criterion to 0.4 2022-11-04 13:26:29 +08:00
PSeitz
509a265659 add docstore version (#1652) 
* add docstore version

closes #1589

* assert for docstore version
 2022-11-04 10:19:16 +09:00
PSeitz
5b2cea1b97 Merge pull request #1656 from quickwit-oss/multival_offset_index 
move multivalue index to own file
 2022-11-02 14:03:06 +08:00
PSeitz
a5a80ffaea Update fastfield_codecs/src/column.rs 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-11-02 06:37:27 +01:00
PSeitz
0f98d91a39 Merge pull request #1646 from quickwit-oss/no_score_calls 
No score calls if score is not requested
 2022-11-01 20:09:32 +08:00
PSeitz
2af6b01c17 Update src/query/boolean_query/boolean_weight.rs 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-11-01 16:13:00 +08:00
Adam Reichold
c32ab66bbd Small improvements to StopWorldFilter (#1657) 
* Do not copy the whole set of stop words for each stream

* Make construction of StopWordFilter more flexible.
 2022-11-01 16:47:34 +09:00
PSeitz
3f3a6f9990 Merge pull request #1653 from quickwit-oss/faster_hash 
switch to fx hashmap
 2022-11-01 14:53:18 +08:00
Pascal Seitz
83325d8f3f move multivalue index to own file 
start_doc parameter in positions to docids
 2022-11-01 10:36:13 +08:00
PSeitz
4e46f4f8c4 Merge pull request #1649 from adamreichold/split-compound-words 
RFC: Add dictionary-based SplitCompoundWords token filter.
 2022-10-27 17:12:48 +08:00
Pascal Seitz
43df356010 rename to docset 2022-10-27 16:53:38 +08:00
PSeitz
6647362464 Merge pull request #1648 from adamreichold/stemmer-todo-alloc 
Avoid unconditional allocation in StemmerTokenStream.
 2022-10-27 16:50:41 +08:00
Pascal Seitz
279b1b28d3 switch to fx hashmap 2022-10-27 16:19:59 +08:00
PSeitz
7a80851e36 Merge pull request #1645 from quickwit-oss/ip_field_range_query 
add ip range query benchmark, add seek behaviour
 2022-10-27 16:13:52 +08:00
Adam Reichold
cd952429d2 Add dictionary-based SplitCompoundWords token filter. 2022-10-27 08:30:33 +02:00
PSeitz
d777c964da Merge pull request #1650 from adamreichold/fnv-rustc-hash 
Replace FNV by rustc-hash
 2022-10-27 12:11:26 +08:00
Adam Reichold
bbb058d976 Replace FNV by rustc-hash 
Both construction have similar goals but rustc-hash ist better suited for
contemporary CPU as it works one word at a time instead of byte per byte.
 2022-10-27 00:35:09 +02:00
Adam Reichold
5f7d027a52 Avoid unconditional allocation in StemmerTokenStream. 
This fixes the TODO in two ways: If the stemmer already yields an owned string,
it is used directly as the new text of the token. Otherwise, a temporary buffer
is used to copy the stemmed text (just as before) and then swapping it into the
token to reuse its existing buffer.
 2022-10-26 18:11:15 +02:00
Pascal Seitz
dfab201191 for_each_docset to iterate without score 2022-10-26 17:25:05 +08:00
PSeitz
0c2bd36fe3 Panic on duplicate field names (#1647) 
fixes #1601
 2022-10-26 16:17:33 +09:00
Pascal Seitz
af839753e0 No score calls if score is not requested 2022-10-26 12:18:35 +08:00
Pascal Seitz
fec2b63571 improve bench by adding more blanks in compact space 2022-10-25 22:09:01 +08:00
Pascal Seitz
6213ea476a pass positions parameter 2022-10-25 17:44:51 +08:00
Pascal Seitz
5e159c26bf add ip range query benchmark, add seek behaviour 2022-10-25 15:57:19 +08:00
PSeitz
a5e59ab598 Merge pull request #1644 from quickwit-oss/get_val_u32 
switch get_val() to u32
 2022-10-24 19:30:03 +08:00
Pascal Seitz
e772d3170d switch get_val() to u32 
Fixes #1638
 2022-10-24 19:05:57 +08:00
PSeitz
8c2ba7bd55 Merge pull request #1637 from quickwit-oss/ip_field_range_query 
add range query via ip fast field
 2022-10-24 18:10:47 +08:00
Pascal Seitz
02328b0151 fix proptest 2022-10-24 17:46:06 +08:00
Pascal Seitz
7cc775256c add comments, rename 2022-10-24 17:08:37 +08:00
Pascal Seitz
07b40f8b8b add proptest 2022-10-24 16:52:55 +08:00
PSeitz
9b6b6be5b9 Apply suggestions from code review 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-10-24 16:00:38 +08:00
Pascal Seitz
6bb73a527f add range query via ip fast field 2022-10-24 16:00:38 +08:00
PSeitz
03885d0f3c Merge pull request #1643 from quickwit-oss/range_query_parser 
allow more characters in range query
 2022-10-24 15:09:47 +08:00
Pascal Seitz
f2e5135870 allow more characters in range query 
closes #1642
 2022-10-21 18:05:15 +08:00
Paul Masurel
c24157f28b Bumping version format. (#1640) 
The docstore format has changed in a non-compatible manner.
 2022-10-21 15:35:35 +09:00
PSeitz
873382cdcb Merge pull request #1639 from quickwit-oss/num_vals_u32 
switch num_vals() to u32
 2022-10-21 12:36:50 +08:00
Pascal Seitz
791350091c switch num_vals() to u32 
fixes #1630
 2022-10-20 19:44:28 +08:00
Paul Masurel
483b1d13d4 Added unit test for long tokens (#1635) 
* Bugfix on long tokens and multivalue text fields.

Fixes a minor bug for the strong edge case
in which a tokenizer would emit tokens where
the last token does not cover the last position.

More importantly, this adds unit tests.

Closes #1634

* Update src/indexer/segment_writer.rs

Co-authored-by: PSeitz <PSeitz@users.noreply.github.com>

Co-authored-by: PSeitz <PSeitz@users.noreply.github.com>
 2022-10-20 15:05:37 +09:00
PSeitz
8de7fa9d95 Merge pull request #1631 from quickwit-oss/high_positions 
add test for phrase search on multi text field
 2022-10-20 10:26:00 +08:00
Paul Masurel
94313b62f8 Hotfix issue/1629 - position broken (#1633) 
* Bugfix position broken.

For Field with several FieldValues, with a
value that contained no token at all, the token position
was reinitialized to 0.

As a result, PhraseQueries can show some false positives.
In addition, after the computation of the position delta, we can
underflow u32, and end up with gigantic delta.

We haven't been able to actually explain the bug in 1629, but it
is assumed that in some corner case these delta can cause a panic.

Closes #1629
 2022-10-20 11:03:55 +09:00
Pascal Seitz
f2b2628feb add test for phrase search on multi text field 2022-10-19 16:29:56 +08:00
PSeitz
449f595832 Merge pull request #1628 from quickwit-oss/skip_index_deser 
faster skipindex deserialization, larger blocksize on sort
 2022-10-19 11:05:20 +08:00
PSeitz
c9235df059 Merge pull request #1627 from quickwit-oss/ip_field_range_query 
add range query handling for ip via term dictionary
 2022-10-19 10:53:00 +08:00
Pascal Seitz
a4485f7611 faster skipindex deserialization, larger blocksize on sort 2022-10-18 19:32:23 +08:00
Pascal Seitz
1082ff60f9 add range query handling for ip via term dictionary 
since IPs are mapped monotonically we can use the term dictionary for range queries
 2022-10-18 13:08:27 +08:00
PSeitz
491854155c Merge pull request #1625 from quickwit-oss/index_ip_field 
index ip field
 2022-10-18 11:18:17 +08:00
Christoph Herzog
96c3d54ac7 fix: Fix power of two computation on 32bit architectures (#1624) 
The current `compute_previous_power_of_two()` implementation used for
TermHashmap takes and returns `usize` , but actually only works
correclty on 64 bit architectures (aka usize == u64)

On other architectures the leading_zeros computation is run on the wrong
type (must be u64), and leads to overflows.

Fixed simply computing the leading_zeros based on a u64 value.
 2022-10-18 11:55:02 +09:00
Pascal Seitz
6800fdec9d add indexing for ip field 
Closes #1595
 2022-10-18 10:07:48 +08:00
PSeitz
c9cf9c952a Merge pull request #1614 from quickwit-oss/remove_superfluous_steps 
refactor Term
 2022-10-17 18:25:31 +08:00
Pascal Seitz
024e53a99c remove truncate 2022-10-17 12:14:35 +08:00
Pascal Seitz
8d75e451bd fix truncate, remove mutable access from term 2022-10-17 12:14:35 +08:00
Pascal Seitz
fcfd76ec55 refactor Term 
fixes some issues with Term
Remove duplicate calls to truncate or resize
Replace Magic Number 5 with constant
Enforce minimum size of 5 for metadata
Fix broken truncate docs
use constructor instead new + set calls
normalize constructor stack
replace assert on internal behavior fixes #1585
 2022-10-17 12:14:34 +08:00
PSeitz
6b7b1cc4fa Merge pull request #1623 from quickwit-oss/remove_unused_buffer 
remove unused buffer
 2022-10-14 20:36:00 +08:00
Pascal Seitz
129f7422f5 remove unused buffer 2022-10-14 20:01:10 +08:00
PSeitz
f39cce2c8b Merge pull request #1622 from quickwit-oss/term_aggregation 
add term aggregation clarification
 2022-10-14 18:09:18 +08:00
PSeitz
d2478fac8a Merge pull request #1621 from quickwit-oss/changelog 
update CHANGELOG
 2022-10-14 18:08:57 +08:00
Pascal Seitz
952b048341 add term aggregation clarification 2022-10-14 16:12:19 +08:00
PSeitz
80f9596ec8 Merge pull request #1611 from quickwit-oss/remove_token_stream_alloc 
remove tokenstream vec alloc
 2022-10-14 15:12:30 +08:00
Pascal Seitz
84f9e77e1d update CHANGELOG 2022-10-14 15:10:33 +08:00
PSeitz
a602c248fb Merge pull request #1590 from waywardmonkeys/fix-doc-warnings-quickwit 
Fix missing doc warnings when enabling feature "quickwit".
 2022-10-14 14:09:25 +08:00
PSeitz
4b9d1fe828 Merge pull request #1620 from quickwit-oss/fix_fieldnorms_indexing 
Fix missing fieldnorm indexing
 2022-10-14 13:41:38 +08:00
Pascal Seitz
63bc390b02 Fix missing fieldnorm indexing 
Fixes broken search (no results) with BM25 for u64, i64, f64, bool, bytes and date after deletion and merge.
There were no fieldnorms recorded for those field. After merge InvertedIndexReader::total_num_tokens returns 0 (Sum over the fieldnorms is 0). BM25 does not work when total_num_tokens is 0.
Fixes #1617
 2022-10-14 12:44:40 +08:00
Paul Masurel
07393c2fa0 Attempt to fix race condition in test. (#1619) 
Close #1550
 2022-10-14 10:56:37 +09:00
PSeitz
77a415cbe4 rename NothingRecorder to DocIdRecorder (#1615) 2022-10-13 15:43:40 +09:00
PSeitz
4b4c231bba Merge pull request #1612 from quickwit-oss/no_panic_please 
return Error instead panic in fastfields
 2022-10-11 18:33:00 +08:00
PSeitz
11d3409286 add missing docs for fastfield_codecs crate (#1613) 
closes #1603
 2022-10-11 18:54:24 +09:00
Pascal Seitz
9cb8cfbea8 return Error instead panic in fastfields 
fixes #1572
 2022-10-11 14:15:22 +08:00
PSeitz
8b69aab0fc avoid prepare_doc allocation (#1610) 
avoid prepare_doc allocation, ~10% more thoughput best case
 2022-10-11 14:15:55 +09:00
PSeitz
3650d1f36a Merge pull request #1553 from quickwit-oss/ip_field 
ip field
 2022-10-11 13:09:47 +08:00
Pascal Seitz
2efebdb1bb remove tokenstream vec alloc 2022-10-11 10:30:56 +08:00
François Massot
e443ca63aa Merge pull request #1608 from quickwit-oss/nigel/serialise-bytes-as-b64-#2042 
Serialise bytes as base64 strings instead of arrays.
 2022-10-10 11:51:23 +02:00
Pascal Seitz
5c9cbee29d handle IpV4 serialization case 2022-10-07 19:52:00 +08:00
Pascal Seitz
b2ca83a93c switch to ipv6, add monotonic_mapping tests 2022-10-07 18:47:55 +08:00
Nigel Andrews
3b189080d4 Use raw string literals in tests 2022-10-07 12:28:25 +02:00
Nigel Andrews
00a6586efe Replaced String::serialize for serializer.serialize_str 2022-10-07 11:55:05 +02:00
Pascal Seitz
b9b913510e fmt 2022-10-07 16:56:19 +08:00
PSeitz
534b1d33c3 use ipv6 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-10-07 16:56:00 +08:00
PSeitz
f465173872 Apply suggestions from code review 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-10-07 16:55:53 +08:00
Pascal Seitz
96315df20d use idx part only for positions_to_docid 2022-10-07 16:54:04 +08:00
Pascal Seitz
9a1609d364 add test 2022-10-07 16:25:01 +08:00
Pascal Seitz
39f4e58450 improve comment 2022-10-07 16:25:01 +08:00
Pascal Seitz
a8a36b62cd enable test 2022-10-07 16:25:01 +08:00
Pascal Seitz
226a49338f add StrictlyMonotonicFn 2022-10-07 16:25:01 +08:00
Pascal Seitz
2864bf7123 use serializer for u128 2022-10-07 16:25:01 +08:00
Pascal Seitz
5171ff611b serialize ip as u128, add test for positions_to_docid 2022-10-07 16:25:01 +08:00
Pascal Seitz
e50e74acf8 remove u128 type 2022-10-07 16:25:01 +08:00
Pascal Seitz
0b86658389 rename ip addr, use buffer 2022-10-07 16:25:01 +08:00
Pascal Seitz
5d6602a8d9 mark null handling TODO 2022-10-07 16:25:01 +08:00
Pascal Seitz
4d29ff4d01 finalize ip addr rename 2022-10-07 16:25:01 +08:00
Pascal Seitz
cdc8e3a8be group montonic mapping and inverse 
fix mapping inverse
remove ip indexing
add get_between_vals test
 2022-10-07 16:25:01 +08:00
Pascal Seitz
67f453b534 rename to iter_gen 2022-10-07 16:25:01 +08:00
Pascal Seitz
787a37bacf expect instead of unwrap 2022-10-07 16:25:01 +08:00
Pascal Seitz
f5039f1846 remove roaring 2022-10-07 16:25:01 +08:00
Pascal Seitz
eeb1f19093 rename to iter_gen 2022-10-07 16:25:01 +08:00
Pascal Seitz
087beaf328 remove null handling 2022-10-07 16:25:01 +08:00
Pascal Seitz
309449dba3 rename to IpAddr 2022-10-07 16:25:01 +08:00
Pascal Seitz
5a76e6c5d3 fix get_between_vals forwarding 
fix get_between_vals forwarding in monotonicmapping column by adding an additional conversion function Output->Input
 2022-10-07 16:25:01 +08:00
Pascal Seitz
c8713a01ed use iter api 2022-10-07 16:25:01 +08:00
Pascal Seitz
6113e0408c remove comment 2022-10-07 16:25:01 +08:00
Pascal Seitz
400a20b7af add ip field 
add u128 multivalue reader and writer
add ip to schema
add ip writers, handle merge
 2022-10-07 16:25:01 +08:00
PSeitz
5f565e77de Merge pull request #1604 from quickwit-oss/replace_cbor 
replace cbor with cborium
 2022-10-07 14:42:55 +08:00
Pascal Seitz
516e60900d remove unwrap 2022-10-07 14:22:37 +08:00
Pascal Seitz
36e1c79f37 replace cbor with cborium 
closes #1526
 2022-10-07 13:23:39 +08:00
Bruce Mitchener
c2f1c250f9 doc: Remove reference to Searcher pool. (#1598) 
The pool of searchers was removed in 23fe73a6 as part of #1411.
 2022-10-06 00:04:11 +09:00
Bruce Mitchener
c694bc039a Fix missing doc warnings when enabling feature "quickwit". 2022-10-05 20:17:10 +07:00
PSeitz
2063f1717f Merge pull request #1591 from quickwit-oss/ff_refact 
disable linear codec for multivalue values
 2022-10-05 19:39:36 +08:00
Pascal Seitz
d742275048 renames 2022-10-05 19:16:49 +08:00
PSeitz
b9f06bc287 Update src/fastfield/multivalued/mod.rs 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-10-05 19:09:19 +08:00
Pascal Seitz
8b42c4c126 disable linear codec for multivalue value index 
don't materialize index column on merge
use simpler chain() variant
 2022-10-05 19:09:17 +08:00
PSeitz
7905965800 Merge pull request #1594 from quickwit-oss/flat_map_with_buffer 
Removing alloc on all .next() in MultiValueColumn
 2022-10-05 18:34:15 +08:00
Pascal Seitz
f60a551890 add flat_map_with_buffer to Iterator trait 2022-10-05 17:44:26 +08:00
Paul Masurel
7baa6e3ec5 Removing alloc on all .next() in MultiValueColumn 2022-10-05 17:12:06 +09:00
PSeitz
2100ec5d26 Merge pull request #1593 from waywardmonkeys/doc-improvements 
Documentation improvements.
 2022-10-05 15:50:08 +08:00
Bruce Mitchener
b3bf9a5716 Documentation improvements. 2022-10-05 14:18:10 +07:00
Paul Masurel
0dc8c458e0 Flaky unit test. (#1592) 2022-10-05 16:15:48 +09:00
Nigel Andrews
e5043d78d2 added a couple of tests + make fmt 2022-10-04 12:52:44 +02:00
Nigel Andrews
6d0bb82bd2 Fix issue 1576: serialize bytes as base64 strings 2022-10-04 12:18:13 +02:00
trinity-1686a
5945dbf0bd change format for store to make it faster with small documents (#1569) 
* use new format for docstore blocks

* move index to end of block

it makes writing the block faster due to one less memcopy
 2022-10-04 09:58:55 +02:00
PSeitz
4cf911d56a Merge pull request #1587 from quickwit-oss/no_get_val_in_serialize 
remove get_val in serialization
 2022-10-04 12:56:48 +08:00
Pascal Seitz
0f5cff762f move enumerate and remove computation 2022-10-04 12:30:19 +08:00
Pascal Seitz
6d9a123cf2 remove get_val in serialization 
remove get_val in serialization and mark as unimplemented!()
replace get_val with iter in linear codec
remove MultivalueStartIndexRandomSeeker
replace MultivalueStartIndexIter with closure
Sample 100 values in linear codec
 2022-10-04 12:01:25 +08:00
PSeitz
0f4a47816a Merge pull request #1582 from quickwit-oss/faster_sorted_field_values 
use groupby instead of vec allocation
 2022-10-04 09:36:24 +08:00
Pascal Seitz
b062ab2196 use groupby instead of vec allocation 2022-10-04 09:26:26 +08:00
Bruce Mitchener
a9d2f3db23 Tantivy requires Rust 1.62 or later. (#1583) 
Tantivy needs the `total_cmp` feature to compile, which was stabilized
in Rust 1.62.
 2022-10-03 18:31:07 +09:00
Bruce Mitchener
44e03791f9 Fix warnings when doc'ing private items. (#1579) 
This also fixes a couple of typos, but plenty remain!
 2022-10-03 14:24:00 +09:00
Bruce Mitchener
2d23763e9f Use u64::from boolean more. (#1580) 
This case is inverted from the previous cases fixed.

This is from nightly clippy.
 2022-10-03 14:17:50 +09:00
Bruce Mitchener
a24ae8d924 clippy: Fix needless-borrow warnings. (#1581) 
These show on nightly clippy.
 2022-10-03 14:15:09 +09:00
PSeitz
927dff5262 Merge pull request #1578 from quickwit-oss/dead_code 
remove dead indexing code
 2022-10-03 11:25:10 +08:00
Pascal Seitz
a695edcc95 remove dead indexing code 2022-10-03 09:44:02 +08:00
Paul Masurel
b4b4f3fa73 Removing default features for zstd (#1574) 2022-09-30 13:02:46 +09:00
PSeitz
b50e4b7c20 Merge pull request #1566 from quickwit-oss/fix_docstore_sorting 
fix docstore settings for temp docstore
 2022-09-30 10:10:36 +08:00
PSeitz
f8686ab1ec improve comments 
Co-authored-by: Paul Masurel <paul@quickwit.io>
 2022-09-30 10:06:34 +08:00
PSeitz
2fe42719d8 Merge pull request #1570 from quickwit-oss/no_sort_on_multi 
validate index settings on create
 2022-09-30 09:17:03 +08:00
PSeitz
fadd784a25 log improvements (#1564) 2022-09-30 09:39:26 +09:00
Pascal Seitz
0e94213af0 validate index settings on create 2022-09-29 18:58:09 +08:00
PSeitz
0da2a2e70d Merge pull request #1567 from quickwit-oss/dependabot/cargo/tantivy-fst-0.4.0 
Update tantivy-fst requirement from 0.3.0 to 0.4.0
 2022-09-29 10:00:16 +08:00
dependabot[bot]
0bcdf3cbbf Update tantivy-fst requirement from 0.3.0 to 0.4.0 
Updates the requirements on [tantivy-fst](https://github.com/tantivy-search/fst) to permit the latest version.
- [Release notes](https://github.com/tantivy-search/fst/releases)
- [Commits](https://github.com/tantivy-search/fst/commits)

---
updated-dependencies:
- dependency-name: tantivy-fst
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
 2022-09-28 20:50:43 +00:00
Pascal Seitz
8f647b817f fix docstore settings for temp docstore 
fixes #1565
 2022-09-28 17:53:59 +08:00
trinity-1686a
a86b0df6f4 Add query matching terms in a set (#1539) 2022-09-28 09:43:18 +02:00
155 changed files with 8909 additions and 1864 deletions
Expand all files Collapse all files

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

@@ -48,7 +48,7 @@ jobs:
strategy:
matrix:
features: [
{ label: "all", flags: "mmap,brotli-compression,lz4-compression,snappy-compression,zstd-compression,failpoints" },
{ label: "all", flags: "mmap,stopwords,brotli-compression,lz4-compression,snappy-compression,zstd-compression,failpoints" },
{ label: "quickwit", flags: "mmap,quickwit,failpoints" }
]

36
CHANGELOG.md
View File

@@ -1,10 +1,38 @@
Tantivy 0.19
================================
#### Bugfixes
- Fix missing fieldnorms for u64, i64, f64, bool, bytes and date [#1620](https://github.com/quickwit-oss/tantivy/pull/1620) (@PSeitz)
- Fix interpolation overflow in linear interpolation fastfield codec [#1480](https://github.com/quickwit-oss/tantivy/pull/1480 (@PSeitz @fulmicoton)
#### Features/Improvements
- Add support for `IN` in queryparser , e.g. `field: IN [val1 val2 val3]` [#1683](https://github.com/quickwit-oss/tantivy/pull/1683) (@trinity-1686a)
- Skip score calculation, when no scoring is required [#1646](https://github.com/quickwit-oss/tantivy/pull/1646) (@PSeitz)
- Limit fast fields to u32 (`get_val(u32)`) [#1644](https://github.com/quickwit-oss/tantivy/pull/1644) (@PSeitz)
- Updated [Date Field Type](https://github.com/quickwit-oss/tantivy/pull/1396)
The `DateTime` type has been updated to hold timestamps with microseconds precision.
`DateOptions` and `DatePrecision` have been added to configure Date fields. The precision is used to hint on fast values compression. Otherwise, seconds precision is used everywhere else (i.e terms, indexing).
- Remove Searcher pool and make `Searcher` cloneable.
`DateOptions` and `DatePrecision` have been added to configure Date fields. The precision is used to hint on fast values compression. Otherwise, seconds precision is used everywhere else (i.e terms, indexing). (@evanxg852000)
- Add IP address field type [#1553](https://github.com/quickwit-oss/tantivy/pull/1553) (@PSeitz)
- Add boolean field type [#1382](https://github.com/quickwit-oss/tantivy/pull/1382) (@boraarslan)
- Remove Searcher pool and make `Searcher` cloneable. (@PSeitz)
- Validate settings on create [#1570](https://github.com/quickwit-oss/tantivy/pull/1570 (@PSeitz)
- Detect and apply gcd on fastfield codecs [#1418](https://github.com/quickwit-oss/tantivy/pull/1418) (@PSeitz)
- Doc store
- use separate thread to compress block store [#1389](https://github.com/quickwit-oss/tantivy/pull/1389) [#1510](https://github.com/quickwit-oss/tantivy/pull/1510 (@PSeitz @fulmicoton)
- Expose doc store cache size [#1403](https://github.com/quickwit-oss/tantivy/pull/1403) (@PSeitz)
- Enable compression levels for doc store [#1378](https://github.com/quickwit-oss/tantivy/pull/1378) (@PSeitz)
- Make block size configurable [#1374](https://github.com/quickwit-oss/tantivy/pull/1374) (@kryesh)
- Make `tantivy::TantivyError` cloneable [#1402](https://github.com/quickwit-oss/tantivy/pull/1402) (@PSeitz)
- Add support for phrase slop in query language [#1393](https://github.com/quickwit-oss/tantivy/pull/1393) (@saroh)
- Aggregation
- Add aggregation support for date type [#1693](https://github.com/quickwit-oss/tantivy/pull/1693)(@PSeitz)
- Add support for keyed parameter in range and histgram aggregations [#1424](https://github.com/quickwit-oss/tantivy/pull/1424) (@k-yomo)
- Add aggregation bucket limit [#1363](https://github.com/quickwit-oss/tantivy/pull/1363) (@PSeitz)
- Faster indexing
- [#1610](https://github.com/quickwit-oss/tantivy/pull/1610) (@PSeitz)
- [#1594](https://github.com/quickwit-oss/tantivy/pull/1594) (@PSeitz)
- [#1582](https://github.com/quickwit-oss/tantivy/pull/1582) (@PSeitz)
- [#1611](https://github.com/quickwit-oss/tantivy/pull/1611) (@PSeitz)
- Added a pre-configured stop word filter for various language [#1666](https://github.com/quickwit-oss/tantivy/pull/1666) (@adamreichold)
Tantivy 0.18
================================
@@ -22,6 +50,10 @@ Tantivy 0.18
- Add terms aggregation (@PSeitz)
- Add support for zstd compression (@kryesh)
Tantivy 0.18.1
================================
- Hotfix: positions computation. #1629 (@fmassot, @fulmicoton, @PSeitz)
Tantivy 0.17
================================

36
Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "tantivy"
version = "0.18.0"
version = "0.19.0"
authors = ["Paul Masurel <paul.masurel@gmail.com>"]
license = "MIT"
categories = ["database-implementations", "data-structures"]
@@ -11,19 +11,21 @@ repository = "https://github.com/quickwit-oss/tantivy"
readme = "README.md"
keywords = ["search", "information", "retrieval"]
edition = "2021"
rust-version = "1.62"
[dependencies]
oneshot = "0.1.3"
oneshot = "0.1.5"
base64 = "0.13.0"
byteorder = "1.4.3"
crc32fast = "1.3.2"
once_cell = "1.10.0"
regex = { version = "1.5.5", default-features = false, features = ["std", "unicode"] }
tantivy-fst = "0.3.0"
aho-corasick = "0.7"
tantivy-fst = "0.4.0"
memmap2 = { version = "0.5.3", optional = true }
lz4_flex = { version = "0.9.2", default-features = false, features = ["checked-decode"], optional = true }
brotli = { version = "3.3.4", optional = true }
zstd = { version = "0.11", optional = true }
zstd = { version = "0.12", optional = true, default-features = false }
snap = { version = "1.0.5", optional = true }
tempfile = { version = "3.3.0", optional = true }
log = "0.4.16"
@@ -34,17 +36,12 @@ fs2 = { version = "0.4.3", optional = true }
levenshtein_automata = "0.2.1"
uuid = { version = "1.0.0", features = ["v4", "serde"] }
crossbeam-channel = "0.5.4"
tantivy-query-grammar = { version="0.18.0", path="./query-grammar" }
tantivy-bitpacker = { version="0.2", path="./bitpacker" }
common = { version = "0.3", path = "./common/", package = "tantivy-common" }
fastfield_codecs = { version="0.2", path="./fastfield_codecs", default-features = false }
ownedbytes = { version="0.3", path="./ownedbytes" }
stable_deref_trait = "1.2.0"
rust-stemmers = "1.2.0"
downcast-rs = "1.2.0"
bitpacking = { version = "0.8.4", default-features = false, features = ["bitpacker4x"] }
census = "0.4.0"
fnv = "1.0.7"
rustc-hash = "1.1.0"
thiserror = "1.0.30"
htmlescape = "0.3.1"
fail = "0.5.0"
@@ -56,10 +53,16 @@ lru = "0.7.5"
fastdivide = "0.4.0"
itertools = "0.10.3"
measure_time = "0.8.2"
serde_cbor = { version = "0.11.2", optional = true }
ciborium = { version = "0.2", optional = true}
async-trait = "0.1.53"
arc-swap = "1.5.0"
tantivy-query-grammar = { version= "0.19.0", path="./query-grammar" }
tantivy-bitpacker = { version= "0.3", path="./bitpacker" }
common = { version= "0.4", path = "./common/", package = "tantivy-common" }
fastfield_codecs = { version= "0.3", path="./fastfield_codecs", default-features = false }
ownedbytes = { version= "0.4", path="./ownedbytes" }
[target.'cfg(windows)'.dependencies]
winapi = "0.3.9"
@@ -69,10 +72,10 @@ maplit = "1.0.2"
matches = "0.1.9"
pretty_assertions = "1.2.1"
proptest = "1.0.0"
criterion = "0.3.5"
criterion = "0.4"
test-log = "0.2.10"
env_logger = "0.9.0"
pprof = { version = "0.10.0", features = ["flamegraph", "criterion"] }
env_logger = "0.10.0"
pprof = { version = "0.11.0", features = ["flamegraph", "criterion"] }
futures = "0.3.21"
[dev-dependencies.fail]
@@ -89,8 +92,9 @@ debug-assertions = true
overflow-checks = true
[features]
default = ["mmap", "lz4-compression" ]
default = ["mmap", "stopwords", "lz4-compression"]
mmap = ["fs2", "tempfile", "memmap2"]
stopwords = []
brotli-compression = ["brotli"]
lz4-compression = ["lz4_flex"]
@@ -100,7 +104,7 @@ zstd-compression = ["zstd"]
failpoints = ["fail/failpoints"]
unstable = [] # useful for benches.
quickwit = ["serde_cbor"]
quickwit = ["ciborium"]
[workspace]
members = ["query-grammar", "bitpacker", "common", "fastfield_codecs", "ownedbytes"]

8
README.md
View File

@@ -58,7 +58,7 @@ Distributed search is out of the scope of Tantivy, but if you are looking for th
# Getting started
Tantivy works on stable Rust (>= 1.27) and supports Linux, macOS, and Windows.
Tantivy works on stable Rust 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/quickwit-oss/tantivy-cli) - `tantivy-cli` is an actual command-line interface that makes it easy for you to create a search engine,
@@ -81,9 +81,13 @@ There are many ways to support this project.
We use the GitHub Pull Request workflow: reference a GitHub ticket and/or include a comprehensive commit message when opening a PR.
## Minimum supported Rust version
Tantivy currently requires at least Rust 1.62 or later to compile.
## Clone and build locally
Tantivy compiles on stable Rust but requires `Rust >= 1.27`.
Tantivy compiles on stable Rust.
To check out and run tests, you can simply run:
```bash

4
bitpacker/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "tantivy-bitpacker"
version = "0.2.0"
version = "0.3.0"
edition = "2021"
authors = ["Paul Masurel <paul.masurel@gmail.com>"]
license = "MIT"
@@ -8,6 +8,8 @@ categories = []
description = """Tantivy-sub crate: bitpacking"""
repository = "https://github.com/quickwit-oss/tantivy"
keywords = []
documentation = "https://docs.rs/tantivy-bitpacker/latest/tantivy_bitpacker"
homepage = "https://github.com/quickwit-oss/tantivy"
# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html

8
bitpacker/src/bitpacker.rs
View File

@@ -87,15 +87,15 @@ impl BitUnpacker {
}
#[inline]
pub fn get(&self, idx: u64, data: &[u8]) -> u64 {
pub fn get(&self, idx: u32, data: &[u8]) -> u64 {
if self.num_bits == 0 {
return 0u64;
}
let addr_in_bits = idx * self.num_bits;
let addr_in_bits = idx * self.num_bits as u32;
let addr = addr_in_bits >> 3;
let bit_shift = addr_in_bits & 7;
debug_assert!(
addr + 8 <= data.len() as u64,
addr + 8 <= data.len() as u32,
"The fast field field should have been padded with 7 bytes."
);
let bytes: [u8; 8] = (&data[(addr as usize)..(addr as usize) + 8])
@@ -130,7 +130,7 @@ mod test {
fn test_bitpacker_util(len: usize, num_bits: u8) {
let (bitunpacker, vals, data) = create_fastfield_bitpacker(len, num_bits);
for (i, val) in vals.iter().enumerate() {
assert_eq!(bitunpacker.get(i as u64, &data), *val);
assert_eq!(bitunpacker.get(i as u32, &data), *val);
}
}

2
bitpacker/src/blocked_bitpacker.rs
View File

@@ -130,7 +130,7 @@ impl BlockedBitpacker {
let pos_in_block = idx % BLOCK_SIZE as usize;
if let Some(metadata) = self.offset_and_bits.get(metadata_pos) {
let unpacked = BitUnpacker::new(metadata.num_bits()).get(
pos_in_block as u64,
pos_in_block as u32,
&self.compressed_blocks[metadata.offset() as usize..],
);
unpacked + metadata.base_value()

8
common/Cargo.toml
View File

@@ -1,16 +1,20 @@
[package]
name = "tantivy-common"
version = "0.3.0"
version = "0.4.0"
authors = ["Paul Masurel <paul@quickwit.io>", "Pascal Seitz <pascal@quickwit.io>"]
license = "MIT"
edition = "2021"
description = "common traits and utility functions used by multiple tantivy subcrates"
documentation = "https://docs.rs/tantivy_common/"
homepage = "https://github.com/quickwit-oss/tantivy"
repository = "https://github.com/quickwit-oss/tantivy"
# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
[dependencies]
byteorder = "1.4.3"
ownedbytes = { version="0.3", path="../ownedbytes" }
ownedbytes = { version= "0.4", path="../ownedbytes" }
[dev-dependencies]
proptest = "1.0.0"

27
common/src/serialize.rs
View File

@@ -94,6 +94,20 @@ impl FixedSize for u32 {
const SIZE_IN_BYTES: usize = 4;
}
impl BinarySerializable for u16 {
fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
writer.write_u16::<Endianness>(*self)
}
fn deserialize<R: Read>(reader: &mut R) -> io::Result<u16> {
reader.read_u16::<Endianness>()
}
}
impl FixedSize for u16 {
const SIZE_IN_BYTES: usize = 2;
}
impl BinarySerializable for u64 {
fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
writer.write_u64::<Endianness>(*self)
@@ -107,6 +121,19 @@ impl FixedSize for u64 {
const SIZE_IN_BYTES: usize = 8;
}
impl BinarySerializable for u128 {
fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
writer.write_u128::<Endianness>(*self)
}
fn deserialize<R: Read>(reader: &mut R) -> io::Result<Self> {
reader.read_u128::<Endianness>()
}
}
impl FixedSize for u128 {
const SIZE_IN_BYTES: usize = 16;
}
impl BinarySerializable for f32 {
fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
writer.write_f32::<Endianness>(*self)

2
common/src/vint.rs
View File

@@ -157,7 +157,7 @@ fn vint_len(data: &[u8]) -> usize {
/// If the buffer does not start by a valid
/// vint payload
pub fn read_u32_vint(data: &mut &[u8]) -> u32 {
let (result, vlen) = read_u32_vint_no_advance(*data);
let (result, vlen) = read_u32_vint_no_advance(data);
*data = &data[vlen..];
result
}

2
examples/aggregation.rs
View File

@@ -118,7 +118,7 @@ fn main() -> tantivy::Result<()> {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults = searcher.search(&term_query, &collector).unwrap();

2
examples/custom_collector.rs
View File

@@ -105,7 +105,7 @@ impl SegmentCollector for StatsSegmentCollector {
type Fruit = Option<Stats>;
fn collect(&mut self, doc: u32, _score: Score) {
let value = self.fast_field_reader.get_val(doc as u64) as f64;
let value = self.fast_field_reader.get_val(doc) as f64;
self.stats.count += 1;
self.stats.sum += value;
self.stats.squared_sum += value * value;

2
examples/warmer.rs
View File

@@ -51,7 +51,7 @@ impl Warmer for DynamicPriceColumn {
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_val(doc as u64))
.map(|doc| product_id_reader.get_val(doc))
.collect();
let mut prices_it = self.price_fetcher.fetch_prices(&product_ids).into_iter();
let mut price_vals: Vec<Price> = Vec::new();

12
fastfield_codecs/Cargo.toml
View File

@@ -1,24 +1,26 @@
[package]
name = "fastfield_codecs"
version = "0.2.0"
version = "0.3.0"
authors = ["Pascal Seitz <pascal@quickwit.io>"]
license = "MIT"
edition = "2021"
description = "Fast field codecs used by tantivy"
documentation = "https://docs.rs/fastfield_codecs/"
homepage = "https://github.com/quickwit-oss/tantivy"
repository = "https://github.com/quickwit-oss/tantivy"
# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
[dependencies]
common = { version = "0.3", path = "../common/", package = "tantivy-common" }
tantivy-bitpacker = { version="0.2", path = "../bitpacker/" }
ownedbytes = { version = "0.3.0", path = "../ownedbytes" }
common = { version = "0.4", path = "../common/", package = "tantivy-common" }
tantivy-bitpacker = { version= "0.3", path = "../bitpacker/" }
ownedbytes = { version = "0.4.0", path = "../ownedbytes" }
prettytable-rs = {version="0.9.0", optional= true}
rand = {version="0.8.3", optional= true}
fastdivide = "0.4"
log = "0.4"
itertools = { version = "0.10.3" }
measure_time = { version="0.8.2", optional=true}
roaring = "0.10.1"
[dev-dependencies]
more-asserts = "0.3.0"

117
fastfield_codecs/benches/bench.rs
View File

@@ -44,15 +44,6 @@ mod tests {
open(OwnedBytes::new(buffer)).unwrap()
}
pub fn serialize_and_load_dense<T: MonotonicallyMappableToU64 + Ord + Default>(
column: &[T],
fill_ratio: u32,
) -> Arc<dyn Column<T>> {
let mut buffer = Vec::new();
serialize(VecColumn::from(&column), &mut buffer, &ALL_CODEC_TYPES).unwrap();
open_dense(OwnedBytes::new(buffer), fill_ratio).unwrap()
}
#[bench]
fn bench_intfastfield_jumpy_veclookup(b: &mut Bencher) {
let permutation = generate_permutation();
@@ -74,7 +65,7 @@ mod tests {
b.iter(|| {
let mut a = 0u64;
for _ in 0..n {
a = column.get_val(a as u64);
a = column.get_val(a as u32);
}
a
});
@@ -109,9 +100,10 @@ mod tests {
fn get_u128_column_from_data(data: &[u128]) -> Arc<dyn Column<u128>> {
let mut out = vec![];
serialize_u128(VecColumn::from(&data), &mut out).unwrap();
let iter_gen = || data.iter().cloned();
serialize_u128(iter_gen, data.len() as u32, &mut out).unwrap();
let out = OwnedBytes::new(out);
open_u128(out).unwrap()
open_u128::<u128>(out).unwrap()
}
#[bench]
@@ -119,7 +111,15 @@ mod tests {
let (major_item, _minor_item, data) = get_data_50percent_item();
let column = get_u128_column_from_data(&data);
b.iter(|| column.get_between_vals(major_item..=major_item));
b.iter(|| {
let mut positions = Vec::new();
column.get_docids_for_value_range(
major_item..=major_item,
0..data.len() as u32,
&mut positions,
);
positions
});
}
#[bench]
@@ -127,7 +127,15 @@ mod tests {
let (_major_item, minor_item, data) = get_data_50percent_item();
let column = get_u128_column_from_data(&data);
b.iter(|| column.get_between_vals(minor_item..=minor_item));
b.iter(|| {
let mut positions = Vec::new();
column.get_docids_for_value_range(
minor_item..=minor_item,
0..data.len() as u32,
&mut positions,
);
positions
});
}
#[bench]
@@ -135,7 +143,11 @@ mod tests {
let (_major_item, _minor_item, data) = get_data_50percent_item();
let column = get_u128_column_from_data(&data);
b.iter(|| column.get_between_vals(0..=u128::MAX));
b.iter(|| {
let mut positions = Vec::new();
column.get_docids_for_value_range(0..=u128::MAX, 0..data.len() as u32, &mut positions);
positions
});
}
#[bench]
@@ -145,7 +157,7 @@ mod tests {
b.iter(|| {
let mut a = 0u128;
for i in 0u64..column.num_vals() as u64 {
a += column.get_val(i);
a += column.get_val(i as u32);
}
a
});
@@ -159,7 +171,7 @@ mod tests {
let n = column.num_vals();
let mut a = 0u128;
for i in (0..n / 5).map(|val| val * 5) {
a += column.get_val(i as u64);
a += column.get_val(i);
}
a
});
@@ -184,70 +196,9 @@ mod tests {
let n = permutation.len();
let column: Arc<dyn Column<u64>> = serialize_and_load(&permutation);
b.iter(|| {
let mut a = 0u64;
let mut a = 0;
for i in (0..n / 7).map(|val| val * 7) {
a += column.get_val(i as u64);
}
a
});
}
#[bench]
fn bench_intfastfield_stride7_fflookup_sparse_roaring(b: &mut Bencher) {
let permutation = generate_permutation();
let n = permutation.len();
let column: Arc<dyn Column<u64>> = serialize_and_load(&permutation);
let column = SparseCodecRoaringBitmap::with_full(column);
b.iter(|| {
let mut a = 0u64;
for i in (0..n / 7).map(|val| val * 7) {
a += column.get_val(i as u64);
}
a
});
}
#[bench]
fn bench_intfastfield_stride7_fflookup_dense_bitmap_with_offset(b: &mut Bencher) {
let permutation = generate_permutation();
let n = permutation.len();
let column: Arc<dyn Column<u64>> = serialize_and_load_dense(&permutation, 1000);
b.iter(|| {
let mut a = 0u64;
for i in (0..n / 7).map(|val| val * 7) {
a += column.get_val(i as u64);
}
a
});
}
#[bench]
fn bench_intfastfield_stride7_fflookup_dense_bitmap_with_offset_70percent_dense(
b: &mut Bencher,
) {
let permutation = generate_permutation();
let n = permutation.len();
let column: Arc<dyn Column<u64>> = serialize_and_load_dense(&permutation, 700);
b.iter(|| {
let mut a = 0u64;
for i in (0..n / 7).map(|val| val * 7) {
a += column.get_val(i as u64);
}
a
});
}
#[bench]
fn bench_intfastfield_stride7_fflookup_dense_bitmap_with_offset_20percent_dense(
b: &mut Bencher,
) {
let permutation = generate_permutation();
let n = permutation.len();
let column: Arc<dyn Column<u64>> = serialize_and_load_dense(&permutation, 200);
b.iter(|| {
let mut a = 0u64;
for i in (0..n / 7).map(|val| val * 7) {
a += column.get_val(i as u64);
a += column.get_val(i as u32);
}
a
});
@@ -260,7 +211,7 @@ mod tests {
let column: Arc<dyn Column<u64>> = serialize_and_load(&permutation);
b.iter(|| {
let mut a = 0u64;
for i in 0u64..n as u64 {
for i in 0u32..n as u32 {
a += column.get_val(i);
}
a
@@ -274,8 +225,8 @@ mod tests {
let column: Arc<dyn Column<u64>> = serialize_and_load(&permutation);
b.iter(|| {
let mut a = 0u64;
for i in 0..n as u64 {
a += column.get_val(i);
for i in 0..n {
a += column.get_val(i as u32);
}
a
});

4
fastfield_codecs/src/bitpacked.rs
View File

@@ -17,7 +17,7 @@ pub struct BitpackedReader {
impl Column for BitpackedReader {
#[inline]
fn get_val(&self, doc: u64) -> u64 {
fn get_val(&self, doc: u32) -> u64 {
self.bit_unpacker.get(doc, &self.data)
}
#[inline]
@@ -30,7 +30,7 @@ impl Column for BitpackedReader {
self.normalized_header.max_value
}
#[inline]
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.normalized_header.num_vals
}
}

18
fastfield_codecs/src/blockwise_linear.rs
View File

@@ -36,7 +36,7 @@ impl BinarySerializable for Block {
}
}
fn compute_num_blocks(num_vals: u64) -> usize {
fn compute_num_blocks(num_vals: u32) -> usize {
(num_vals as usize + CHUNK_SIZE - 1) / CHUNK_SIZE
}
@@ -72,13 +72,13 @@ impl FastFieldCodec for BlockwiseLinearCodec {
// Estimate first_chunk and extrapolate
fn estimate(column: &dyn crate::Column) -> Option<f32> {
if column.num_vals() < 10 * CHUNK_SIZE as u64 {
if column.num_vals() < 10 * CHUNK_SIZE as u32 {
return None;
}
let mut first_chunk: Vec<u64> = column.iter().take(CHUNK_SIZE as usize).collect();
let line = Line::train(&VecColumn::from(&first_chunk));
for (i, buffer_val) in first_chunk.iter_mut().enumerate() {
let interpolated_val = line.eval(i as u64);
let interpolated_val = line.eval(i as u32);
*buffer_val = buffer_val.wrapping_sub(interpolated_val);
}
let estimated_bit_width = first_chunk
@@ -95,7 +95,7 @@ impl FastFieldCodec for BlockwiseLinearCodec {
};
let num_bits = estimated_bit_width as u64 * column.num_vals() as u64
// function metadata per block
+ metadata_per_block as u64 * (column.num_vals() / CHUNK_SIZE as u64);
+ metadata_per_block as u64 * (column.num_vals() as u64 / CHUNK_SIZE as u64);
let num_bits_uncompressed = 64 * column.num_vals();
Some(num_bits as f32 / num_bits_uncompressed as f32)
}
@@ -121,7 +121,7 @@ impl FastFieldCodec for BlockwiseLinearCodec {
assert!(!buffer.is_empty());
for (i, buffer_val) in buffer.iter_mut().enumerate() {
let interpolated_val = line.eval(i as u64);
let interpolated_val = line.eval(i as u32);
*buffer_val = buffer_val.wrapping_sub(interpolated_val);
}
let bit_width = buffer.iter().copied().map(compute_num_bits).max().unwrap();
@@ -161,9 +161,9 @@ pub struct BlockwiseLinearReader {
impl Column for BlockwiseLinearReader {
#[inline(always)]
fn get_val(&self, idx: u64) -> u64 {
let block_id = (idx / CHUNK_SIZE as u64) as usize;
let idx_within_block = idx % (CHUNK_SIZE as u64);
fn get_val(&self, idx: u32) -> u64 {
let block_id = (idx / CHUNK_SIZE as u32) as usize;
let idx_within_block = idx % (CHUNK_SIZE as u32);
let block = &self.blocks[block_id];
let interpoled_val: u64 = block.line.eval(idx_within_block);
let block_bytes = &self.data[block.data_start_offset..];
@@ -180,7 +180,7 @@ impl Column for BlockwiseLinearReader {
self.normalized_header.max_value
}
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.normalized_header.num_vals
}
}

168
fastfield_codecs/src/column.rs
View File

@@ -1,8 +1,11 @@
use std::marker::PhantomData;
use std::ops::RangeInclusive;
use std::ops::{Range, RangeInclusive};
use tantivy_bitpacker::minmax;
use crate::monotonic_mapping::StrictlyMonotonicFn;
/// `Column` provides columnar access on a field.
pub trait Column<T: PartialOrd = u64>: Send + Sync {
/// Return the value associated with the given idx.
///
@@ -11,7 +14,7 @@ pub trait Column<T: PartialOrd = u64>: Send + Sync {
/// # Panics
///
/// May panic if `idx` is greater than the column length.
fn get_val(&self, idx: u64) -> T;
fn get_val(&self, idx: u32) -> T;
/// Fills an output buffer with the fast field values
/// associated with the `DocId` going from
@@ -24,21 +27,28 @@ pub trait Column<T: PartialOrd = u64>: Send + Sync {
#[inline]
fn get_range(&self, start: u64, output: &mut [T]) {
for (out, idx) in output.iter_mut().zip(start..) {
*out = self.get_val(idx);
*out = self.get_val(idx as u32);
}
}
/// Return the positions of values which are in the provided range.
/// Get the positions of values which are in the provided value range.
///
/// Note that position == docid for single value fast fields
#[inline]
fn get_between_vals(&self, range: RangeInclusive<T>) -> Vec<u64> {
let mut vals = Vec::new();
for idx in 0..self.num_vals() {
fn get_docids_for_value_range(
&self,
value_range: RangeInclusive<T>,
doc_id_range: Range<u32>,
positions: &mut Vec<u32>,
) {
let doc_id_range = doc_id_range.start..doc_id_range.end.min(self.num_vals());
for idx in doc_id_range.start..doc_id_range.end {
let val = self.get_val(idx);
if range.contains(&val) {
vals.push(idx);
if value_range.contains(&val) {
positions.push(idx);
}
}
vals
}
/// Returns the minimum value for this fast field.
@@ -57,7 +67,8 @@ pub trait Column<T: PartialOrd = u64>: Send + Sync {
/// `.max_value()`.
fn max_value(&self) -> T;
fn num_vals(&self) -> u64;
/// The number of values in the column.
fn num_vals(&self) -> u32;
/// Returns a iterator over the data
fn iter<'a>(&'a self) -> Box<dyn Iterator<Item = T> + 'a> {
@@ -65,6 +76,7 @@ pub trait Column<T: PartialOrd = u64>: Send + Sync {
}
}
/// VecColumn provides `Column` over a slice.
pub struct VecColumn<'a, T = u64> {
values: &'a [T],
min_value: T,
@@ -72,7 +84,7 @@ pub struct VecColumn<'a, T = u64> {
}
impl<'a, C: Column<T>, T: Copy + PartialOrd> Column<T> for &'a C {
fn get_val(&self, idx: u64) -> T {
fn get_val(&self, idx: u32) -> T {
(*self).get_val(idx)
}
@@ -84,7 +96,7 @@ impl<'a, C: Column<T>, T: Copy + PartialOrd> Column<T> for &'a C {
(*self).max_value()
}
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
(*self).num_vals()
}
@@ -98,7 +110,7 @@ impl<'a, C: Column<T>, T: Copy + PartialOrd> Column<T> for &'a C {
}
impl<'a, T: Copy + PartialOrd + Send + Sync> Column<T> for VecColumn<'a, T> {
fn get_val(&self, position: u64) -> T {
fn get_val(&self, position: u32) -> T {
self.values[position as usize]
}
@@ -114,8 +126,8 @@ impl<'a, T: Copy + PartialOrd + Send + Sync> Column<T> for VecColumn<'a, T> {
self.max_value
}
fn num_vals(&self) -> u64 {
self.values.len() as u64
fn num_vals(&self) -> u32 {
self.values.len() as u32
}
fn get_range(&self, start: u64, output: &mut [T]) {
@@ -143,16 +155,30 @@ struct MonotonicMappingColumn<C, T, Input> {
_phantom: PhantomData<Input>,
}
/// Creates a view of a column transformed by a monotonic mapping.
pub fn monotonic_map_column<C, T, Input: PartialOrd, Output: PartialOrd>(
/// Creates a view of a column transformed by a strictly monotonic mapping. See
/// [`StrictlyMonotonicFn`].
///
/// E.g. apply a gcd monotonic_mapping([100, 200, 300]) == [1, 2, 3]
/// monotonic_mapping.mapping() is expected to be injective, and we should always have
/// monotonic_mapping.inverse(monotonic_mapping.mapping(el)) == el
///
/// The inverse of the mapping is required for:
/// `fn get_positions_for_value_range(&self, range: RangeInclusive<T>) -> Vec<u64> `
/// The user provides the original value range and we need to monotonic map them in the same way the
/// serialization does before calling the underlying column.
///
/// Note that when opening a codec, the monotonic_mapping should be the inverse of the mapping
/// during serialization. And therefore the monotonic_mapping_inv when opening is the same as
/// monotonic_mapping during serialization.
pub fn monotonic_map_column<C, T, Input, Output>(
from_column: C,
monotonic_mapping: T,
) -> impl Column<Output>
where
C: Column<Input>,
T: Fn(Input) -> Output + Send + Sync,
Input: Send + Sync,
Output: Send + Sync,
T: StrictlyMonotonicFn<Input, Output> + Send + Sync,
Input: PartialOrd + Send + Sync + Clone,
Output: PartialOrd + Send + Sync + Clone,
{
MonotonicMappingColumn {
from_column,
@@ -161,42 +187,60 @@ where
}
}
impl<C, T, Input: PartialOrd, Output: PartialOrd> Column<Output>
for MonotonicMappingColumn<C, T, Input>
impl<C, T, Input, Output> Column<Output> for MonotonicMappingColumn<C, T, Input>
where
C: Column<Input>,
T: Fn(Input) -> Output + Send + Sync,
Input: Send + Sync,
Output: Send + Sync,
T: StrictlyMonotonicFn<Input, Output> + Send + Sync,
Input: PartialOrd + Send + Sync + Clone,
Output: PartialOrd + Send + Sync + Clone,
{
#[inline]
fn get_val(&self, idx: u64) -> Output {
fn get_val(&self, idx: u32) -> Output {
let from_val = self.from_column.get_val(idx);
(self.monotonic_mapping)(from_val)
self.monotonic_mapping.mapping(from_val)
}
fn min_value(&self) -> Output {
let from_min_value = self.from_column.min_value();
(self.monotonic_mapping)(from_min_value)
self.monotonic_mapping.mapping(from_min_value)
}
fn max_value(&self) -> Output {
let from_max_value = self.from_column.max_value();
(self.monotonic_mapping)(from_max_value)
self.monotonic_mapping.mapping(from_max_value)
}
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.from_column.num_vals()
}
fn iter(&self) -> Box<dyn Iterator<Item = Output> + '_> {
Box::new(self.from_column.iter().map(&self.monotonic_mapping))
Box::new(
self.from_column
.iter()
.map(|el| self.monotonic_mapping.mapping(el)),
)
}
fn get_docids_for_value_range(
&self,
range: RangeInclusive<Output>,
doc_id_range: Range<u32>,
positions: &mut Vec<u32>,
) {
self.from_column.get_docids_for_value_range(
self.monotonic_mapping.inverse(range.start().clone())
..=self.monotonic_mapping.inverse(range.end().clone()),
doc_id_range,
positions,
)
}
// We voluntarily do not implement get_range as it yields a regression,
// and we do not have any specialized implementation anyway.
}
/// Wraps an iterator into a `Column`.
pub struct IterColumn<T>(T);
impl<T> From<T> for IterColumn<T>
@@ -212,7 +256,7 @@ where
T: Iterator + Clone + ExactSizeIterator + Send + Sync,
T::Item: PartialOrd,
{
fn get_val(&self, idx: u64) -> T::Item {
fn get_val(&self, idx: u32) -> T::Item {
self.0.clone().nth(idx as usize).unwrap()
}
@@ -224,8 +268,8 @@ where
self.0.clone().last().unwrap()
}
fn num_vals(&self) -> u64 {
self.0.len() as u64
fn num_vals(&self) -> u32 {
self.0.len() as u32
}
fn iter(&self) -> Box<dyn Iterator<Item = T::Item> + '_> {
@@ -236,19 +280,22 @@ where
#[cfg(test)]
mod tests {
use super::*;
use crate::MonotonicallyMappableToU64;
use crate::monotonic_mapping::{
StrictlyMonotonicMappingInverter, StrictlyMonotonicMappingToInternalBaseval,
StrictlyMonotonicMappingToInternalGCDBaseval,
};
#[test]
fn test_monotonic_mapping() {
let vals = &[1u64, 3u64][..];
let vals = &[3u64, 5u64][..];
let col = VecColumn::from(vals);
let mapped = monotonic_map_column(col, |el| el + 4);
assert_eq!(mapped.min_value(), 5u64);
assert_eq!(mapped.max_value(), 7u64);
let mapped = monotonic_map_column(col, StrictlyMonotonicMappingToInternalBaseval::new(2));
assert_eq!(mapped.min_value(), 1u64);
assert_eq!(mapped.max_value(), 3u64);
assert_eq!(mapped.num_vals(), 2);
assert_eq!(mapped.num_vals(), 2);
assert_eq!(mapped.get_val(0), 5);
assert_eq!(mapped.get_val(1), 7);
assert_eq!(mapped.get_val(0), 1);
assert_eq!(mapped.get_val(1), 3);
}
#[test]
@@ -260,10 +307,15 @@ mod tests {
#[test]
fn test_monotonic_mapping_iter() {
let vals: Vec<u64> = (-1..99).map(i64::to_u64).collect();
let vals: Vec<u64> = (10..110u64).map(|el| el * 10).collect();
let col = VecColumn::from(&vals);
let mapped = monotonic_map_column(col, |el| i64::from_u64(el) * 10i64);
let val_i64s: Vec<i64> = mapped.iter().collect();
let mapped = monotonic_map_column(
col,
StrictlyMonotonicMappingInverter::from(
StrictlyMonotonicMappingToInternalGCDBaseval::new(10, 100),
),
);
let val_i64s: Vec<u64> = mapped.iter().collect();
for i in 0..100 {
assert_eq!(val_i64s[i as usize], mapped.get_val(i));
}
@@ -271,20 +323,26 @@ mod tests {
#[test]
fn test_monotonic_mapping_get_range() {
let vals: Vec<u64> = (-1..99).map(i64::to_u64).collect();
let vals: Vec<u64> = (0..100u64).map(|el| el * 10).collect();
let col = VecColumn::from(&vals);
let mapped = monotonic_map_column(col, |el| i64::from_u64(el) * 10i64);
assert_eq!(mapped.min_value(), -10i64);
assert_eq!(mapped.max_value(), 980i64);
let mapped = monotonic_map_column(
col,
StrictlyMonotonicMappingInverter::from(
StrictlyMonotonicMappingToInternalGCDBaseval::new(10, 0),
),
);
assert_eq!(mapped.min_value(), 0u64);
assert_eq!(mapped.max_value(), 9900u64);
assert_eq!(mapped.num_vals(), 100);
let val_i64s: Vec<i64> = mapped.iter().collect();
assert_eq!(val_i64s.len(), 100);
let val_u64s: Vec<u64> = mapped.iter().collect();
assert_eq!(val_u64s.len(), 100);
for i in 0..100 {
assert_eq!(val_i64s[i as usize], mapped.get_val(i));
assert_eq!(val_i64s[i as usize], i64::from_u64(vals[i as usize]) * 10);
assert_eq!(val_u64s[i as usize], mapped.get_val(i));
assert_eq!(val_u64s[i as usize], vals[i as usize] * 10);
}
let mut buf = [0i64; 20];
let mut buf = [0u64; 20];
mapped.get_range(7, &mut buf[..]);
assert_eq!(&val_i64s[7..][..20], &buf);
assert_eq!(&val_u64s[7..][..20], &buf);
}
}

2
fastfield_codecs/src/compact_space/build_compact_space.rs
View File

@@ -57,7 +57,7 @@ fn num_bits(val: u128) -> u8 {
/// metadata.
pub fn get_compact_space(
values_deduped_sorted: &BTreeSet<u128>,
total_num_values: u64,
total_num_values: u32,
cost_per_blank: usize,
) -> CompactSpace {
let mut compact_space_builder = CompactSpaceBuilder::new();

285
fastfield_codecs/src/compact_space/mod.rs
View File

@@ -14,7 +14,7 @@ use std::{
cmp::Ordering,
collections::BTreeSet,
io::{self, Write},
ops::RangeInclusive,
ops::{Range, RangeInclusive},
};
use common::{BinarySerializable, CountingWriter, VInt, VIntU128};
@@ -165,16 +165,16 @@ pub struct IPCodecParams {
bit_unpacker: BitUnpacker,
min_value: u128,
max_value: u128,
num_vals: u64,
num_vals: u32,
num_bits: u8,
}
impl CompactSpaceCompressor {
/// Taking the vals as Vec may cost a lot of memory. It is used to sort the vals.
pub fn train_from(column: &impl Column<u128>) -> Self {
pub fn train_from(iter: impl Iterator<Item = u128>, num_vals: u32) -> Self {
let mut values_sorted = BTreeSet::new();
values_sorted.extend(column.iter());
let total_num_values = column.num_vals();
values_sorted.extend(iter);
let total_num_values = num_vals;
let compact_space =
get_compact_space(&values_sorted, total_num_values, COST_PER_BLANK_IN_BITS);
@@ -200,7 +200,7 @@ impl CompactSpaceCompressor {
bit_unpacker: BitUnpacker::new(num_bits),
min_value,
max_value,
num_vals: total_num_values as u64,
num_vals: total_num_values,
num_bits,
},
}
@@ -267,7 +267,7 @@ impl BinarySerializable for IPCodecParams {
let _header_flags = u64::deserialize(reader)?;
let min_value = VIntU128::deserialize(reader)?.0;
let max_value = VIntU128::deserialize(reader)?.0;
let num_vals = VIntU128::deserialize(reader)?.0 as u64;
let num_vals = VIntU128::deserialize(reader)?.0 as u32;
let num_bits = u8::deserialize(reader)?;
let compact_space = CompactSpace::deserialize(reader)?;
@@ -284,7 +284,7 @@ impl BinarySerializable for IPCodecParams {
impl Column<u128> for CompactSpaceDecompressor {
#[inline]
fn get_val(&self, doc: u64) -> u128 {
fn get_val(&self, doc: u32) -> u128 {
self.get(doc)
}
@@ -296,7 +296,7 @@ impl Column<u128> for CompactSpaceDecompressor {
self.max_value()
}
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.params.num_vals
}
@@ -304,8 +304,15 @@ impl Column<u128> for CompactSpaceDecompressor {
fn iter(&self) -> Box<dyn Iterator<Item = u128> + '_> {
Box::new(self.iter())
}
fn get_between_vals(&self, range: RangeInclusive<u128>) -> Vec<u64> {
self.get_between_vals(range)
#[inline]
fn get_docids_for_value_range(
&self,
value_range: RangeInclusive<u128>,
positions_range: Range<u32>,
positions: &mut Vec<u32>,
) {
self.get_positions_for_value_range(value_range, positions_range, positions)
}
}
@@ -340,12 +347,19 @@ impl CompactSpaceDecompressor {
/// Comparing on compact space: Real dataset 1.08 GElements/s
///
/// Comparing on original space: Real dataset .06 GElements/s (not completely optimized)
pub fn get_between_vals(&self, range: RangeInclusive<u128>) -> Vec<u64> {
if range.start() > range.end() {
return Vec::new();
#[inline]
pub fn get_positions_for_value_range(
&self,
value_range: RangeInclusive<u128>,
position_range: Range<u32>,
positions: &mut Vec<u32>,
) {
if value_range.start() > value_range.end() {
return;
}
let from_value = *range.start();
let to_value = *range.end();
let position_range = position_range.start..position_range.end.min(self.num_vals());
let from_value = *value_range.start();
let to_value = *value_range.end();
assert!(to_value >= from_value);
let compact_from = self.u128_to_compact(from_value);
let compact_to = self.u128_to_compact(to_value);
@@ -353,7 +367,7 @@ impl CompactSpaceDecompressor {
// Quick return, if both ranges fall into the same non-mapped space, the range can't cover
// any values, so we can early exit
match (compact_to, compact_from) {
(Err(pos1), Err(pos2)) if pos1 == pos2 => return Vec::new(),
(Err(pos1), Err(pos2)) if pos1 == pos2 => return,
_ => {}
}
@@ -375,27 +389,28 @@ impl CompactSpaceDecompressor {
});
let range = compact_from..=compact_to;
let mut positions = Vec::new();
let scan_num_docs = position_range.end - position_range.start;
let step_size = 4;
let cutoff = self.params.num_vals - self.params.num_vals % step_size;
let cutoff = position_range.start + scan_num_docs - scan_num_docs % step_size;
let mut push_if_in_range = |idx, val| {
if range.contains(&val) {
positions.push(idx);
}
};
let get_val = |idx| self.params.bit_unpacker.get(idx as u64, &self.data);
let get_val = |idx| self.params.bit_unpacker.get(idx, &self.data);
// unrolled loop
for idx in (0..cutoff).step_by(step_size as usize) {
for idx in (position_range.start..cutoff).step_by(step_size as usize) {
let idx1 = idx;
let idx2 = idx + 1;
let idx3 = idx + 2;
let idx4 = idx + 3;
let val1 = get_val(idx1);
let val2 = get_val(idx2);
let val3 = get_val(idx3);
let val4 = get_val(idx4);
let val1 = get_val(idx1 as u32);
let val2 = get_val(idx2 as u32);
let val3 = get_val(idx3 as u32);
let val4 = get_val(idx4 as u32);
push_if_in_range(idx1, val1);
push_if_in_range(idx2, val2);
push_if_in_range(idx3, val3);
@@ -403,17 +418,15 @@ impl CompactSpaceDecompressor {
}
// handle rest
for idx in cutoff..self.params.num_vals {
push_if_in_range(idx, get_val(idx));
for idx in cutoff..position_range.end {
push_if_in_range(idx, get_val(idx as u32));
}
positions
}
#[inline]
fn iter_compact(&self) -> impl Iterator<Item = u64> + '_ {
(0..self.params.num_vals)
.map(move |idx| self.params.bit_unpacker.get(idx as u64, &self.data) as u64)
.map(move |idx| self.params.bit_unpacker.get(idx, &self.data) as u64)
}
#[inline]
@@ -425,7 +438,7 @@ impl CompactSpaceDecompressor {
}
#[inline]
pub fn get(&self, idx: u64) -> u128 {
pub fn get(&self, idx: u32) -> u128 {
let compact = self.params.bit_unpacker.get(idx, &self.data);
self.compact_to_u128(compact)
}
@@ -443,7 +456,10 @@ impl CompactSpaceDecompressor {
mod tests {
use super::*;
use crate::{open_u128, serialize_u128, VecColumn};
use crate::format_version::read_format_version;
use crate::null_index_footer::read_null_index_footer;
use crate::serialize::U128Header;
use crate::{open_u128, serialize_u128};
#[test]
fn compact_space_test() {
@@ -452,7 +468,7 @@ mod tests {
]
.into_iter()
.collect();
let compact_space = get_compact_space(ips, ips.len() as u64, 11);
let compact_space = get_compact_space(ips, ips.len() as u32, 11);
let amplitude = compact_space.amplitude_compact_space();
assert_eq!(amplitude, 17);
assert_eq!(1, compact_space.u128_to_compact(2).unwrap());
@@ -483,24 +499,30 @@ mod tests {
#[test]
fn compact_space_amplitude_test() {
let ips = &[100000u128, 1000000].into_iter().collect();
let compact_space = get_compact_space(ips, ips.len() as u64, 1);
let compact_space = get_compact_space(ips, ips.len() as u32, 1);
let amplitude = compact_space.amplitude_compact_space();
assert_eq!(amplitude, 2);
}
fn test_all(data: OwnedBytes, expected: &[u128]) {
fn test_all(mut data: OwnedBytes, expected: &[u128]) {
let _header = U128Header::deserialize(&mut data);
let decompressor = CompactSpaceDecompressor::open(data).unwrap();
for (idx, expected_val) in expected.iter().cloned().enumerate() {
let val = decompressor.get(idx as u64);
let val = decompressor.get(idx as u32);
assert_eq!(val, expected_val);
let test_range = |range: RangeInclusive<u128>| {
let expected_positions = expected
.iter()
.positions(|val| range.contains(val))
.map(|pos| pos as u64)
.map(|pos| pos as u32)
.collect::<Vec<_>>();
let positions = decompressor.get_between_vals(range);
let mut positions = Vec::new();
decompressor.get_positions_for_value_range(
range,
0..decompressor.num_vals(),
&mut positions,
);
assert_eq!(positions, expected_positions);
};
@@ -513,10 +535,18 @@ mod tests {
fn test_aux_vals(u128_vals: &[u128]) -> OwnedBytes {
let mut out = Vec::new();
serialize_u128(VecColumn::from(u128_vals), &mut out).unwrap();
serialize_u128(
|| u128_vals.iter().cloned(),
u128_vals.len() as u32,
&mut out,
)
.unwrap();
let data = OwnedBytes::new(out);
let (data, _format_version) = read_format_version(data).unwrap();
let (data, _null_index_footer) = read_null_index_footer(data).unwrap();
test_all(data.clone(), u128_vals);
data
}
@@ -533,26 +563,111 @@ mod tests {
4_000_211_222u128,
333u128,
];
let data = test_aux_vals(vals);
let mut data = test_aux_vals(vals);
let _header = U128Header::deserialize(&mut data);
let decomp = CompactSpaceDecompressor::open(data).unwrap();
let positions = decomp.get_between_vals(0..=1);
let complete_range = 0..vals.len() as u32;
for (pos, val) in vals.iter().enumerate() {
let val = *val as u128;
let pos = pos as u32;
let mut positions = Vec::new();
decomp.get_positions_for_value_range(val..=val, pos..pos + 1, &mut positions);
assert_eq!(positions, vec![pos]);
}
// handle docid range out of bounds
let positions = get_positions_for_value_range_helper(&decomp, 0..=1, 1..u32::MAX);
assert_eq!(positions, vec![]);
let positions =
get_positions_for_value_range_helper(&decomp, 0..=1, complete_range.clone());
assert_eq!(positions, vec![0]);
let positions = decomp.get_between_vals(0..=2);
let positions =
get_positions_for_value_range_helper(&decomp, 0..=2, complete_range.clone());
assert_eq!(positions, vec![0]);
let positions = decomp.get_between_vals(0..=3);
let positions =
get_positions_for_value_range_helper(&decomp, 0..=3, complete_range.clone());
assert_eq!(positions, vec![0, 2]);
assert_eq!(decomp.get_between_vals(99999u128..=99999u128), vec![3]);
assert_eq!(decomp.get_between_vals(99999u128..=100000u128), vec![3, 4]);
assert_eq!(decomp.get_between_vals(99998u128..=100000u128), vec![3, 4]);
assert_eq!(decomp.get_between_vals(99998u128..=99999u128), vec![3]);
assert_eq!(decomp.get_between_vals(99998u128..=99998u128), vec![]);
assert_eq!(decomp.get_between_vals(333u128..=333u128), vec![8]);
assert_eq!(decomp.get_between_vals(332u128..=333u128), vec![8]);
assert_eq!(decomp.get_between_vals(332u128..=334u128), vec![8]);
assert_eq!(decomp.get_between_vals(333u128..=334u128), vec![8]);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
99999u128..=99999u128,
complete_range.clone()
),
vec![3]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
99999u128..=100000u128,
complete_range.clone()
),
vec![3, 4]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
99998u128..=100000u128,
complete_range.clone()
),
vec![3, 4]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
99998u128..=99999u128,
complete_range.clone()
),
vec![3]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
99998u128..=99998u128,
complete_range.clone()
),
vec![]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
333u128..=333u128,
complete_range.clone()
),
vec![8]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
332u128..=333u128,
complete_range.clone()
),
vec![8]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
332u128..=334u128,
complete_range.clone()
),
vec![8]
);
assert_eq!(
get_positions_for_value_range_helper(
&decomp,
333u128..=334u128,
complete_range.clone()
),
vec![8]
);
assert_eq!(
decomp.get_between_vals(4_000_211_221u128..=5_000_000_000u128),
get_positions_for_value_range_helper(
&decomp,
4_000_211_221u128..=5_000_000_000u128,
complete_range.clone()
),
vec![6, 7]
);
}
@@ -575,14 +690,32 @@ mod tests {
4_000_211_222u128,
333u128,
];
let data = test_aux_vals(vals);
let mut data = test_aux_vals(vals);
let _header = U128Header::deserialize(&mut data);
let decomp = CompactSpaceDecompressor::open(data).unwrap();
let positions = decomp.get_between_vals(0..=5);
assert_eq!(positions, vec![]);
let positions = decomp.get_between_vals(0..=100);
assert_eq!(positions, vec![0]);
let positions = decomp.get_between_vals(0..=105);
assert_eq!(positions, vec![0]);
let complete_range = 0..vals.len() as u32;
assert_eq!(
get_positions_for_value_range_helper(&decomp, 0..=5, complete_range.clone()),
vec![]
);
assert_eq!(
get_positions_for_value_range_helper(&decomp, 0..=100, complete_range.clone()),
vec![0]
);
assert_eq!(
get_positions_for_value_range_helper(&decomp, 0..=105, complete_range.clone()),
vec![0]
);
}
fn get_positions_for_value_range_helper<C: Column<T> + ?Sized, T: PartialOrd>(
column: &C,
value_range: RangeInclusive<T>,
doc_id_range: Range<u32>,
) -> Vec<u32> {
let mut positions = Vec::new();
column.get_docids_for_value_range(value_range, doc_id_range, &mut positions);
positions
}
#[test]
@@ -603,13 +736,33 @@ mod tests {
5_000_000_000,
];
let mut out = Vec::new();
serialize_u128(VecColumn::from(vals), &mut out).unwrap();
let decomp = open_u128(OwnedBytes::new(out)).unwrap();
serialize_u128(|| vals.iter().cloned(), vals.len() as u32, &mut out).unwrap();
let decomp = open_u128::<u128>(OwnedBytes::new(out)).unwrap();
let complete_range = 0..vals.len() as u32;
assert_eq!(decomp.get_between_vals(199..=200), vec![0]);
assert_eq!(decomp.get_between_vals(199..=201), vec![0, 1]);
assert_eq!(decomp.get_between_vals(200..=200), vec![0]);
assert_eq!(decomp.get_between_vals(1_000_000..=1_000_000), vec![11]);
assert_eq!(
get_positions_for_value_range_helper(&*decomp, 199..=200, complete_range.clone()),
vec![0]
);
assert_eq!(
get_positions_for_value_range_helper(&*decomp, 199..=201, complete_range.clone()),
vec![0, 1]
);
assert_eq!(
get_positions_for_value_range_helper(&*decomp, 200..=200, complete_range.clone()),
vec![0]
);
assert_eq!(
get_positions_for_value_range_helper(
&*decomp,
1_000_000..=1_000_000,
complete_range.clone()
),
vec![11]
);
}
#[test]

39
fastfield_codecs/src/format_version.rs Normal file
View File

@@ -0,0 +1,39 @@
use std::io;
use common::BinarySerializable;
use ownedbytes::OwnedBytes;
const MAGIC_NUMBER: u16 = 4335u16;
const FASTFIELD_FORMAT_VERSION: u8 = 1;
pub(crate) fn append_format_version(output: &mut impl io::Write) -> io::Result<()> {
FASTFIELD_FORMAT_VERSION.serialize(output)?;
MAGIC_NUMBER.serialize(output)?;
Ok(())
}
pub(crate) fn read_format_version(data: OwnedBytes) -> io::Result<(OwnedBytes, u8)> {
let (data, magic_number_bytes) = data.rsplit(2);
let magic_number = u16::deserialize(&mut magic_number_bytes.as_slice())?;
if magic_number != MAGIC_NUMBER {
return Err(io::Error::new(
io::ErrorKind::InvalidData,
format!("magic number mismatch {} != {}", magic_number, MAGIC_NUMBER),
));
}
let (data, format_version_bytes) = data.rsplit(1);
let format_version = u8::deserialize(&mut format_version_bytes.as_slice())?;
if format_version > FASTFIELD_FORMAT_VERSION {
return Err(io::Error::new(
io::ErrorKind::InvalidData,
format!(
"Unsupported fastfield format version: {}. Max supported version: {}",
format_version, FASTFIELD_FORMAT_VERSION
),
));
}
Ok((data, format_version))
}

159
fastfield_codecs/src/lib.rs
View File

@@ -1,5 +1,12 @@
#![warn(missing_docs)]
#![cfg_attr(all(feature = "unstable", test), feature(test))]
//! # `fastfield_codecs`
//!
//! - Columnar storage of data for tantivy [`Column`].
//! - Encode data in different codecs.
//! - Monotonically map values to u64/u128
#[cfg(test)]
#[macro_use]
extern crate more_asserts;
@@ -13,16 +20,24 @@ use std::sync::Arc;
use common::BinarySerializable;
use compact_space::CompactSpaceDecompressor;
use format_version::read_format_version;
use monotonic_mapping::{
StrictlyMonotonicMappingInverter, StrictlyMonotonicMappingToInternal,
StrictlyMonotonicMappingToInternalBaseval, StrictlyMonotonicMappingToInternalGCDBaseval,
};
use null_index_footer::read_null_index_footer;
use ownedbytes::OwnedBytes;
use serialize::Header;
use serialize::{Header, U128Header};
mod bitpacked;
mod blockwise_linear;
mod compact_space;
mod format_version;
mod line;
mod linear;
mod monotonic_mapping;
mod sparse_codec_wrapper;
mod monotonic_mapping_u128;
mod null_index_footer;
mod column;
mod gcd;
@@ -30,20 +45,26 @@ mod serialize;
use self::bitpacked::BitpackedCodec;
use self::blockwise_linear::BlockwiseLinearCodec;
pub use self::column::{monotonic_map_column, Column, VecColumn};
pub use self::column::{monotonic_map_column, Column, IterColumn, VecColumn};
use self::linear::LinearCodec;
pub use self::monotonic_mapping::MonotonicallyMappableToU64;
pub use self::monotonic_mapping::{MonotonicallyMappableToU64, StrictlyMonotonicFn};
pub use self::monotonic_mapping_u128::MonotonicallyMappableToU128;
pub use self::serialize::{
estimate, serialize, serialize_and_load, serialize_u128, NormalizedHeader,
};
pub use sparse_codec_wrapper::DenseCodec;
pub use sparse_codec_wrapper::SparseCodecRoaringBitmap;
#[derive(PartialEq, Eq, PartialOrd, Ord, Debug, Clone, Copy)]
#[repr(u8)]
/// Available codecs to use to encode the u64 (via [`MonotonicallyMappableToU64`]) converted data.
pub enum FastFieldCodecType {
/// Bitpack all values in the value range. The number of bits is defined by the amplitude
/// `column.max_value() - column.min_value()`
Bitpacked = 1,
/// Linear interpolation puts a line between the first and last value and then bitpacks the
/// values by the offset from the line. The number of bits is defined by the max deviation from
/// the line.
Linear = 2,
/// Same as [`FastFieldCodecType::Linear`], but encodes in blocks of 512 elements.
BlockwiseLinear = 3,
}
@@ -61,11 +82,11 @@ impl BinarySerializable for FastFieldCodecType {
}
impl FastFieldCodecType {
pub fn to_code(self) -> u8 {
pub(crate) fn to_code(self) -> u8 {
self as u8
}
pub fn from_code(code: u8) -> Option<Self> {
pub(crate) fn from_code(code: u8) -> Option<Self> {
match code {
1 => Some(Self::Bitpacked),
2 => Some(Self::Linear),
@@ -75,53 +96,59 @@ impl FastFieldCodecType {
}
}
/// Returns the correct codec reader wrapped in the `Arc` for the data.
pub fn open_u128(bytes: OwnedBytes) -> io::Result<Arc<dyn Column<u128>>> {
Ok(Arc::new(CompactSpaceDecompressor::open(bytes)?))
#[derive(PartialEq, Eq, PartialOrd, Ord, Debug, Clone, Copy)]
#[repr(u8)]
/// Available codecs to use to encode the u128 (via [`MonotonicallyMappableToU128`]) converted data.
pub enum U128FastFieldCodecType {
/// This codec takes a large number space (u128) and reduces it to a compact number space, by
/// removing the holes.
CompactSpace = 1,
}
//DenseCodec
//
/// Returns the correct codec reader wrapped in the `Arc` for the data.
pub fn open_dense<T: MonotonicallyMappableToU64>(
mut bytes: OwnedBytes,
fill_ratio: u32,
) -> io::Result<Arc<dyn Column<T>>> {
let header = Header::deserialize(&mut bytes)?;
match header.codec_type {
FastFieldCodecType::Bitpacked => {
open_specific_codec_dense::<BitpackedCodec, _>(bytes, &header, fill_ratio)
}
FastFieldCodecType::Linear => {
open_specific_codec_dense::<LinearCodec, _>(bytes, &header, fill_ratio)
}
FastFieldCodecType::BlockwiseLinear => {
open_specific_codec_dense::<BlockwiseLinearCodec, _>(bytes, &header, fill_ratio)
impl BinarySerializable for U128FastFieldCodecType {
fn serialize<W: Write>(&self, wrt: &mut W) -> io::Result<()> {
self.to_code().serialize(wrt)
}
fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
let code = u8::deserialize(reader)?;
let codec_type: Self = Self::from_code(code)
.ok_or_else(|| io::Error::new(io::ErrorKind::InvalidData, "Unknown code `{code}.`"))?;
Ok(codec_type)
}
}
impl U128FastFieldCodecType {
pub(crate) fn to_code(self) -> u8 {
self as u8
}
pub(crate) fn from_code(code: u8) -> Option<Self> {
match code {
1 => Some(Self::CompactSpace),
_ => None,
}
}
}
fn open_specific_codec_dense<C: FastFieldCodec, Item: MonotonicallyMappableToU64>(
/// Returns the correct codec reader wrapped in the `Arc` for the data.
pub fn open_u128<Item: MonotonicallyMappableToU128>(
bytes: OwnedBytes,
header: &Header,
fill_ratio: u32,
) -> io::Result<Arc<dyn Column<Item>>> {
let normalized_header = header.normalized();
let reader = C::open_from_bytes(bytes, normalized_header)?;
let reader = DenseCodec::with_fill_ratio(reader, fill_ratio);
let min_value = header.min_value;
if let Some(gcd) = header.gcd {
let monotonic_mapping = move |val: u64| Item::from_u64(min_value + val * gcd.get());
Ok(Arc::new(monotonic_map_column(reader, monotonic_mapping)))
} else {
let monotonic_mapping = move |val: u64| Item::from_u64(min_value + val);
Ok(Arc::new(monotonic_map_column(reader, monotonic_mapping)))
}
let (bytes, _format_version) = read_format_version(bytes)?;
let (mut bytes, _null_index_footer) = read_null_index_footer(bytes)?;
let header = U128Header::deserialize(&mut bytes)?;
assert_eq!(header.codec_type, U128FastFieldCodecType::CompactSpace);
let reader = CompactSpaceDecompressor::open(bytes)?;
let inverted: StrictlyMonotonicMappingInverter<StrictlyMonotonicMappingToInternal<Item>> =
StrictlyMonotonicMappingToInternal::<Item>::new().into();
Ok(Arc::new(monotonic_map_column(reader, inverted)))
}
/// Returns the correct codec reader wrapped in the `Arc` for the data.
pub fn open<T: MonotonicallyMappableToU64>(
mut bytes: OwnedBytes,
) -> io::Result<Arc<dyn Column<T>>> {
pub fn open<T: MonotonicallyMappableToU64>(bytes: OwnedBytes) -> io::Result<Arc<dyn Column<T>>> {
let (bytes, _format_version) = read_format_version(bytes)?;
let (mut bytes, _null_index_footer) = read_null_index_footer(bytes)?;
let header = Header::deserialize(&mut bytes)?;
match header.codec_type {
FastFieldCodecType::Bitpacked => open_specific_codec::<BitpackedCodec, _>(bytes, &header),
@@ -140,11 +167,15 @@ fn open_specific_codec<C: FastFieldCodec, Item: MonotonicallyMappableToU64>(
let reader = C::open_from_bytes(bytes, normalized_header)?;
let min_value = header.min_value;
if let Some(gcd) = header.gcd {
let monotonic_mapping = move |val: u64| Item::from_u64(min_value + val * gcd.get());
Ok(Arc::new(monotonic_map_column(reader, monotonic_mapping)))
let mapping = StrictlyMonotonicMappingInverter::from(
StrictlyMonotonicMappingToInternalGCDBaseval::new(gcd.get(), min_value),
);
Ok(Arc::new(monotonic_map_column(reader, mapping)))
} else {
let monotonic_mapping = move |val: u64| Item::from_u64(min_value + val);
Ok(Arc::new(monotonic_map_column(reader, monotonic_mapping)))
let mapping = StrictlyMonotonicMappingInverter::from(
StrictlyMonotonicMappingToInternalBaseval::new(min_value),
);
Ok(Arc::new(monotonic_map_column(reader, mapping)))
}
}
@@ -176,6 +207,7 @@ trait FastFieldCodec: 'static {
fn estimate(column: &dyn Column) -> Option<f32>;
}
/// The list of all available codecs for u64 convertible data.
pub const ALL_CODEC_TYPES: [FastFieldCodecType; 3] = [
FastFieldCodecType::Bitpacked,
FastFieldCodecType::BlockwiseLinear,
@@ -184,6 +216,7 @@ pub const ALL_CODEC_TYPES: [FastFieldCodecType; 3] = [
#[cfg(test)]
mod tests {
use proptest::prelude::*;
use proptest::strategy::Strategy;
use proptest::{prop_oneof, proptest};
@@ -209,15 +242,32 @@ mod tests {
let actual_compression = out.len() as f32 / (data.len() as f32 * 8.0);
let reader = crate::open::<u64>(OwnedBytes::new(out)).unwrap();
assert_eq!(reader.num_vals(), data.len() as u64);
assert_eq!(reader.num_vals(), data.len() as u32);
for (doc, orig_val) in data.iter().copied().enumerate() {
let val = reader.get_val(doc as u64);
let val = reader.get_val(doc as u32);
assert_eq!(
val, orig_val,
"val `{val}` does not match orig_val {orig_val:?}, in data set {name}, data \
`{data:?}`",
);
}
if !data.is_empty() {
let test_rand_idx = rand::thread_rng().gen_range(0..=data.len() - 1);
let expected_positions: Vec<u32> = data
.iter()
.enumerate()
.filter(|(_, el)| **el == data[test_rand_idx])
.map(|(pos, _)| pos as u32)
.collect();
let mut positions = Vec::new();
reader.get_docids_for_value_range(
data[test_rand_idx]..=data[test_rand_idx],
0..data.len() as u32,
&mut positions,
);
assert_eq!(expected_positions, positions);
}
Some((estimation, actual_compression))
}
@@ -353,7 +403,7 @@ mod tests {
#[test]
fn estimation_test_bad_interpolation_case_monotonically_increasing() {
let mut data: Vec<u64> = (200..=20000_u64).collect();
let mut data: Vec<u64> = (201..=20000_u64).collect();
data.push(1_000_000);
let data: VecColumn = data.as_slice().into();
@@ -427,7 +477,7 @@ mod bench {
b.iter(|| {
let mut sum = 0u64;
for pos in value_iter() {
let val = col.get_val(pos as u64);
let val = col.get_val(pos as u32);
sum = sum.wrapping_add(val);
}
sum
@@ -439,7 +489,7 @@ mod bench {
b.iter(|| {
let mut sum = 0u64;
for pos in value_iter() {
let val = col.get_val(pos as u64);
let val = col.get_val(pos as u32);
sum = sum.wrapping_add(val);
}
sum
@@ -481,7 +531,6 @@ mod bench {
let data: Vec<_> = get_data();
bench_get::<BitpackedCodec>(b, &data);
}
#[bench]
fn bench_fastfield_bitpack_get_dynamic(b: &mut Bencher) {
let data: Vec<_> = get_data();

52
fastfield_codecs/src/line.rs
View File

@@ -1,5 +1,5 @@
use std::io;
use std::num::NonZeroU64;
use std::num::NonZeroU32;
use common::{BinarySerializable, VInt};
@@ -29,7 +29,7 @@ pub struct Line {
/// compute_slope(y0, y1)
/// = compute_slope(y0 + X % 2^64, y1 + X % 2^64)
/// `
fn compute_slope(y0: u64, y1: u64, num_vals: NonZeroU64) -> u64 {
fn compute_slope(y0: u64, y1: u64, num_vals: NonZeroU32) -> u64 {
let dy = y1.wrapping_sub(y0);
let sign = dy <= (1 << 63);
let abs_dy = if sign {
@@ -43,7 +43,7 @@ fn compute_slope(y0: u64, y1: u64, num_vals: NonZeroU64) -> u64 {
return 0u64;
}
let abs_slope = (abs_dy << 32) / num_vals.get();
let abs_slope = (abs_dy << 32) / num_vals.get() as u64;
if sign {
abs_slope
} else {
@@ -62,35 +62,43 @@ fn compute_slope(y0: u64, y1: u64, num_vals: NonZeroU64) -> u64 {
impl Line {
#[inline(always)]
pub fn eval(&self, x: u64) -> u64 {
let linear_part = (x.wrapping_mul(self.slope) >> 32) as i32 as u64;
pub fn eval(&self, x: u32) -> u64 {
let linear_part = ((x as u64).wrapping_mul(self.slope) >> 32) as i32 as u64;
self.intercept.wrapping_add(linear_part)
}
// Same as train, but the intercept is only estimated from provided sample positions
pub fn estimate(ys: &dyn Column, sample_positions: &[u64]) -> Self {
pub fn estimate(sample_positions_and_values: &[(u64, u64)]) -> Self {
let first_val = sample_positions_and_values[0].1;
let last_val = sample_positions_and_values[sample_positions_and_values.len() - 1].1;
let num_vals = sample_positions_and_values[sample_positions_and_values.len() - 1].0 + 1;
Self::train_from(
ys,
sample_positions
.iter()
.cloned()
.map(|pos| (pos, ys.get_val(pos))),
first_val,
last_val,
num_vals as u32,
sample_positions_and_values.iter().cloned(),
)
}
// Intercept is only computed from provided positions
fn train_from(ys: &dyn Column, positions_and_values: impl Iterator<Item = (u64, u64)>) -> Self {
let num_vals = if let Some(num_vals) = NonZeroU64::new(ys.num_vals() - 1) {
num_vals
fn train_from(
first_val: u64,
last_val: u64,
num_vals: u32,
positions_and_values: impl Iterator<Item = (u64, u64)>,
) -> Self {
// TODO replace with let else
let idx_last_val = if let Some(idx_last_val) = NonZeroU32::new(num_vals - 1) {
idx_last_val
} else {
return Line::default();
};
let y0 = ys.get_val(0);
let y1 = ys.get_val(num_vals.get());
let y0 = first_val;
let y1 = last_val;
// We first independently pick our slope.
let slope = compute_slope(y0, y1, num_vals);
let slope = compute_slope(y0, y1, idx_last_val);
// We picked our slope. Note that it does not have to be perfect.
// Now we need to compute the best intercept.
@@ -121,7 +129,7 @@ impl Line {
};
let heuristic_shift = y0.wrapping_sub(MID_POINT);
line.intercept = positions_and_values
.map(|(pos, y)| y.wrapping_sub(line.eval(pos)))
.map(|(pos, y)| y.wrapping_sub(line.eval(pos as u32)))
.min_by_key(|&val| val.wrapping_sub(heuristic_shift))
.unwrap_or(0u64); //< Never happens.
line
@@ -138,8 +146,12 @@ impl Line {
/// This function is only invariable by translation if all of the
/// `ys` are packaged into half of the space. (See heuristic below)
pub fn train(ys: &dyn Column) -> Self {
let first_val = ys.iter().next().unwrap();
let last_val = ys.iter().nth(ys.num_vals() as usize - 1).unwrap();
Self::train_from(
ys,
first_val,
last_val,
ys.num_vals(),
ys.iter().enumerate().map(|(pos, val)| (pos as u64, val)),
)
}
@@ -187,7 +199,7 @@ mod tests {
let line = Line::train(&VecColumn::from(&ys));
ys.iter()
.enumerate()
.map(|(x, y)| y.wrapping_sub(line.eval(x as u64)))
.map(|(x, y)| y.wrapping_sub(line.eval(x as u32)))
.max()
}

31
fastfield_codecs/src/linear.rs
View File

@@ -19,7 +19,7 @@ pub struct LinearReader {
impl Column for LinearReader {
#[inline]
fn get_val(&self, doc: u64) -> u64 {
fn get_val(&self, doc: u32) -> u64 {
let interpoled_val: u64 = self.linear_params.line.eval(doc);
let bitpacked_diff = self.linear_params.bit_unpacker.get(doc, &self.data);
interpoled_val.wrapping_add(bitpacked_diff)
@@ -37,7 +37,7 @@ impl Column for LinearReader {
}
#[inline]
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.header.num_vals
}
}
@@ -93,7 +93,7 @@ impl FastFieldCodec for LinearCodec {
.iter()
.enumerate()
.map(|(pos, actual_value)| {
let calculated_value = line.eval(pos as u64);
let calculated_value = line.eval(pos as u32);
actual_value.wrapping_sub(calculated_value)
})
.max()
@@ -108,7 +108,7 @@ impl FastFieldCodec for LinearCodec {
let mut bit_packer = BitPacker::new();
for (pos, actual_value) in column.iter().enumerate() {
let calculated_value = line.eval(pos as u64);
let calculated_value = line.eval(pos as u32);
let offset = actual_value.wrapping_sub(calculated_value);
bit_packer.write(offset, num_bits, write)?;
}
@@ -126,19 +126,21 @@ impl FastFieldCodec for LinearCodec {
return None; // disable compressor for this case
}
// let's sample at 0%, 5%, 10% .. 95%, 100%
let num_vals = column.num_vals() as f32 / 100.0;
let sample_positions = (0..20)
.map(|pos| (num_vals * pos as f32 * 5.0) as u64)
.collect::<Vec<_>>();
let limit_num_vals = column.num_vals().min(100_000);
let line = Line::estimate(column, &sample_positions);
let num_samples = 100;
let step_size = (limit_num_vals / num_samples).max(1); // 20 samples
let mut sample_positions_and_values: Vec<_> = Vec::new();
for (pos, val) in column.iter().enumerate().step_by(step_size as usize) {
sample_positions_and_values.push((pos as u64, val));
}
let estimated_bit_width = sample_positions
let line = Line::estimate(&sample_positions_and_values);
let estimated_bit_width = sample_positions_and_values
.into_iter()
.map(|pos| {
let actual_value = column.get_val(pos);
let interpolated_val = line.eval(pos as u64);
.map(|(pos, actual_value)| {
let interpolated_val = line.eval(pos as u32);
actual_value.wrapping_sub(interpolated_val)
})
.map(|diff| ((diff as f32 * 1.5) * 2.0) as u64)
@@ -146,6 +148,7 @@ impl FastFieldCodec for LinearCodec {
.max()
.unwrap_or(0);
// Extrapolate to whole column
let num_bits = (estimated_bit_width as u64 * column.num_vals() as u64) + 64;
let num_bits_uncompressed = 64 * column.num_vals();
Some(num_bits as f32 / num_bits_uncompressed as f32)

17
fastfield_codecs/src/main.rs
View File

@@ -90,7 +90,7 @@ fn bench_ip() {
{
let mut data = vec![];
for dataset in dataset.chunks(500_000) {
serialize_u128(VecColumn::from(dataset), &mut data).unwrap();
serialize_u128(|| dataset.iter().cloned(), dataset.len() as u32, &mut data).unwrap();
}
let compression = data.len() as f64 / (dataset.len() * 16) as f64;
println!("Compression 50_000 chunks {:.4}", compression);
@@ -101,7 +101,10 @@ fn bench_ip() {
}
let mut data = vec![];
serialize_u128(VecColumn::from(&dataset), &mut data).unwrap();
{
print_time!("creation");
serialize_u128(|| dataset.iter().cloned(), dataset.len() as u32, &mut data).unwrap();
}
let compression = data.len() as f64 / (dataset.len() * 16) as f64;
println!("Compression {:.2}", compression);
@@ -110,11 +113,17 @@ fn bench_ip() {
(data.len() * 8) as f32 / dataset.len() as f32
);
let decompressor = open_u128(OwnedBytes::new(data)).unwrap();
let decompressor = open_u128::<u128>(OwnedBytes::new(data)).unwrap();
// Sample some ranges
let mut doc_values = Vec::new();
for value in dataset.iter().take(1110).skip(1100).cloned() {
doc_values.clear();
print_time!("get range");
let doc_values = decompressor.get_between_vals(value..=value);
decompressor.get_docids_for_value_range(
value..=value,
0..decompressor.num_vals(),
&mut doc_values,
);
println!("{:?}", doc_values.len());
}
}

177
fastfield_codecs/src/monotonic_mapping.rs
View File

@@ -1,3 +1,11 @@
use std::marker::PhantomData;
use fastdivide::DividerU64;
use crate::MonotonicallyMappableToU128;
/// Monotonic maps a value to u64 value space.
/// Monotonic mapping enables `PartialOrd` on u64 space without conversion to original space.
pub trait MonotonicallyMappableToU64: 'static + PartialOrd + Copy + Send + Sync {
/// Converts a value to u64.
///
@@ -11,6 +19,145 @@ pub trait MonotonicallyMappableToU64: 'static + PartialOrd + Copy + Send + Sync
fn from_u64(val: u64) -> Self;
}
/// Values need to be strictly monotonic mapped to a `Internal` value (u64 or u128) that can be
/// used in fast field codecs.
///
/// The monotonic mapping is required so that `PartialOrd` can be used on `Internal` without
/// converting to `External`.
///
/// All strictly monotonic functions are invertible because they are guaranteed to have a one-to-one
/// mapping from their range to their domain. The `inverse` method is required when opening a codec,
/// so a value can be converted back to its original domain (e.g. ip address or f64) from its
/// internal representation.
pub trait StrictlyMonotonicFn<External, Internal> {
/// Strictly monotonically maps the value from External to Internal.
fn mapping(&self, inp: External) -> Internal;
/// Inverse of `mapping`. Maps the value from Internal to External.
fn inverse(&self, out: Internal) -> External;
}
/// Inverts a strictly monotonic mapping from `StrictlyMonotonicFn<A, B>` to
/// `StrictlyMonotonicFn<B, A>`.
///
/// # Warning
///
/// This type comes with a footgun. A type being strictly monotonic does not impose that the inverse
/// mapping is strictly monotonic over the entire space External. e.g. a -> a * 2. Use at your own
/// risks.
pub(crate) struct StrictlyMonotonicMappingInverter<T> {
orig_mapping: T,
}
impl<T> From<T> for StrictlyMonotonicMappingInverter<T> {
fn from(orig_mapping: T) -> Self {
Self { orig_mapping }
}
}
impl<From, To, T> StrictlyMonotonicFn<To, From> for StrictlyMonotonicMappingInverter<T>
where T: StrictlyMonotonicFn<From, To>
{
fn mapping(&self, val: To) -> From {
self.orig_mapping.inverse(val)
}
fn inverse(&self, val: From) -> To {
self.orig_mapping.mapping(val)
}
}
/// Applies the strictly monotonic mapping from `T` without any additional changes.
pub(crate) struct StrictlyMonotonicMappingToInternal<T> {
_phantom: PhantomData<T>,
}
impl<T> StrictlyMonotonicMappingToInternal<T> {
pub(crate) fn new() -> StrictlyMonotonicMappingToInternal<T> {
Self {
_phantom: PhantomData,
}
}
}
impl<External: MonotonicallyMappableToU128, T: MonotonicallyMappableToU128>
StrictlyMonotonicFn<External, u128> for StrictlyMonotonicMappingToInternal<T>
where T: MonotonicallyMappableToU128
{
fn mapping(&self, inp: External) -> u128 {
External::to_u128(inp)
}
fn inverse(&self, out: u128) -> External {
External::from_u128(out)
}
}
impl<External: MonotonicallyMappableToU64, T: MonotonicallyMappableToU64>
StrictlyMonotonicFn<External, u64> for StrictlyMonotonicMappingToInternal<T>
where T: MonotonicallyMappableToU64
{
fn mapping(&self, inp: External) -> u64 {
External::to_u64(inp)
}
fn inverse(&self, out: u64) -> External {
External::from_u64(out)
}
}
/// Mapping dividing by gcd and a base value.
///
/// The function is assumed to be only called on values divided by passed
/// gcd value. (It is necessary for the function to be monotonic.)
pub(crate) struct StrictlyMonotonicMappingToInternalGCDBaseval {
gcd_divider: DividerU64,
gcd: u64,
min_value: u64,
}
impl StrictlyMonotonicMappingToInternalGCDBaseval {
pub(crate) fn new(gcd: u64, min_value: u64) -> Self {
let gcd_divider = DividerU64::divide_by(gcd);
Self {
gcd_divider,
gcd,
min_value,
}
}
}
impl<External: MonotonicallyMappableToU64> StrictlyMonotonicFn<External, u64>
for StrictlyMonotonicMappingToInternalGCDBaseval
{
fn mapping(&self, inp: External) -> u64 {
self.gcd_divider
.divide(External::to_u64(inp) - self.min_value)
}
fn inverse(&self, out: u64) -> External {
External::from_u64(self.min_value + out * self.gcd)
}
}
/// Strictly monotonic mapping with a base value.
pub(crate) struct StrictlyMonotonicMappingToInternalBaseval {
min_value: u64,
}
impl StrictlyMonotonicMappingToInternalBaseval {
pub(crate) fn new(min_value: u64) -> Self {
Self { min_value }
}
}
impl<External: MonotonicallyMappableToU64> StrictlyMonotonicFn<External, u64>
for StrictlyMonotonicMappingToInternalBaseval
{
fn mapping(&self, val: External) -> u64 {
External::to_u64(val) - self.min_value
}
fn inverse(&self, val: u64) -> External {
External::from_u64(self.min_value + val)
}
}
impl MonotonicallyMappableToU64 for u64 {
fn to_u64(self) -> u64 {
self
@@ -54,3 +201,33 @@ impl MonotonicallyMappableToU64 for f64 {
common::u64_to_f64(val)
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn strictly_monotonic_test() {
// identity mapping
test_round_trip(&StrictlyMonotonicMappingToInternal::<u64>::new(), 100u64);
// round trip to i64
test_round_trip(&StrictlyMonotonicMappingToInternal::<i64>::new(), 100u64);
// identity mapping
test_round_trip(&StrictlyMonotonicMappingToInternal::<u128>::new(), 100u128);
// base value to i64 round trip
let mapping = StrictlyMonotonicMappingToInternalBaseval::new(100);
test_round_trip::<_, _, u64>(&mapping, 100i64);
// base value and gcd to u64 round trip
let mapping = StrictlyMonotonicMappingToInternalGCDBaseval::new(10, 100);
test_round_trip::<_, _, u64>(&mapping, 100u64);
}
fn test_round_trip<T: StrictlyMonotonicFn<K, L>, K: std::fmt::Debug + Eq + Copy, L>(
mapping: &T,
test_val: K,
) {
assert_eq!(mapping.inverse(mapping.mapping(test_val)), test_val);
}
}

16
fastfield_codecs/src/monotonic_mapping_u128.rs
View File

@@ -1,5 +1,7 @@
use std::net::{IpAddr, Ipv6Addr};
use std::net::Ipv6Addr;
/// Montonic maps a value to u128 value space
/// Monotonic mapping enables `PartialOrd` on u128 space without conversion to original space.
pub trait MonotonicallyMappableToU128: 'static + PartialOrd + Copy + Send + Sync {
/// Converts a value to u128.
///
@@ -23,20 +25,16 @@ impl MonotonicallyMappableToU128 for u128 {
}
}
impl MonotonicallyMappableToU128 for IpAddr {
impl MonotonicallyMappableToU128 for Ipv6Addr {
fn to_u128(self) -> u128 {
ip_to_u128(self)
}
fn from_u128(val: u128) -> Self {
IpAddr::from(val.to_be_bytes())
Ipv6Addr::from(val.to_be_bytes())
}
}
fn ip_to_u128(ip_addr: IpAddr) -> u128 {
let ip_addr_v6: Ipv6Addr = match ip_addr {
IpAddr::V4(v4) => v4.to_ipv6_mapped(),
IpAddr::V6(v6) => v6,
};
u128::from_be_bytes(ip_addr_v6.octets())
fn ip_to_u128(ip_addr: Ipv6Addr) -> u128 {
u128::from_be_bytes(ip_addr.octets())
}

144
fastfield_codecs/src/null_index_footer.rs Normal file
View File

@@ -0,0 +1,144 @@
use std::io::{self, Write};
use std::ops::Range;
use common::{BinarySerializable, CountingWriter, VInt};
use ownedbytes::OwnedBytes;
#[derive(Debug, Clone, Copy, Eq, PartialEq)]
pub(crate) enum FastFieldCardinality {
Single = 1,
}
impl BinarySerializable for FastFieldCardinality {
fn serialize<W: Write>(&self, wrt: &mut W) -> io::Result<()> {
self.to_code().serialize(wrt)
}
fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
let code = u8::deserialize(reader)?;
let codec_type: Self = Self::from_code(code)
.ok_or_else(|| io::Error::new(io::ErrorKind::InvalidData, "Unknown code `{code}.`"))?;
Ok(codec_type)
}
}
impl FastFieldCardinality {
pub(crate) fn to_code(self) -> u8 {
self as u8
}
pub(crate) fn from_code(code: u8) -> Option<Self> {
match code {
1 => Some(Self::Single),
_ => None,
}
}
}
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub(crate) enum NullIndexCodec {
Full = 1,
}
impl BinarySerializable for NullIndexCodec {
fn serialize<W: Write>(&self, wrt: &mut W) -> io::Result<()> {
self.to_code().serialize(wrt)
}
fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
let code = u8::deserialize(reader)?;
let codec_type: Self = Self::from_code(code)
.ok_or_else(|| io::Error::new(io::ErrorKind::InvalidData, "Unknown code `{code}.`"))?;
Ok(codec_type)
}
}
impl NullIndexCodec {
pub(crate) fn to_code(self) -> u8 {
self as u8
}
pub(crate) fn from_code(code: u8) -> Option<Self> {
match code {
1 => Some(Self::Full),
_ => None,
}
}
}
#[derive(Debug, Clone, Eq, PartialEq)]
pub(crate) struct NullIndexFooter {
pub(crate) cardinality: FastFieldCardinality,
pub(crate) null_index_codec: NullIndexCodec,
// Unused for NullIndexCodec::Full
pub(crate) null_index_byte_range: Range<u64>,
}
impl BinarySerializable for NullIndexFooter {
fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
self.cardinality.serialize(writer)?;
self.null_index_codec.serialize(writer)?;
VInt(self.null_index_byte_range.start).serialize(writer)?;
VInt(self.null_index_byte_range.end - self.null_index_byte_range.start)
.serialize(writer)?;
Ok(())
}
fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
let cardinality = FastFieldCardinality::deserialize(reader)?;
let null_index_codec = NullIndexCodec::deserialize(reader)?;
let null_index_byte_range_start = VInt::deserialize(reader)?.0;
let null_index_byte_range_end = VInt::deserialize(reader)?.0 + null_index_byte_range_start;
Ok(Self {
cardinality,
null_index_codec,
null_index_byte_range: null_index_byte_range_start..null_index_byte_range_end,
})
}
}
pub(crate) fn append_null_index_footer(
output: &mut impl io::Write,
null_index_footer: NullIndexFooter,
) -> io::Result<()> {
let mut counting_write = CountingWriter::wrap(output);
null_index_footer.serialize(&mut counting_write)?;
let footer_payload_len = counting_write.written_bytes();
BinarySerializable::serialize(&(footer_payload_len as u16), &mut counting_write)?;
Ok(())
}
pub(crate) fn read_null_index_footer(
data: OwnedBytes,
) -> io::Result<(OwnedBytes, NullIndexFooter)> {
let (data, null_footer_length_bytes) = data.rsplit(2);
let footer_length = u16::deserialize(&mut null_footer_length_bytes.as_slice())?;
let (data, null_index_footer_bytes) = data.rsplit(footer_length as usize);
let null_index_footer = NullIndexFooter::deserialize(&mut null_index_footer_bytes.as_ref())?;
Ok((data, null_index_footer))
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn null_index_footer_deser_test() {
let null_index_footer = NullIndexFooter {
cardinality: FastFieldCardinality::Single,
null_index_codec: NullIndexCodec::Full,
null_index_byte_range: 100..120,
};
let mut out = vec![];
null_index_footer.serialize(&mut out).unwrap();
assert_eq!(
null_index_footer,
NullIndexFooter::deserialize(&mut &out[..]).unwrap()
);
}
}

141
fastfield_codecs/src/serialize.rs
View File

@@ -22,33 +22,42 @@ use std::num::NonZeroU64;
use std::sync::Arc;
use common::{BinarySerializable, VInt};
use fastdivide::DividerU64;
use log::warn;
use ownedbytes::OwnedBytes;
use crate::bitpacked::BitpackedCodec;
use crate::blockwise_linear::BlockwiseLinearCodec;
use crate::compact_space::CompactSpaceCompressor;
use crate::format_version::append_format_version;
use crate::linear::LinearCodec;
use crate::monotonic_mapping::{
StrictlyMonotonicFn, StrictlyMonotonicMappingToInternal,
StrictlyMonotonicMappingToInternalGCDBaseval,
};
use crate::null_index_footer::{
append_null_index_footer, FastFieldCardinality, NullIndexCodec, NullIndexFooter,
};
use crate::{
monotonic_map_column, Column, FastFieldCodec, FastFieldCodecType, MonotonicallyMappableToU64,
VecColumn, ALL_CODEC_TYPES,
U128FastFieldCodecType, VecColumn, ALL_CODEC_TYPES,
};
/// The normalized header gives some parameters after applying the following
/// normalization of the vector:
/// val -> (val - min_value) / gcd
/// `val -> (val - min_value) / gcd`
///
/// By design, after normalization, `min_value = 0` and `gcd = 1`.
#[derive(Debug, Copy, Clone)]
pub struct NormalizedHeader {
pub num_vals: u64,
/// The number of values in the underlying column.
pub num_vals: u32,
/// The max value of the underlying column.
pub max_value: u64,
}
#[derive(Debug, Copy, Clone)]
pub(crate) struct Header {
pub num_vals: u64,
pub num_vals: u32,
pub min_value: u64,
pub max_value: u64,
pub gcd: Option<NonZeroU64>,
@@ -57,8 +66,11 @@ pub(crate) struct Header {
impl Header {
pub fn normalized(self) -> NormalizedHeader {
let max_value =
(self.max_value - self.min_value) / self.gcd.map(|gcd| gcd.get()).unwrap_or(1);
let gcd = self.gcd.map(|gcd| gcd.get()).unwrap_or(1);
let gcd_min_val_mapping =
StrictlyMonotonicMappingToInternalGCDBaseval::new(gcd, self.min_value);
let max_value = gcd_min_val_mapping.mapping(self.max_value);
NormalizedHeader {
num_vals: self.num_vals,
max_value,
@@ -66,10 +78,7 @@ impl Header {
}
pub fn normalize_column<C: Column>(&self, from_column: C) -> impl Column {
let min_value = self.min_value;
let gcd = self.gcd.map(|gcd| gcd.get()).unwrap_or(1);
let divider = DividerU64::divide_by(gcd);
monotonic_map_column(from_column, move |val| divider.divide(val - min_value))
normalize_column(from_column, self.min_value, self.gcd)
}
pub fn compute_header(
@@ -81,9 +90,8 @@ impl Header {
let max_value = column.max_value();
let gcd = crate::gcd::find_gcd(column.iter().map(|val| val - min_value))
.filter(|gcd| gcd.get() > 1u64);
let divider = DividerU64::divide_by(gcd.map(|gcd| gcd.get()).unwrap_or(1u64));
let shifted_column = monotonic_map_column(&column, |val| divider.divide(val - min_value));
let codec_type = detect_codec(shifted_column, codecs)?;
let normalized_column = normalize_column(column, min_value, gcd);
let codec_type = detect_codec(normalized_column, codecs)?;
Some(Header {
num_vals,
min_value,
@@ -94,9 +102,42 @@ impl Header {
}
}
#[derive(Debug, Copy, Clone, PartialEq, Eq)]
pub(crate) struct U128Header {
pub num_vals: u32,
pub codec_type: U128FastFieldCodecType,
}
impl BinarySerializable for U128Header {
fn serialize<W: io::Write>(&self, writer: &mut W) -> io::Result<()> {
VInt(self.num_vals as u64).serialize(writer)?;
self.codec_type.serialize(writer)?;
Ok(())
}
fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
let num_vals = VInt::deserialize(reader)?.0 as u32;
let codec_type = U128FastFieldCodecType::deserialize(reader)?;
Ok(U128Header {
num_vals,
codec_type,
})
}
}
pub fn normalize_column<C: Column>(
from_column: C,
min_value: u64,
gcd: Option<NonZeroU64>,
) -> impl Column {
let gcd = gcd.map(|gcd| gcd.get()).unwrap_or(1);
let mapping = StrictlyMonotonicMappingToInternalGCDBaseval::new(gcd, min_value);
monotonic_map_column(from_column, mapping)
}
impl BinarySerializable for Header {
fn serialize<W: io::Write>(&self, writer: &mut W) -> io::Result<()> {
VInt(self.num_vals).serialize(writer)?;
VInt(self.num_vals as u64).serialize(writer)?;
VInt(self.min_value).serialize(writer)?;
VInt(self.max_value - self.min_value).serialize(writer)?;
if let Some(gcd) = self.gcd {
@@ -109,7 +150,7 @@ impl BinarySerializable for Header {
}
fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
let num_vals = VInt::deserialize(reader)?.0;
let num_vals = VInt::deserialize(reader)?.0 as u32;
let min_value = VInt::deserialize(reader)?.0;
let amplitude = VInt::deserialize(reader)?.0;
let max_value = min_value + amplitude;
@@ -125,16 +166,21 @@ impl BinarySerializable for Header {
}
}
/// Return estimated compression for given codec in the value range [0.0..1.0], where 1.0 means no
/// compression.
pub fn estimate<T: MonotonicallyMappableToU64>(
typed_column: impl Column<T>,
codec_type: FastFieldCodecType,
) -> Option<f32> {
let column = monotonic_map_column(typed_column, T::to_u64);
let column = monotonic_map_column(typed_column, StrictlyMonotonicMappingToInternal::<T>::new());
let min_value = column.min_value();
let gcd = crate::gcd::find_gcd(column.iter().map(|val| val - min_value))
.filter(|gcd| gcd.get() > 1u64);
let divider = DividerU64::divide_by(gcd.map(|gcd| gcd.get()).unwrap_or(1u64));
let normalized_column = monotonic_map_column(&column, |val| divider.divide(val - min_value));
let mapping = StrictlyMonotonicMappingToInternalGCDBaseval::new(
gcd.map(|gcd| gcd.get()).unwrap_or(1u64),
min_value,
);
let normalized_column = monotonic_map_column(&column, mapping);
match codec_type {
FastFieldCodecType::Bitpacked => BitpackedCodec::estimate(&normalized_column),
FastFieldCodecType::Linear => LinearCodec::estimate(&normalized_column),
@@ -142,25 +188,38 @@ pub fn estimate<T: MonotonicallyMappableToU64>(
}
}
pub fn serialize_u128(
typed_column: impl Column<u128>,
/// Serializes u128 values with the compact space codec.
pub fn serialize_u128<F: Fn() -> I, I: Iterator<Item = u128>>(
iter_gen: F,
num_vals: u32,
output: &mut impl io::Write,
) -> io::Result<()> {
// TODO write header, to later support more codecs
let compressor = CompactSpaceCompressor::train_from(&typed_column);
compressor
.compress_into(typed_column.iter(), output)
.unwrap();
let header = U128Header {
num_vals,
codec_type: U128FastFieldCodecType::CompactSpace,
};
header.serialize(output)?;
let compressor = CompactSpaceCompressor::train_from(iter_gen(), num_vals);
compressor.compress_into(iter_gen(), output).unwrap();
let null_index_footer = NullIndexFooter {
cardinality: FastFieldCardinality::Single,
null_index_codec: NullIndexCodec::Full,
null_index_byte_range: 0..0,
};
append_null_index_footer(output, null_index_footer)?;
append_format_version(output)?;
Ok(())
}
/// Serializes the column with the codec with the best estimate on the data.
pub fn serialize<T: MonotonicallyMappableToU64>(
typed_column: impl Column<T>,
output: &mut impl io::Write,
codecs: &[FastFieldCodecType],
) -> io::Result<()> {
let column = monotonic_map_column(typed_column, T::to_u64);
let column = monotonic_map_column(typed_column, StrictlyMonotonicMappingToInternal::<T>::new());
let header = Header::compute_header(&column, codecs).ok_or_else(|| {
io::Error::new(
io::ErrorKind::InvalidInput,
@@ -174,6 +233,15 @@ pub fn serialize<T: MonotonicallyMappableToU64>(
let normalized_column = header.normalize_column(column);
assert_eq!(normalized_column.min_value(), 0u64);
serialize_given_codec(normalized_column, header.codec_type, output)?;
let null_index_footer = NullIndexFooter {
cardinality: FastFieldCardinality::Single,
null_index_codec: NullIndexCodec::Full,
null_index_byte_range: 0..0,
};
append_null_index_footer(output, null_index_footer)?;
append_format_version(output)?;
Ok(())
}
@@ -225,6 +293,7 @@ fn serialize_given_codec(
Ok(())
}
/// Helper function to serialize a column (autodetect from all codecs) and then open it
pub fn serialize_and_load<T: MonotonicallyMappableToU64 + Ord + Default>(
column: &[T],
) -> Arc<dyn Column<T>> {
@@ -237,6 +306,18 @@ pub fn serialize_and_load<T: MonotonicallyMappableToU64 + Ord + Default>(
mod tests {
use super::*;
#[test]
fn test_serialize_deserialize_u128_header() {
let original = U128Header {
num_vals: 11,
codec_type: U128FastFieldCodecType::CompactSpace,
};
let mut out = Vec::new();
original.serialize(&mut out).unwrap();
let restored = U128Header::deserialize(&mut &out[..]).unwrap();
assert_eq!(restored, original);
}
#[test]
fn test_serialize_deserialize() {
let original = [1u64, 5u64, 10u64];
@@ -250,7 +331,7 @@ mod tests {
let col = VecColumn::from(&[false, true][..]);
serialize(col, &mut buffer, &ALL_CODEC_TYPES).unwrap();
// 5 bytes of header, 1 byte of value, 7 bytes of padding.
assert_eq!(buffer.len(), 5 + 8);
assert_eq!(buffer.len(), 3 + 5 + 8 + 4 + 2);
}
#[test]
@@ -259,7 +340,7 @@ mod tests {
let col = VecColumn::from(&[true][..]);
serialize(col, &mut buffer, &ALL_CODEC_TYPES).unwrap();
// 5 bytes of header, 0 bytes of value, 7 bytes of padding.
assert_eq!(buffer.len(), 5 + 7);
assert_eq!(buffer.len(), 3 + 5 + 7 + 4 + 2);
}
#[test]
@@ -269,6 +350,6 @@ mod tests {
let col = VecColumn::from(&vals[..]);
serialize(col, &mut buffer, &[FastFieldCodecType::Bitpacked]).unwrap();
// Values are stored over 3 bits.
assert_eq!(buffer.len(), 7 + (3 * 80 / 8) + 7);
assert_eq!(buffer.len(), 3 + 7 + (3 * 80 / 8) + 7 + 4 + 2);
}
}

157
fastfield_codecs/src/sparse_codec_wrapper.rs
View File

@@ -1,157 +0,0 @@
use std::sync::Arc;
use rand::{thread_rng, Rng};
use roaring::RoaringBitmap;
use crate::Column;
pub struct SparseCodecRoaringBitmap {
null: RoaringBitmap,
column: Arc<dyn Column<u64>>, // column: C,
}
impl SparseCodecRoaringBitmap {
pub fn with_full(column: Arc<dyn Column<u64>>) -> Self {
let mut rb = RoaringBitmap::new();
rb.insert_range(0..column.num_vals() as u32 + 1);
Self { null: rb, column }
}
}
impl Column for SparseCodecRoaringBitmap {
fn get_val(&self, idx: u64) -> u64 {
let position_of_val = self.null.rank(idx as u32);
self.column.get_val(position_of_val) // TODO this does not handle null!
// self.null.select(num_vals)
}
fn min_value(&self) -> u64 {
todo!()
}
fn max_value(&self) -> u64 {
todo!()
}
fn num_vals(&self) -> u64 {
todo!()
}
}
pub struct DenseCodec<C> {
// the bitmap blocks of length 64 bit each
blocks: Vec<u64>,
// the offset for each block
offsets: Vec<u32>,
column: C, // column: C,
}
impl<C: Column> DenseCodec<C> {
// fill ratio valid range 0..1000 1000 == all elements, 1 == every 1000th element
pub fn with_fill_ratio(column: C, fill_ratio: u32) -> Self {
let mut rng = thread_rng();
let num_blocks = (column.num_vals() as usize / 64) + 1;
let mut blocks = Vec::with_capacity(num_blocks);
let mut offsets = Vec::with_capacity(num_blocks);
// fill all blocks
let mut offset = 0;
for _block_num in 0..num_blocks {
let mut block = 0;
for n in 0..64 {
if rng.gen_range(0..=1000) <= fill_ratio {
set_bit_at(&mut block, n);
}
}
blocks.push(block);
offsets.push(offset);
offset += block.count_ones();
}
Self {
blocks,
offsets,
column,
}
}
pub fn with_full(column: C) -> Self {
let num_blocks = (column.num_vals() as usize / 64) + 1;
let mut blocks = Vec::with_capacity(num_blocks);
let mut offsets = Vec::with_capacity(num_blocks);
// fill all blocks
let mut offset = 0;
for _block_num in 0..num_blocks {
let block = u64::MAX;
blocks.push(block);
offsets.push(offset);
offset += block.count_ones();
}
Self {
blocks,
offsets,
column,
}
}
}
fn gen_mask(msb: u64) -> u64 {
let src = 1 << msb;
src - 1
}
fn get_bit_at(input: u64, n: u64) -> bool {
input & (1 << n) != 0
}
fn set_bit_at(input: &mut u64, n: u64) {
*input |= 1 << n;
}
impl<C: Column> Column for DenseCodec<C> {
fn get_val(&self, idx: u64) -> u64 {
let block_pos = idx / 64;
let pos_in_block = idx % 64;
let offset = self.offsets[block_pos as usize];
let bitvec = self.blocks[block_pos as usize];
let offset_in_block = (bitvec & gen_mask(pos_in_block)).count_ones();
let dense_idx = offset as u64 + offset_in_block as u64;
if get_bit_at(bitvec, pos_in_block) {
self.column.get_val(dense_idx)
} else {
0 // TODO null
}
}
fn min_value(&self) -> u64 {
todo!()
}
fn max_value(&self) -> u64 {
todo!()
}
fn num_vals(&self) -> u64 {
todo!()
}
}
#[cfg(test)]
mod tests {
//use itertools::Itertools;
//use super::*;
//use crate::serialize_and_load;
//#[test]
//fn dense_test() {
//let data = (0..100u64).collect_vec();
//{
//let column = serialize_and_load(&data);
//let dense = DenseCodec::with_full(column);
//for i in 0..100 {
//dense.get_val(i);
//}
//}
//}
}

6
ownedbytes/Cargo.toml
View File

@@ -1,10 +1,14 @@
[package]
authors = ["Paul Masurel <paul@quickwit.io>", "Pascal Seitz <pascal@quickwit.io>"]
name = "ownedbytes"
version = "0.3.0"
version = "0.4.0"
edition = "2021"
description = "Expose data as static slice"
license = "MIT"
documentation = "https://docs.rs/ownedbytes/"
homepage = "https://github.com/quickwit-oss/tantivy"
repository = "https://github.com/quickwit-oss/tantivy"
# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
[dependencies]

15
ownedbytes/src/lib.rs
View File

@@ -80,6 +80,21 @@ impl OwnedBytes {
(left, right)
}
/// Splits the OwnedBytes into two OwnedBytes `(left, right)`.
///
/// Right will hold `split_len` bytes.
///
/// This operation is cheap and does not require to copy any memory.
/// 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 rsplit(self, split_len: usize) -> (OwnedBytes, OwnedBytes) {
let data_len = self.data.len();
self.split(data_len - split_len)
}
/// Splits the right part of the `OwnedBytes` at the given offset.
///
/// `self` is truncated to `split_len`, left with the remaining bytes.

2
query-grammar/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "tantivy-query-grammar"
version = "0.18.0"
version = "0.19.0"
authors = ["Paul Masurel <paul.masurel@gmail.com>"]
license = "MIT"
categories = ["database-implementations", "data-structures"]

67
query-grammar/src/query_grammar.rs
View File

@@ -5,7 +5,8 @@ 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,
attempt, between, choice, eof, many, many1, one_of, optional, parser, satisfy, sep_by,
skip_many1, value,
};
use once_cell::sync::Lazy;
use regex::Regex;
@@ -62,6 +63,20 @@ fn word<'a>() -> impl Parser<&'a str, Output = String> {
})
}
// word variant that allows more characters, e.g. for range queries that don't allow field
// specifier
fn relaxed_word<'a>() -> impl Parser<&'a str, Output = String> {
(
satisfy(|c: char| {
!c.is_whitespace() && !['`', '{', '}', '"', '[', ']', '(', ')'].contains(&c)
}),
many(satisfy(|c: char| {
!c.is_whitespace() && !['{', '}', '"', '[', ']', '(', ')'].contains(&c)
})),
)
.map(|(s1, s2): (char, String)| format!("{}{}", s1, s2))
}
/// Parses a date time according to rfc3339
/// 2015-08-02T18:54:42+02
/// 2021-04-13T19:46:26.266051969+00:00
@@ -181,8 +196,8 @@ fn spaces1<'a>() -> impl Parser<&'a str, Output = ()> {
fn range<'a>() -> impl Parser<&'a str, Output = UserInputLeaf> {
let range_term_val = || {
attempt(date_time())
.or(word())
.or(negative_number())
.or(relaxed_word())
.or(char('*').with(value("*".to_string())))
};
@@ -250,6 +265,17 @@ fn range<'a>() -> impl Parser<&'a str, Output = UserInputLeaf> {
})
}
/// Function that parses a set out of a Stream
/// Supports ranges like: `IN [val1 val2 val3]`
fn set<'a>() -> impl Parser<&'a str, Output = UserInputLeaf> {
let term_list = between(char('['), char(']'), sep_by(term_val(), spaces()));
let set_content = ((string("IN"), spaces()), term_list).map(|(_, elements)| elements);
(optional(attempt(field_name().skip(spaces()))), set_content)
.map(|(field, elements)| UserInputLeaf::Set { field, elements })
}
fn negate(expr: UserInputAst) -> UserInputAst {
expr.unary(Occur::MustNot)
}
@@ -264,6 +290,7 @@ fn leaf<'a>() -> impl Parser<&'a str, Output = UserInputAst> {
string("NOT").skip(spaces1()).with(leaf()).map(negate),
))
.or(attempt(range().map(UserInputAst::from)))
.or(attempt(set().map(UserInputAst::from)))
.or(literal().map(UserInputAst::from))
.parse_stream(input)
.into_result()
@@ -649,6 +676,34 @@ mod test {
.expect("Cannot parse date range")
.0;
assert_eq!(res6, expected_flexible_dates);
// IP Range Unbounded
let expected_weight = UserInputLeaf::Range {
field: Some("ip".to_string()),
lower: UserInputBound::Inclusive("::1".to_string()),
upper: UserInputBound::Unbounded,
};
let res1 = range()
.parse("ip: >=::1")
.expect("Cannot parse ip v6 format")
.0;
let res2 = range()
.parse("ip:[::1 TO *}")
.expect("Cannot parse ip v6 format")
.0;
assert_eq!(res1, expected_weight);
assert_eq!(res2, expected_weight);
// IP Range Bounded
let expected_weight = UserInputLeaf::Range {
field: Some("ip".to_string()),
lower: UserInputBound::Inclusive("::0.0.0.50".to_string()),
upper: UserInputBound::Exclusive("::0.0.0.52".to_string()),
};
let res1 = range()
.parse("ip:[::0.0.0.50 TO ::0.0.0.52}")
.expect("Cannot parse ip v6 format")
.0;
assert_eq!(res1, expected_weight);
}
#[test]
@@ -705,6 +760,14 @@ mod test {
test_parse_query_to_ast_helper("+(a b) +d", "(+(*\"a\" *\"b\") +\"d\")");
}
#[test]
fn test_parse_test_query_set() {
test_parse_query_to_ast_helper("abc: IN [a b c]", r#""abc": IN ["a" "b" "c"]"#);
test_parse_query_to_ast_helper("abc: IN [1]", r#""abc": IN ["1"]"#);
test_parse_query_to_ast_helper("abc: IN []", r#""abc": IN []"#);
test_parse_query_to_ast_helper("IN [1 2]", r#"IN ["1" "2"]"#);
}
#[test]
fn test_parse_test_query_other() {
test_parse_query_to_ast_helper("(+a +b) d", "(*(+\"a\" +\"b\") *\"d\")");

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

@@ -12,6 +12,10 @@ pub enum UserInputLeaf {
lower: UserInputBound,
upper: UserInputBound,
},
Set {
field: Option<String>,
elements: Vec<String>,
},
}
impl Debug for UserInputLeaf {
@@ -31,6 +35,19 @@ impl Debug for UserInputLeaf {
upper.display_upper(formatter)?;
Ok(())
}
UserInputLeaf::Set { field, elements } => {
if let Some(ref field) = field {
write!(formatter, "\"{}\": ", field)?;
}
write!(formatter, "IN [")?;
for (i, element) in elements.iter().enumerate() {
if i != 0 {
write!(formatter, " ")?;
}
write!(formatter, "\"{}\"", element)?;
}
write!(formatter, "]")
}
UserInputLeaf::All => write!(formatter, "*"),
}
}

10
src/aggregation/agg_req_with_accessor.rs
View File

@@ -11,7 +11,7 @@ use super::bucket::{HistogramAggregation, RangeAggregation, TermsAggregation};
use super::metric::{AverageAggregation, StatsAggregation};
use super::segment_agg_result::BucketCount;
use super::VecWithNames;
use crate::fastfield::{type_and_cardinality, FastType, MultiValuedFastFieldReader};
use crate::fastfield::{type_and_cardinality, MultiValuedFastFieldReader};
use crate::schema::{Cardinality, Type};
use crate::{InvertedIndexReader, SegmentReader, TantivyError};
@@ -194,13 +194,7 @@ fn get_ff_reader_and_validate(
.ok_or_else(|| TantivyError::FieldNotFound(field_name.to_string()))?;
let field_type = reader.schema().get_field_entry(field).field_type();
if let Some((ff_type, field_cardinality)) = type_and_cardinality(field_type) {
if ff_type == FastType::Date {
return Err(TantivyError::InvalidArgument(
"Unsupported field type date in aggregation".to_string(),
));
}
if let Some((_ff_type, field_cardinality)) = type_and_cardinality(field_type) {
if cardinality != field_cardinality {
return Err(TantivyError::InvalidArgument(format!(
"Invalid field cardinality on field {} expected {:?}, but got {:?}",

25
src/aggregation/agg_result.rs
View File

@@ -4,9 +4,7 @@
//! intermediate average results, which is the sum and the number of values. The actual average is
//! calculated on the step from intermediate to final aggregation result tree.
use std::collections::HashMap;
use fnv::FnvHashMap;
use rustc_hash::FxHashMap;
use serde::{Deserialize, Serialize};
use super::agg_req::BucketAggregationInternal;
@@ -14,11 +12,12 @@ use super::bucket::GetDocCount;
use super::intermediate_agg_result::{IntermediateBucketResult, IntermediateMetricResult};
use super::metric::{SingleMetricResult, Stats};
use super::Key;
use crate::schema::Schema;
use crate::TantivyError;
#[derive(Clone, Default, Debug, PartialEq, Serialize, Deserialize)]
/// The final aggegation result.
pub struct AggregationResults(pub HashMap<String, AggregationResult>);
pub struct AggregationResults(pub FxHashMap<String, AggregationResult>);
impl AggregationResults {
pub(crate) fn get_value_from_aggregation(
@@ -131,9 +130,12 @@ pub enum BucketResult {
}
impl BucketResult {
pub(crate) fn empty_from_req(req: &BucketAggregationInternal) -> crate::Result<Self> {
pub(crate) fn empty_from_req(
req: &BucketAggregationInternal,
schema: &Schema,
) -> crate::Result<Self> {
let empty_bucket = IntermediateBucketResult::empty_from_req(&req.bucket_agg);
empty_bucket.into_final_bucket_result(req)
empty_bucket.into_final_bucket_result(req, schema)
}
}
@@ -145,7 +147,7 @@ pub enum BucketEntries<T> {
/// Vector format bucket entries
Vec(Vec<T>),
/// HashMap format bucket entries
HashMap(FnvHashMap<String, T>),
HashMap(FxHashMap<String, T>),
}
/// This is the default entry for a bucket, which contains a key, count, and optionally
@@ -176,6 +178,9 @@ pub enum BucketEntries<T> {
/// ```
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize)]
pub struct BucketEntry {
#[serde(skip_serializing_if = "Option::is_none")]
/// The string representation of the bucket.
pub key_as_string: Option<String>,
/// The identifier of the bucket.
pub key: Key,
/// Number of documents in the bucket.
@@ -240,4 +245,10 @@ pub struct RangeBucketEntry {
/// The to range of the bucket. Equals `f64::MAX` when `None`.
#[serde(skip_serializing_if = "Option::is_none")]
pub to: Option<f64>,
/// The optional string representation for the `from` range.
#[serde(skip_serializing_if = "Option::is_none")]
pub from_as_string: Option<String>,
/// The optional string representation for the `to` range.
#[serde(skip_serializing_if = "Option::is_none")]
pub to_as_string: Option<String>,
}

172
src/aggregation/bucket/histogram/histogram.rs
View File

@@ -10,12 +10,12 @@ use crate::aggregation::agg_req_with_accessor::{
AggregationsWithAccessor, BucketAggregationWithAccessor,
};
use crate::aggregation::agg_result::BucketEntry;
use crate::aggregation::f64_from_fastfield_u64;
use crate::aggregation::intermediate_agg_result::{
IntermediateAggregationResults, IntermediateBucketResult, IntermediateHistogramBucketEntry,
};
use crate::aggregation::segment_agg_result::SegmentAggregationResultsCollector;
use crate::schema::Type;
use crate::aggregation::{f64_from_fastfield_u64, format_date};
use crate::schema::{Schema, Type};
use crate::{DocId, TantivyError};
/// Histogram is a bucket aggregation, where buckets are created dynamically for given `interval`.
@@ -206,6 +206,7 @@ pub struct SegmentHistogramCollector {
field_type: Type,
interval: f64,
offset: f64,
min_doc_count: u64,
first_bucket_num: i64,
bounds: HistogramBounds,
}
@@ -215,6 +216,30 @@ impl SegmentHistogramCollector {
self,
agg_with_accessor: &BucketAggregationWithAccessor,
) -> crate::Result<IntermediateBucketResult> {
// Compute the number of buckets to validate against max num buckets
// Note: We use min_doc_count here, but it's only an lowerbound here, since were are on the
// intermediate level and after merging the number of documents of a bucket could exceed
// `min_doc_count`.
{
let cut_off_buckets_front = self
.buckets
.iter()
.take_while(|bucket| bucket.doc_count <= self.min_doc_count)
.count();
let cut_off_buckets_back = self.buckets[cut_off_buckets_front..]
.iter()
.rev()
.take_while(|bucket| bucket.doc_count <= self.min_doc_count)
.count();
let estimate_num_buckets =
self.buckets.len() - cut_off_buckets_front - cut_off_buckets_back;
agg_with_accessor
.bucket_count
.add_count(estimate_num_buckets as u32);
agg_with_accessor.bucket_count.validate_bucket_count()?;
}
let mut buckets = Vec::with_capacity(
self.buckets
.iter()
@@ -251,11 +276,6 @@ impl SegmentHistogramCollector {
);
};
agg_with_accessor
.bucket_count
.add_count(buckets.len() as u32);
agg_with_accessor.bucket_count.validate_bucket_count()?;
Ok(IntermediateBucketResult::Histogram { buckets })
}
@@ -308,6 +328,7 @@ impl SegmentHistogramCollector {
first_bucket_num,
bounds,
sub_aggregations,
min_doc_count: req.min_doc_count(),
})
}
@@ -331,10 +352,10 @@ impl SegmentHistogramCollector {
.expect("unexpected fast field cardinatility");
let mut iter = doc.chunks_exact(4);
for docs in iter.by_ref() {
let val0 = self.f64_from_fastfield_u64(accessor.get_val(docs[0] as u64));
let val1 = self.f64_from_fastfield_u64(accessor.get_val(docs[1] as u64));
let val2 = self.f64_from_fastfield_u64(accessor.get_val(docs[2] as u64));
let val3 = self.f64_from_fastfield_u64(accessor.get_val(docs[3] as u64));
let val0 = self.f64_from_fastfield_u64(accessor.get_val(docs[0]));
let val1 = self.f64_from_fastfield_u64(accessor.get_val(docs[1]));
let val2 = self.f64_from_fastfield_u64(accessor.get_val(docs[2]));
let val3 = self.f64_from_fastfield_u64(accessor.get_val(docs[3]));
let bucket_pos0 = get_bucket_num(val0);
let bucket_pos1 = get_bucket_num(val1);
@@ -371,7 +392,7 @@ impl SegmentHistogramCollector {
)?;
}
for &doc in iter.remainder() {
let val = f64_from_fastfield_u64(accessor.get_val(doc as u64), &self.field_type);
let val = f64_from_fastfield_u64(accessor.get_val(doc), &self.field_type);
if !bounds.contains(val) {
continue;
}
@@ -451,8 +472,9 @@ fn intermediate_buckets_to_final_buckets_fill_gaps(
buckets: Vec<IntermediateHistogramBucketEntry>,
histogram_req: &HistogramAggregation,
sub_aggregation: &AggregationsInternal,
schema: &Schema,
) -> crate::Result<Vec<BucketEntry>> {
// Generate the the full list of buckets without gaps.
// Generate the full list of buckets without gaps.
//
// The bounds are the min max from the current buckets, optionally extended by
// extended_bounds from the request
@@ -491,7 +513,9 @@ fn intermediate_buckets_to_final_buckets_fill_gaps(
sub_aggregation: empty_sub_aggregation.clone(),
},
})
.map(|intermediate_bucket| intermediate_bucket.into_final_bucket_entry(sub_aggregation))
.map(|intermediate_bucket| {
intermediate_bucket.into_final_bucket_entry(sub_aggregation, schema)
})
.collect::<crate::Result<Vec<_>>>()
}
@@ -500,20 +524,43 @@ pub(crate) fn intermediate_histogram_buckets_to_final_buckets(
buckets: Vec<IntermediateHistogramBucketEntry>,
histogram_req: &HistogramAggregation,
sub_aggregation: &AggregationsInternal,
schema: &Schema,
) -> crate::Result<Vec<BucketEntry>> {
if histogram_req.min_doc_count() == 0 {
let mut buckets = if histogram_req.min_doc_count() == 0 {
// With min_doc_count != 0, we may need to add buckets, so that there are no
// gaps, since intermediate result does not contain empty buckets (filtered to
// reduce serialization size).
intermediate_buckets_to_final_buckets_fill_gaps(buckets, histogram_req, sub_aggregation)
intermediate_buckets_to_final_buckets_fill_gaps(
buckets,
histogram_req,
sub_aggregation,
schema,
)?
} else {
buckets
.into_iter()
.filter(|histogram_bucket| histogram_bucket.doc_count >= histogram_req.min_doc_count())
.map(|histogram_bucket| histogram_bucket.into_final_bucket_entry(sub_aggregation))
.collect::<crate::Result<Vec<_>>>()
.map(|histogram_bucket| {
histogram_bucket.into_final_bucket_entry(sub_aggregation, schema)
})
.collect::<crate::Result<Vec<_>>>()?
};
// If we have a date type on the histogram buckets, we add the `key_as_string` field as rfc339
let field = schema
.get_field(&histogram_req.field)
.ok_or_else(|| TantivyError::FieldNotFound(histogram_req.field.to_string()))?;
if schema.get_field_entry(field).field_type().is_date() {
for bucket in buckets.iter_mut() {
if let crate::aggregation::Key::F64(val) = bucket.key {
let key_as_string = format_date(val as i64)?;
bucket.key_as_string = Some(key_as_string);
}
}
}
Ok(buckets)
}
/// Applies req extended_bounds/hard_bounds on the min_max value
@@ -1372,6 +1419,63 @@ mod tests {
Ok(())
}
#[test]
fn histogram_date_test_single_segment() -> crate::Result<()> {
histogram_date_test_with_opt(true)
}
#[test]
fn histogram_date_test_multi_segment() -> crate::Result<()> {
histogram_date_test_with_opt(false)
}
fn histogram_date_test_with_opt(merge_segments: bool) -> crate::Result<()> {
let index = get_test_index_2_segments(merge_segments)?;
let agg_req: Aggregations = vec![(
"histogram".to_string(),
Aggregation::Bucket(BucketAggregation {
bucket_agg: BucketAggregationType::Histogram(HistogramAggregation {
field: "date".to_string(),
interval: 86400000000.0, // one day in microseconds
..Default::default()
}),
sub_aggregation: Default::default(),
}),
)]
.into_iter()
.collect();
let agg_res = exec_request(agg_req, &index)?;
let res: Value = serde_json::from_str(&serde_json::to_string(&agg_res)?)?;
assert_eq!(res["histogram"]["buckets"][0]["key"], 1546300800000000.0);
assert_eq!(
res["histogram"]["buckets"][0]["key_as_string"],
"2019-01-01T00:00:00Z"
);
assert_eq!(res["histogram"]["buckets"][0]["doc_count"], 1);
assert_eq!(res["histogram"]["buckets"][1]["key"], 1546387200000000.0);
assert_eq!(
res["histogram"]["buckets"][1]["key_as_string"],
"2019-01-02T00:00:00Z"
);
assert_eq!(res["histogram"]["buckets"][1]["doc_count"], 5);
assert_eq!(res["histogram"]["buckets"][2]["key"], 1546473600000000.0);
assert_eq!(
res["histogram"]["buckets"][2]["key_as_string"],
"2019-01-03T00:00:00Z"
);
assert_eq!(res["histogram"]["buckets"][3], Value::Null);
Ok(())
}
#[test]
fn histogram_invalid_request() -> crate::Result<()> {
let index = get_test_index_2_segments(true)?;
@@ -1438,4 +1542,36 @@ mod tests {
Ok(())
}
#[test]
fn histogram_test_max_buckets_segments() -> crate::Result<()> {
let values = vec![0.0, 70000.0];
let index = get_test_index_from_values(true, &values)?;
let agg_req: Aggregations = vec![(
"my_interval".to_string(),
Aggregation::Bucket(BucketAggregation {
bucket_agg: BucketAggregationType::Histogram(HistogramAggregation {
field: "score_f64".to_string(),
interval: 1.0,
..Default::default()
}),
sub_aggregation: Default::default(),
}),
)]
.into_iter()
.collect();
let res = exec_request(agg_req, &index);
assert_eq!(
res.unwrap_err().to_string(),
"An invalid argument was passed: 'Aborting aggregation because too many buckets were \
created'"
.to_string()
);
Ok(())
}
}

126
src/aggregation/bucket/range.rs
View File

@@ -1,7 +1,8 @@
use std::fmt::Debug;
use std::ops::Range;
use fnv::FnvHashMap;
use fastfield_codecs::MonotonicallyMappableToU64;
use rustc_hash::FxHashMap;
use serde::{Deserialize, Serialize};
use crate::aggregation::agg_req_with_accessor::{
@@ -11,7 +12,9 @@ use crate::aggregation::intermediate_agg_result::{
IntermediateBucketResult, IntermediateRangeBucketEntry, IntermediateRangeBucketResult,
};
use crate::aggregation::segment_agg_result::{BucketCount, SegmentAggregationResultsCollector};
use crate::aggregation::{f64_from_fastfield_u64, f64_to_fastfield_u64, Key, SerializedKey};
use crate::aggregation::{
f64_from_fastfield_u64, f64_to_fastfield_u64, format_date, Key, SerializedKey,
};
use crate::schema::Type;
use crate::{DocId, TantivyError};
@@ -176,12 +179,12 @@ impl SegmentRangeCollector {
) -> crate::Result<IntermediateBucketResult> {
let field_type = self.field_type;
let buckets: FnvHashMap<SerializedKey, IntermediateRangeBucketEntry> = self
let buckets: FxHashMap<SerializedKey, IntermediateRangeBucketEntry> = self
.buckets
.into_iter()
.map(move |range_bucket| {
Ok((
range_to_string(&range_bucket.range, &field_type),
range_to_string(&range_bucket.range, &field_type)?,
range_bucket
.bucket
.into_intermediate_bucket_entry(&agg_with_accessor.sub_aggregation)?,
@@ -209,8 +212,8 @@ impl SegmentRangeCollector {
let key = range
.key
.clone()
.map(Key::Str)
.unwrap_or_else(|| range_to_key(&range.range, &field_type));
.map(|key| Ok(Key::Str(key)))
.unwrap_or_else(|| range_to_key(&range.range, &field_type))?;
let to = if range.range.end == u64::MAX {
None
} else {
@@ -228,6 +231,7 @@ impl SegmentRangeCollector {
sub_aggregation,
)?)
};
Ok(SegmentRangeAndBucketEntry {
range: range.range.clone(),
bucket: SegmentRangeBucketEntry {
@@ -263,10 +267,10 @@ impl SegmentRangeCollector {
.as_single()
.expect("unexpected fast field cardinality");
for docs in iter.by_ref() {
let val1 = accessor.get_val(docs[0] as u64);
let val2 = accessor.get_val(docs[1] as u64);
let val3 = accessor.get_val(docs[2] as u64);
let val4 = accessor.get_val(docs[3] as u64);
let val1 = accessor.get_val(docs[0]);
let val2 = accessor.get_val(docs[1]);
let val3 = accessor.get_val(docs[2]);
let val4 = accessor.get_val(docs[3]);
let bucket_pos1 = self.get_bucket_pos(val1);
let bucket_pos2 = self.get_bucket_pos(val2);
let bucket_pos3 = self.get_bucket_pos(val3);
@@ -278,7 +282,7 @@ impl SegmentRangeCollector {
self.increment_bucket(bucket_pos4, docs[3], &bucket_with_accessor.sub_aggregation)?;
}
for &doc in iter.remainder() {
let val = accessor.get_val(doc as u64);
let val = accessor.get_val(doc);
let bucket_pos = self.get_bucket_pos(val);
self.increment_bucket(bucket_pos, doc, &bucket_with_accessor.sub_aggregation)?;
}
@@ -323,8 +327,8 @@ impl SegmentRangeCollector {
/// Converts the user provided f64 range value to fast field value space.
///
/// Internally fast field values are always stored as u64.
/// If the fast field has u64 [1,2,5], these values are stored as is in the fast field.
/// A fast field with f64 [1.0, 2.0, 5.0] is converted to u64 space, using a
/// If the fast field has u64 `[1, 2, 5]`, these values are stored as is in the fast field.
/// A fast field with f64 `[1.0, 2.0, 5.0]` is converted to u64 space, using a
/// monotonic mapping function, so the order is preserved.
///
/// Consequently, a f64 user range 1.0..3.0 needs to be converted to fast field value space using
@@ -402,34 +406,45 @@ fn extend_validate_ranges(
Ok(converted_buckets)
}
pub(crate) fn range_to_string(range: &Range<u64>, field_type: &Type) -> String {
pub(crate) fn range_to_string(range: &Range<u64>, field_type: &Type) -> crate::Result<String> {
// is_start is there for malformed requests, e.g. ig the user passes the range u64::MIN..0.0,
// it should be rendered as "*-0" and not "*-*"
let to_str = |val: u64, is_start: bool| {
if (is_start && val == u64::MIN) || (!is_start && val == u64::MAX) {
"*".to_string()
Ok("*".to_string())
} else if *field_type == Type::Date {
let val = i64::from_u64(val);
format_date(val)
} else {
f64_from_fastfield_u64(val, field_type).to_string()
Ok(f64_from_fastfield_u64(val, field_type).to_string())
}
};
format!("{}-{}", to_str(range.start, true), to_str(range.end, false))
Ok(format!(
"{}-{}",
to_str(range.start, true)?,
to_str(range.end, false)?
))
}
pub(crate) fn range_to_key(range: &Range<u64>, field_type: &Type) -> Key {
Key::Str(range_to_string(range, field_type))
pub(crate) fn range_to_key(range: &Range<u64>, field_type: &Type) -> crate::Result<Key> {
Ok(Key::Str(range_to_string(range, field_type)?))
}
#[cfg(test)]
mod tests {
use fastfield_codecs::MonotonicallyMappableToU64;
use serde_json::Value;
use super::*;
use crate::aggregation::agg_req::{
Aggregation, Aggregations, BucketAggregation, BucketAggregationType,
};
use crate::aggregation::tests::{exec_request_with_query, get_test_index_with_num_docs};
use crate::aggregation::tests::{
exec_request, exec_request_with_query, get_test_index_2_segments,
get_test_index_with_num_docs,
};
pub fn get_collector_from_ranges(
ranges: Vec<RangeAggregationRange>,
@@ -567,6 +582,77 @@ mod tests {
Ok(())
}
#[test]
fn range_date_test_single_segment() -> crate::Result<()> {
range_date_test_with_opt(true)
}
#[test]
fn range_date_test_multi_segment() -> crate::Result<()> {
range_date_test_with_opt(false)
}
fn range_date_test_with_opt(merge_segments: bool) -> crate::Result<()> {
let index = get_test_index_2_segments(merge_segments)?;
let agg_req: Aggregations = vec![(
"date_ranges".to_string(),
Aggregation::Bucket(BucketAggregation {
bucket_agg: BucketAggregationType::Range(RangeAggregation {
field: "date".to_string(),
ranges: vec![
RangeAggregationRange {
key: None,
from: None,
to: Some(1546300800000000.0f64),
},
RangeAggregationRange {
key: None,
from: Some(1546300800000000.0f64),
to: Some(1546387200000000.0f64),
},
],
keyed: false,
}),
sub_aggregation: Default::default(),
}),
)]
.into_iter()
.collect();
let agg_res = exec_request(agg_req, &index)?;
let res: Value = serde_json::from_str(&serde_json::to_string(&agg_res)?)?;
assert_eq!(
res["date_ranges"]["buckets"][0]["from_as_string"],
Value::Null
);
assert_eq!(
res["date_ranges"]["buckets"][0]["key"],
"*-2019-01-01T00:00:00Z"
);
assert_eq!(
res["date_ranges"]["buckets"][1]["from_as_string"],
"2019-01-01T00:00:00Z"
);
assert_eq!(
res["date_ranges"]["buckets"][1]["to_as_string"],
"2019-01-02T00:00:00Z"
);
assert_eq!(
res["date_ranges"]["buckets"][2]["from_as_string"],
"2019-01-02T00:00:00Z"
);
assert_eq!(
res["date_ranges"]["buckets"][2]["to_as_string"],
Value::Null
);
Ok(())
}
#[test]
fn range_custom_key_keyed_buckets_test() -> crate::Result<()> {
let index = get_test_index_with_num_docs(false, 100)?;

67
src/aggregation/bucket/term_agg.rs
View File

@@ -1,7 +1,7 @@
use std::fmt::Debug;
use fnv::FnvHashMap;
use itertools::Itertools;
use rustc_hash::FxHashMap;
use serde::{Deserialize, Serialize};
use super::{CustomOrder, Order, OrderTarget};
@@ -17,7 +17,11 @@ use crate::fastfield::MultiValuedFastFieldReader;
use crate::schema::Type;
use crate::{DocId, TantivyError};
/// Creates a bucket for every unique term
/// Creates a bucket for every unique term and counts the number of occurences.
/// Note that doc_count in the response buckets equals term count here.
///
/// If the text is untokenized and single value, that means one term per document and therefore it
/// is in fact doc count.
///
/// ### Terminology
/// Shard parameters are supposed to be equivalent to elasticsearch shard parameter.
@@ -64,6 +68,25 @@ use crate::{DocId, TantivyError};
/// }
/// }
/// ```
///
/// /// # Response JSON Format
/// ```json
/// {
/// ...
/// "aggregations": {
/// "genres": {
/// "doc_count_error_upper_bound": 0,
/// "sum_other_doc_count": 0,
/// "buckets": [
/// { "key": "drumnbass", "doc_count": 6 },
/// { "key": "raggae", "doc_count": 4 },
/// { "key": "jazz", "doc_count": 2 }
/// ]
/// }
/// }
/// }
/// ```
#[derive(Clone, Debug, Default, PartialEq, Serialize, Deserialize)]
pub struct TermsAggregation {
/// The field to aggregate on.
@@ -176,7 +199,7 @@ impl TermsAggregationInternal {
#[derive(Clone, Debug, PartialEq)]
/// Container to store term_ids and their buckets.
struct TermBuckets {
pub(crate) entries: FnvHashMap<u32, TermBucketEntry>,
pub(crate) entries: FxHashMap<u32, TermBucketEntry>,
blueprint: Option<SegmentAggregationResultsCollector>,
}
@@ -374,7 +397,7 @@ impl SegmentTermCollector {
.expect("internal error: inverted index not loaded for term aggregation");
let term_dict = inverted_index.terms();
let mut dict: FnvHashMap<String, IntermediateTermBucketEntry> = Default::default();
let mut dict: FxHashMap<String, IntermediateTermBucketEntry> = Default::default();
let mut buffer = vec![];
for (term_id, entry) in entries {
term_dict
@@ -1106,9 +1129,9 @@ mod tests {
assert_eq!(res["my_texts"]["buckets"][0]["key"], "terma");
assert_eq!(res["my_texts"]["buckets"][0]["doc_count"], 4);
assert_eq!(res["my_texts"]["buckets"][1]["key"], "termb");
assert_eq!(res["my_texts"]["buckets"][1]["key"], "termc");
assert_eq!(res["my_texts"]["buckets"][1]["doc_count"], 0);
assert_eq!(res["my_texts"]["buckets"][2]["key"], "termc");
assert_eq!(res["my_texts"]["buckets"][2]["key"], "termb");
assert_eq!(res["my_texts"]["buckets"][2]["doc_count"], 0);
assert_eq!(res["my_texts"]["sum_other_doc_count"], 0);
assert_eq!(res["my_texts"]["doc_count_error_upper_bound"], 0);
@@ -1206,11 +1229,43 @@ mod tests {
.collect();
let res = exec_request_with_query(agg_req, &index, None);
assert!(res.is_err());
Ok(())
}
#[test]
fn terms_aggregation_multi_token_per_doc() -> crate::Result<()> {
let terms = vec!["Hello Hello", "Hallo Hallo"];
let index = get_test_index_from_terms(true, &[terms])?;
let agg_req: Aggregations = vec![(
"my_texts".to_string(),
Aggregation::Bucket(BucketAggregation {
bucket_agg: BucketAggregationType::Terms(TermsAggregation {
field: "text_id".to_string(),
min_doc_count: Some(0),
..Default::default()
}),
sub_aggregation: Default::default(),
}),
)]
.into_iter()
.collect();
let res = exec_request_with_query(agg_req, &index, None).unwrap();
assert_eq!(res["my_texts"]["buckets"][0]["key"], "hello");
assert_eq!(res["my_texts"]["buckets"][0]["doc_count"], 2);
assert_eq!(res["my_texts"]["buckets"][1]["key"], "hallo");
assert_eq!(res["my_texts"]["buckets"][1]["doc_count"], 2);
Ok(())
}
#[test]
fn test_json_format() -> crate::Result<()> {
let agg_req: Aggregations = vec![(

7
src/aggregation/collector.rs
View File

@@ -7,6 +7,7 @@ use super::intermediate_agg_result::IntermediateAggregationResults;
use super::segment_agg_result::SegmentAggregationResultsCollector;
use crate::aggregation::agg_req_with_accessor::get_aggs_with_accessor_and_validate;
use crate::collector::{Collector, SegmentCollector};
use crate::schema::Schema;
use crate::{SegmentReader, TantivyError};
/// The default max bucket count, before the aggregation fails.
@@ -16,6 +17,7 @@ pub const MAX_BUCKET_COUNT: u32 = 65000;
///
/// The collector collects all aggregations by the underlying aggregation request.
pub struct AggregationCollector {
schema: Schema,
agg: Aggregations,
max_bucket_count: u32,
}
@@ -25,8 +27,9 @@ impl AggregationCollector {
///
/// Aggregation fails when the total bucket count is higher than max_bucket_count.
/// max_bucket_count will default to `MAX_BUCKET_COUNT` (65000) when unset
pub fn from_aggs(agg: Aggregations, max_bucket_count: Option<u32>) -> Self {
pub fn from_aggs(agg: Aggregations, max_bucket_count: Option<u32>, schema: Schema) -> Self {
Self {
schema,
agg,
max_bucket_count: max_bucket_count.unwrap_or(MAX_BUCKET_COUNT),
}
@@ -113,7 +116,7 @@ impl Collector for AggregationCollector {
segment_fruits: Vec<<Self::Child as SegmentCollector>::Fruit>,
) -> crate::Result<Self::Fruit> {
let res = merge_fruits(segment_fruits)?;
res.into_final_bucket_result(self.agg.clone())
res.into_final_bucket_result(self.agg.clone(), &self.schema)
}
}

18
src/aggregation/date.rs Normal file
View File

@@ -0,0 +1,18 @@
use time::format_description::well_known::Rfc3339;
use time::OffsetDateTime;
use crate::TantivyError;
pub(crate) fn format_date(val: i64) -> crate::Result<String> {
let datetime =
OffsetDateTime::from_unix_timestamp_nanos(1_000 * (val as i128)).map_err(|err| {
TantivyError::InvalidArgument(format!(
"Could not convert {:?} to OffsetDateTime, err {:?}",
val, err
))
})?;
let key_as_string = datetime
.format(&Rfc3339)
.map_err(|_err| TantivyError::InvalidArgument("Could not serialize date".to_string()))?;
Ok(key_as_string)
}

103
src/aggregation/intermediate_agg_result.rs
View File

@@ -3,15 +3,14 @@
//! indices.
use std::cmp::Ordering;
use std::collections::HashMap;
use fnv::FnvHashMap;
use itertools::Itertools;
use rustc_hash::FxHashMap;
use serde::{Deserialize, Serialize};
use super::agg_req::{
Aggregations, AggregationsInternal, BucketAggregationInternal, BucketAggregationType,
MetricAggregation,
MetricAggregation, RangeAggregation,
};
use super::agg_result::{AggregationResult, BucketResult, RangeBucketEntry};
use super::bucket::{
@@ -20,9 +19,11 @@ use super::bucket::{
};
use super::metric::{IntermediateAverage, IntermediateStats};
use super::segment_agg_result::SegmentMetricResultCollector;
use super::{Key, SerializedKey, VecWithNames};
use super::{format_date, Key, SerializedKey, VecWithNames};
use crate::aggregation::agg_result::{AggregationResults, BucketEntries, BucketEntry};
use crate::aggregation::bucket::TermsAggregationInternal;
use crate::schema::Schema;
use crate::TantivyError;
/// Contains the intermediate aggregation result, which is optimized to be merged with other
/// intermediate results.
@@ -36,8 +37,12 @@ pub struct IntermediateAggregationResults {
impl IntermediateAggregationResults {
/// Convert intermediate result and its aggregation request to the final result.
pub fn into_final_bucket_result(self, req: Aggregations) -> crate::Result<AggregationResults> {
self.into_final_bucket_result_internal(&(req.into()))
pub fn into_final_bucket_result(
self,
req: Aggregations,
schema: &Schema,
) -> crate::Result<AggregationResults> {
self.into_final_bucket_result_internal(&(req.into()), schema)
}
/// Convert intermediate result and its aggregation request to the final result.
@@ -47,18 +52,19 @@ impl IntermediateAggregationResults {
pub(crate) fn into_final_bucket_result_internal(
self,
req: &AggregationsInternal,
schema: &Schema,
) -> crate::Result<AggregationResults> {
// Important assumption:
// When the tree contains buckets/metric, we expect it to have all buckets/metrics from the
// request
let mut results: HashMap<String, AggregationResult> = HashMap::new();
let mut results: FxHashMap<String, AggregationResult> = FxHashMap::default();
if let Some(buckets) = self.buckets {
convert_and_add_final_buckets_to_result(&mut results, buckets, &req.buckets)?
convert_and_add_final_buckets_to_result(&mut results, buckets, &req.buckets, schema)?
} else {
// When there are no buckets, we create empty buckets, so that the serialized json
// format is constant
add_empty_final_buckets_to_result(&mut results, &req.buckets)?
add_empty_final_buckets_to_result(&mut results, &req.buckets, schema)?
};
if let Some(metrics) = self.metrics {
@@ -132,7 +138,7 @@ impl IntermediateAggregationResults {
}
fn convert_and_add_final_metrics_to_result(
results: &mut HashMap<String, AggregationResult>,
results: &mut FxHashMap<String, AggregationResult>,
metrics: VecWithNames<IntermediateMetricResult>,
) {
results.extend(
@@ -143,7 +149,7 @@ fn convert_and_add_final_metrics_to_result(
}
fn add_empty_final_metrics_to_result(
results: &mut HashMap<String, AggregationResult>,
results: &mut FxHashMap<String, AggregationResult>,
req_metrics: &VecWithNames<MetricAggregation>,
) -> crate::Result<()> {
results.extend(req_metrics.iter().map(|(key, req)| {
@@ -157,27 +163,30 @@ fn add_empty_final_metrics_to_result(
}
fn add_empty_final_buckets_to_result(
results: &mut HashMap<String, AggregationResult>,
results: &mut FxHashMap<String, AggregationResult>,
req_buckets: &VecWithNames<BucketAggregationInternal>,
schema: &Schema,
) -> crate::Result<()> {
let requested_buckets = req_buckets.iter();
for (key, req) in requested_buckets {
let empty_bucket = AggregationResult::BucketResult(BucketResult::empty_from_req(req)?);
let empty_bucket =
AggregationResult::BucketResult(BucketResult::empty_from_req(req, schema)?);
results.insert(key.to_string(), empty_bucket);
}
Ok(())
}
fn convert_and_add_final_buckets_to_result(
results: &mut HashMap<String, AggregationResult>,
results: &mut FxHashMap<String, AggregationResult>,
buckets: VecWithNames<IntermediateBucketResult>,
req_buckets: &VecWithNames<BucketAggregationInternal>,
schema: &Schema,
) -> crate::Result<()> {
assert_eq!(buckets.len(), req_buckets.len());
let buckets_with_request = buckets.into_iter().zip(req_buckets.values());
for ((key, bucket), req) in buckets_with_request {
let result = AggregationResult::BucketResult(bucket.into_final_bucket_result(req)?);
let result = AggregationResult::BucketResult(bucket.into_final_bucket_result(req, schema)?);
results.insert(key, result);
}
Ok(())
@@ -267,13 +276,21 @@ impl IntermediateBucketResult {
pub(crate) fn into_final_bucket_result(
self,
req: &BucketAggregationInternal,
schema: &Schema,
) -> crate::Result<BucketResult> {
match self {
IntermediateBucketResult::Range(range_res) => {
let mut buckets: Vec<RangeBucketEntry> = range_res
.buckets
.into_iter()
.map(|(_, bucket)| bucket.into_final_bucket_entry(&req.sub_aggregation))
.map(|(_, bucket)| {
bucket.into_final_bucket_entry(
&req.sub_aggregation,
schema,
req.as_range()
.expect("unexpected aggregation, expected histogram aggregation"),
)
})
.collect::<crate::Result<Vec<_>>>()?;
buckets.sort_by(|left, right| {
@@ -288,7 +305,7 @@ impl IntermediateBucketResult {
.keyed;
let buckets = if is_keyed {
let mut bucket_map =
FnvHashMap::with_capacity_and_hasher(buckets.len(), Default::default());
FxHashMap::with_capacity_and_hasher(buckets.len(), Default::default());
for bucket in buckets {
bucket_map.insert(bucket.key.to_string(), bucket);
}
@@ -304,11 +321,12 @@ impl IntermediateBucketResult {
req.as_histogram()
.expect("unexpected aggregation, expected histogram aggregation"),
&req.sub_aggregation,
schema,
)?;
let buckets = if req.as_histogram().unwrap().keyed {
let mut bucket_map =
FnvHashMap::with_capacity_and_hasher(buckets.len(), Default::default());
FxHashMap::with_capacity_and_hasher(buckets.len(), Default::default());
for bucket in buckets {
bucket_map.insert(bucket.key.to_string(), bucket);
}
@@ -322,6 +340,7 @@ impl IntermediateBucketResult {
req.as_term()
.expect("unexpected aggregation, expected term aggregation"),
&req.sub_aggregation,
schema,
),
}
}
@@ -396,13 +415,13 @@ impl IntermediateBucketResult {
#[derive(Default, Clone, Debug, PartialEq, Serialize, Deserialize)]
/// Range aggregation including error counts
pub struct IntermediateRangeBucketResult {
pub(crate) buckets: FnvHashMap<SerializedKey, IntermediateRangeBucketEntry>,
pub(crate) buckets: FxHashMap<SerializedKey, IntermediateRangeBucketEntry>,
}
#[derive(Default, Clone, Debug, PartialEq, Serialize, Deserialize)]
/// Term aggregation including error counts
pub struct IntermediateTermBucketResult {
pub(crate) entries: FnvHashMap<String, IntermediateTermBucketEntry>,
pub(crate) entries: FxHashMap<String, IntermediateTermBucketEntry>,
pub(crate) sum_other_doc_count: u64,
pub(crate) doc_count_error_upper_bound: u64,
}
@@ -412,6 +431,7 @@ impl IntermediateTermBucketResult {
self,
req: &TermsAggregation,
sub_aggregation_req: &AggregationsInternal,
schema: &Schema,
) -> crate::Result<BucketResult> {
let req = TermsAggregationInternal::from_req(req);
let mut buckets: Vec<BucketEntry> = self
@@ -420,11 +440,12 @@ impl IntermediateTermBucketResult {
.filter(|bucket| bucket.1.doc_count >= req.min_doc_count)
.map(|(key, entry)| {
Ok(BucketEntry {
key_as_string: None,
key: Key::Str(key),
doc_count: entry.doc_count,
sub_aggregation: entry
.sub_aggregation
.into_final_bucket_result_internal(sub_aggregation_req)?,
.into_final_bucket_result_internal(sub_aggregation_req, schema)?,
})
})
.collect::<crate::Result<_>>()?;
@@ -499,8 +520,8 @@ trait MergeFruits {
}
fn merge_maps<V: MergeFruits + Clone>(
entries_left: &mut FnvHashMap<SerializedKey, V>,
mut entries_right: FnvHashMap<SerializedKey, V>,
entries_left: &mut FxHashMap<SerializedKey, V>,
mut entries_right: FxHashMap<SerializedKey, V>,
) {
for (name, entry_left) in entries_left.iter_mut() {
if let Some(entry_right) = entries_right.remove(name) {
@@ -529,13 +550,15 @@ impl IntermediateHistogramBucketEntry {
pub(crate) fn into_final_bucket_entry(
self,
req: &AggregationsInternal,
schema: &Schema,
) -> crate::Result<BucketEntry> {
Ok(BucketEntry {
key_as_string: None,
key: Key::F64(self.key),
doc_count: self.doc_count,
sub_aggregation: self
.sub_aggregation
.into_final_bucket_result_internal(req)?,
.into_final_bucket_result_internal(req, schema)?,
})
}
}
@@ -572,16 +595,38 @@ impl IntermediateRangeBucketEntry {
pub(crate) fn into_final_bucket_entry(
self,
req: &AggregationsInternal,
schema: &Schema,
range_req: &RangeAggregation,
) -> crate::Result<RangeBucketEntry> {
Ok(RangeBucketEntry {
let mut range_bucket_entry = RangeBucketEntry {
key: self.key,
doc_count: self.doc_count,
sub_aggregation: self
.sub_aggregation
.into_final_bucket_result_internal(req)?,
.into_final_bucket_result_internal(req, schema)?,
to: self.to,
from: self.from,
})
to_as_string: None,
from_as_string: None,
};
// If we have a date type on the histogram buckets, we add the `key_as_string` field as
// rfc339
let field = schema
.get_field(&range_req.field)
.ok_or_else(|| TantivyError::FieldNotFound(range_req.field.to_string()))?;
if schema.get_field_entry(field).field_type().is_date() {
if let Some(val) = range_bucket_entry.to {
let key_as_string = format_date(val as i64)?;
range_bucket_entry.to_as_string = Some(key_as_string);
}
if let Some(val) = range_bucket_entry.from {
let key_as_string = format_date(val as i64)?;
range_bucket_entry.from_as_string = Some(key_as_string);
}
}
Ok(range_bucket_entry)
}
}
@@ -626,7 +671,7 @@ mod tests {
fn get_sub_test_tree(data: &[(String, u64)]) -> IntermediateAggregationResults {
let mut map = HashMap::new();
let mut buckets = FnvHashMap::default();
let mut buckets = FxHashMap::default();
for (key, doc_count) in data {
buckets.insert(
key.to_string(),
@@ -653,7 +698,7 @@ mod tests {
data: &[(String, u64, String, u64)],
) -> IntermediateAggregationResults {
let mut map = HashMap::new();
let mut buckets: FnvHashMap<_, _> = Default::default();
let mut buckets: FxHashMap<_, _> = Default::default();
for (key, doc_count, sub_aggregation_key, sub_aggregation_count) in data {
buckets.insert(
key.to_string(),

10
src/aggregation/metric/average.rs
View File

@@ -60,10 +60,10 @@ impl SegmentAverageCollector {
pub(crate) fn collect_block(&mut self, doc: &[DocId], field: &dyn Column<u64>) {
let mut iter = doc.chunks_exact(4);
for docs in iter.by_ref() {
let val1 = field.get_val(docs[0] as u64);
let val2 = field.get_val(docs[1] as u64);
let val3 = field.get_val(docs[2] as u64);
let val4 = field.get_val(docs[3] as u64);
let val1 = field.get_val(docs[0]);
let val2 = field.get_val(docs[1]);
let val3 = field.get_val(docs[2]);
let val4 = field.get_val(docs[3]);
let val1 = f64_from_fastfield_u64(val1, &self.field_type);
let val2 = f64_from_fastfield_u64(val2, &self.field_type);
let val3 = f64_from_fastfield_u64(val3, &self.field_type);
@@ -74,7 +74,7 @@ impl SegmentAverageCollector {
self.data.collect(val4);
}
for &doc in iter.remainder() {
let val = field.get_val(doc as u64);
let val = field.get_val(doc);
let val = f64_from_fastfield_u64(val, &self.field_type);
self.data.collect(val);
}

14
src/aggregation/metric/stats.rs
View File

@@ -166,10 +166,10 @@ impl SegmentStatsCollector {
pub(crate) fn collect_block(&mut self, doc: &[DocId], field: &dyn Column<u64>) {
let mut iter = doc.chunks_exact(4);
for docs in iter.by_ref() {
let val1 = field.get_val(docs[0] as u64);
let val2 = field.get_val(docs[1] as u64);
let val3 = field.get_val(docs[2] as u64);
let val4 = field.get_val(docs[3] as u64);
let val1 = field.get_val(docs[0]);
let val2 = field.get_val(docs[1]);
let val3 = field.get_val(docs[2]);
let val4 = field.get_val(docs[3]);
let val1 = f64_from_fastfield_u64(val1, &self.field_type);
let val2 = f64_from_fastfield_u64(val2, &self.field_type);
let val3 = f64_from_fastfield_u64(val3, &self.field_type);
@@ -180,7 +180,7 @@ impl SegmentStatsCollector {
self.stats.collect(val4);
}
for &doc in iter.remainder() {
let val = field.get_val(doc as u64);
let val = field.get_val(doc);
let val = f64_from_fastfield_u64(val, &self.field_type);
self.stats.collect(val);
}
@@ -222,7 +222,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let reader = index.reader()?;
let searcher = reader.searcher();
@@ -300,7 +300,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults = searcher.search(&term_query, &collector).unwrap();

84
src/aggregation/mod.rs
View File

@@ -10,21 +10,19 @@
//!
//! There are two categories: [Metrics](metric) and [Buckets](bucket).
//!
//! # Usage
//!
//! ## Prerequisite
//! Currently aggregations work only on [fast fields](`crate::fastfield`). Single value fast fields
//! of type `u64`, `f64`, `i64`, `date` and fast fields on text fields.
//!
//! ## Usage
//! To use aggregations, build an aggregation request by constructing
//! [`Aggregations`](agg_req::Aggregations).
//! Create an [`AggregationCollector`] from this request. `AggregationCollector` implements the
//! [`Collector`](crate::collector::Collector) trait and can be passed as collector into
//! [`Searcher::search()`](crate::Searcher::search).
//!
//! #### Limitations
//!
//! Currently aggregations work only on single value fast fields of type `u64`, `f64`, `i64` and
//! fast fields on text fields.
//!
//! # JSON Format
//! ## JSON Format
//! Aggregations request and result structures de/serialize into elasticsearch compatible JSON.
//!
//! ```verbatim
@@ -35,7 +33,7 @@
//! let json_response_string: String = &serde_json::to_string(&agg_res)?;
//! ```
//!
//! # Supported Aggregations
//! ## Supported Aggregations
//! - [Bucket](bucket)
//! - [Histogram](bucket::HistogramAggregation)
//! - [Range](bucket::RangeAggregation)
@@ -55,9 +53,10 @@
//! use tantivy::query::AllQuery;
//! use tantivy::aggregation::agg_result::AggregationResults;
//! use tantivy::IndexReader;
//! use tantivy::schema::Schema;
//!
//! # #[allow(dead_code)]
//! fn aggregate_on_index(reader: &IndexReader) {
//! fn aggregate_on_index(reader: &IndexReader, schema: Schema) {
//! let agg_req: Aggregations = vec![
//! (
//! "average".to_string(),
@@ -69,7 +68,7 @@
//! .into_iter()
//! .collect();
//!
//! let collector = AggregationCollector::from_aggs(agg_req, None);
//! let collector = AggregationCollector::from_aggs(agg_req, None, schema);
//!
//! let searcher = reader.searcher();
//! let agg_res: AggregationResults = searcher.search(&AllQuery, &collector).unwrap();
@@ -159,6 +158,7 @@ mod agg_req_with_accessor;
pub mod agg_result;
pub mod bucket;
mod collector;
mod date;
pub mod intermediate_agg_result;
pub mod metric;
mod segment_agg_result;
@@ -169,6 +169,7 @@ pub use collector::{
AggregationCollector, AggregationSegmentCollector, DistributedAggregationCollector,
MAX_BUCKET_COUNT,
};
pub(crate) use date::format_date;
use fastfield_codecs::MonotonicallyMappableToU64;
use itertools::Itertools;
use serde::{Deserialize, Serialize};
@@ -285,11 +286,11 @@ impl Display for Key {
/// Inverse of `to_fastfield_u64`. Used to convert to `f64` for metrics.
///
/// # Panics
/// Only `u64`, `f64`, and `i64` are supported.
/// Only `u64`, `f64`, `date`, and `i64` are supported.
pub(crate) fn f64_from_fastfield_u64(val: u64, field_type: &Type) -> f64 {
match field_type {
Type::U64 => val as f64,
Type::I64 => i64::from_u64(val) as f64,
Type::I64 | Type::Date => i64::from_u64(val) as f64,
Type::F64 => f64::from_u64(val),
_ => {
panic!("unexpected type {:?}. This should not happen", field_type)
@@ -297,10 +298,9 @@ pub(crate) fn f64_from_fastfield_u64(val: u64, field_type: &Type) -> f64 {
}
}
/// Converts the `f64` value to fast field value space.
/// Converts the `f64` value to fast field value space, which is always u64.
///
/// If the fast field has `u64`, values are stored as `u64` in the fast field.
/// A `f64` value of e.g. `2.0` therefore needs to be converted to `1u64`.
/// If the fast field has `u64`, values are stored unchanged as `u64` in the fast field.
///
/// If the fast field has `f64` values are converted and stored to `u64` using a
/// monotonic mapping.
@@ -310,7 +310,7 @@ pub(crate) fn f64_from_fastfield_u64(val: u64, field_type: &Type) -> f64 {
pub(crate) fn f64_to_fastfield_u64(val: f64, field_type: &Type) -> Option<u64> {
match field_type {
Type::U64 => Some(val as u64),
Type::I64 => Some((val as i64).to_u64()),
Type::I64 | Type::Date => Some((val as i64).to_u64()),
Type::F64 => Some(val.to_u64()),
_ => None,
}
@@ -319,6 +319,7 @@ pub(crate) fn f64_to_fastfield_u64(val: f64, field_type: &Type) -> Option<u64> {
#[cfg(test)]
mod tests {
use serde_json::Value;
use time::OffsetDateTime;
use super::agg_req::{Aggregation, Aggregations, BucketAggregation};
use super::bucket::RangeAggregation;
@@ -334,7 +335,7 @@ mod tests {
use crate::aggregation::DistributedAggregationCollector;
use crate::query::{AllQuery, TermQuery};
use crate::schema::{Cardinality, IndexRecordOption, Schema, TextFieldIndexing, FAST, STRING};
use crate::{Index, Term};
use crate::{DateTime, Index, Term};
fn get_avg_req(field_name: &str) -> Aggregation {
Aggregation::Metric(MetricAggregation::Average(
@@ -360,7 +361,7 @@ mod tests {
index: &Index,
query: Option<(&str, &str)>,
) -> crate::Result<Value> {
let collector = AggregationCollector::from_aggs(agg_req, None);
let collector = AggregationCollector::from_aggs(agg_req, None, index.schema());
let reader = index.reader()?;
let searcher = reader.searcher();
@@ -554,10 +555,10 @@ mod tests {
let searcher = reader.searcher();
let intermediate_agg_result = searcher.search(&AllQuery, &collector).unwrap();
intermediate_agg_result
.into_final_bucket_result(agg_req)
.into_final_bucket_result(agg_req, &index.schema())
.unwrap()
} else {
let collector = AggregationCollector::from_aggs(agg_req, None);
let collector = AggregationCollector::from_aggs(agg_req, None, index.schema());
let searcher = reader.searcher();
searcher.search(&AllQuery, &collector).unwrap()
@@ -650,6 +651,7 @@ mod tests {
.set_fast()
.set_stored();
let text_field = schema_builder.add_text_field("text", text_fieldtype);
let date_field = schema_builder.add_date_field("date", FAST);
schema_builder.add_text_field("dummy_text", STRING);
let score_fieldtype =
crate::schema::NumericOptions::default().set_fast(Cardinality::SingleValue);
@@ -667,6 +669,7 @@ mod tests {
// writing the segment
index_writer.add_document(doc!(
text_field => "cool",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800).unwrap()),
score_field => 1u64,
score_field_f64 => 1f64,
score_field_i64 => 1i64,
@@ -675,6 +678,7 @@ mod tests {
))?;
index_writer.add_document(doc!(
text_field => "cool",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400).unwrap()),
score_field => 3u64,
score_field_f64 => 3f64,
score_field_i64 => 3i64,
@@ -683,18 +687,21 @@ mod tests {
))?;
index_writer.add_document(doc!(
text_field => "cool",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400).unwrap()),
score_field => 5u64,
score_field_f64 => 5f64,
score_field_i64 => 5i64,
))?;
index_writer.add_document(doc!(
text_field => "nohit",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400).unwrap()),
score_field => 6u64,
score_field_f64 => 6f64,
score_field_i64 => 6i64,
))?;
index_writer.add_document(doc!(
text_field => "cool",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400).unwrap()),
score_field => 7u64,
score_field_f64 => 7f64,
score_field_i64 => 7i64,
@@ -702,12 +709,14 @@ mod tests {
index_writer.commit()?;
index_writer.add_document(doc!(
text_field => "cool",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400).unwrap()),
score_field => 11u64,
score_field_f64 => 11f64,
score_field_i64 => 11i64,
))?;
index_writer.add_document(doc!(
text_field => "cool",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400 + 86400).unwrap()),
score_field => 14u64,
score_field_f64 => 14f64,
score_field_i64 => 14i64,
@@ -715,6 +724,7 @@ mod tests {
index_writer.add_document(doc!(
text_field => "cool",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400 + 86400).unwrap()),
score_field => 44u64,
score_field_f64 => 44.5f64,
score_field_i64 => 44i64,
@@ -725,6 +735,7 @@ mod tests {
// no hits segment
index_writer.add_document(doc!(
text_field => "nohit",
date_field => DateTime::from_utc(OffsetDateTime::from_unix_timestamp(1_546_300_800 + 86400 + 86400).unwrap()),
score_field => 44u64,
score_field_f64 => 44.5f64,
score_field_i64 => 44i64,
@@ -797,7 +808,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults = searcher.search(&term_query, &collector).unwrap();
@@ -997,9 +1008,10 @@ mod tests {
// Test de/serialization roundtrip on intermediate_agg_result
let res: IntermediateAggregationResults =
serde_json::from_str(&serde_json::to_string(&res).unwrap()).unwrap();
res.into_final_bucket_result(agg_req.clone()).unwrap()
res.into_final_bucket_result(agg_req.clone(), &index.schema())
.unwrap()
} else {
let collector = AggregationCollector::from_aggs(agg_req.clone(), None);
let collector = AggregationCollector::from_aggs(agg_req.clone(), None, index.schema());
let searcher = reader.searcher();
searcher.search(&term_query, &collector).unwrap()
@@ -1057,7 +1069,7 @@ mod tests {
);
// Test empty result set
let collector = AggregationCollector::from_aggs(agg_req, None);
let collector = AggregationCollector::from_aggs(agg_req, None, index.schema());
let searcher = reader.searcher();
searcher.search(&query_with_no_hits, &collector).unwrap();
@@ -1122,7 +1134,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
@@ -1235,7 +1247,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1266,7 +1278,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1297,7 +1309,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1336,7 +1348,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1365,7 +1377,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req, None);
let collector = AggregationCollector::from_aggs(agg_req, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1394,7 +1406,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req, None);
let collector = AggregationCollector::from_aggs(agg_req, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1431,7 +1443,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1466,7 +1478,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1505,7 +1517,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1535,7 +1547,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =
@@ -1592,7 +1604,7 @@ mod tests {
.into_iter()
.collect();
let collector = AggregationCollector::from_aggs(agg_req_1, None);
let collector = AggregationCollector::from_aggs(agg_req_1, None, index.schema());
let searcher = reader.searcher();
let agg_res: AggregationResults =

8
src/collector/facet_collector.rs
View File

@@ -338,11 +338,7 @@ impl SegmentCollector for FacetSegmentCollector {
let mut previous_collapsed_ord: usize = usize::MAX;
for &facet_ord in &self.facet_ords_buf {
let collapsed_ord = self.collapse_mapping[facet_ord as usize];
self.counts[collapsed_ord] += if collapsed_ord == previous_collapsed_ord {
0
} else {
1
};
self.counts[collapsed_ord] += u64::from(collapsed_ord != previous_collapsed_ord);
previous_collapsed_ord = collapsed_ord;
}
}
@@ -620,7 +616,7 @@ mod tests {
.map(|mut doc| {
doc.add_facet(
facet_field,
&format!("/facet/{}", thread_rng().sample(&uniform)),
&format!("/facet/{}", thread_rng().sample(uniform)),
);
doc
})

2
src/collector/filter_collector_wrapper.rs
View File

@@ -177,7 +177,7 @@ where
type Fruit = TSegmentCollector::Fruit;
fn collect(&mut self, doc: u32, score: Score) {
let value = self.fast_field_reader.get_val(doc as u64);
let value = self.fast_field_reader.get_val(doc);
if (self.predicate)(value) {
self.segment_collector.collect(doc, score)
}

2
src/collector/histogram_collector.rs
View File

@@ -94,7 +94,7 @@ impl SegmentCollector for SegmentHistogramCollector {
type Fruit = Vec<u64>;
fn collect(&mut self, doc: DocId, _score: Score) {
let value = self.ff_reader.get_val(doc as u64);
let value = self.ff_reader.get_val(doc);
self.histogram_computer.add_value(value);
}

34
src/collector/mod.rs
View File

@@ -172,17 +172,33 @@ pub trait Collector: Sync + Send {
) -> crate::Result<<Self::Child as SegmentCollector>::Fruit> {
let mut segment_collector = self.for_segment(segment_ord as u32, reader)?;
if let Some(alive_bitset) = reader.alive_bitset() {
weight.for_each(reader, &mut |doc, score| {
if alive_bitset.is_alive(doc) {
match (reader.alive_bitset(), self.requires_scoring()) {
(Some(alive_bitset), true) => {
weight.for_each(reader, &mut |doc, score| {
if alive_bitset.is_alive(doc) {
segment_collector.collect(doc, score);
}
})?;
}
(Some(alive_bitset), false) => {
weight.for_each_no_score(reader, &mut |doc| {
if alive_bitset.is_alive(doc) {
segment_collector.collect(doc, 0.0);
}
})?;
}
(None, true) => {
weight.for_each(reader, &mut |doc, score| {
segment_collector.collect(doc, score);
}
})?;
} else {
weight.for_each(reader, &mut |doc, score| {
segment_collector.collect(doc, score);
})?;
})?;
}
(None, false) => {
weight.for_each_no_score(reader, &mut |doc| {
segment_collector.collect(doc, 0.0);
})?;
}
}
Ok(segment_collector.harvest())
}
}

2
src/collector/tests.rs
View File

@@ -201,7 +201,7 @@ impl SegmentCollector for FastFieldSegmentCollector {
type Fruit = Vec<u64>;
fn collect(&mut self, doc: DocId, _score: Score) {
let val = self.reader.get_val(doc as u64);
let val = self.reader.get_val(doc);
self.vals.push(val);
}

8
src/collector/top_score_collector.rs
View File

@@ -137,7 +137,7 @@ struct ScorerByFastFieldReader {
impl CustomSegmentScorer<u64> for ScorerByFastFieldReader {
fn score(&mut self, doc: DocId) -> u64 {
self.ff_reader.get_val(doc as u64)
self.ff_reader.get_val(doc)
}
}
@@ -458,7 +458,7 @@ impl TopDocs {
///
/// // We can now define our actual scoring function
/// move |doc: DocId, original_score: Score| {
/// let popularity: u64 = popularity_reader.get_val(doc as u64);
/// let popularity: u64 = popularity_reader.get_val(doc);
/// // Well.. For the sake of the example we use a simple logarithm
/// // function.
/// let popularity_boost_score = ((2u64 + popularity) as Score).log2();
@@ -567,8 +567,8 @@ impl TopDocs {
///
/// // We can now define our actual scoring function
/// move |doc: DocId| {
/// let popularity: u64 = popularity_reader.get_val(doc as u64);
/// let boosted: u64 = boosted_reader.get_val(doc as u64);
/// let popularity: u64 = popularity_reader.get_val(doc);
/// let boosted: u64 = boosted_reader.get_val(doc);
/// // Score do not have to be `f64` in tantivy.
/// // Here we return a couple to get lexicographical order
/// // for free.

43
src/core/index.rs
View File

@@ -19,7 +19,7 @@ use crate::error::{DataCorruption, TantivyError};
use crate::indexer::index_writer::{MAX_NUM_THREAD, MEMORY_ARENA_NUM_BYTES_MIN};
use crate::indexer::segment_updater::save_metas;
use crate::reader::{IndexReader, IndexReaderBuilder};
use crate::schema::{Field, FieldType, Schema};
use crate::schema::{Cardinality, Field, FieldType, Schema};
use crate::tokenizer::{TextAnalyzer, TokenizerManager};
use crate::IndexWriter;
@@ -149,12 +149,11 @@ impl IndexBuilder {
/// Creates a new index using the [`RamDirectory`].
///
/// The index will be allocated in anonymous memory.
/// This should only be used for unit tests.
/// This is useful for indexing small set of documents
/// for instances like unit test or temporary in memory index.
pub fn create_in_ram(self) -> Result<Index, TantivyError> {
let ram_directory = RamDirectory::create();
Ok(self
.create(ram_directory)
.expect("Creating a RamDirectory should never fail"))
self.create(ram_directory)
}
/// Creates a new index in a given filepath.
@@ -228,10 +227,44 @@ impl IndexBuilder {
))
}
}
fn validate(&self) -> crate::Result<()> {
if let Some(schema) = self.schema.as_ref() {
if let Some(sort_by_field) = self.index_settings.sort_by_field.as_ref() {
let schema_field = schema.get_field(&sort_by_field.field).ok_or_else(|| {
TantivyError::InvalidArgument(format!(
"Field to sort index {} not found in schema",
sort_by_field.field
))
})?;
let entry = schema.get_field_entry(schema_field);
if !entry.is_fast() {
return Err(TantivyError::InvalidArgument(format!(
"Field {} is no fast field. Field needs to be a single value fast field \
to be used to sort an index",
sort_by_field.field
)));
}
if entry.field_type().fastfield_cardinality() != Some(Cardinality::SingleValue) {
return Err(TantivyError::InvalidArgument(format!(
"Only single value fast field Cardinality supported for sorting index {}",
sort_by_field.field
)));
}
}
Ok(())
} else {
Err(TantivyError::InvalidArgument(
"no schema passed".to_string(),
))
}
}
/// Creates a new index given an implementation of the trait `Directory`.
///
/// If a directory previously existed, it will be erased.
fn create<T: Into<Box<dyn Directory>>>(self, dir: T) -> crate::Result<Index> {
self.validate()?;
let dir = dir.into();
let directory = ManagedDirectory::wrap(dir)?;
save_new_metas(

2
src/core/index_meta.rs
View File

@@ -133,7 +133,7 @@ impl SegmentMeta {
/// associated with a segment component.
pub fn relative_path(&self, component: SegmentComponent) -> PathBuf {
let mut path = self.id().uuid_string();
path.push_str(&*match component {
path.push_str(&match component {
SegmentComponent::Postings => ".idx".to_string(),
SegmentComponent::Positions => ".pos".to_string(),
SegmentComponent::Terms => ".term".to_string(),

35
src/core/inverted_index_reader.rs
View File

@@ -15,12 +15,11 @@ use crate::termdict::TermDictionary;
///
/// It is safe to delete the segment associated with
/// an `InvertedIndexReader`. As long as it is open,
/// the `FileSlice` it is relying on should
/// the [`FileSlice`] it is relying on should
/// stay available.
///
///
/// `InvertedIndexReader` are created by calling
/// the `SegmentReader`'s [`.inverted_index(...)`] method
/// [`SegmentReader::inverted_index()`](crate::SegmentReader::inverted_index).
pub struct InvertedIndexReader {
termdict: TermDictionary,
postings_file_slice: FileSlice,
@@ -75,7 +74,7 @@ impl InvertedIndexReader {
///
/// This is useful for enumerating through a list of terms,
/// and consuming the associated posting lists while avoiding
/// reallocating a `BlockSegmentPostings`.
/// reallocating a [`BlockSegmentPostings`].
///
/// # Warning
///
@@ -96,7 +95,7 @@ impl InvertedIndexReader {
/// Returns a block postings given a `Term`.
/// This method is for an advanced usage only.
///
/// Most user should prefer using `read_postings` instead.
/// Most users should prefer using [`Self::read_postings()`] instead.
pub fn read_block_postings(
&self,
term: &Term,
@@ -110,7 +109,7 @@ impl InvertedIndexReader {
/// Returns a block postings given a `term_info`.
/// This method is for an advanced usage only.
///
/// Most user should prefer using `read_postings` instead.
/// Most users should prefer using [`Self::read_postings()`] instead.
pub fn read_block_postings_from_terminfo(
&self,
term_info: &TermInfo,
@@ -130,7 +129,7 @@ impl InvertedIndexReader {
/// Returns a posting object given a `term_info`.
/// This method is for an advanced usage only.
///
/// Most user should prefer using `read_postings` instead.
/// Most users should prefer using [`Self::read_postings()`] instead.
pub fn read_postings_from_terminfo(
&self,
term_info: &TermInfo,
@@ -164,12 +163,12 @@ impl InvertedIndexReader {
/// or `None` if the term has never been encountered and indexed.
///
/// If the field was not indexed with the indexing options that cover
/// the requested options, the returned `SegmentPostings` the method does not fail
/// the requested options, the returned [`SegmentPostings`] the method does not fail
/// and returns a `SegmentPostings` with as much information as possible.
///
/// For instance, requesting `IndexRecordOption::Freq` for a
/// `TextIndexingOptions` that does not index position will return a `SegmentPostings`
/// with `DocId`s and frequencies.
/// For instance, requesting [`IndexRecordOption::WithFreqs`] for a
/// [`TextOptions`](crate::schema::TextOptions) that does not index position
/// will return a [`SegmentPostings`] with `DocId`s and frequencies.
pub fn read_postings(
&self,
term: &Term,
@@ -211,7 +210,7 @@ impl InvertedIndexReader {
/// Returns a block postings given a `Term`.
/// This method is for an advanced usage only.
///
/// Most user should prefer using `read_postings` instead.
/// Most users should prefer using [`Self::read_postings()`] instead.
pub async fn warm_postings(
&self,
term: &Term,
@@ -231,6 +230,18 @@ impl InvertedIndexReader {
Ok(())
}
/// Read the block postings for all terms.
/// This method is for an advanced usage only.
///
/// If you know which terms to pre-load, prefer using [`Self::warm_postings`] instead.
pub async fn warm_postings_full(&self, with_positions: bool) -> crate::AsyncIoResult<()> {
self.postings_file_slice.read_bytes_async().await?;
if with_positions {
self.positions_file_slice.read_bytes_async().await?;
}
Ok(())
}
/// Returns the number of documents containing the term asynchronously.
pub async fn doc_freq_async(&self, term: &Term) -> crate::AsyncIoResult<u32> {
Ok(self

9
src/core/searcher.rs
View File

@@ -4,7 +4,7 @@ use std::{fmt, io};
use crate::collector::Collector;
use crate::core::{Executor, SegmentReader};
use crate::query::Query;
use crate::query::{EnableScoring, Query};
use crate::schema::{Document, Schema, Term};
use crate::space_usage::SearcherSpaceUsage;
use crate::store::{CacheStats, StoreReader};
@@ -199,7 +199,12 @@ impl Searcher {
executor: &Executor,
) -> crate::Result<C::Fruit> {
let scoring_enabled = collector.requires_scoring();
let weight = query.weight(self, scoring_enabled)?;
let enabled_scoring = if scoring_enabled {
EnableScoring::Enabled(self)
} else {
EnableScoring::Disabled(self.schema())
};
let weight = query.weight(enabled_scoring)?;
let segment_readers = self.segment_readers();
let fruits = executor.map(
|(segment_ord, segment_reader)| {

6
src/core/segment_reader.rs
View File

@@ -216,10 +216,10 @@ impl SegmentReader {
/// term dictionary associated with a specific field,
/// and opening the posting list associated with any term.
///
/// If the field is not marked as index, a warn is logged and an empty `InvertedIndexReader`
/// If the field is not marked as index, a warning is logged and an empty `InvertedIndexReader`
/// is returned.
/// Similarly if the field is marked as indexed but no term has been indexed for the given
/// index. an empty `InvertedIndexReader` is returned (but no warning is logged).
/// Similarly, if the field is marked as indexed but no term has been indexed for the given
/// index, an empty `InvertedIndexReader` is returned (but no warning is logged).
pub fn inverted_index(&self, field: Field) -> crate::Result<Arc<InvertedIndexReader>> {
if let Some(inv_idx_reader) = self
.inv_idx_reader_cache

2
src/directory/directory.rs
View File

@@ -55,7 +55,7 @@ impl<T: Send + Sync + 'static> From<Box<T>> for DirectoryLock {
impl Drop for DirectoryLockGuard {
fn drop(&mut self) {
if let Err(e) = self.directory.delete(&*self.path) {
if let Err(e) = self.directory.delete(&self.path) {
error!("Failed to remove the lock file. {:?}", e);
}
}

4
src/directory/file_slice.rs
View File

@@ -13,8 +13,8 @@ use crate::directory::OwnedBytes;
/// By contract, whatever happens to the directory file, as long as a FileHandle
/// is alive, the data associated with it cannot be altered or destroyed.
///
/// The underlying behavior is therefore specific to the `Directory` that created it.
/// Despite its name, a `FileSlice` may or may not directly map to an actual file
/// The underlying behavior is therefore specific to the [`Directory`](crate::Directory) that
/// created it. Despite its name, a [`FileSlice`] may or may not directly map to an actual file
/// on the filesystem.
#[async_trait]

65
src/directory/mmap_directory.rs
View File

@@ -304,7 +304,7 @@ pub(crate) fn atomic_write(path: &Path, content: &[u8]) -> io::Result<()> {
"Path {:?} does not have parent directory.",
)
})?;
let mut tempfile = tempfile::Builder::new().tempfile_in(&parent_path)?;
let mut tempfile = tempfile::Builder::new().tempfile_in(parent_path)?;
tempfile.write_all(content)?;
tempfile.flush()?;
tempfile.as_file_mut().sync_data()?;
@@ -571,9 +571,21 @@ mod tests {
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 0);
}
fn assert_eventually<P: Fn() -> Option<String>>(predicate: P) {
for _ in 0..30 {
if predicate().is_none() {
break;
}
std::thread::sleep(Duration::from_millis(200));
}
if let Some(error_msg) = predicate() {
panic!("{}", error_msg);
}
}
#[test]
fn test_mmap_released() -> crate::Result<()> {
let mmap_directory = MmapDirectory::create_from_tempdir()?;
fn test_mmap_released() {
let mmap_directory = MmapDirectory::create_from_tempdir().unwrap();
let mut schema_builder: SchemaBuilder = Schema::builder();
let text_field = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
@@ -582,47 +594,56 @@ mod tests {
let index =
Index::create(mmap_directory.clone(), schema, IndexSettings::default()).unwrap();
let mut index_writer = index.writer_for_tests()?;
let mut index_writer = index.writer_for_tests().unwrap();
let mut log_merge_policy = LogMergePolicy::default();
log_merge_policy.set_min_num_segments(3);
index_writer.set_merge_policy(Box::new(log_merge_policy));
for _num_commits in 0..10 {
for _ in 0..10 {
index_writer.add_document(doc!(text_field=>"abc"))?;
index_writer.add_document(doc!(text_field=>"abc")).unwrap();
}
index_writer.commit()?;
index_writer.commit().unwrap();
}
let reader = index
.reader_builder()
.reload_policy(ReloadPolicy::Manual)
.try_into()?;
.try_into()
.unwrap();
for _ in 0..4 {
index_writer.add_document(doc!(text_field=>"abc"))?;
index_writer.commit()?;
reader.reload()?;
index_writer.add_document(doc!(text_field=>"abc")).unwrap();
index_writer.commit().unwrap();
reader.reload().unwrap();
}
index_writer.wait_merging_threads()?;
index_writer.wait_merging_threads().unwrap();
reader.reload()?;
reader.reload().unwrap();
let num_segments = reader.searcher().segment_readers().len();
assert!(num_segments <= 4);
let num_components_except_deletes_and_tempstore =
crate::core::SegmentComponent::iterator().len() - 2;
assert_eq!(
num_segments * num_components_except_deletes_and_tempstore,
mmap_directory.get_cache_info().mmapped.len()
);
let max_num_mmapped = num_components_except_deletes_and_tempstore * num_segments;
assert_eventually(|| {
let num_mmapped = mmap_directory.get_cache_info().mmapped.len();
if num_mmapped > max_num_mmapped {
Some(format!(
"Expected at most {max_num_mmapped} mmapped files, got {num_mmapped}"
))
} else {
None
}
});
}
// This test failed on CI. The last Mmap is dropped from the merging thread so there might
// be a race condition indeed.
for _ in 0..10 {
if mmap_directory.get_cache_info().mmapped.is_empty() {
return Ok(());
assert_eventually(|| {
let num_mmapped = mmap_directory.get_cache_info().mmapped.len();
if num_mmapped > 0 {
Some(format!("Expected no mmapped files, got {num_mmapped}"))
} else {
None
}
std::thread::sleep(Duration::from_millis(200));
}
panic!("The cache still contains information. One of the Mmap has not been dropped.");
});
}
}

7
src/fastfield/bytes/mod.rs
View File

@@ -6,7 +6,7 @@ pub use self::writer::BytesFastFieldWriter;
#[cfg(test)]
mod tests {
use crate::query::TermQuery;
use crate::query::{EnableScoring, TermQuery};
use crate::schema::{BytesOptions, IndexRecordOption, Schema, Value, FAST, INDEXED, STORED};
use crate::{DocAddress, DocSet, Index, Searcher, Term};
@@ -82,7 +82,7 @@ mod tests {
let field = searcher.schema().get_field("string_bytes").unwrap();
let term = Term::from_field_bytes(field, b"lucene".as_ref());
let term_query = TermQuery::new(term, IndexRecordOption::Basic);
let term_weight = term_query.specialized_weight(&searcher, true)?;
let term_weight = term_query.specialized_weight(EnableScoring::Enabled(&searcher))?;
let term_scorer = term_weight.specialized_scorer(searcher.segment_reader(0), 1.0)?;
assert_eq!(term_scorer.doc(), 0u32);
Ok(())
@@ -95,7 +95,8 @@ mod tests {
let field = searcher.schema().get_field("string_bytes").unwrap();
let term = Term::from_field_bytes(field, b"lucene".as_ref());
let term_query = TermQuery::new(term, IndexRecordOption::Basic);
let term_weight_err = term_query.specialized_weight(&searcher, false);
let term_weight_err =
term_query.specialized_weight(EnableScoring::Disabled(searcher.schema()));
assert!(matches!(
term_weight_err,
Err(crate::TantivyError::SchemaError(_))

40
src/fastfield/bytes/reader.rs
View File

@@ -1,10 +1,9 @@
use std::ops::Range;
use std::sync::Arc;
use fastfield_codecs::Column;
use crate::directory::{FileSlice, OwnedBytes};
use crate::fastfield::MultiValueLength;
use crate::fastfield::MultiValueIndex;
use crate::DocId;
/// Reader for byte array fast fields
@@ -19,7 +18,7 @@ use crate::DocId;
/// and the start index for the next document, and keeping the bytes in between.
#[derive(Clone)]
pub struct BytesFastFieldReader {
idx_reader: Arc<dyn Column<u64>>,
idx_reader: MultiValueIndex,
values: OwnedBytes,
}
@@ -29,42 +28,31 @@ impl BytesFastFieldReader {
values_file: FileSlice,
) -> crate::Result<BytesFastFieldReader> {
let values = values_file.read_bytes()?;
Ok(BytesFastFieldReader { idx_reader, values })
Ok(BytesFastFieldReader {
idx_reader: MultiValueIndex::new(idx_reader),
values,
})
}
fn range(&self, doc: DocId) -> Range<u64> {
let idx = doc as u64;
let start = self.idx_reader.get_val(idx);
let end = self.idx_reader.get_val(idx + 1);
start..end
/// returns the multivalue index
pub fn get_index_reader(&self) -> &MultiValueIndex {
&self.idx_reader
}
/// Returns the bytes associated with the given `doc`
pub fn get_bytes(&self, doc: DocId) -> &[u8] {
let range = self.range(doc);
let range = self.idx_reader.range(doc);
&self.values.as_slice()[range.start as usize..range.end as usize]
}
/// Returns the length of the bytes associated with the given `doc`
pub fn num_bytes(&self, doc: DocId) -> u64 {
let range = self.range(doc);
range.end - range.start
let range = self.idx_reader.range(doc);
(range.end - range.start) as u64
}
/// Returns the overall number of bytes in this bytes fast field.
pub fn total_num_bytes(&self) -> u64 {
self.values.len() as u64
}
}
impl MultiValueLength for BytesFastFieldReader {
fn get_range(&self, doc_id: DocId) -> std::ops::Range<u64> {
self.range(doc_id)
}
fn get_len(&self, doc_id: DocId) -> u64 {
self.num_bytes(doc_id)
}
fn get_total_len(&self) -> u64 {
self.total_num_bytes()
pub fn total_num_bytes(&self) -> u32 {
self.values.len() as u32
}
}

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

@@ -57,14 +57,15 @@ impl BytesFastFieldWriter {
/// Shift to the next document and add all of the
/// matching field values present in the document.
pub fn add_document(&mut self, doc: &Document) {
pub fn add_document(&mut self, doc: &Document) -> crate::Result<()> {
self.next_doc();
for field_value in doc.get_all(self.field) {
if let Value::Bytes(ref bytes) = field_value {
self.vals.extend_from_slice(bytes);
return;
return Ok(());
}
}
Ok(())
}
/// Register the bytes associated with a document.

165
src/fastfield/mod.rs
View File

@@ -7,16 +7,15 @@
//! 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.
//! Fast fields are useful when a field is required for all or most of
//! the `DocSet`: for instance for scoring, grouping, aggregation, filtering, or faceting.
//!
//!
//! Fields have to be declared as `FAST` in the schema.
//! Currently supported fields are: u64, i64, f64 and bytes.
//! Fields have to be declared as `FAST` in the schema.
//! Currently supported fields are: u64, i64, f64, bytes and text.
//!
//! u64, i64 and f64 fields are stored in a bit-packed fashion so that
//! their memory usage is directly linear with the amplitude of the
//! values stored.
//! Fast fields are stored in with [different codecs](fastfield_codecs). The best codec is detected
//! automatically, when serializing.
//!
//! Read access performance is comparable to that of an array lookup.
@@ -26,14 +25,18 @@ pub use self::alive_bitset::{intersect_alive_bitsets, write_alive_bitset, AliveB
pub use self::bytes::{BytesFastFieldReader, BytesFastFieldWriter};
pub use self::error::{FastFieldNotAvailableError, Result};
pub use self::facet_reader::FacetReader;
pub(crate) use self::multivalued::MultivalueStartIndex;
pub use self::multivalued::{MultiValuedFastFieldReader, MultiValuedFastFieldWriter};
pub(crate) use self::multivalued::{get_fastfield_codecs_for_multivalue, MultivalueStartIndex};
pub use self::multivalued::{
MultiValueIndex, MultiValueU128FastFieldWriter, MultiValuedFastFieldReader,
MultiValuedFastFieldWriter, MultiValuedU128FastFieldReader,
};
pub(crate) use self::readers::type_and_cardinality;
pub use self::readers::FastFieldReaders;
pub(crate) use self::readers::{type_and_cardinality, FastType};
pub use self::serializer::{Column, CompositeFastFieldSerializer};
use self::writer::unexpected_value;
pub use self::writer::{FastFieldsWriter, IntFastFieldWriter};
use crate::schema::{Type, Value};
use crate::{DateTime, DocId};
use crate::DateTime;
mod alive_bitset;
mod bytes;
@@ -44,17 +47,6 @@ mod readers;
mod serializer;
mod writer;
/// Trait for `BytesFastFieldReader` and `MultiValuedFastFieldReader` to return the length of data
/// for a doc_id
pub trait MultiValueLength {
/// returns the positions for a docid
fn get_range(&self, doc_id: DocId) -> std::ops::Range<u64>;
/// returns the num of values associated with a doc_id
fn get_len(&self, doc_id: DocId) -> u64;
/// returns the sum of num values for all doc_ids
fn get_total_len(&self) -> u64;
}
/// Trait for types that are allowed for fast fields:
/// (u64, i64 and f64, bool, DateTime).
pub trait FastValue:
@@ -117,15 +109,16 @@ impl FastValue for DateTime {
}
}
fn value_to_u64(value: &Value) -> u64 {
match value {
fn value_to_u64(value: &Value) -> crate::Result<u64> {
let value = match value {
Value::U64(val) => val.to_u64(),
Value::I64(val) => val.to_u64(),
Value::F64(val) => val.to_u64(),
Value::Bool(val) => val.to_u64(),
Value::Date(val) => val.to_u64(),
_ => panic!("Expected a u64/i64/f64/bool/date field, got {:?} ", value),
}
_ => return Err(unexpected_value("u64/i64/f64/bool/date", value)),
};
Ok(value)
}
/// The fast field type
@@ -180,9 +173,9 @@ mod tests {
#[test]
pub fn test_fastfield() {
let test_fastfield = fastfield_codecs::serialize_and_load(&[100u64, 200u64, 300u64][..]);
assert_eq!(test_fastfield.get_val(0u64), 100);
assert_eq!(test_fastfield.get_val(1u64), 200);
assert_eq!(test_fastfield.get_val(2u64), 300);
assert_eq!(test_fastfield.get_val(0), 100);
assert_eq!(test_fastfield.get_val(1), 200);
assert_eq!(test_fastfield.get_val(2), 300);
}
#[test]
@@ -199,16 +192,22 @@ mod tests {
let write: WritePtr = directory.open_write(Path::new("test")).unwrap();
let mut serializer = CompositeFastFieldSerializer::from_write(write).unwrap();
let mut fast_field_writers = FastFieldsWriter::from_schema(&SCHEMA);
fast_field_writers.add_document(&doc!(*FIELD=>13u64));
fast_field_writers.add_document(&doc!(*FIELD=>14u64));
fast_field_writers.add_document(&doc!(*FIELD=>2u64));
fast_field_writers
.add_document(&doc!(*FIELD=>13u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>14u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>2u64))
.unwrap();
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
.unwrap();
serializer.close().unwrap();
}
let file = directory.open_read(path).unwrap();
assert_eq!(file.len(), 25);
assert_eq!(file.len(), 34);
let composite_file = CompositeFile::open(&file)?;
let fast_field_bytes = composite_file.open_read(*FIELD).unwrap().read_bytes()?;
let fast_field_reader = open::<u64>(fast_field_bytes)?;
@@ -226,20 +225,38 @@ mod tests {
let write: WritePtr = directory.open_write(Path::new("test"))?;
let mut serializer = CompositeFastFieldSerializer::from_write(write)?;
let mut fast_field_writers = FastFieldsWriter::from_schema(&SCHEMA);
fast_field_writers.add_document(&doc!(*FIELD=>4u64));
fast_field_writers.add_document(&doc!(*FIELD=>14_082_001u64));
fast_field_writers.add_document(&doc!(*FIELD=>3_052u64));
fast_field_writers.add_document(&doc!(*FIELD=>9_002u64));
fast_field_writers.add_document(&doc!(*FIELD=>15_001u64));
fast_field_writers.add_document(&doc!(*FIELD=>777u64));
fast_field_writers.add_document(&doc!(*FIELD=>1_002u64));
fast_field_writers.add_document(&doc!(*FIELD=>1_501u64));
fast_field_writers.add_document(&doc!(*FIELD=>215u64));
fast_field_writers
.add_document(&doc!(*FIELD=>4u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>14_082_001u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>3_052u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>9_002u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>15_001u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>777u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>1_002u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>1_501u64))
.unwrap();
fast_field_writers
.add_document(&doc!(*FIELD=>215u64))
.unwrap();
fast_field_writers.serialize(&mut serializer, &HashMap::new(), None)?;
serializer.close()?;
}
let file = directory.open_read(path)?;
assert_eq!(file.len(), 53);
assert_eq!(file.len(), 62);
{
let fast_fields_composite = CompositeFile::open(&file)?;
let data = fast_fields_composite
@@ -270,7 +287,9 @@ mod tests {
let mut serializer = CompositeFastFieldSerializer::from_write(write).unwrap();
let mut fast_field_writers = FastFieldsWriter::from_schema(&SCHEMA);
for _ in 0..10_000 {
fast_field_writers.add_document(&doc!(*FIELD=>100_000u64));
fast_field_writers
.add_document(&doc!(*FIELD=>100_000u64))
.unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
@@ -278,7 +297,7 @@ mod tests {
serializer.close().unwrap();
}
let file = directory.open_read(path).unwrap();
assert_eq!(file.len(), 26);
assert_eq!(file.len(), 35);
{
let fast_fields_composite = CompositeFile::open(&file).unwrap();
let data = fast_fields_composite
@@ -303,9 +322,13 @@ mod tests {
let mut serializer = CompositeFastFieldSerializer::from_write(write).unwrap();
let mut fast_field_writers = FastFieldsWriter::from_schema(&SCHEMA);
// forcing the amplitude to be high
fast_field_writers.add_document(&doc!(*FIELD=>0u64));
fast_field_writers
.add_document(&doc!(*FIELD=>0u64))
.unwrap();
for i in 0u64..10_000u64 {
fast_field_writers.add_document(&doc!(*FIELD=>5_000_000_000_000_000_000u64 + i));
fast_field_writers
.add_document(&doc!(*FIELD=>5_000_000_000_000_000_000u64 + i))
.unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
@@ -313,7 +336,7 @@ mod tests {
serializer.close().unwrap();
}
let file = directory.open_read(path).unwrap();
assert_eq!(file.len(), 80040);
assert_eq!(file.len(), 80049);
{
let fast_fields_composite = CompositeFile::open(&file)?;
let data = fast_fields_composite
@@ -347,7 +370,7 @@ mod tests {
for i in -100i64..10_000i64 {
let mut doc = Document::default();
doc.add_i64(i64_field, i);
fast_field_writers.add_document(&doc);
fast_field_writers.add_document(&doc).unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
@@ -355,7 +378,7 @@ mod tests {
serializer.close().unwrap();
}
let file = directory.open_read(path).unwrap();
assert_eq!(file.len(), 40_usize);
assert_eq!(file.len(), 49_usize);
{
let fast_fields_composite = CompositeFile::open(&file)?;
@@ -368,7 +391,7 @@ mod tests {
assert_eq!(fast_field_reader.min_value(), -100i64);
assert_eq!(fast_field_reader.max_value(), 9_999i64);
for (doc, i) in (-100i64..10_000i64).enumerate() {
assert_eq!(fast_field_reader.get_val(doc as u64), i);
assert_eq!(fast_field_reader.get_val(doc as u32), i);
}
let mut buffer = vec![0i64; 100];
fast_field_reader.get_range(53, &mut buffer[..]);
@@ -392,7 +415,7 @@ mod tests {
let mut serializer = CompositeFastFieldSerializer::from_write(write).unwrap();
let mut fast_field_writers = FastFieldsWriter::from_schema(&schema);
let doc = Document::default();
fast_field_writers.add_document(&doc);
fast_field_writers.add_document(&doc).unwrap();
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
.unwrap();
@@ -435,7 +458,7 @@ mod tests {
let mut serializer = CompositeFastFieldSerializer::from_write(write)?;
let mut fast_field_writers = FastFieldsWriter::from_schema(&SCHEMA);
for &x in &permutation {
fast_field_writers.add_document(&doc!(*FIELD=>x));
fast_field_writers.add_document(&doc!(*FIELD=>x)).unwrap();
}
fast_field_writers.serialize(&mut serializer, &HashMap::new(), None)?;
serializer.close()?;
@@ -450,7 +473,7 @@ mod tests {
let fast_field_reader = open::<u64>(data)?;
for a in 0..n {
assert_eq!(fast_field_reader.get_val(a as u64), permutation[a as usize]);
assert_eq!(fast_field_reader.get_val(a as u32), permutation[a as usize]);
}
}
Ok(())
@@ -785,17 +808,21 @@ mod tests {
let write: WritePtr = directory.open_write(path).unwrap();
let mut serializer = CompositeFastFieldSerializer::from_write(write).unwrap();
let mut fast_field_writers = FastFieldsWriter::from_schema(&schema);
fast_field_writers.add_document(&doc!(field=>true));
fast_field_writers.add_document(&doc!(field=>false));
fast_field_writers.add_document(&doc!(field=>true));
fast_field_writers.add_document(&doc!(field=>false));
fast_field_writers.add_document(&doc!(field=>true)).unwrap();
fast_field_writers
.add_document(&doc!(field=>false))
.unwrap();
fast_field_writers.add_document(&doc!(field=>true)).unwrap();
fast_field_writers
.add_document(&doc!(field=>false))
.unwrap();
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
.unwrap();
serializer.close().unwrap();
}
let file = directory.open_read(path).unwrap();
assert_eq!(file.len(), 24);
assert_eq!(file.len(), 33);
let composite_file = CompositeFile::open(&file)?;
let data = composite_file.open_read(field).unwrap().read_bytes()?;
let fast_field_reader = open::<bool>(data)?;
@@ -822,8 +849,10 @@ mod tests {
let mut serializer = CompositeFastFieldSerializer::from_write(write).unwrap();
let mut fast_field_writers = FastFieldsWriter::from_schema(&schema);
for _ in 0..50 {
fast_field_writers.add_document(&doc!(field=>true));
fast_field_writers.add_document(&doc!(field=>false));
fast_field_writers.add_document(&doc!(field=>true)).unwrap();
fast_field_writers
.add_document(&doc!(field=>false))
.unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
@@ -831,7 +860,7 @@ mod tests {
serializer.close().unwrap();
}
let file = directory.open_read(path).unwrap();
assert_eq!(file.len(), 36);
assert_eq!(file.len(), 45);
let composite_file = CompositeFile::open(&file)?;
let data = composite_file.open_read(field).unwrap().read_bytes()?;
let fast_field_reader = open::<bool>(data)?;
@@ -857,13 +886,13 @@ mod tests {
let mut serializer = CompositeFastFieldSerializer::from_write(write)?;
let mut fast_field_writers = FastFieldsWriter::from_schema(&schema);
let doc = Document::default();
fast_field_writers.add_document(&doc);
fast_field_writers.add_document(&doc).unwrap();
fast_field_writers.serialize(&mut serializer, &HashMap::new(), None)?;
serializer.close()?;
}
let file = directory.open_read(path).unwrap();
let composite_file = CompositeFile::open(&file)?;
assert_eq!(file.len(), 23);
assert_eq!(file.len(), 32);
let data = composite_file.open_read(field).unwrap().read_bytes()?;
let fast_field_reader = open::<bool>(data)?;
assert_eq!(fast_field_reader.get_val(0), false);
@@ -883,7 +912,7 @@ mod tests {
CompositeFastFieldSerializer::from_write_with_codec(write, codec_types).unwrap();
let mut fast_field_writers = FastFieldsWriter::from_schema(schema);
for doc in docs {
fast_field_writers.add_document(doc);
fast_field_writers.add_document(doc).unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
@@ -897,10 +926,10 @@ mod tests {
pub fn test_gcd_date() -> crate::Result<()> {
let size_prec_sec =
test_gcd_date_with_codec(FastFieldCodecType::Bitpacked, DatePrecision::Seconds)?;
assert_eq!(size_prec_sec, 28 + (1_000 * 13) / 8); // 13 bits per val = ceil(log_2(number of seconds in 2hours);
assert_eq!(size_prec_sec, 5 + 4 + 28 + (1_000 * 13) / 8); // 13 bits per val = ceil(log_2(number of seconds in 2hours);
let size_prec_micro =
test_gcd_date_with_codec(FastFieldCodecType::Bitpacked, DatePrecision::Microseconds)?;
assert_eq!(size_prec_micro, 26 + (1_000 * 33) / 8); // 33 bits per val = ceil(log_2(number of microsecsseconds in 2hours);
assert_eq!(size_prec_micro, 5 + 4 + 26 + (1_000 * 33) / 8); // 33 bits per val = ceil(log_2(number of microsecsseconds in 2hours);
Ok(())
}
@@ -936,7 +965,7 @@ mod tests {
let test_fastfield = open::<DateTime>(file.read_bytes()?)?;
for (i, time) in times.iter().enumerate() {
assert_eq!(test_fastfield.get_val(i as u64), time.truncate(precision));
assert_eq!(test_fastfield.get_val(i as u32), time.truncate(precision));
}
Ok(len)
}

148
src/fastfield/multivalued/index.rs Normal file
View File

@@ -0,0 +1,148 @@
use std::ops::Range;
use std::sync::Arc;
use fastfield_codecs::Column;
use crate::DocId;
#[derive(Clone)]
/// Index to resolve value range for given doc_id.
/// Starts at 0.
pub struct MultiValueIndex {
idx: Arc<dyn Column<u64>>,
}
impl MultiValueIndex {
pub(crate) fn new(idx: Arc<dyn Column<u64>>) -> Self {
Self { idx }
}
/// Returns `[start, end)`, such that the values associated with
/// the given document are `start..end`.
#[inline]
pub(crate) fn range(&self, doc: DocId) -> Range<u32> {
let start = self.idx.get_val(doc) as u32;
let end = self.idx.get_val(doc + 1) as u32;
start..end
}
/// Given a range of documents, returns the Range of value offsets fo
/// these documents.
///
/// For instance, `given start_doc..end_doc`,
/// if we assume Document #start_doc end #end_doc both
/// have values, this function returns `start..end`
/// such that `value_column.get(start_doc)` is the first value of
/// `start_doc` (well, if there is one), and `value_column.get(end_doc - 1)`
/// is the last value of `end_doc`.
///
/// The passed end range is allowed to be out of bounds, in which case
/// it will be clipped to make it valid.
#[inline]
pub(crate) fn docid_range_to_position_range(&self, range: Range<DocId>) -> Range<u32> {
let end_docid = range.end.min(self.num_docs() - 1) + 1;
let start_docid = range.start.min(end_docid);
let start = self.idx.get_val(start_docid) as u32;
let end = self.idx.get_val(end_docid) as u32;
assert!(start <= end);
start..end
}
/// returns the num of values associated with a doc_id
pub(crate) fn num_vals_for_doc(&self, doc: DocId) -> u32 {
let range = self.range(doc);
range.end - range.start
}
/// Returns the overall number of values in this field.
#[inline]
pub fn total_num_vals(&self) -> u32 {
self.idx.max_value() as u32
}
/// Returns the number of documents in the index.
#[inline]
pub fn num_docs(&self) -> u32 {
self.idx.num_vals() - 1
}
/// Converts a list of positions of values in a 1:n index to the corresponding list of DocIds.
/// Positions are converted inplace to docids.
///
/// Since there is no index for value pos -> docid, but docid -> value pos range, we scan the
/// index.
///
/// Correctness: positions needs to be sorted. idx_reader needs to contain monotonically
/// increasing positions.
///
///
/// TODO: Instead of a linear scan we can employ a exponential search into binary search to
/// match a docid to its value position.
pub(crate) fn positions_to_docids(&self, doc_id_range: Range<u32>, positions: &mut Vec<u32>) {
if positions.is_empty() {
return;
}
let mut cur_doc = doc_id_range.start;
let mut last_doc = None;
assert!(self.idx.get_val(doc_id_range.start) as u32 <= positions[0]);
let mut write_doc_pos = 0;
for i in 0..positions.len() {
let pos = positions[i];
loop {
let end = self.idx.get_val(cur_doc + 1) as u32;
if end > pos {
positions[write_doc_pos] = cur_doc;
write_doc_pos += if last_doc == Some(cur_doc) { 0 } else { 1 };
last_doc = Some(cur_doc);
break;
}
cur_doc += 1;
}
}
positions.truncate(write_doc_pos);
}
}
#[cfg(test)]
mod tests {
use std::ops::Range;
use std::sync::Arc;
use fastfield_codecs::IterColumn;
use crate::fastfield::MultiValueIndex;
fn index_to_pos_helper(
index: &MultiValueIndex,
doc_id_range: Range<u32>,
positions: &[u32],
) -> Vec<u32> {
let mut positions = positions.to_vec();
index.positions_to_docids(doc_id_range, &mut positions);
positions
}
#[test]
fn test_positions_to_docid() {
let offsets = vec![0, 10, 12, 15, 22, 23]; // docid values are [0..10, 10..12, 12..15, etc.]
let column = IterColumn::from(offsets.into_iter());
let index = MultiValueIndex::new(Arc::new(column));
assert_eq!(index.num_docs(), 5);
{
let positions = vec![10u32, 11, 15, 20, 21, 22];
assert_eq!(index_to_pos_helper(&index, 0..5, &positions), vec![1, 3, 4]);
assert_eq!(index_to_pos_helper(&index, 1..5, &positions), vec![1, 3, 4]);
assert_eq!(index_to_pos_helper(&index, 0..5, &[9]), vec![0]);
assert_eq!(index_to_pos_helper(&index, 1..5, &[10]), vec![1]);
assert_eq!(index_to_pos_helper(&index, 1..5, &[11]), vec![1]);
assert_eq!(index_to_pos_helper(&index, 2..5, &[12]), vec![2]);
assert_eq!(index_to_pos_helper(&index, 2..5, &[12, 14]), vec![2]);
assert_eq!(index_to_pos_helper(&index, 2..5, &[12, 14, 15]), vec![2, 3]);
}
}
}

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

@@ -1,9 +1,23 @@
mod index;
mod reader;
mod writer;
pub use self::reader::MultiValuedFastFieldReader;
pub use self::writer::MultiValuedFastFieldWriter;
use fastfield_codecs::FastFieldCodecType;
pub use index::MultiValueIndex;
pub use self::reader::{MultiValuedFastFieldReader, MultiValuedU128FastFieldReader};
pub(crate) use self::writer::MultivalueStartIndex;
pub use self::writer::{MultiValueU128FastFieldWriter, MultiValuedFastFieldWriter};
/// The valid codecs for multivalue values excludes the linear interpolation codec.
///
/// This limitation is only valid for the values, not the offset index of the multivalue index.
pub(crate) fn get_fastfield_codecs_for_multivalue() -> [FastFieldCodecType; 2] {
[
FastFieldCodecType::Bitpacked,
FastFieldCodecType::BlockwiseLinear,
]
}
#[cfg(test)]
mod tests {
@@ -503,7 +517,7 @@ mod bench {
for val in block {
doc.add_u64(field, *val);
}
fast_field_writers.add_document(&doc);
fast_field_writers.add_document(&doc).unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
@@ -561,7 +575,7 @@ mod bench {
for val in block {
doc.add_u64(field, *val);
}
fast_field_writers.add_document(&doc);
fast_field_writers.add_document(&doc).unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), None)
@@ -594,7 +608,7 @@ mod bench {
for val in block {
doc.add_u64(field, *val);
}
fast_field_writers.add_document(&doc);
fast_field_writers.add_document(&doc).unwrap();
}
fast_field_writers
.serialize(&mut serializer, &HashMap::new(), Some(&doc_id_mapping))

161
src/fastfield/multivalued/reader.rs
View File

@@ -1,9 +1,10 @@
use std::ops::Range;
use std::ops::{Range, RangeInclusive};
use std::sync::Arc;
use fastfield_codecs::Column;
use fastfield_codecs::{Column, MonotonicallyMappableToU128};
use crate::fastfield::{FastValue, MultiValueLength};
use super::MultiValueIndex;
use crate::fastfield::FastValue;
use crate::DocId;
/// Reader for a multivalued `u64` fast field.
@@ -13,9 +14,10 @@ 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.
/// Stores the start position for each document.
#[derive(Clone)]
pub struct MultiValuedFastFieldReader<Item: FastValue> {
idx_reader: Arc<dyn Column<u64>>,
idx_reader: MultiValueIndex,
vals_reader: Arc<dyn Column<Item>>,
}
@@ -25,36 +27,32 @@ impl<Item: FastValue> MultiValuedFastFieldReader<Item> {
vals_reader: Arc<dyn Column<Item>>,
) -> MultiValuedFastFieldReader<Item> {
MultiValuedFastFieldReader {
idx_reader,
idx_reader: MultiValueIndex::new(idx_reader),
vals_reader,
}
}
/// Returns `[start, end)`, such that the values associated with
/// the given document are `start..end`.
#[inline]
fn range(&self, doc: DocId) -> Range<u64> {
let idx = doc as u64;
let start = self.idx_reader.get_val(idx);
let end = self.idx_reader.get_val(idx + 1);
start..end
}
/// Returns the array of values associated with the given `doc`.
#[inline]
fn get_vals_for_range(&self, range: Range<u64>, vals: &mut Vec<Item>) {
fn get_vals_for_range(&self, range: Range<u32>, vals: &mut Vec<Item>) {
let len = (range.end - range.start) as usize;
vals.resize(len, Item::make_zero());
self.vals_reader.get_range(range.start, &mut vals[..]);
self.vals_reader
.get_range(range.start as u64, &mut vals[..]);
}
/// Returns the array of values associated with the given `doc`.
#[inline]
pub fn get_vals(&self, doc: DocId, vals: &mut Vec<Item>) {
let range = self.range(doc);
let range = self.idx_reader.range(doc);
self.get_vals_for_range(range, vals);
}
/// returns the multivalue index
pub fn get_index_reader(&self) -> &MultiValueIndex {
&self.idx_reader
}
/// Returns the minimum value for this fast field.
///
/// The min value does not take in account of possible
@@ -75,30 +73,129 @@ impl<Item: FastValue> MultiValuedFastFieldReader<Item> {
/// Returns the number of values associated with the document `DocId`.
#[inline]
pub fn num_vals(&self, doc: DocId) -> usize {
let range = self.range(doc);
(range.end - range.start) as usize
pub fn num_vals(&self, doc: DocId) -> u32 {
self.idx_reader.num_vals_for_doc(doc)
}
/// Returns the overall number of values in this field .
/// Returns the overall number of values in this field.
#[inline]
pub fn total_num_vals(&self) -> u64 {
self.idx_reader.max_value()
pub fn total_num_vals(&self) -> u32 {
self.idx_reader.total_num_vals()
}
}
impl<Item: FastValue> MultiValueLength for MultiValuedFastFieldReader<Item> {
fn get_range(&self, doc_id: DocId) -> Range<u64> {
self.range(doc_id)
}
fn get_len(&self, doc_id: DocId) -> u64 {
self.num_vals(doc_id) as u64
/// Reader for a multivalued `u128` fast field.
///
/// The reader is implemented as a `u64` fast field for the index and a `u128` fast field.
///
/// 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 MultiValuedU128FastFieldReader<T: MonotonicallyMappableToU128> {
idx_reader: MultiValueIndex,
vals_reader: Arc<dyn Column<T>>,
}
impl<T: MonotonicallyMappableToU128> MultiValuedU128FastFieldReader<T> {
pub(crate) fn open(
idx_reader: Arc<dyn Column<u64>>,
vals_reader: Arc<dyn Column<T>>,
) -> MultiValuedU128FastFieldReader<T> {
Self {
idx_reader: MultiValueIndex::new(idx_reader),
vals_reader,
}
}
fn get_total_len(&self) -> u64 {
self.total_num_vals() as u64
/// Returns the array of values associated to the given `doc`.
#[inline]
pub fn get_first_val(&self, doc: DocId) -> Option<T> {
let range = self.idx_reader.range(doc);
if range.is_empty() {
return None;
}
Some(self.vals_reader.get_val(range.start))
}
/// Returns the array of values associated to the given `doc`.
#[inline]
fn get_vals_for_range(&self, range: Range<u32>, vals: &mut Vec<T>) {
let len = (range.end - range.start) as usize;
vals.resize(len, T::from_u128(0));
self.vals_reader
.get_range(range.start as u64, &mut vals[..]);
}
/// Returns the index reader
pub fn get_index_reader(&self) -> &MultiValueIndex {
&self.idx_reader
}
/// Returns the array of values associated to the given `doc`.
#[inline]
pub fn get_vals(&self, doc: DocId, vals: &mut Vec<T>) {
let range = self.idx_reader.range(doc);
self.get_vals_for_range(range, vals);
}
/// Iterates over all elements in the fast field
pub fn iter(&self) -> impl Iterator<Item = T> + '_ {
self.vals_reader.iter()
}
/// Returns the minimum value for this fast field.
///
/// The min value does not take in account of possible
/// deleted document, and should be considered as a lower bound
/// of the actual mimimum value.
pub fn min_value(&self) -> T {
self.vals_reader.min_value()
}
/// Returns the maximum value for this fast field.
///
/// The max value does not take in account of possible
/// deleted document, and should be considered as an upper bound
/// of the actual maximum value.
pub fn max_value(&self) -> T {
self.vals_reader.max_value()
}
/// Returns the number of values associated with the document `DocId`.
#[inline]
pub fn num_vals(&self, doc: DocId) -> u32 {
self.idx_reader.num_vals_for_doc(doc)
}
/// Returns the overall number of values in this field. It does not include deletes.
#[inline]
pub fn total_num_vals(&self) -> u32 {
assert_eq!(
self.vals_reader.num_vals(),
self.get_index_reader().total_num_vals()
);
self.idx_reader.total_num_vals()
}
/// Returns the docids matching given doc_id_range and value_range.
#[inline]
pub fn get_docids_for_value_range(
&self,
value_range: RangeInclusive<T>,
doc_id_range: Range<u32>,
positions: &mut Vec<u32>,
) {
let position_range = self
.get_index_reader()
.docid_range_to_position_range(doc_id_range.clone());
self.vals_reader
.get_docids_for_value_range(value_range, position_range, positions);
self.idx_reader.positions_to_docids(doc_id_range, positions);
}
}
#[cfg(test)]
mod tests {

266
src/fastfield/multivalued/writer.rs
View File

@@ -1,9 +1,12 @@
use std::io;
use std::sync::Mutex;
use fastfield_codecs::{Column, MonotonicallyMappableToU64, VecColumn};
use fnv::FnvHashMap;
use fastfield_codecs::{
Column, MonotonicallyMappableToU128, MonotonicallyMappableToU64, VecColumn,
};
use rustc_hash::FxHashMap;
use super::get_fastfield_codecs_for_multivalue;
use crate::fastfield::writer::unexpected_value;
use crate::fastfield::{value_to_u64, CompositeFastFieldSerializer, FastFieldType};
use crate::indexer::doc_id_mapping::DocIdMapping;
use crate::postings::UnorderedTermId;
@@ -79,11 +82,11 @@ impl MultiValuedFastFieldWriter {
/// Shift to the next document and adds
/// all of the matching field values present in the document.
pub fn add_document(&mut self, doc: &Document) {
pub fn add_document(&mut self, doc: &Document) -> crate::Result<()> {
self.next_doc();
// facets/texts are indexed in the `SegmentWriter` as we encode their unordered id.
if self.fast_field_type.is_storing_term_ids() {
return;
return Ok(());
}
for field_value in doc.field_values() {
if field_value.field == self.field {
@@ -92,11 +95,12 @@ impl MultiValuedFastFieldWriter {
(Some(precision), Value::Date(date_val)) => {
date_val.truncate(precision).to_u64()
}
_ => value_to_u64(value),
_ => value_to_u64(value)?,
};
self.add_val(value_u64);
}
}
Ok(())
}
/// Returns an iterator over values per doc_id in ascending doc_id order.
@@ -140,7 +144,7 @@ impl MultiValuedFastFieldWriter {
pub fn serialize(
mut self,
serializer: &mut CompositeFastFieldSerializer,
term_mapping_opt: Option<&FnvHashMap<UnorderedTermId, TermOrdinal>>,
term_mapping_opt: Option<&FxHashMap<UnorderedTermId, TermOrdinal>>,
doc_id_map: Option<&DocIdMapping>,
) -> io::Result<()> {
{
@@ -195,7 +199,12 @@ impl MultiValuedFastFieldWriter {
}
}
let col = VecColumn::from(&values[..]);
serializer.create_auto_detect_u64_fast_field_with_idx(self.field, col, 1)?;
serializer.create_auto_detect_u64_fast_field_with_idx_and_codecs(
self.field,
col,
1,
&get_fastfield_codecs_for_multivalue(),
)?;
}
Ok(())
}
@@ -204,112 +213,197 @@ impl MultiValuedFastFieldWriter {
pub(crate) struct MultivalueStartIndex<'a, C: Column> {
column: &'a C,
doc_id_map: &'a DocIdMapping,
min_max_opt: Mutex<Option<(u64, u64)>>,
random_seeker: Mutex<MultivalueStartIndexRandomSeeker<'a, C>>,
}
struct MultivalueStartIndexRandomSeeker<'a, C: Column> {
seek_head: MultivalueStartIndexIter<'a, C>,
seek_next_id: u64,
}
impl<'a, C: Column> MultivalueStartIndexRandomSeeker<'a, C> {
fn new(column: &'a C, doc_id_map: &'a DocIdMapping) -> Self {
Self {
seek_head: MultivalueStartIndexIter {
column,
doc_id_map,
new_doc_id: 0,
offset: 0u64,
},
seek_next_id: 0u64,
}
}
min: u64,
max: u64,
}
impl<'a, C: Column> MultivalueStartIndex<'a, C> {
pub fn new(column: &'a C, doc_id_map: &'a DocIdMapping) -> Self {
assert_eq!(column.num_vals(), doc_id_map.num_old_doc_ids() as u64 + 1);
assert_eq!(column.num_vals(), doc_id_map.num_old_doc_ids() as u32 + 1);
let (min, max) =
tantivy_bitpacker::minmax(iter_remapped_multivalue_index(doc_id_map, column))
.unwrap_or((0u64, 0u64));
MultivalueStartIndex {
column,
doc_id_map,
min_max_opt: Mutex::default(),
random_seeker: Mutex::new(MultivalueStartIndexRandomSeeker::new(column, doc_id_map)),
min,
max,
}
}
fn minmax(&self) -> (u64, u64) {
if let Some((min, max)) = *self.min_max_opt.lock().unwrap() {
return (min, max);
}
let (min, max) = tantivy_bitpacker::minmax(self.iter()).unwrap_or((0u64, 0u64));
*self.min_max_opt.lock().unwrap() = Some((min, max));
(min, max)
}
}
impl<'a, C: Column> Column for MultivalueStartIndex<'a, C> {
fn get_val(&self, idx: u64) -> u64 {
let mut random_seeker_lock = self.random_seeker.lock().unwrap();
if random_seeker_lock.seek_next_id > idx {
*random_seeker_lock =
MultivalueStartIndexRandomSeeker::new(self.column, self.doc_id_map);
}
let to_skip = idx - random_seeker_lock.seek_next_id;
random_seeker_lock.seek_next_id = idx + 1;
random_seeker_lock.seek_head.nth(to_skip as usize).unwrap()
fn get_val(&self, _idx: u32) -> u64 {
unimplemented!()
}
fn min_value(&self) -> u64 {
self.minmax().0
self.min
}
fn max_value(&self) -> u64 {
self.minmax().1
self.max
}
fn num_vals(&self) -> u64 {
(self.doc_id_map.num_new_doc_ids() + 1) as u64
fn num_vals(&self) -> u32 {
(self.doc_id_map.num_new_doc_ids() + 1) as u32
}
fn iter<'b>(&'b self) -> Box<dyn Iterator<Item = u64> + 'b> {
Box::new(MultivalueStartIndexIter::new(self.column, self.doc_id_map))
fn iter(&self) -> Box<dyn Iterator<Item = u64> + '_> {
Box::new(iter_remapped_multivalue_index(
self.doc_id_map,
&self.column,
))
}
}
struct MultivalueStartIndexIter<'a, C: Column> {
pub column: &'a C,
pub doc_id_map: &'a DocIdMapping,
pub new_doc_id: usize,
pub offset: u64,
fn iter_remapped_multivalue_index<'a, C: Column>(
doc_id_map: &'a DocIdMapping,
column: &'a C,
) -> impl Iterator<Item = u64> + 'a {
let mut offset = 0;
std::iter::once(0).chain(doc_id_map.iter_old_doc_ids().map(move |old_doc| {
let num_vals_for_doc = column.get_val(old_doc + 1) - column.get_val(old_doc);
offset += num_vals_for_doc;
offset as u64
}))
}
impl<'a, C: Column> MultivalueStartIndexIter<'a, C> {
fn new(column: &'a C, doc_id_map: &'a DocIdMapping) -> Self {
Self {
column,
doc_id_map,
new_doc_id: 0,
offset: 0,
/// Writer for multi-valued (as in, more than one value per document)
/// int fast field.
///
/// This `Writer` is only useful for advanced users.
/// The normal way to get your multivalued int in your index
/// is to
/// - declare your field with fast set to `Cardinality::MultiValues`
/// in your schema
/// - add your document simply by calling `.add_document(...)`.
///
/// The `MultiValuedFastFieldWriter` can be acquired from the
pub struct MultiValueU128FastFieldWriter {
field: Field,
vals: Vec<u128>,
doc_index: Vec<u64>,
}
impl MultiValueU128FastFieldWriter {
/// Creates a new `U128MultiValueFastFieldWriter`
pub(crate) fn new(field: Field) -> Self {
MultiValueU128FastFieldWriter {
field,
vals: Vec::new(),
doc_index: Vec::new(),
}
}
/// The memory used (inclusive childs)
pub fn mem_usage(&self) -> usize {
self.vals.capacity() * std::mem::size_of::<UnorderedTermId>()
+ self.doc_index.capacity() * std::mem::size_of::<u64>()
}
/// Finalize the current document.
pub(crate) fn next_doc(&mut self) {
self.doc_index.push(self.vals.len() as u64);
}
/// Pushes a new value to the current document.
pub(crate) fn add_val(&mut self, val: u128) {
self.vals.push(val);
}
/// Shift to the next document and adds
/// all of the matching field values present in the document.
pub fn add_document(&mut self, doc: &Document) -> crate::Result<()> {
self.next_doc();
for field_value in doc.field_values() {
if field_value.field == self.field {
let value = field_value.value();
let ip_addr = value
.as_ip_addr()
.ok_or_else(|| unexpected_value("ip", value))?;
let ip_addr_u128 = ip_addr.to_u128();
self.add_val(ip_addr_u128);
}
}
Ok(())
}
/// Returns an iterator over values per doc_id in ascending doc_id order.
///
/// Normally the order is simply iterating self.doc_id_index.
/// With doc_id_map it accounts for the new mapping, returning values in the order of the
/// new doc_ids.
fn get_ordered_values<'a: 'b, 'b>(
&'a self,
doc_id_map: Option<&'b DocIdMapping>,
) -> impl Iterator<Item = &'b [u128]> {
get_ordered_values(&self.vals, &self.doc_index, doc_id_map)
}
/// Serializes fast field values.
pub fn serialize(
mut self,
serializer: &mut CompositeFastFieldSerializer,
doc_id_map: Option<&DocIdMapping>,
) -> io::Result<()> {
{
// writing the offset index
//
self.doc_index.push(self.vals.len() as u64);
let col = VecColumn::from(&self.doc_index[..]);
if let Some(doc_id_map) = doc_id_map {
let multi_value_start_index = MultivalueStartIndex::new(&col, doc_id_map);
serializer.create_auto_detect_u64_fast_field_with_idx(
self.field,
multi_value_start_index,
0,
)?;
} else {
serializer.create_auto_detect_u64_fast_field_with_idx(self.field, col, 0)?;
}
}
{
let iter_gen = || self.get_ordered_values(doc_id_map).flatten().cloned();
serializer.create_u128_fast_field_with_idx(
self.field,
iter_gen,
self.vals.len() as u32,
1,
)?;
}
Ok(())
}
}
impl<'a, C: Column> Iterator for MultivalueStartIndexIter<'a, C> {
type Item = u64;
/// Returns an iterator over values per doc_id in ascending doc_id order.
///
/// Normally the order is simply iterating self.doc_id_index.
/// With doc_id_map it accounts for the new mapping, returning values in the order of the
/// new doc_ids.
fn get_ordered_values<'a: 'b, 'b, T>(
vals: &'a [T],
doc_index: &'a [u64],
doc_id_map: Option<&'b DocIdMapping>,
) -> impl Iterator<Item = &'b [T]> {
let doc_id_iter: Box<dyn Iterator<Item = u32>> = if let Some(doc_id_map) = doc_id_map {
Box::new(doc_id_map.iter_old_doc_ids())
} else {
let max_doc = doc_index.len() as DocId;
Box::new(0..max_doc)
};
doc_id_iter.map(move |doc_id| get_values_for_doc_id(doc_id, vals, doc_index))
}
fn next(&mut self) -> Option<Self::Item> {
if self.new_doc_id > self.doc_id_map.num_new_doc_ids() {
return None;
}
let new_doc_id = self.new_doc_id;
self.new_doc_id += 1;
let start_offset = self.offset;
if new_doc_id < self.doc_id_map.num_new_doc_ids() {
let old_doc = self.doc_id_map.get_old_doc_id(new_doc_id as u32) as u64;
let num_vals_for_doc = self.column.get_val(old_doc + 1) - self.column.get_val(old_doc);
self.offset += num_vals_for_doc;
}
Some(start_offset)
}
/// returns all values for a doc_id
fn get_values_for_doc_id<'a, T>(doc_id: u32, vals: &'a [T], doc_index: &'a [u64]) -> &'a [T] {
let start_pos = doc_index[doc_id as usize] as usize;
let end_pos = doc_index
.get(doc_id as usize + 1)
.cloned()
.unwrap_or(vals.len() as u64) as usize; // special case, last doc_id has no offset information
&vals[start_pos..end_pos]
}
#[cfg(test)]
@@ -344,11 +438,5 @@ mod tests {
vec![0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55]
);
assert_eq!(multivalue_start_index.num_vals(), 11);
assert_eq!(multivalue_start_index.get_val(3), 2);
assert_eq!(multivalue_start_index.get_val(5), 5);
assert_eq!(multivalue_start_index.get_val(8), 21);
assert_eq!(multivalue_start_index.get_val(4), 3);
assert_eq!(multivalue_start_index.get_val(0), 0);
assert_eq!(multivalue_start_index.get_val(10), 55);
}
}

61
src/fastfield/readers.rs
View File

@@ -1,7 +1,9 @@
use std::net::Ipv6Addr;
use std::sync::Arc;
use fastfield_codecs::{open, Column};
use fastfield_codecs::{open, open_u128, Column};
use super::multivalued::MultiValuedU128FastFieldReader;
use crate::directory::{CompositeFile, FileSlice};
use crate::fastfield::{
BytesFastFieldReader, FastFieldNotAvailableError, FastValue, MultiValuedFastFieldReader,
@@ -23,6 +25,7 @@ pub struct FastFieldReaders {
pub(crate) enum FastType {
I64,
U64,
U128,
F64,
Bool,
Date,
@@ -49,6 +52,9 @@ pub(crate) fn type_and_cardinality(field_type: &FieldType) -> Option<(FastType,
FieldType::Str(options) if options.is_fast() => {
Some((FastType::U64, Cardinality::MultiValues))
}
FieldType::IpAddr(options) => options
.get_fastfield_cardinality()
.map(|cardinality| (FastType::U128, cardinality)),
_ => None,
}
}
@@ -143,6 +149,59 @@ impl FastFieldReaders {
self.typed_fast_field_reader(field)
}
/// Returns the `ip` fast field reader reader associated to `field`.
///
/// If `field` is not a u128 fast field, this method returns an Error.
pub fn ip_addr(&self, field: Field) -> crate::Result<Arc<dyn Column<Ipv6Addr>>> {
self.check_type(field, FastType::U128, Cardinality::SingleValue)?;
let bytes = self.fast_field_data(field, 0)?.read_bytes()?;
Ok(open_u128::<Ipv6Addr>(bytes)?)
}
/// Returns the `ip` fast field reader reader associated to `field`.
///
/// If `field` is not a u128 fast field, this method returns an Error.
pub fn ip_addrs(
&self,
field: Field,
) -> crate::Result<MultiValuedU128FastFieldReader<Ipv6Addr>> {
self.check_type(field, FastType::U128, Cardinality::MultiValues)?;
let idx_reader: Arc<dyn Column<u64>> = self.typed_fast_field_reader(field)?;
let bytes = self.fast_field_data(field, 1)?.read_bytes()?;
let vals_reader = open_u128::<Ipv6Addr>(bytes)?;
Ok(MultiValuedU128FastFieldReader::open(
idx_reader,
vals_reader,
))
}
/// Returns the `u128` fast field reader reader associated to `field`.
///
/// If `field` is not a u128 fast field, this method returns an Error.
pub(crate) fn u128(&self, field: Field) -> crate::Result<Arc<dyn Column<u128>>> {
self.check_type(field, FastType::U128, Cardinality::SingleValue)?;
let bytes = self.fast_field_data(field, 0)?.read_bytes()?;
Ok(open_u128::<u128>(bytes)?)
}
/// Returns the `u128` multi-valued fast field reader reader associated to `field`.
///
/// If `field` is not a u128 multi-valued fast field, this method returns an Error.
pub fn u128s(&self, field: Field) -> crate::Result<MultiValuedU128FastFieldReader<u128>> {
self.check_type(field, FastType::U128, Cardinality::MultiValues)?;
let idx_reader: Arc<dyn Column<u64>> = self.typed_fast_field_reader(field)?;
let bytes = self.fast_field_data(field, 1)?.read_bytes()?;
let vals_reader = open_u128::<u128>(bytes)?;
Ok(MultiValuedU128FastFieldReader::open(
idx_reader,
vals_reader,
))
}
/// Returns the `u64` fast field reader reader associated with `field`, regardless of whether
/// the given field is effectively of type `u64` or not.
///

29
src/fastfield/serializer/mod.rs
View File

@@ -70,6 +70,35 @@ impl CompositeFastFieldSerializer {
Ok(())
}
/// Serialize data into a new u64 fast field. The best compression codec of the the provided
/// will be chosen.
pub fn create_auto_detect_u64_fast_field_with_idx_and_codecs<T: MonotonicallyMappableToU64>(
&mut self,
field: Field,
fastfield_accessor: impl Column<T>,
idx: usize,
codec_types: &[FastFieldCodecType],
) -> io::Result<()> {
let field_write = self.composite_write.for_field_with_idx(field, idx);
fastfield_codecs::serialize(fastfield_accessor, field_write, codec_types)?;
Ok(())
}
/// Serialize data into a new u128 fast field. The codec will be compact space compressor,
/// which is optimized for scanning the fast field for a given range.
pub fn create_u128_fast_field_with_idx<F: Fn() -> I, I: Iterator<Item = u128>>(
&mut self,
field: Field,
iter_gen: F,
num_vals: u32,
idx: usize,
) -> io::Result<()> {
let field_write = self.composite_write.for_field_with_idx(field, idx);
fastfield_codecs::serialize_u128(iter_gen, num_vals, field_write)?;
Ok(())
}
/// Start serializing a new [u8] fast field. Use the returned writer to write data into the
/// bytes field. To associate the bytes with documents a seperate index must be created on
/// index 0. See bytes/writer.rs::serialize for an example.

190
src/fastfield/writer.rs
View File

@@ -2,11 +2,11 @@ use std::collections::HashMap;
use std::io;
use common;
use fastfield_codecs::{Column, MonotonicallyMappableToU64};
use fnv::FnvHashMap;
use fastfield_codecs::{Column, MonotonicallyMappableToU128, MonotonicallyMappableToU64};
use rustc_hash::FxHashMap;
use tantivy_bitpacker::BlockedBitpacker;
use super::multivalued::MultiValuedFastFieldWriter;
use super::multivalued::{MultiValueU128FastFieldWriter, MultiValuedFastFieldWriter};
use super::FastFieldType;
use crate::fastfield::{BytesFastFieldWriter, CompositeFastFieldSerializer};
use crate::indexer::doc_id_mapping::DocIdMapping;
@@ -19,10 +19,19 @@ use crate::DatePrecision;
pub struct FastFieldsWriter {
term_id_writers: Vec<MultiValuedFastFieldWriter>,
single_value_writers: Vec<IntFastFieldWriter>,
u128_value_writers: Vec<U128FastFieldWriter>,
u128_multi_value_writers: Vec<MultiValueU128FastFieldWriter>,
multi_values_writers: Vec<MultiValuedFastFieldWriter>,
bytes_value_writers: Vec<BytesFastFieldWriter>,
}
pub(crate) fn unexpected_value(expected: &str, actual: &Value) -> crate::TantivyError {
crate::TantivyError::SchemaError(format!(
"Expected a {:?} in fast field, but got {:?}",
expected, actual
))
}
fn fast_field_default_value(field_entry: &FieldEntry) -> u64 {
match *field_entry.field_type() {
FieldType::I64(_) | FieldType::Date(_) => common::i64_to_u64(0i64),
@@ -34,6 +43,8 @@ fn fast_field_default_value(field_entry: &FieldEntry) -> u64 {
impl FastFieldsWriter {
/// Create all `FastFieldWriter` required by the schema.
pub fn from_schema(schema: &Schema) -> FastFieldsWriter {
let mut u128_value_writers = Vec::new();
let mut u128_multi_value_writers = Vec::new();
let mut single_value_writers = Vec::new();
let mut term_id_writers = Vec::new();
let mut multi_values_writers = Vec::new();
@@ -97,10 +108,27 @@ impl FastFieldsWriter {
bytes_value_writers.push(fast_field_writer);
}
}
FieldType::IpAddr(opt) => {
if opt.is_fast() {
match opt.get_fastfield_cardinality() {
Some(Cardinality::SingleValue) => {
let fast_field_writer = U128FastFieldWriter::new(field);
u128_value_writers.push(fast_field_writer);
}
Some(Cardinality::MultiValues) => {
let fast_field_writer = MultiValueU128FastFieldWriter::new(field);
u128_multi_value_writers.push(fast_field_writer);
}
None => {}
}
}
}
FieldType::Str(_) | FieldType::JsonObject(_) => {}
}
}
FastFieldsWriter {
u128_value_writers,
u128_multi_value_writers,
term_id_writers,
single_value_writers,
multi_values_writers,
@@ -129,6 +157,16 @@ impl FastFieldsWriter {
.iter()
.map(|w| w.mem_usage())
.sum::<usize>()
+ self
.u128_value_writers
.iter()
.map(|w| w.mem_usage())
.sum::<usize>()
+ self
.u128_multi_value_writers
.iter()
.map(|w| w.mem_usage())
.sum::<usize>()
}
/// Get the `FastFieldWriter` associated with a field.
@@ -190,21 +228,27 @@ impl FastFieldsWriter {
.iter_mut()
.find(|field_writer| field_writer.field() == field)
}
/// Indexes all of the fastfields of a new document.
pub fn add_document(&mut self, doc: &Document) {
pub fn add_document(&mut self, doc: &Document) -> crate::Result<()> {
for field_writer in &mut self.term_id_writers {
field_writer.add_document(doc);
field_writer.add_document(doc)?;
}
for field_writer in &mut self.single_value_writers {
field_writer.add_document(doc);
field_writer.add_document(doc)?;
}
for field_writer in &mut self.multi_values_writers {
field_writer.add_document(doc);
field_writer.add_document(doc)?;
}
for field_writer in &mut self.bytes_value_writers {
field_writer.add_document(doc);
field_writer.add_document(doc)?;
}
for field_writer in &mut self.u128_value_writers {
field_writer.add_document(doc)?;
}
for field_writer in &mut self.u128_multi_value_writers {
field_writer.add_document(doc)?;
}
Ok(())
}
/// Serializes all of the `FastFieldWriter`s by pushing them in
@@ -212,7 +256,7 @@ impl FastFieldsWriter {
pub fn serialize(
self,
serializer: &mut CompositeFastFieldSerializer,
mapping: &HashMap<Field, FnvHashMap<UnorderedTermId, TermOrdinal>>,
mapping: &HashMap<Field, FxHashMap<UnorderedTermId, TermOrdinal>>,
doc_id_map: Option<&DocIdMapping>,
) -> io::Result<()> {
for field_writer in self.term_id_writers {
@@ -230,6 +274,108 @@ impl FastFieldsWriter {
for field_writer in self.bytes_value_writers {
field_writer.serialize(serializer, doc_id_map)?;
}
for field_writer in self.u128_value_writers {
field_writer.serialize(serializer, doc_id_map)?;
}
for field_writer in self.u128_multi_value_writers {
field_writer.serialize(serializer, doc_id_map)?;
}
Ok(())
}
}
/// Fast field writer for u128 values.
/// The fast field writer just keeps the values in memory.
///
/// Only when the segment writer can be closed and
/// persisted on disk, the fast field writer is
/// sent to a `FastFieldSerializer` via the `.serialize(...)`
/// method.
///
/// We cannot serialize earlier as the values are
/// compressed to a compact number space and the number of
/// bits required for bitpacking can only been known once
/// we have seen all of the values.
pub struct U128FastFieldWriter {
field: Field,
vals: Vec<u128>,
val_count: u32,
}
impl U128FastFieldWriter {
/// Creates a new `IntFastFieldWriter`
pub fn new(field: Field) -> Self {
Self {
field,
vals: vec![],
val_count: 0,
}
}
/// The memory used (inclusive childs)
pub fn mem_usage(&self) -> usize {
self.vals.len() * 16
}
/// Records a new value.
///
/// The n-th value being recorded is implicitely
/// associated to the document with the `DocId` n.
/// (Well, `n-1` actually because of 0-indexing)
pub fn add_val(&mut self, val: u128) {
self.vals.push(val);
}
/// Extract the fast field value from the document
/// (or use the default value) and records it.
///
/// Extract the value associated to the fast field for
/// this document.
pub fn add_document(&mut self, doc: &Document) -> crate::Result<()> {
match doc.get_first(self.field) {
Some(v) => {
let ip_addr = v.as_ip_addr().ok_or_else(|| unexpected_value("ip", v))?;
let value = ip_addr.to_u128();
self.add_val(value);
}
None => {
self.add_val(0); // TODO fix null handling
}
};
self.val_count += 1;
Ok(())
}
/// Push the fast fields value to the `FastFieldWriter`.
pub fn serialize(
&self,
serializer: &mut CompositeFastFieldSerializer,
doc_id_map: Option<&DocIdMapping>,
) -> io::Result<()> {
if let Some(doc_id_map) = doc_id_map {
let iter_gen = || {
doc_id_map
.iter_old_doc_ids()
.map(|idx| self.vals[idx as usize])
};
serializer.create_u128_fast_field_with_idx(
self.field,
iter_gen,
self.val_count as u32,
0,
)?;
} else {
let iter_gen = || self.vals.iter().cloned();
serializer.create_u128_fast_field_with_idx(
self.field,
iter_gen,
self.val_count as u32,
0,
)?;
}
Ok(())
}
}
@@ -238,7 +384,7 @@ impl FastFieldsWriter {
/// The fast field writer just keeps the values in memory.
///
/// Only when the segment writer can be closed and
/// persisted on disc, the fast field writer is
/// persisted on disk, the fast field writer is
/// sent to a `FastFieldSerializer` via the `.serialize(...)`
/// method.
///
@@ -325,14 +471,14 @@ impl IntFastFieldWriter {
/// only the first one is taken in account.
///
/// Values on text fast fields are skipped.
pub fn add_document(&mut self, doc: &Document) {
pub fn add_document(&mut self, doc: &Document) -> crate::Result<()> {
match doc.get_first(self.field) {
Some(v) => {
let value = match (self.precision_opt, v) {
(Some(precision), Value::Date(date_val)) => {
date_val.truncate(precision).to_u64()
}
_ => super::value_to_u64(v),
_ => super::value_to_u64(v)?,
};
self.add_val(value);
}
@@ -340,6 +486,7 @@ impl IntFastFieldWriter {
self.add_val(self.val_if_missing);
}
};
Ok(())
}
/// get iterator over the data
@@ -364,7 +511,7 @@ impl IntFastFieldWriter {
vals: &self.vals,
min_value: min,
max_value: max,
num_vals: self.val_count as u64,
num_vals: self.val_count as u32,
};
serializer.create_auto_detect_u64_fast_field(self.field, fastfield_accessor)?;
@@ -379,7 +526,7 @@ struct WriterFastFieldAccessProvider<'map, 'bitp> {
vals: &'bitp BlockedBitpacker,
min_value: u64,
max_value: u64,
num_vals: u64,
num_vals: u32,
}
impl<'map, 'bitp> Column for WriterFastFieldAccessProvider<'map, 'bitp> {
@@ -391,15 +538,8 @@ impl<'map, 'bitp> Column for WriterFastFieldAccessProvider<'map, 'bitp> {
/// # Panics
///
/// May panic if `doc` is greater than the index.
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
} else {
self.vals.get(doc as usize)
}
fn get_val(&self, _doc: u32) -> u64 {
unimplemented!()
}
fn iter(&self) -> Box<dyn Iterator<Item = u64> + '_> {
@@ -422,7 +562,7 @@ impl<'map, 'bitp> Column for WriterFastFieldAccessProvider<'map, 'bitp> {
self.max_value
}
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.num_vals
}
}

6
src/fieldnorm/mod.rs
View File

@@ -34,7 +34,7 @@ mod tests {
use crate::directory::{CompositeFile, Directory, RamDirectory, WritePtr};
use crate::fieldnorm::{FieldNormReader, FieldNormsSerializer, FieldNormsWriter};
use crate::query::{Query, TermQuery};
use crate::query::{EnableScoring, Query, TermQuery};
use crate::schema::{
Field, IndexRecordOption, Schema, TextFieldIndexing, TextOptions, STORED, TEXT,
};
@@ -112,7 +112,7 @@ mod tests {
Term::from_field_text(text, "hello"),
IndexRecordOption::WithFreqs,
);
let weight = query.weight(&searcher, true)?;
let weight = query.weight(EnableScoring::Enabled(&searcher))?;
let mut scorer = weight.scorer(searcher.segment_reader(0), 1.0f32)?;
assert_eq!(scorer.doc(), 0);
assert!((scorer.score() - 0.22920431).abs() < 0.001f32);
@@ -141,7 +141,7 @@ mod tests {
Term::from_field_text(text, "hello"),
IndexRecordOption::WithFreqs,
);
let weight = query.weight(&searcher, true)?;
let weight = query.weight(EnableScoring::Enabled(&searcher))?;
let mut scorer = weight.scorer(searcher.segment_reader(0), 1.0f32)?;
assert_eq!(scorer.doc(), 0);
assert!((scorer.score() - 0.22920431).abs() < 0.001f32);

2
src/fieldnorm/writer.rs
View File

@@ -9,7 +9,7 @@ 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
/// `FieldNormsWriter` stores a `Vec<u8>` for each tracked field, using a
/// byte per document per field.
pub struct FieldNormsWriter {
fieldnorms_buffers: Vec<Option<Vec<u8>>>,

4
src/indexer/doc_id_mapping.rs
View File

@@ -34,10 +34,6 @@ impl SegmentDocIdMapping {
self.new_doc_id_to_old_doc_addr.len()
}
pub(crate) fn get_old_doc_addr(&self, new_doc_id: DocId) -> DocAddress {
self.new_doc_id_to_old_doc_addr[new_doc_id as usize]
}
/// This flags means the segments are simply stacked in the order of their ordinal.
/// e.g. [(0, 1), .. (n, 1), (0, 2)..., (m, 2)]
///

69
src/indexer/flat_map_with_buffer.rs Normal file
View File

@@ -0,0 +1,69 @@
pub struct FlatMapWithBuffer<T, F, Iter> {
buffer: Vec<T>,
fill_buffer: F,
underlying_it: Iter,
}
impl<T, F, Iter, I> Iterator for FlatMapWithBuffer<T, F, Iter>
where
Iter: Iterator<Item = I>,
F: Fn(I, &mut Vec<T>),
{
type Item = T;
fn next(&mut self) -> Option<Self::Item> {
while self.buffer.is_empty() {
let next_el = self.underlying_it.next()?;
(self.fill_buffer)(next_el, &mut self.buffer);
// We will pop elements, so we reverse the buffer first.
self.buffer.reverse();
}
self.buffer.pop()
}
}
pub trait FlatMapWithBufferIter: Iterator {
/// Function similar to `flat_map`, but allows reusing a shared `Vec`.
fn flat_map_with_buffer<F, T>(self, fill_buffer: F) -> FlatMapWithBuffer<T, F, Self>
where
F: Fn(Self::Item, &mut Vec<T>),
Self: Sized,
{
FlatMapWithBuffer {
buffer: Vec::with_capacity(10),
fill_buffer,
underlying_it: self,
}
}
}
impl<T: ?Sized> FlatMapWithBufferIter for T where T: Iterator {}
#[cfg(test)]
mod tests {
use crate::indexer::flat_map_with_buffer::FlatMapWithBufferIter;
#[test]
fn test_flat_map_with_buffer_empty() {
let mut empty_iter = std::iter::empty::<usize>()
.flat_map_with_buffer(|_val: usize, _buffer: &mut Vec<usize>| {});
assert!(empty_iter.next().is_none());
}
#[test]
fn test_flat_map_with_buffer_simple() {
let vals: Vec<usize> = (1..5)
.flat_map_with_buffer(|val: usize, buffer: &mut Vec<usize>| buffer.extend(0..val))
.collect();
assert_eq!(&[0, 0, 1, 0, 1, 2, 0, 1, 2, 3], &vals[..]);
}
#[test]
fn test_flat_map_filling_no_elements_does_not_stop_iterator() {
let vals: Vec<usize> = [2, 0, 0, 3]
.into_iter()
.flat_map_with_buffer(|val: usize, buffer: &mut Vec<usize>| buffer.extend(0..val))
.collect();
assert_eq!(&[0, 1, 0, 1, 2], &vals[..]);
}
}

577
src/indexer/index_writer.rs
View File

@@ -19,9 +19,9 @@ use crate::indexer::index_writer_status::IndexWriterStatus;
use crate::indexer::operation::DeleteOperation;
use crate::indexer::stamper::Stamper;
use crate::indexer::{MergePolicy, SegmentEntry, SegmentWriter};
use crate::query::{Query, TermQuery};
use crate::query::{EnableScoring, Query, TermQuery};
use crate::schema::{Document, IndexRecordOption, Term};
use crate::{FutureResult, IndexReader, Opstamp};
use crate::{FutureResult, Opstamp};
// 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.
@@ -57,7 +57,6 @@ pub struct IndexWriter {
_directory_lock: Option<DirectoryLock>,
index: Index,
index_reader: IndexReader,
memory_arena_in_bytes_per_thread: usize,
@@ -95,7 +94,7 @@ fn compute_deleted_bitset(
// document that were inserted before it.
delete_op
.target
.for_each(segment_reader, &mut |doc_matching_delete_query, _| {
.for_each_no_score(segment_reader, &mut |doc_matching_delete_query| {
if doc_opstamps.is_deleted(doc_matching_delete_query, delete_op.opstamp) {
alive_bitset.remove(doc_matching_delete_query);
might_have_changed = true;
@@ -298,8 +297,6 @@ impl IndexWriter {
memory_arena_in_bytes_per_thread,
index: index.clone(),
index_reader: index.reader()?,
index_writer_status: IndexWriterStatus::from(document_receiver),
operation_sender: document_sender,
@@ -681,8 +678,7 @@ impl IndexWriter {
/// only after calling `commit()`.
#[doc(hidden)]
pub fn delete_query(&self, query: Box<dyn Query>) -> crate::Result<Opstamp> {
let weight = query.weight(&self.index_reader.searcher(), false)?;
let weight = query.weight(EnableScoring::Disabled(&self.index.schema()))?;
let opstamp = self.stamper.stamp();
let delete_operation = DeleteOperation {
opstamp,
@@ -763,8 +759,7 @@ impl IndexWriter {
match user_op {
UserOperation::Delete(term) => {
let query = TermQuery::new(term, IndexRecordOption::Basic);
let weight = query.weight(&self.index_reader.searcher(), false)?;
let weight = query.weight(EnableScoring::Disabled(&self.index.schema()))?;
let delete_operation = DeleteOperation {
opstamp,
target: weight,
@@ -803,7 +798,9 @@ impl Drop for IndexWriter {
#[cfg(test)]
mod tests {
use std::collections::{HashMap, HashSet};
use std::net::Ipv6Addr;
use fastfield_codecs::MonotonicallyMappableToU128;
use proptest::prelude::*;
use proptest::prop_oneof;
use proptest::strategy::Strategy;
@@ -815,11 +812,13 @@ mod tests {
use crate::indexer::NoMergePolicy;
use crate::query::{BooleanQuery, Occur, Query, QueryParser, TermQuery};
use crate::schema::{
self, Cardinality, Facet, FacetOptions, IndexRecordOption, NumericOptions,
self, Cardinality, Facet, FacetOptions, IndexRecordOption, IpAddrOptions, NumericOptions,
TextFieldIndexing, TextOptions, FAST, INDEXED, STORED, STRING, TEXT,
};
use crate::store::DOCSTORE_CACHE_CAPACITY;
use crate::{DocAddress, Index, IndexSettings, IndexSortByField, Order, ReloadPolicy, Term};
use crate::{
DateTime, 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 \
@@ -1395,6 +1394,35 @@ mod tests {
assert!(commit_again.is_ok());
}
#[test]
fn test_sort_by_multivalue_field_error() -> crate::Result<()> {
let mut schema_builder = schema::Schema::builder();
let options = NumericOptions::default().set_fast(Cardinality::MultiValues);
schema_builder.add_u64_field("id", options);
let schema = schema_builder.build();
let settings = IndexSettings {
sort_by_field: Some(IndexSortByField {
field: "id".to_string(),
order: Order::Desc,
}),
..Default::default()
};
let err = Index::builder()
.schema(schema)
.settings(settings)
.create_in_ram()
.unwrap_err();
assert_eq!(
err.to_string(),
"An invalid argument was passed: 'Only single value fast field Cardinality supported \
for sorting index id'"
);
Ok(())
}
#[test]
fn test_delete_with_sort_by_field() -> crate::Result<()> {
let mut schema_builder = schema::Schema::builder();
@@ -1439,7 +1467,7 @@ mod tests {
let fast_field_reader = segment_reader.fast_fields().u64(id_field)?;
let in_order_alive_ids: Vec<u64> = segment_reader
.doc_ids_alive()
.map(|doc| fast_field_reader.get_val(doc as u64))
.map(|doc| fast_field_reader.get_val(doc))
.collect();
assert_eq!(&in_order_alive_ids[..], &[9, 8, 7, 6, 5, 4, 1, 0]);
Ok(())
@@ -1500,7 +1528,7 @@ mod tests {
let fast_field_reader = segment_reader.fast_fields().u64(id_field)?;
let in_order_alive_ids: Vec<u64> = segment_reader
.doc_ids_alive()
.map(|doc| fast_field_reader.get_val(doc as u64))
.map(|doc| fast_field_reader.get_val(doc))
.collect();
assert_eq!(&in_order_alive_ids[..], &[9, 8, 7, 6, 5, 4, 2, 0]);
Ok(())
@@ -1558,13 +1586,42 @@ mod tests {
(existing_ids, deleted_ids)
}
fn get_id_list(ops: &[IndexingOp]) -> Vec<u64> {
let mut id_list = Vec::new();
for &op in ops {
match op {
IndexingOp::AddDoc { id } => {
id_list.push(id);
}
IndexingOp::DeleteDoc { id } => {
id_list.retain(|el| *el != id);
}
IndexingOp::DeleteDocQuery { id } => {
id_list.retain(|el| *el != id);
}
_ => {}
}
}
id_list
}
fn test_operation_strategy(
ops: &[IndexingOp],
sort_index: bool,
force_end_merge: bool,
) -> crate::Result<()> {
let mut schema_builder = schema::Schema::builder();
let ip_field = schema_builder.add_ip_addr_field("ip", FAST | INDEXED | STORED);
let ips_field = schema_builder.add_ip_addr_field(
"ips",
IpAddrOptions::default()
.set_fast(Cardinality::MultiValues)
.set_indexed(),
);
let id_field = schema_builder.add_u64_field("id", FAST | INDEXED | STORED);
let i64_field = schema_builder.add_i64_field("i64", INDEXED);
let f64_field = schema_builder.add_f64_field("f64", INDEXED);
let date_field = schema_builder.add_date_field("date", INDEXED);
let bytes_field = schema_builder.add_bytes_field("bytes", FAST | INDEXED | STORED);
let bool_field = schema_builder.add_bool_field("bool", FAST | INDEXED | STORED);
let text_field = schema_builder.add_text_field(
@@ -1578,6 +1635,7 @@ mod tests {
);
let large_text_field = schema_builder.add_text_field("large_text_field", TEXT | STORED);
let multi_text_fields = schema_builder.add_text_field("multi_text_fields", TEXT | STORED);
let multi_numbers = schema_builder.add_u64_field(
"multi_numbers",
@@ -1615,21 +1673,63 @@ mod tests {
let old_reader = index.reader()?;
let ip_exists = |id| id % 3 != 0; // 0 does not exist
let multi_text_field_text1 = "test1 test2 test3 test1 test2 test3";
// rotate left
let multi_text_field_text2 = "test2 test3 test1 test2 test3 test1";
// rotate right
let multi_text_field_text3 = "test3 test1 test2 test3 test1 test2";
let ip_from_id = |id| Ipv6Addr::from_u128(id as u128);
for &op in ops {
match op {
IndexingOp::AddDoc { id } => {
let facet = Facet::from(&("/cola/".to_string() + &id.to_string()));
index_writer.add_document(doc!(id_field=>id,
bytes_field => id.to_le_bytes().as_slice(),
multi_numbers=> id,
multi_numbers => id,
bool_field => (id % 2u64) != 0,
multi_bools => (id % 2u64) != 0,
multi_bools => (id % 2u64) == 0,
text_field => id.to_string(),
facet_field => facet,
large_text_field=> LOREM
))?;
let ip = ip_from_id(id);
if !ip_exists(id) {
// every 3rd doc has no ip field
index_writer.add_document(doc!(id_field=>id,
bytes_field => id.to_le_bytes().as_slice(),
multi_numbers=> id,
multi_numbers => id,
bool_field => (id % 2u64) != 0,
i64_field => id as i64,
f64_field => id as f64,
date_field => DateTime::from_timestamp_secs(id as i64),
multi_bools => (id % 2u64) != 0,
multi_bools => (id % 2u64) == 0,
text_field => id.to_string(),
facet_field => facet,
large_text_field => LOREM,
multi_text_fields => multi_text_field_text1,
multi_text_fields => multi_text_field_text2,
multi_text_fields => multi_text_field_text3,
))?;
} else {
index_writer.add_document(doc!(id_field=>id,
bytes_field => id.to_le_bytes().as_slice(),
ip_field => ip,
ips_field => ip,
ips_field => ip,
multi_numbers=> id,
multi_numbers => id,
bool_field => (id % 2u64) != 0,
i64_field => id as i64,
f64_field => id as f64,
date_field => DateTime::from_timestamp_secs(id as i64),
multi_bools => (id % 2u64) != 0,
multi_bools => (id % 2u64) == 0,
text_field => id.to_string(),
facet_field => facet,
large_text_field => LOREM,
multi_text_fields => multi_text_field_text1,
multi_text_fields => multi_text_field_text2,
multi_text_fields => multi_text_field_text3,
))?;
}
}
IndexingOp::DeleteDoc { id } => {
index_writer.delete_term(Term::from_field_u64(id_field, id));
@@ -1656,6 +1756,7 @@ mod tests {
index_writer.commit()?;
let searcher = index.reader()?.searcher();
let num_segments_before_merge = searcher.segment_readers().len();
if force_end_merge {
index_writer.wait_merging_threads()?;
let mut index_writer = index.writer_for_tests()?;
@@ -1667,6 +1768,7 @@ mod tests {
assert!(index_writer.wait_merging_threads().is_ok());
}
}
let num_segments_after_merge = searcher.segment_readers().len();
old_reader.reload()?;
let old_searcher = old_reader.searcher();
@@ -1678,7 +1780,7 @@ mod tests {
let ff_reader = segment_reader.fast_fields().u64(id_field).unwrap();
segment_reader
.doc_ids_alive()
.map(move |doc| ff_reader.get_val(doc as u64))
.map(move |doc| ff_reader.get_val(doc))
})
.collect();
@@ -1689,11 +1791,27 @@ mod tests {
let ff_reader = segment_reader.fast_fields().u64(id_field).unwrap();
segment_reader
.doc_ids_alive()
.map(move |doc| ff_reader.get_val(doc as u64))
.map(move |doc| ff_reader.get_val(doc))
})
.collect();
let (expected_ids_and_num_occurrences, deleted_ids) = expected_ids(ops);
let id_list = get_id_list(ops);
// multivalue fast field content
let mut all_ips = Vec::new();
let mut num_ips = 0;
for segment_reader in searcher.segment_readers().iter() {
let ip_reader = segment_reader.fast_fields().ip_addrs(ips_field).unwrap();
for doc in segment_reader.doc_ids_alive() {
let mut vals = vec![];
ip_reader.get_vals(doc, &mut vals);
all_ips.extend_from_slice(&vals);
}
num_ips += ip_reader.total_num_vals();
}
let num_docs_expected = expected_ids_and_num_occurrences
.iter()
.map(|(_, id_occurrences)| *id_occurrences as usize)
@@ -1715,6 +1833,84 @@ mod tests {
.collect::<HashSet<_>>()
);
if force_end_merge && num_segments_before_merge > 1 && num_segments_after_merge == 1 {
let mut expected_multi_ips: Vec<_> = id_list
.iter()
.filter(|id| ip_exists(**id))
.flat_map(|id| vec![ip_from_id(*id), ip_from_id(*id)])
.collect();
assert_eq!(num_ips, expected_multi_ips.len() as u32);
expected_multi_ips.sort();
all_ips.sort();
assert_eq!(expected_multi_ips, all_ips);
// Test fastfield num_docs
let num_docs: usize = searcher
.segment_readers()
.iter()
.map(|segment_reader| {
let ff_reader = segment_reader.fast_fields().ip_addrs(ips_field).unwrap();
ff_reader.get_index_reader().num_docs() as usize
})
.sum();
assert_eq!(num_docs, num_docs_expected);
}
// Load all ips addr
let ips: HashSet<Ipv6Addr> = searcher
.segment_readers()
.iter()
.flat_map(|segment_reader| {
let ff_reader = segment_reader.fast_fields().ip_addr(ip_field).unwrap();
segment_reader.doc_ids_alive().flat_map(move |doc| {
let val = ff_reader.get_val(doc);
if val == Ipv6Addr::from_u128(0) {
// TODO Fix null handling
None
} else {
Some(val)
}
})
})
.collect();
let expected_ips = expected_ids_and_num_occurrences
.keys()
.flat_map(|id| {
if !ip_exists(*id) {
None
} else {
Some(Ipv6Addr::from_u128(*id as u128))
}
})
.collect::<HashSet<_>>();
assert_eq!(ips, expected_ips);
let expected_ips = expected_ids_and_num_occurrences
.keys()
.filter_map(|id| {
if !ip_exists(*id) {
None
} else {
Some(Ipv6Addr::from_u128(*id as u128))
}
})
.collect::<HashSet<_>>();
let ips: HashSet<Ipv6Addr> = searcher
.segment_readers()
.iter()
.flat_map(|segment_reader| {
let ff_reader = segment_reader.fast_fields().ip_addrs(ips_field).unwrap();
segment_reader.doc_ids_alive().flat_map(move |doc| {
let mut vals = vec![];
ff_reader.get_vals(doc, &mut vals);
vals.into_iter().filter(|val| val.to_u128() != 0) // TODO Fix null handling
})
})
.collect();
assert_eq!(ips, expected_ips);
// multivalue fast field tests
for segment_reader in searcher.segment_readers().iter() {
let id_reader = segment_reader.fast_fields().u64(id_field).unwrap();
@@ -1725,7 +1921,7 @@ mod tests {
ff_reader.get_vals(doc, &mut vals);
assert_eq!(vals.len(), 2);
assert_eq!(vals[0], vals[1]);
assert_eq!(id_reader.get_val(doc as u64), vals[0]);
assert_eq!(id_reader.get_val(doc), vals[0]);
let mut bool_vals = vec![];
bool_ff_reader.get_vals(doc, &mut bool_vals);
@@ -1779,10 +1975,8 @@ mod tests {
}
}
// test search
let my_text_field = index.schema().get_field("text_field").unwrap();
let do_search = |term: &str| {
let query = QueryParser::for_index(&index, vec![my_text_field])
let do_search = |term: &str, field| {
let query = QueryParser::for_index(&index, vec![field])
.parse_query(term)
.unwrap();
let top_docs: Vec<(f32, DocAddress)> =
@@ -1791,12 +1985,126 @@ mod tests {
top_docs.iter().map(|el| el.1).collect::<Vec<_>>()
};
for (existing_id, count) in expected_ids_and_num_occurrences {
assert_eq!(do_search(&existing_id.to_string()).len() as u64, count);
let do_search2 = |term: Term| {
let query = TermQuery::new(term, IndexRecordOption::Basic);
let top_docs: Vec<(f32, DocAddress)> =
searcher.search(&query, &TopDocs::with_limit(1000)).unwrap();
top_docs.iter().map(|el| el.1).collect::<Vec<_>>()
};
for (existing_id, count) in &expected_ids_and_num_occurrences {
let (existing_id, count) = (*existing_id, *count);
let get_num_hits = |field| do_search(&existing_id.to_string(), field).len() as u64;
assert_eq!(get_num_hits(text_field), count);
assert_eq!(get_num_hits(i64_field), count);
assert_eq!(get_num_hits(f64_field), count);
assert_eq!(get_num_hits(id_field), count);
// Test multi text
assert_eq!(
do_search("\"test1 test2\"", multi_text_fields).len(),
num_docs_expected
);
assert_eq!(
do_search("\"test2 test3\"", multi_text_fields).len(),
num_docs_expected
);
// Test bytes
let term = Term::from_field_bytes(bytes_field, existing_id.to_le_bytes().as_slice());
assert_eq!(do_search2(term).len() as u64, count);
// Test date
let term = Term::from_field_date(
date_field,
DateTime::from_timestamp_secs(existing_id as i64),
);
assert_eq!(do_search2(term).len() as u64, count);
}
for existing_id in deleted_ids {
assert_eq!(do_search(&existing_id.to_string()).len(), 0);
for deleted_id in deleted_ids {
let assert_field = |field| {
assert_eq!(do_search(&deleted_id.to_string(), field).len() as u64, 0);
};
assert_field(text_field);
assert_field(f64_field);
assert_field(i64_field);
assert_field(id_field);
// Test bytes
let term = Term::from_field_bytes(bytes_field, deleted_id.to_le_bytes().as_slice());
assert_eq!(do_search2(term).len() as u64, 0);
// Test date
let term =
Term::from_field_date(date_field, DateTime::from_timestamp_secs(deleted_id as i64));
assert_eq!(do_search2(term).len() as u64, 0);
}
// search ip address
//
for (existing_id, count) in &expected_ids_and_num_occurrences {
let (existing_id, count) = (*existing_id, *count);
if !ip_exists(existing_id) {
continue;
}
let do_search_ip_field = |term: &str| do_search(term, ip_field).len() as u64;
let ip_addr = Ipv6Addr::from_u128(existing_id as u128);
// Test incoming ip as ipv6
assert_eq!(do_search_ip_field(&format!("\"{}\"", ip_addr)), count);
let term = Term::from_field_ip_addr(ip_field, ip_addr);
assert_eq!(do_search2(term).len() as u64, count);
// Test incoming ip as ipv4
if let Some(ip_addr) = ip_addr.to_ipv4_mapped() {
assert_eq!(do_search_ip_field(&format!("\"{}\"", ip_addr)), count);
}
}
// assert data is like expected
//
for (existing_id, count) in expected_ids_and_num_occurrences.iter().take(10) {
let (existing_id, count) = (*existing_id, *count);
if !ip_exists(existing_id) {
continue;
}
let gen_query_inclusive = |field: &str, from: Ipv6Addr, to: Ipv6Addr| {
format!("{}:[{} TO {}]", field, &from.to_string(), &to.to_string())
};
let ip = ip_from_id(existing_id);
let do_search_ip_field = |term: &str| do_search(term, ip_field).len() as u64;
// Range query on single value field
// let query = gen_query_inclusive("ip", ip, ip);
// assert_eq!(do_search_ip_field(&query), count);
// Range query on multi value field
let query = gen_query_inclusive("ips", ip, ip);
assert_eq!(do_search_ip_field(&query), count);
}
// ip range query on fast field
//
for (existing_id, count) in expected_ids_and_num_occurrences.iter().take(10) {
let (existing_id, count) = (*existing_id, *count);
if !ip_exists(existing_id) {
continue;
}
let gen_query_inclusive = |field: &str, from: Ipv6Addr, to: Ipv6Addr| {
format!("{}:[{} TO {}]", field, &from.to_string(), &to.to_string())
};
let ip = ip_from_id(existing_id);
let do_search_ip_field = |term: &str| do_search(term, ip_field).len() as u64;
// Range query on single value field
// let query = gen_query_inclusive("ip", ip, ip);
// assert_eq!(do_search_ip_field(&query), count);
// Range query on multi value field
let query = gen_query_inclusive("ips", ip, ip);
assert_eq!(do_search_ip_field(&query), count);
}
// test facets
for segment_reader in searcher.segment_readers().iter() {
let mut facet_reader = segment_reader.facet_reader(facet_field).unwrap();
@@ -1809,7 +2117,7 @@ mod tests {
facet_reader
.facet_from_ord(facet_ords[0], &mut facet)
.unwrap();
let id = ff_reader.get_val(doc_id as u64);
let id = ff_reader.get_val(doc_id);
let facet_expected = Facet::from(&("/cola/".to_string() + &id.to_string()));
assert_eq!(facet, facet_expected);
@@ -1818,6 +2126,70 @@ mod tests {
Ok(())
}
#[test]
fn test_ip_range_query_multivalue_bug() {
assert!(test_operation_strategy(
&[
IndexingOp::AddDoc { id: 2 },
IndexingOp::Commit,
IndexingOp::AddDoc { id: 1 },
IndexingOp::AddDoc { id: 1 },
IndexingOp::Commit,
IndexingOp::Merge
],
true,
false
)
.is_ok());
}
#[test]
fn test_ff_num_ips_regression() {
assert!(test_operation_strategy(
&[
IndexingOp::AddDoc { id: 13 },
IndexingOp::AddDoc { id: 1 },
IndexingOp::Commit,
IndexingOp::DeleteDocQuery { id: 13 },
IndexingOp::AddDoc { id: 1 },
IndexingOp::Commit,
],
false,
true
)
.is_ok());
}
#[test]
fn test_minimal() {
assert!(test_operation_strategy(
&[
IndexingOp::AddDoc { id: 23 },
IndexingOp::AddDoc { id: 13 },
IndexingOp::DeleteDoc { id: 13 }
],
true,
true
)
.is_ok());
assert!(test_operation_strategy(
&[
IndexingOp::AddDoc { id: 23 },
IndexingOp::AddDoc { id: 13 },
IndexingOp::DeleteDoc { id: 13 }
],
false,
false
)
.is_ok());
}
#[test]
fn test_minimal_sort_merge() {
assert!(test_operation_strategy(&[IndexingOp::AddDoc { id: 3 },], true, true).is_ok());
}
proptest! {
#![proptest_config(ProptestConfig::with_cases(20))]
#[test]
@@ -1910,4 +2282,135 @@ mod tests {
index_writer.commit()?;
Ok(())
}
#[test]
fn test_bug_1617_3() {
assert!(test_operation_strategy(
&[
IndexingOp::DeleteDoc { id: 0 },
IndexingOp::AddDoc { id: 6 },
IndexingOp::DeleteDocQuery { id: 11 },
IndexingOp::Commit,
IndexingOp::Merge,
IndexingOp::Commit,
IndexingOp::Commit
],
false,
false
)
.is_ok());
}
#[test]
fn test_bug_1617_2() {
assert!(test_operation_strategy(
&[
IndexingOp::AddDoc { id: 13 },
IndexingOp::DeleteDoc { id: 13 },
IndexingOp::Commit,
IndexingOp::AddDoc { id: 30 },
IndexingOp::Commit,
IndexingOp::Merge,
],
false,
true
)
.is_ok());
}
#[test]
fn test_bug_1617() -> crate::Result<()> {
let mut schema_builder = schema::Schema::builder();
let id_field = schema_builder.add_u64_field("id", INDEXED);
let schema = schema_builder.build();
let index = Index::builder().schema(schema).create_in_ram()?;
let mut index_writer = index.writer_for_tests()?;
index_writer.set_merge_policy(Box::new(NoMergePolicy));
let existing_id = 16u64;
let deleted_id = 13u64;
index_writer.add_document(doc!(
id_field=>existing_id,
))?;
index_writer.add_document(doc!(
id_field=>deleted_id,
))?;
index_writer.delete_term(Term::from_field_u64(id_field, deleted_id));
index_writer.commit()?;
// Merge
{
assert!(index_writer.wait_merging_threads().is_ok());
let mut index_writer = index.writer_for_tests()?;
let segment_ids = index
.searchable_segment_ids()
.expect("Searchable segments failed.");
index_writer.merge(&segment_ids).wait().unwrap();
assert!(index_writer.wait_merging_threads().is_ok());
}
let searcher = index.reader()?.searcher();
let query = TermQuery::new(
Term::from_field_u64(id_field, existing_id),
IndexRecordOption::Basic,
);
let top_docs: Vec<(f32, DocAddress)> =
searcher.search(&query, &TopDocs::with_limit(10)).unwrap();
assert_eq!(top_docs.len(), 1); // Fails
Ok(())
}
#[test]
fn test_bug_1618() -> crate::Result<()> {
let mut schema_builder = schema::Schema::builder();
let id_field = schema_builder.add_i64_field("id", INDEXED);
let schema = schema_builder.build();
let index = Index::builder().schema(schema).create_in_ram()?;
let mut index_writer = index.writer_for_tests()?;
index_writer.set_merge_policy(Box::new(NoMergePolicy));
index_writer.add_document(doc!(
id_field=>10i64,
))?;
index_writer.add_document(doc!(
id_field=>30i64,
))?;
index_writer.commit()?;
// Merge
{
assert!(index_writer.wait_merging_threads().is_ok());
let mut index_writer = index.writer_for_tests()?;
let segment_ids = index
.searchable_segment_ids()
.expect("Searchable segments failed.");
index_writer.merge(&segment_ids).wait().unwrap();
assert!(index_writer.wait_merging_threads().is_ok());
}
let searcher = index.reader()?.searcher();
let query = TermQuery::new(
Term::from_field_i64(id_field, 10i64),
IndexRecordOption::Basic,
);
let top_docs: Vec<(f32, DocAddress)> =
searcher.search(&query, &TopDocs::with_limit(10)).unwrap();
assert_eq!(top_docs.len(), 1); // Fails
let query = TermQuery::new(
Term::from_field_i64(id_field, 30i64),
IndexRecordOption::Basic,
);
let top_docs: Vec<(f32, DocAddress)> =
searcher.search(&query, &TopDocs::with_limit(10)).unwrap();
assert_eq!(top_docs.len(), 1); // Fails
Ok(())
}
}

219
src/indexer/json_term_writer.rs
View File

@@ -1,6 +1,6 @@
use fastfield_codecs::MonotonicallyMappableToU64;
use fnv::FnvHashMap;
use murmurhash32::murmurhash2;
use rustc_hash::FxHashMap;
use crate::fastfield::FastValue;
use crate::postings::{IndexingContext, IndexingPosition, PostingsWriter};
@@ -52,7 +52,7 @@ use crate::{DatePrecision, DateTime, DocId, Term};
/// path map to the same index position as long as the probability is relatively low.
#[derive(Default)]
struct IndexingPositionsPerPath {
positions_per_path: FnvHashMap<u32, IndexingPosition>,
positions_per_path: FxHashMap<u32, IndexingPosition>,
}
impl IndexingPositionsPerPath {
@@ -67,11 +67,12 @@ pub(crate) fn index_json_values<'a>(
doc: DocId,
json_values: impl Iterator<Item = crate::Result<&'a serde_json::Map<String, serde_json::Value>>>,
text_analyzer: &TextAnalyzer,
expand_dots_enabled: bool,
term_buffer: &mut Term,
postings_writer: &mut dyn PostingsWriter,
ctx: &mut IndexingContext,
) -> crate::Result<()> {
let mut json_term_writer = JsonTermWriter::wrap(term_buffer);
let mut json_term_writer = JsonTermWriter::wrap(term_buffer, expand_dots_enabled);
let mut positions_per_path: IndexingPositionsPerPath = Default::default();
for json_value_res in json_values {
let json_value = json_value_res?;
@@ -242,10 +243,12 @@ pub(crate) fn set_string_and_get_terms(
) -> Vec<(usize, Term)> {
let mut positions_and_terms = Vec::<(usize, Term)>::new();
json_term_writer.close_path_and_set_type(Type::Str);
let term_num_bytes = json_term_writer.term_buffer.as_slice().len();
let term_num_bytes = json_term_writer.term_buffer.len_bytes();
let mut token_stream = text_analyzer.token_stream(value);
token_stream.process(&mut |token| {
json_term_writer.term_buffer.truncate(term_num_bytes);
json_term_writer
.term_buffer
.truncate_value_bytes(term_num_bytes);
json_term_writer
.term_buffer
.append_bytes(token.text.as_bytes());
@@ -257,56 +260,105 @@ pub(crate) fn set_string_and_get_terms(
pub struct JsonTermWriter<'a> {
term_buffer: &'a mut Term,
path_stack: Vec<usize>,
expand_dots_enabled: bool,
}
/// Splits a json path supplied to the query parser in such a way that
/// `.` can be escaped.
///
/// In other words,
/// - `k8s.node` ends up as `["k8s", "node"]`.
/// - `k8s\.node` ends up as `["k8s.node"]`.
fn split_json_path(json_path: &str) -> Vec<String> {
let mut escaped_state: bool = false;
let mut json_path_segments = Vec::new();
let mut buffer = String::new();
for ch in json_path.chars() {
if escaped_state {
buffer.push(ch);
escaped_state = false;
continue;
}
match ch {
'\\' => {
escaped_state = true;
}
'.' => {
let new_segment = std::mem::take(&mut buffer);
json_path_segments.push(new_segment);
}
_ => {
buffer.push(ch);
}
}
}
json_path_segments.push(buffer);
json_path_segments
}
impl<'a> JsonTermWriter<'a> {
pub fn from_field_and_json_path(
field: Field,
json_path: &str,
expand_dots_enabled: bool,
term_buffer: &'a mut Term,
) -> Self {
term_buffer.set_field(Type::Json, field);
let mut json_term_writer = Self::wrap(term_buffer);
for segment in json_path.split('.') {
json_term_writer.push_path_segment(segment);
term_buffer.set_field_and_type(field, Type::Json);
let mut json_term_writer = Self::wrap(term_buffer, expand_dots_enabled);
for segment in split_json_path(json_path) {
json_term_writer.push_path_segment(&segment);
}
json_term_writer
}
pub fn wrap(term_buffer: &'a mut Term) -> Self {
pub fn wrap(term_buffer: &'a mut Term, expand_dots_enabled: bool) -> Self {
term_buffer.clear_with_type(Type::Json);
let mut path_stack = Vec::with_capacity(10);
path_stack.push(5);
path_stack.push(0);
Self {
term_buffer,
path_stack,
expand_dots_enabled,
}
}
fn trim_to_end_of_path(&mut self) {
let end_of_path = *self.path_stack.last().unwrap();
self.term_buffer.truncate(end_of_path);
self.term_buffer.truncate_value_bytes(end_of_path);
}
pub fn close_path_and_set_type(&mut self, typ: Type) {
self.trim_to_end_of_path();
let buffer = self.term_buffer.as_mut();
let buffer = self.term_buffer.value_bytes_mut();
let buffer_len = buffer.len();
buffer[buffer_len - 1] = JSON_END_OF_PATH;
buffer.push(typ.to_code());
self.term_buffer.append_bytes(&[typ.to_code()]);
}
pub fn push_path_segment(&mut self, segment: &str) {
// the path stack should never be empty.
self.trim_to_end_of_path();
let buffer = self.term_buffer.as_mut();
let buffer = self.term_buffer.value_bytes_mut();
let buffer_len = buffer.len();
if self.path_stack.len() > 1 {
buffer[buffer_len - 1] = JSON_PATH_SEGMENT_SEP;
}
buffer.extend(segment.as_bytes());
buffer.push(JSON_PATH_SEGMENT_SEP);
self.path_stack.push(buffer.len());
if self.expand_dots_enabled && segment.as_bytes().contains(&b'.') {
// We need to replace `.` by JSON_PATH_SEGMENT_SEP.
self.term_buffer
.append_bytes(segment.as_bytes())
.iter_mut()
.for_each(|byte| {
if *byte == b'.' {
*byte = JSON_PATH_SEGMENT_SEP;
}
});
} else {
self.term_buffer.append_bytes(segment.as_bytes());
}
self.term_buffer.push_byte(JSON_PATH_SEGMENT_SEP);
self.path_stack.push(self.term_buffer.len_bytes());
}
pub fn pop_path_segment(&mut self) {
@@ -318,8 +370,8 @@ impl<'a> JsonTermWriter<'a> {
/// Returns the json path of the term being currently built.
#[cfg(test)]
pub(crate) fn path(&self) -> &[u8] {
let end_of_path = self.path_stack.last().cloned().unwrap_or(6);
&self.term().as_slice()[5..end_of_path - 1]
let end_of_path = self.path_stack.last().cloned().unwrap_or(1);
&self.term().value_bytes()[..end_of_path - 1]
}
pub fn set_fast_value<T: FastValue>(&mut self, val: T) {
@@ -332,14 +384,13 @@ impl<'a> JsonTermWriter<'a> {
val.to_u64()
};
self.term_buffer
.as_mut()
.extend_from_slice(value.to_be_bytes().as_slice());
.append_bytes(value.to_be_bytes().as_slice());
}
#[cfg(test)]
pub(crate) fn set_str(&mut self, text: &str) {
self.close_path_and_set_type(Type::Str);
self.term_buffer.as_mut().extend_from_slice(text.as_bytes());
self.term_buffer.append_bytes(text.as_bytes());
}
pub fn term(&self) -> &Term {
@@ -349,16 +400,15 @@ impl<'a> JsonTermWriter<'a> {
#[cfg(test)]
mod tests {
use super::JsonTermWriter;
use super::{split_json_path, JsonTermWriter};
use crate::schema::{Field, Type};
use crate::Term;
#[test]
fn test_json_writer() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("attributes");
json_writer.push_path_segment("color");
json_writer.set_str("red");
@@ -391,9 +441,8 @@ mod tests {
#[test]
fn test_string_term() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color");
json_writer.set_str("red");
assert_eq!(
@@ -405,9 +454,8 @@ mod tests {
#[test]
fn test_i64_term() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color");
json_writer.set_fast_value(-4i64);
assert_eq!(
@@ -419,9 +467,8 @@ mod tests {
#[test]
fn test_u64_term() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color");
json_writer.set_fast_value(4u64);
assert_eq!(
@@ -433,9 +480,8 @@ mod tests {
#[test]
fn test_f64_term() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color");
json_writer.set_fast_value(4.0f64);
assert_eq!(
@@ -447,9 +493,8 @@ mod tests {
#[test]
fn test_bool_term() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color");
json_writer.set_fast_value(true);
assert_eq!(
@@ -461,9 +506,8 @@ mod tests {
#[test]
fn test_push_after_set_path_segment() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("attribute");
json_writer.set_str("something");
json_writer.push_path_segment("color");
@@ -477,9 +521,8 @@ mod tests {
#[test]
fn test_pop_segment() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color");
json_writer.push_path_segment("hue");
json_writer.pop_path_segment();
@@ -493,9 +536,8 @@ mod tests {
#[test]
fn test_json_writer_path() {
let field = Field::from_field_id(1);
let mut term = Term::new();
term.set_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color");
assert_eq!(json_writer.path(), b"color");
json_writer.push_path_segment("hue");
@@ -503,4 +545,79 @@ mod tests {
json_writer.set_str("pink");
assert_eq!(json_writer.path(), b"color\x01hue");
}
#[test]
fn test_json_path_expand_dots_disabled() {
let field = Field::from_field_id(1);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, false);
json_writer.push_path_segment("color.hue");
assert_eq!(json_writer.path(), b"color.hue");
}
#[test]
fn test_json_path_expand_dots_enabled() {
let field = Field::from_field_id(1);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, true);
json_writer.push_path_segment("color.hue");
assert_eq!(json_writer.path(), b"color\x01hue");
}
#[test]
fn test_json_path_expand_dots_enabled_pop_segment() {
let field = Field::from_field_id(1);
let mut term = Term::with_type_and_field(Type::Json, field);
let mut json_writer = JsonTermWriter::wrap(&mut term, true);
json_writer.push_path_segment("hello");
assert_eq!(json_writer.path(), b"hello");
json_writer.push_path_segment("color.hue");
assert_eq!(json_writer.path(), b"hello\x01color\x01hue");
json_writer.pop_path_segment();
assert_eq!(json_writer.path(), b"hello");
}
#[test]
fn test_split_json_path_simple() {
let json_path = split_json_path("titi.toto");
assert_eq!(&json_path, &["titi", "toto"]);
}
#[test]
fn test_split_json_path_single_segment() {
let json_path = split_json_path("toto");
assert_eq!(&json_path, &["toto"]);
}
#[test]
fn test_split_json_path_trailing_dot() {
let json_path = split_json_path("toto.");
assert_eq!(&json_path, &["toto", ""]);
}
#[test]
fn test_split_json_path_heading_dot() {
let json_path = split_json_path(".toto");
assert_eq!(&json_path, &["", "toto"]);
}
#[test]
fn test_split_json_path_escaped_dot() {
let json_path = split_json_path(r#"toto\.titi"#);
assert_eq!(&json_path, &["toto.titi"]);
let json_path_2 = split_json_path(r#"k8s\.container\.name"#);
assert_eq!(&json_path_2, &["k8s.container.name"]);
}
#[test]
fn test_split_json_path_escaped_backslash() {
let json_path = split_json_path(r#"toto\\titi"#);
assert_eq!(&json_path, &[r#"toto\titi"#]);
}
#[test]
fn test_split_json_path_escaped_normal_letter() {
let json_path = split_json_path(r#"toto\titi"#);
assert_eq!(&json_path, &[r#"tototiti"#]);
}
}

215
src/indexer/merger.rs
View File

@@ -6,16 +6,19 @@ use fastfield_codecs::VecColumn;
use itertools::Itertools;
use measure_time::debug_time;
use super::flat_map_with_buffer::FlatMapWithBufferIter;
use super::sorted_doc_id_multivalue_column::RemappedDocIdMultiValueIndexColumn;
use crate::core::{Segment, SegmentReader};
use crate::docset::{DocSet, TERMINATED};
use crate::error::DataCorruption;
use crate::fastfield::{
AliveBitSet, Column, CompositeFastFieldSerializer, MultiValueLength, MultiValuedFastFieldReader,
get_fastfield_codecs_for_multivalue, AliveBitSet, Column, CompositeFastFieldSerializer,
MultiValueIndex, MultiValuedFastFieldReader, MultiValuedU128FastFieldReader,
};
use crate::fieldnorm::{FieldNormReader, FieldNormReaders, FieldNormsSerializer, FieldNormsWriter};
use crate::indexer::doc_id_mapping::{expect_field_id_for_sort_field, SegmentDocIdMapping};
use crate::indexer::sorted_doc_id_column::SortedDocIdColumn;
use crate::indexer::sorted_doc_id_multivalue_column::SortedDocIdMultiValueColumn;
use crate::indexer::sorted_doc_id_column::RemappedDocIdColumn;
use crate::indexer::sorted_doc_id_multivalue_column::RemappedDocIdMultiValueColumn;
use crate::indexer::SegmentSerializer;
use crate::postings::{InvertedIndexSerializer, Postings, SegmentPostings};
use crate::schema::{Cardinality, Field, FieldType, Schema};
@@ -293,6 +296,24 @@ impl IndexMerger {
self.write_bytes_fast_field(field, fast_field_serializer, doc_id_mapping)?;
}
}
FieldType::IpAddr(options) => match options.get_fastfield_cardinality() {
Some(Cardinality::SingleValue) => {
self.write_u128_single_fast_field(
field,
fast_field_serializer,
doc_id_mapping,
)?;
}
Some(Cardinality::MultiValues) => {
self.write_u128_multi_fast_field(
field,
fast_field_serializer,
doc_id_mapping,
)?;
}
None => {}
},
FieldType::JsonObject(_) | FieldType::Facet(_) | FieldType::Str(_) => {
// We don't handle json fast field for the moment
// They can be implemented using what is done
@@ -303,6 +324,106 @@ impl IndexMerger {
Ok(())
}
// used to merge `u128` single fast fields.
fn write_u128_multi_fast_field(
&self,
field: Field,
fast_field_serializer: &mut CompositeFastFieldSerializer,
doc_id_mapping: &SegmentDocIdMapping,
) -> crate::Result<()> {
let segment_and_ff_readers: Vec<(&SegmentReader, MultiValuedU128FastFieldReader<u128>)> =
self.readers
.iter()
.map(|segment_reader| {
let ff_reader: MultiValuedU128FastFieldReader<u128> =
segment_reader.fast_fields().u128s(field).expect(
"Failed to find index for multivalued field. This is a bug in \
tantivy, please report.",
);
(segment_reader, ff_reader)
})
.collect::<Vec<_>>();
Self::write_1_n_fast_field_idx_generic(
field,
fast_field_serializer,
doc_id_mapping,
&segment_and_ff_readers
.iter()
.map(|(segment_reader, u64s_reader)| {
(*segment_reader, u64s_reader.get_index_reader())
})
.collect::<Vec<_>>(),
)?;
let num_vals = segment_and_ff_readers
.iter()
.map(|(segment_reader, reader)| {
// TODO implement generic version, implement reverse scan, all - deletes
if let Some(alive_bitset) = segment_reader.alive_bitset() {
alive_bitset
.iter_alive()
.map(|doc| reader.num_vals(doc))
.sum()
} else {
reader.total_num_vals() as u32
}
})
.sum();
let fast_field_readers = segment_and_ff_readers
.into_iter()
.map(|(_, ff_reader)| ff_reader)
.collect::<Vec<_>>();
let iter_gen = || {
doc_id_mapping
.iter_old_doc_addrs()
.flat_map_with_buffer(|doc_addr, buffer| {
let fast_field_reader = &fast_field_readers[doc_addr.segment_ord as usize];
fast_field_reader.get_vals(doc_addr.doc_id, buffer);
})
};
fast_field_serializer.create_u128_fast_field_with_idx(field, iter_gen, num_vals, 1)?;
Ok(())
}
// used to merge `u128` single fast fields.
fn write_u128_single_fast_field(
&self,
field: Field,
fast_field_serializer: &mut CompositeFastFieldSerializer,
doc_id_mapping: &SegmentDocIdMapping,
) -> crate::Result<()> {
let fast_field_readers = self
.readers
.iter()
.map(|reader| {
let u128_reader: Arc<dyn Column<u128>> = reader.fast_fields().u128(field).expect(
"Failed to find a reader for single fast field. This is a tantivy bug and it \
should never happen.",
);
u128_reader
})
.collect::<Vec<_>>();
let iter_gen = || {
doc_id_mapping.iter_old_doc_addrs().map(|doc_addr| {
let fast_field_reader = &fast_field_readers[doc_addr.segment_ord as usize];
fast_field_reader.get_val(doc_addr.doc_id)
})
};
fast_field_serializer.create_u128_fast_field_with_idx(
field,
iter_gen,
doc_id_mapping.len() as u32,
0,
)?;
Ok(())
}
// used both to merge field norms, `u64/i64` single fast fields.
fn write_single_fast_field(
&self,
@@ -310,7 +431,7 @@ impl IndexMerger {
fast_field_serializer: &mut CompositeFastFieldSerializer,
doc_id_mapping: &SegmentDocIdMapping,
) -> crate::Result<()> {
let fast_field_accessor = SortedDocIdColumn::new(&self.readers, doc_id_mapping, field);
let fast_field_accessor = RemappedDocIdColumn::new(&self.readers, doc_id_mapping, field);
fast_field_serializer.create_auto_detect_u64_fast_field(field, fast_field_accessor)?;
Ok(())
@@ -404,8 +525,8 @@ impl IndexMerger {
doc_id_reader_pair
.into_iter()
.kmerge_by(|a, b| {
let val1 = a.2.get_val(a.0 as u64);
let val2 = b.2.get_val(b.0 as u64);
let val1 = a.2.get_val(a.0);
let val2 = b.2.get_val(b.0);
if sort_by_field.order == Order::Asc {
val1 < val2
} else {
@@ -423,33 +544,17 @@ impl IndexMerger {
// Creating the index file to point into the data, generic over `BytesFastFieldReader` and
// `MultiValuedFastFieldReader`
//
fn write_1_n_fast_field_idx_generic<T: MultiValueLength>(
fn write_1_n_fast_field_idx_generic(
field: Field,
fast_field_serializer: &mut CompositeFastFieldSerializer,
doc_id_mapping: &SegmentDocIdMapping,
reader_and_field_accessors: &[(&SegmentReader, T)],
) -> crate::Result<Vec<u64>> {
// We can now create our `idx` serializer, and in a second pass,
// can effectively push the different indexes.
segment_and_ff_readers: &[(&SegmentReader, &MultiValueIndex)],
) -> crate::Result<()> {
let column =
RemappedDocIdMultiValueIndexColumn::new(segment_and_ff_readers, doc_id_mapping);
// copying into a temp vec is not ideal, but the fast field codec api requires random
// access, which is used in the estimation. It's possible to 1. calculate random
// access on the fly or 2. change the codec api to make random access optional, but
// they both have also major drawbacks.
let mut offsets = Vec::with_capacity(doc_id_mapping.len());
let mut offset = 0;
for old_doc_addr in doc_id_mapping.iter_old_doc_addrs() {
let reader = &reader_and_field_accessors[old_doc_addr.segment_ord as usize].1;
offsets.push(offset);
offset += reader.get_len(old_doc_addr.doc_id) as u64;
}
offsets.push(offset);
let fastfield_accessor = VecColumn::from(&offsets[..]);
fast_field_serializer.create_auto_detect_u64_fast_field(field, fastfield_accessor)?;
Ok(offsets)
fast_field_serializer.create_auto_detect_u64_fast_field(field, column)?;
Ok(())
}
/// Returns the fastfield index (index for the data, not the data).
fn write_multi_value_fast_field_idx(
@@ -457,8 +562,8 @@ impl IndexMerger {
field: Field,
fast_field_serializer: &mut CompositeFastFieldSerializer,
doc_id_mapping: &SegmentDocIdMapping,
) -> crate::Result<Vec<u64>> {
let reader_ordinal_and_field_accessors = self
) -> crate::Result<()> {
let segment_and_ff_readers = self
.readers
.iter()
.map(|reader| {
@@ -477,7 +582,12 @@ impl IndexMerger {
field,
fast_field_serializer,
doc_id_mapping,
&reader_ordinal_and_field_accessors,
&segment_and_ff_readers
.iter()
.map(|(segment_reader, u64s_reader)| {
(*segment_reader, u64s_reader.get_index_reader())
})
.collect::<Vec<_>>(),
)
}
@@ -526,7 +636,12 @@ impl IndexMerger {
}
let col = VecColumn::from(&vals[..]);
fast_field_serializer.create_auto_detect_u64_fast_field_with_idx(field, col, 1)?;
fast_field_serializer.create_auto_detect_u64_fast_field_with_idx_and_codecs(
field,
col,
1,
&get_fastfield_codecs_for_multivalue(),
)?;
}
Ok(())
}
@@ -561,20 +676,21 @@ impl IndexMerger {
fast_field_serializer: &mut CompositeFastFieldSerializer,
doc_id_mapping: &SegmentDocIdMapping,
) -> crate::Result<()> {
// Multifastfield consists in 2 fastfields.
// Multifastfield consists of 2 fastfields.
// The first serves as an index into the second one and is strictly increasing.
// The second contains the actual values.
// First we merge the idx fast field.
let offsets =
self.write_multi_value_fast_field_idx(field, fast_field_serializer, doc_id_mapping)?;
self.write_multi_value_fast_field_idx(field, fast_field_serializer, doc_id_mapping)?;
let fastfield_accessor =
SortedDocIdMultiValueColumn::new(&self.readers, doc_id_mapping, &offsets, field);
fast_field_serializer.create_auto_detect_u64_fast_field_with_idx(
RemappedDocIdMultiValueColumn::new(&self.readers, doc_id_mapping, field);
fast_field_serializer.create_auto_detect_u64_fast_field_with_idx_and_codecs(
field,
fastfield_accessor,
1,
&get_fastfield_codecs_for_multivalue(),
)?;
Ok(())
@@ -586,7 +702,7 @@ impl IndexMerger {
fast_field_serializer: &mut CompositeFastFieldSerializer,
doc_id_mapping: &SegmentDocIdMapping,
) -> crate::Result<()> {
let reader_and_field_accessors = self
let segment_and_ff_readers = self
.readers
.iter()
.map(|reader| {
@@ -597,17 +713,22 @@ impl IndexMerger {
(reader, bytes_reader)
})
.collect::<Vec<_>>();
Self::write_1_n_fast_field_idx_generic(
field,
fast_field_serializer,
doc_id_mapping,
&reader_and_field_accessors,
&segment_and_ff_readers
.iter()
.map(|(segment_reader, u64s_reader)| {
(*segment_reader, u64s_reader.get_index_reader())
})
.collect::<Vec<_>>(),
)?;
let mut serialize_vals = fast_field_serializer.new_bytes_fast_field(field);
for old_doc_addr in doc_id_mapping.iter_old_doc_addrs() {
let bytes_reader = &reader_and_field_accessors[old_doc_addr.segment_ord as usize].1;
let bytes_reader = &segment_and_ff_readers[old_doc_addr.segment_ord as usize].1;
let val = bytes_reader.get_bytes(old_doc_addr.doc_id);
serialize_vals.write_all(val)?;
}
@@ -708,7 +829,7 @@ impl IndexMerger {
// Let's compute the list of non-empty posting lists
for (segment_ord, term_info) in merged_terms.current_segment_ords_and_term_infos() {
let segment_reader = &self.readers[segment_ord];
let inverted_index: &InvertedIndexReader = &*field_readers[segment_ord];
let inverted_index: &InvertedIndexReader = &field_readers[segment_ord];
let segment_postings = inverted_index
.read_postings_from_terminfo(&term_info, segment_postings_option)?;
let alive_bitset_opt = segment_reader.alive_bitset();
@@ -943,7 +1064,7 @@ mod tests {
};
use crate::collector::{Count, FacetCollector};
use crate::core::Index;
use crate::query::{AllQuery, BooleanQuery, Scorer, TermQuery};
use crate::query::{AllQuery, BooleanQuery, EnableScoring, Scorer, TermQuery};
use crate::schema::{
Cardinality, Document, Facet, FacetOptions, IndexRecordOption, NumericOptions, Term,
TextFieldIndexing, INDEXED, TEXT,
@@ -1856,7 +1977,7 @@ mod tests {
let reader = index.reader()?;
let searcher = reader.searcher();
let mut term_scorer = term_query
.specialized_weight(&searcher, true)?
.specialized_weight(EnableScoring::Enabled(&searcher))?
.specialized_scorer(searcher.segment_reader(0u32), 1.0)?;
assert_eq!(term_scorer.doc(), 0);
assert_nearly_equals!(term_scorer.block_max_score(), 0.0079681855);
@@ -1871,7 +1992,7 @@ mod tests {
assert_eq!(searcher.segment_readers().len(), 2);
for segment_reader in searcher.segment_readers() {
let mut term_scorer = term_query
.specialized_weight(&searcher, true)?
.specialized_weight(EnableScoring::Enabled(&searcher))?
.specialized_scorer(segment_reader, 1.0)?;
// the difference compared to before is intrinsic to the bm25 formula. no worries
// there.
@@ -1896,7 +2017,7 @@ mod tests {
let segment_reader = searcher.segment_reader(0u32);
let mut term_scorer = term_query
.specialized_weight(&searcher, true)?
.specialized_weight(EnableScoring::Enabled(&searcher))?
.specialized_scorer(segment_reader, 1.0)?;
// the difference compared to before is intrinsic to the bm25 formula. no worries there.
for doc in segment_reader.doc_ids_alive() {

12
src/indexer/merger_sorted_index_test.rs
View File

@@ -190,13 +190,13 @@ mod tests {
assert_eq!(fast_field.get_val(4), 2u64);
assert_eq!(fast_field.get_val(3), 3u64);
if force_disjunct_segment_sort_values {
assert_eq!(fast_field.get_val(2u64), 20u64);
assert_eq!(fast_field.get_val(1u64), 100u64);
assert_eq!(fast_field.get_val(2), 20u64);
assert_eq!(fast_field.get_val(1), 100u64);
} else {
assert_eq!(fast_field.get_val(2u64), 10u64);
assert_eq!(fast_field.get_val(1u64), 20u64);
assert_eq!(fast_field.get_val(2), 10u64);
assert_eq!(fast_field.get_val(1), 20u64);
}
assert_eq!(fast_field.get_val(0u64), 1_000u64);
assert_eq!(fast_field.get_val(0), 1_000u64);
// test new field norm mapping
{
@@ -545,7 +545,7 @@ mod bench_sorted_index_merge {
// add values in order of the new doc_ids
let mut val = 0;
for (doc_id, _reader, field_reader) in sorted_doc_ids {
val = field_reader.get_val(doc_id as u64);
val = field_reader.get_val(doc_id);
}
val

50
src/indexer/mod.rs
View File

@@ -3,6 +3,7 @@ pub mod delete_queue;
pub mod demuxer;
pub mod doc_id_mapping;
mod doc_opstamp_mapping;
mod flat_map_with_buffer;
pub mod index_writer;
mod index_writer_status;
mod json_term_writer;
@@ -57,13 +58,15 @@ type AddBatchReceiver = channel::Receiver<AddBatch>;
#[cfg(feature = "mmap")]
#[cfg(test)]
mod tests_mmap {
use crate::schema::{self, Schema};
use crate::collector::Count;
use crate::query::QueryParser;
use crate::schema::{JsonObjectOptions, Schema, TEXT};
use crate::{Index, Term};
#[test]
fn test_advance_delete_bug() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let text_field = schema_builder.add_text_field("text", schema::TEXT);
let text_field = schema_builder.add_text_field("text", TEXT);
let index = Index::create_from_tempdir(schema_builder.build())?;
let mut index_writer = index.writer_for_tests()?;
// there must be one deleted document in the segment
@@ -74,7 +77,48 @@ mod tests_mmap {
index_writer.add_document(doc!(text_field=>"c"))?;
}
index_writer.commit()?;
index_writer.commit()?;
Ok(())
}
#[test]
fn test_json_field_expand_dots_disabled_dot_escaped_required() {
let mut schema_builder = Schema::builder();
let json_field = schema_builder.add_json_field("json", TEXT);
let index = Index::create_in_ram(schema_builder.build());
let mut index_writer = index.writer_for_tests().unwrap();
let json = serde_json::json!({"k8s.container.name": "prometheus", "val": "hello"});
index_writer.add_document(doc!(json_field=>json)).unwrap();
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
assert_eq!(searcher.num_docs(), 1);
let parse_query = QueryParser::for_index(&index, Vec::new());
let query = parse_query
.parse_query(r#"json.k8s\.container\.name:prometheus"#)
.unwrap();
let num_docs = searcher.search(&query, &Count).unwrap();
assert_eq!(num_docs, 1);
}
#[test]
fn test_json_field_expand_dots_enabled_dot_escape_not_required() {
let mut schema_builder = Schema::builder();
let json_options: JsonObjectOptions =
JsonObjectOptions::from(TEXT).set_expand_dots_enabled();
let json_field = schema_builder.add_json_field("json", json_options);
let index = Index::create_in_ram(schema_builder.build());
let mut index_writer = index.writer_for_tests().unwrap();
let json = serde_json::json!({"k8s.container.name": "prometheus", "val": "hello"});
index_writer.add_document(doc!(json_field=>json)).unwrap();
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
assert_eq!(searcher.num_docs(), 1);
let parse_query = QueryParser::for_index(&index, Vec::new());
let query = parse_query
.parse_query(r#"json.k8s.container.name:prometheus"#)
.unwrap();
let num_docs = searcher.search(&query, &Count).unwrap();
assert_eq!(num_docs, 1);
}
}

30
src/indexer/segment_serializer.rs
View File

@@ -24,12 +24,27 @@ impl SegmentSerializer {
// In the merge case this is not necessary because we can kmerge the already sorted
// segments
let remapping_required = segment.index().settings().sort_by_field.is_some() && !is_in_merge;
let store_component = if remapping_required {
SegmentComponent::TempStore
let settings = segment.index().settings().clone();
let store_writer = if remapping_required {
let store_write = segment.open_write(SegmentComponent::TempStore)?;
StoreWriter::new(
store_write,
crate::store::Compressor::None,
// We want fast random access on the docs, so we choose a small block size.
// If this is zero, the skip index will contain too many checkpoints and
// therefore will be relatively slow.
16000,
settings.docstore_compress_dedicated_thread,
)?
} else {
SegmentComponent::Store
let store_write = segment.open_write(SegmentComponent::Store)?;
StoreWriter::new(
store_write,
settings.docstore_compression,
settings.docstore_blocksize,
settings.docstore_compress_dedicated_thread,
)?
};
let store_write = segment.open_write(store_component)?;
let fast_field_write = segment.open_write(SegmentComponent::FastFields)?;
let fast_field_serializer = CompositeFastFieldSerializer::from_write(fast_field_write)?;
@@ -38,13 +53,6 @@ impl SegmentSerializer {
let fieldnorms_serializer = FieldNormsSerializer::from_write(fieldnorms_write)?;
let postings_serializer = InvertedIndexSerializer::open(&mut segment)?;
let settings = segment.index().settings();
let store_writer = StoreWriter::new(
store_write,
settings.docstore_compression,
settings.docstore_blocksize,
settings.docstore_compress_dedicated_thread,
)?;
Ok(SegmentSerializer {
segment,
store_writer,

12
src/indexer/segment_updater.rs
View File

@@ -133,15 +133,15 @@ fn merge(
/// Advanced: Merges a list of segments from different indices in a new index.
///
/// Returns `TantivyError` if the the indices list is empty or their
/// Returns `TantivyError` if the indices list is empty or their
/// schemas don't match.
///
/// `output_directory`: is assumed to be empty.
///
/// # Warning
/// This function does NOT check or take the `IndexWriter` is running. It is not
/// meant to work if you have an IndexWriter running for the origin indices, or
/// the destination Index.
/// meant to work if you have an `IndexWriter` running for the origin indices, or
/// the destination `Index`.
#[doc(hidden)]
pub fn merge_indices<T: Into<Box<dyn Directory>>>(
indices: &[Index],
@@ -179,15 +179,15 @@ pub fn merge_indices<T: Into<Box<dyn Directory>>>(
/// Advanced: Merges a list of segments from different indices in a new index.
/// Additional you can provide a delete bitset for each segment to ignore doc_ids.
///
/// Returns `TantivyError` if the the indices list is empty or their
/// Returns `TantivyError` if the indices list is empty or their
/// schemas don't match.
///
/// `output_directory`: is assumed to be empty.
///
/// # Warning
/// This function does NOT check or take the `IndexWriter` is running. It is not
/// meant to work if you have an IndexWriter running for the origin indices, or
/// the destination Index.
/// meant to work if you have an `IndexWriter` running for the origin indices, or
/// the destination `Index`.
#[doc(hidden)]
pub fn merge_filtered_segments<T: Into<Box<dyn Directory>>>(
segments: &[Segment],

301
src/indexer/segment_writer.rs
View File

@@ -1,4 +1,5 @@
use fastfield_codecs::MonotonicallyMappableToU64;
use itertools::Itertools;
use super::doc_id_mapping::{get_doc_id_mapping_from_field, DocIdMapping};
use super::operation::AddOperation;
@@ -11,11 +12,9 @@ use crate::postings::{
compute_table_size, serialize_postings, IndexingContext, IndexingPosition,
PerFieldPostingsWriter, PostingsWriter,
};
use crate::schema::{FieldEntry, FieldType, FieldValue, Schema, Term, Value};
use crate::schema::{FieldEntry, FieldType, Schema, Term, Value};
use crate::store::{StoreReader, StoreWriter};
use crate::tokenizer::{
BoxTokenStream, FacetTokenizer, PreTokenizedStream, TextAnalyzer, Tokenizer,
};
use crate::tokenizer::{FacetTokenizer, PreTokenizedStream, TextAnalyzer, Tokenizer};
use crate::{DatePrecision, DocId, Document, Opstamp, SegmentComponent};
/// Computes the initial size of the hash table.
@@ -115,7 +114,7 @@ impl SegmentWriter {
fast_field_writers: FastFieldsWriter::from_schema(&schema),
doc_opstamps: Vec::with_capacity(1_000),
per_field_text_analyzers,
term_buffer: Term::new(),
term_buffer: Term::with_capacity(16),
schema,
})
}
@@ -157,7 +156,13 @@ impl SegmentWriter {
fn index_document(&mut self, doc: &Document) -> crate::Result<()> {
let doc_id = self.max_doc;
for (field, values) in doc.get_sorted_field_values() {
let vals_grouped_by_field = doc
.field_values()
.iter()
.sorted_by_key(|el| el.field())
.group_by(|el| el.field());
for (field, field_values) in &vals_grouped_by_field {
let values = field_values.map(|field_value| field_value.value());
let field_entry = self.schema.get_field_entry(field);
let make_schema_error = || {
crate::TantivyError::SchemaError(format!(
@@ -169,11 +174,13 @@ impl SegmentWriter {
if !field_entry.is_indexed() {
continue;
}
let (term_buffer, ctx) = (&mut self.term_buffer, &mut self.ctx);
let postings_writer: &mut dyn PostingsWriter =
self.per_field_postings_writers.get_for_field_mut(field);
term_buffer.set_field(field_entry.field_type().value_type(), field);
match *field_entry.field_type() {
term_buffer.clear_with_field_and_type(field_entry.field_type().value_type(), field);
match field_entry.field_type() {
FieldType::Facet(_) => {
for value in values {
let facet = value.as_facet().ok_or_else(make_schema_error)?;
@@ -197,35 +204,23 @@ impl SegmentWriter {
}
}
FieldType::Str(_) => {
let mut token_streams: Vec<BoxTokenStream> = vec![];
let mut offsets = vec![];
let mut total_offset = 0;
let mut indexing_position = IndexingPosition::default();
for value in values {
match value {
let mut token_stream = match value {
Value::PreTokStr(tok_str) => {
offsets.push(total_offset);
if let Some(last_token) = tok_str.tokens.last() {
total_offset += last_token.offset_to;
}
token_streams
.push(PreTokenizedStream::from(tok_str.clone()).into());
PreTokenizedStream::from(tok_str.clone()).into()
}
Value::Str(ref text) => {
let text_analyzer =
&self.per_field_text_analyzers[field.field_id() as usize];
offsets.push(total_offset);
total_offset += text.len();
token_streams.push(text_analyzer.token_stream(text));
text_analyzer.token_stream(text)
}
_ => (),
}
}
_ => {
continue;
}
};
let mut indexing_position = IndexingPosition::default();
for mut token_stream in token_streams {
assert_eq!(term_buffer.as_slice().len(), 5);
assert!(term_buffer.is_empty());
postings_writer.index_text(
doc_id,
&mut *token_stream,
@@ -241,61 +236,103 @@ impl SegmentWriter {
}
}
FieldType::U64(_) => {
let mut num_vals = 0;
for value in values {
num_vals += 1;
let u64_val = value.as_u64().ok_or_else(make_schema_error)?;
term_buffer.set_u64(u64_val);
postings_writer.subscribe(doc_id, 0u32, term_buffer, ctx);
}
if field_entry.has_fieldnorms() {
self.fieldnorms_writer.record(doc_id, field, num_vals);
}
}
FieldType::Date(_) => {
let mut num_vals = 0;
for value in values {
num_vals += 1;
let date_val = value.as_date().ok_or_else(make_schema_error)?;
term_buffer.set_u64(date_val.truncate(DatePrecision::Seconds).to_u64());
postings_writer.subscribe(doc_id, 0u32, term_buffer, ctx);
}
if field_entry.has_fieldnorms() {
self.fieldnorms_writer.record(doc_id, field, num_vals);
}
}
FieldType::I64(_) => {
let mut num_vals = 0;
for value in values {
num_vals += 1;
let i64_val = value.as_i64().ok_or_else(make_schema_error)?;
term_buffer.set_i64(i64_val);
postings_writer.subscribe(doc_id, 0u32, term_buffer, ctx);
}
if field_entry.has_fieldnorms() {
self.fieldnorms_writer.record(doc_id, field, num_vals);
}
}
FieldType::F64(_) => {
let mut num_vals = 0;
for value in values {
num_vals += 1;
let f64_val = value.as_f64().ok_or_else(make_schema_error)?;
term_buffer.set_f64(f64_val);
postings_writer.subscribe(doc_id, 0u32, term_buffer, ctx);
}
if field_entry.has_fieldnorms() {
self.fieldnorms_writer.record(doc_id, field, num_vals);
}
}
FieldType::Bool(_) => {
let mut num_vals = 0;
for value in values {
num_vals += 1;
let bool_val = value.as_bool().ok_or_else(make_schema_error)?;
term_buffer.set_bool(bool_val);
postings_writer.subscribe(doc_id, 0u32, term_buffer, ctx);
}
if field_entry.has_fieldnorms() {
self.fieldnorms_writer.record(doc_id, field, num_vals);
}
}
FieldType::Bytes(_) => {
let mut num_vals = 0;
for value in values {
num_vals += 1;
let bytes = value.as_bytes().ok_or_else(make_schema_error)?;
term_buffer.set_bytes(bytes);
postings_writer.subscribe(doc_id, 0u32, term_buffer, ctx);
}
if field_entry.has_fieldnorms() {
self.fieldnorms_writer.record(doc_id, field, num_vals);
}
}
FieldType::JsonObject(_) => {
FieldType::JsonObject(json_options) => {
let text_analyzer = &self.per_field_text_analyzers[field.field_id() as usize];
let json_values_it = values
.iter()
.map(|value| value.as_json().ok_or_else(make_schema_error));
let json_values_it =
values.map(|value| value.as_json().ok_or_else(make_schema_error));
index_json_values(
doc_id,
json_values_it,
text_analyzer,
json_options.is_expand_dots_enabled(),
term_buffer,
postings_writer,
ctx,
)?;
}
FieldType::IpAddr(_) => {
let mut num_vals = 0;
for value in values {
num_vals += 1;
let ip_addr = value.as_ip_addr().ok_or_else(make_schema_error)?;
term_buffer.set_ip_addr(ip_addr);
postings_writer.subscribe(doc_id, 0u32, term_buffer, ctx);
}
if field_entry.has_fieldnorms() {
self.fieldnorms_writer.record(doc_id, field, num_vals);
}
}
}
}
Ok(())
@@ -307,11 +344,10 @@ impl SegmentWriter {
pub fn add_document(&mut self, add_operation: AddOperation) -> crate::Result<()> {
let doc = add_operation.document;
self.doc_opstamps.push(add_operation.opstamp);
self.fast_field_writers.add_document(&doc);
self.fast_field_writers.add_document(&doc)?;
self.index_document(&doc)?;
let prepared_doc = prepare_doc_for_store(doc, &self.schema);
let doc_writer = self.segment_serializer.get_store_writer();
doc_writer.store(&prepared_doc)?;
doc_writer.store(&doc, &self.schema)?;
self.max_doc += 1;
Ok(())
}
@@ -374,9 +410,9 @@ fn remap_and_write(
doc_id_map,
)?;
debug!("resort-docstore");
// finalize temp docstore and create version, which reflects the doc_id_map
if let Some(doc_id_map) = doc_id_map {
debug!("resort-docstore");
let store_write = serializer
.segment_mut()
.open_write(SegmentComponent::Store)?;
@@ -393,7 +429,8 @@ fn remap_and_write(
serializer
.segment()
.open_read(SegmentComponent::TempStore)?,
50,
1, /* The docstore is configured to have one doc per block, and each doc is accessed
* only once: we don't need caching. */
)?;
for old_doc_id in doc_id_map.iter_old_doc_ids() {
let doc_bytes = store_read.get_document_bytes(old_doc_id)?;
@@ -407,40 +444,24 @@ fn remap_and_write(
Ok(())
}
/// Prepares Document for being stored in the document store
///
/// Method transforms PreTokenizedString values into String
/// values.
pub fn prepare_doc_for_store(doc: Document, schema: &Schema) -> Document {
Document::from(
doc.into_iter()
.filter(|field_value| schema.get_field_entry(field_value.field()).is_stored())
.map(|field_value| match field_value {
FieldValue {
field,
value: Value::PreTokStr(pre_tokenized_text),
} => FieldValue {
field,
value: Value::Str(pre_tokenized_text.text),
},
field_value => field_value,
})
.collect::<Vec<_>>(),
)
}
#[cfg(test)]
mod tests {
use std::path::Path;
use super::compute_initial_table_size;
use crate::collector::Count;
use crate::directory::RamDirectory;
use crate::indexer::json_term_writer::JsonTermWriter;
use crate::postings::TermInfo;
use crate::query::PhraseQuery;
use crate::schema::{IndexRecordOption, Schema, Type, STORED, STRING, TEXT};
use crate::store::{Compressor, StoreReader, StoreWriter};
use crate::time::format_description::well_known::Rfc3339;
use crate::time::OffsetDateTime;
use crate::tokenizer::{PreTokenizedString, Token};
use crate::{DateTime, DocAddress, DocSet, Document, Index, Postings, Term, TERMINATED};
use crate::{
DateTime, Directory, DocAddress, DocSet, Document, Index, Postings, Term, TERMINATED,
};
#[test]
fn test_hashmap_size() {
@@ -470,14 +491,21 @@ mod tests {
doc.add_pre_tokenized_text(text_field, pre_tokenized_text);
doc.add_text(text_field, "title");
let prepared_doc = super::prepare_doc_for_store(doc, &schema);
assert_eq!(prepared_doc.field_values().len(), 2);
assert_eq!(prepared_doc.field_values()[0].value().as_text(), Some("A"));
assert_eq!(
prepared_doc.field_values()[1].value().as_text(),
Some("title")
);
let path = Path::new("store");
let directory = RamDirectory::create();
let store_wrt = directory.open_write(path).unwrap();
let mut store_writer = StoreWriter::new(store_wrt, Compressor::None, 0, false).unwrap();
store_writer.store(&doc, &schema).unwrap();
store_writer.close().unwrap();
let reader = StoreReader::open(directory.open_read(path).unwrap(), 0).unwrap();
let doc = reader.get(0).unwrap();
assert_eq!(doc.field_values().len(), 2);
assert_eq!(doc.field_values()[0].value().as_text(), Some("A"));
assert_eq!(doc.field_values()[1].value().as_text(), Some("title"));
}
#[test]
@@ -527,11 +555,10 @@ mod tests {
let inv_idx = segment_reader.inverted_index(json_field).unwrap();
let term_dict = inv_idx.terms();
let mut term = Term::new();
term.set_field(Type::Json, json_field);
let mut term = Term::with_type_and_field(Type::Json, json_field);
let mut term_stream = term_dict.stream().unwrap();
let mut json_term_writer = JsonTermWriter::wrap(&mut term);
let mut json_term_writer = JsonTermWriter::wrap(&mut term, false);
json_term_writer.push_path_segment("bool");
json_term_writer.set_fast_value(true);
@@ -621,9 +648,8 @@ mod tests {
let searcher = reader.searcher();
let segment_reader = searcher.segment_reader(0u32);
let inv_index = segment_reader.inverted_index(json_field).unwrap();
let mut term = Term::new();
term.set_field(Type::Json, json_field);
let mut json_term_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, json_field);
let mut json_term_writer = JsonTermWriter::wrap(&mut term, false);
json_term_writer.push_path_segment("mykey");
json_term_writer.set_str("token");
let term_info = inv_index
@@ -666,9 +692,8 @@ mod tests {
let searcher = reader.searcher();
let segment_reader = searcher.segment_reader(0u32);
let inv_index = segment_reader.inverted_index(json_field).unwrap();
let mut term = Term::new();
term.set_field(Type::Json, json_field);
let mut json_term_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, json_field);
let mut json_term_writer = JsonTermWriter::wrap(&mut term, false);
json_term_writer.push_path_segment("mykey");
json_term_writer.set_str("two tokens");
let term_info = inv_index
@@ -712,9 +737,8 @@ mod tests {
writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
let mut term = Term::new();
term.set_field(Type::Json, json_field);
let mut json_term_writer = JsonTermWriter::wrap(&mut term);
let mut term = Term::with_type_and_field(Type::Json, json_field);
let mut json_term_writer = JsonTermWriter::wrap(&mut term, false);
json_term_writer.push_path_segment("mykey");
json_term_writer.push_path_segment("field");
json_term_writer.set_str("hello");
@@ -728,4 +752,121 @@ mod tests {
let phrase_query = PhraseQuery::new(vec![nothello_term, happy_term]);
assert_eq!(searcher.search(&phrase_query, &Count).unwrap(), 0);
}
#[test]
fn test_bug_regression_1629_position_when_array_with_a_field_value_that_does_not_contain_any_token(
) {
// We experienced a bug where we would have a position underflow when computing position
// delta in an horrible corner case.
//
// See the commit with this unit test if you want the details.
let mut schema_builder = Schema::builder();
let text = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
let doc = schema
.parse_document(r#"{"text": [ "bbb", "aaa", "", "aaa"]}"#)
.unwrap();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests().unwrap();
index_writer.add_document(doc).unwrap();
// On debug this did panic on the underflow
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
let seg_reader = searcher.segment_reader(0);
let inv_index = seg_reader.inverted_index(text).unwrap();
let term = Term::from_field_text(text, "aaa");
let mut postings = inv_index
.read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
.unwrap()
.unwrap();
assert_eq!(postings.doc(), 0u32);
let mut positions = Vec::new();
postings.positions(&mut positions);
// On release this was [2, 1]. (< note the decreasing values)
assert_eq!(positions, &[2, 5]);
}
#[test]
fn test_multiple_field_value_and_long_tokens() {
let mut schema_builder = Schema::builder();
let text = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
let mut doc = Document::default();
// This is a bit of a contrived example.
let tokens = PreTokenizedString {
text: "roller-coaster".to_string(),
tokens: vec![Token {
offset_from: 0,
offset_to: 14,
position: 0,
text: "rollercoaster".to_string(),
position_length: 2,
}],
};
doc.add_pre_tokenized_text(text, tokens.clone());
doc.add_pre_tokenized_text(text, tokens);
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests().unwrap();
index_writer.add_document(doc).unwrap();
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
let seg_reader = searcher.segment_reader(0);
let inv_index = seg_reader.inverted_index(text).unwrap();
let term = Term::from_field_text(text, "rollercoaster");
let mut postings = inv_index
.read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
.unwrap()
.unwrap();
assert_eq!(postings.doc(), 0u32);
let mut positions = Vec::new();
postings.positions(&mut positions);
assert_eq!(positions, &[0, 3]); //< as opposed to 0, 2 if we had a position length of 1.
}
#[test]
fn test_last_token_not_ending_last() {
let mut schema_builder = Schema::builder();
let text = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
let mut doc = Document::default();
// This is a bit of a contrived example.
let tokens = PreTokenizedString {
text: "contrived-example".to_string(), //< I can't think of a use case where this corner case happens in real life.
tokens: vec![Token { // Not the last token, yet ends after the last token.
offset_from: 0,
offset_to: 14,
position: 0,
text: "long_token".to_string(),
position_length: 3,
},
Token {
offset_from: 0,
offset_to: 14,
position: 1,
text: "short".to_string(),
position_length: 1,
}],
};
doc.add_pre_tokenized_text(text, tokens);
doc.add_text(text, "hello");
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests().unwrap();
index_writer.add_document(doc).unwrap();
index_writer.commit().unwrap();
let reader = index.reader().unwrap();
let searcher = reader.searcher();
let seg_reader = searcher.segment_reader(0);
let inv_index = seg_reader.inverted_index(text).unwrap();
let term = Term::from_field_text(text, "hello");
let mut postings = inv_index
.read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
.unwrap()
.unwrap();
assert_eq!(postings.doc(), 0u32);
let mut positions = Vec::new();
postings.positions(&mut positions);
assert_eq!(positions, &[4]); //< as opposed to 3 if we had a position length of 1.
}
}

28
src/indexer/sorted_doc_id_column.rs
View File

@@ -5,14 +5,14 @@ use itertools::Itertools;
use crate::indexer::doc_id_mapping::SegmentDocIdMapping;
use crate::schema::Field;
use crate::{DocAddress, SegmentReader};
use crate::SegmentReader;
pub(crate) struct SortedDocIdColumn<'a> {
pub(crate) struct RemappedDocIdColumn<'a> {
doc_id_mapping: &'a SegmentDocIdMapping,
fast_field_readers: Vec<Arc<dyn Column<u64>>>,
min_value: u64,
max_value: u64,
num_vals: u64,
num_vals: u32,
}
fn compute_min_max_val(
@@ -32,12 +32,12 @@ fn compute_min_max_val(
// we need to recompute the max / min
segment_reader
.doc_ids_alive()
.map(|doc_id| u64_reader.get_val(doc_id as u64))
.map(|doc_id| u64_reader.get_val(doc_id))
.minmax()
.into_option()
}
impl<'a> SortedDocIdColumn<'a> {
impl<'a> RemappedDocIdColumn<'a> {
pub(crate) fn new(
readers: &'a [SegmentReader],
doc_id_mapping: &'a SegmentDocIdMapping,
@@ -68,23 +68,19 @@ impl<'a> SortedDocIdColumn<'a> {
})
.collect::<Vec<_>>();
SortedDocIdColumn {
RemappedDocIdColumn {
doc_id_mapping,
fast_field_readers,
min_value,
max_value,
num_vals: doc_id_mapping.len() as u64,
num_vals: doc_id_mapping.len() as u32,
}
}
}
impl<'a> Column for SortedDocIdColumn<'a> {
fn get_val(&self, doc: u64) -> u64 {
let DocAddress {
doc_id,
segment_ord,
} = self.doc_id_mapping.get_old_doc_addr(doc as u32);
self.fast_field_readers[segment_ord as usize].get_val(doc_id as u64)
impl<'a> Column for RemappedDocIdColumn<'a> {
fn get_val(&self, _doc: u32) -> u64 {
unimplemented!()
}
fn iter(&self) -> Box<dyn Iterator<Item = u64> + '_> {
@@ -94,7 +90,7 @@ impl<'a> Column for SortedDocIdColumn<'a> {
.map(|old_doc_addr| {
let fast_field_reader =
&self.fast_field_readers[old_doc_addr.segment_ord as usize];
fast_field_reader.get_val(old_doc_addr.doc_id as u64)
fast_field_reader.get_val(old_doc_addr.doc_id)
}),
)
}
@@ -106,7 +102,7 @@ impl<'a> Column for SortedDocIdColumn<'a> {
self.max_value
}
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.num_vals
}
}

125
src/indexer/sorted_doc_id_multivalue_column.rs
View File

@@ -2,26 +2,24 @@ use std::cmp;
use fastfield_codecs::Column;
use crate::fastfield::{MultiValueLength, MultiValuedFastFieldReader};
use super::flat_map_with_buffer::FlatMapWithBufferIter;
use crate::fastfield::{MultiValueIndex, MultiValuedFastFieldReader};
use crate::indexer::doc_id_mapping::SegmentDocIdMapping;
use crate::schema::Field;
use crate::{DocId, SegmentReader};
use crate::{DocAddress, SegmentReader};
// We can now initialize our serializer, and push it the different values
pub(crate) struct SortedDocIdMultiValueColumn<'a> {
pub(crate) struct RemappedDocIdMultiValueColumn<'a> {
doc_id_mapping: &'a SegmentDocIdMapping,
fast_field_readers: Vec<MultiValuedFastFieldReader<u64>>,
offsets: &'a [u64],
min_value: u64,
max_value: u64,
num_vals: u64,
num_vals: u32,
}
impl<'a> SortedDocIdMultiValueColumn<'a> {
impl<'a> RemappedDocIdMultiValueColumn<'a> {
pub(crate) fn new(
readers: &'a [SegmentReader],
doc_id_mapping: &'a SegmentDocIdMapping,
offsets: &'a [u64],
field: Field,
) -> Self {
// Our values are bitpacked and we need to know what should be
@@ -58,48 +56,28 @@ impl<'a> SortedDocIdMultiValueColumn<'a> {
min_value = 0;
max_value = 0;
}
SortedDocIdMultiValueColumn {
RemappedDocIdMultiValueColumn {
doc_id_mapping,
fast_field_readers,
offsets,
min_value,
max_value,
num_vals: num_vals as u64,
num_vals: num_vals as u32,
}
}
}
impl<'a> Column for SortedDocIdMultiValueColumn<'a> {
fn get_val(&self, pos: u64) -> u64 {
// use the offsets index to find the doc_id which will contain the position.
// the offsets are strictly increasing so we can do a binary search on it.
let new_doc_id: DocId = self.offsets.partition_point(|&offset| offset <= pos) as DocId - 1; // Offsets start at 0, so -1 is safe
// now we need to find the position of `pos` in the multivalued bucket
let num_pos_covered_until_now = self.offsets[new_doc_id as usize];
let pos_in_values = pos - num_pos_covered_until_now;
let old_doc_addr = self.doc_id_mapping.get_old_doc_addr(new_doc_id);
let num_vals =
self.fast_field_readers[old_doc_addr.segment_ord as usize].get_len(old_doc_addr.doc_id);
assert!(num_vals >= pos_in_values);
let mut vals = Vec::new();
self.fast_field_readers[old_doc_addr.segment_ord as usize]
.get_vals(old_doc_addr.doc_id, &mut vals);
vals[pos_in_values as usize]
impl<'a> Column for RemappedDocIdMultiValueColumn<'a> {
fn get_val(&self, _pos: u32) -> u64 {
unimplemented!()
}
fn iter(&self) -> Box<dyn Iterator<Item = u64> + '_> {
Box::new(
self.doc_id_mapping
.iter_old_doc_addrs()
.flat_map(|old_doc_addr| {
.flat_map_with_buffer(|old_doc_addr: DocAddress, buffer| {
let ff_reader = &self.fast_field_readers[old_doc_addr.segment_ord as usize];
let mut vals = Vec::new();
ff_reader.get_vals(old_doc_addr.doc_id, &mut vals);
vals.into_iter()
ff_reader.get_vals(old_doc_addr.doc_id, buffer);
}),
)
}
@@ -111,7 +89,82 @@ impl<'a> Column for SortedDocIdMultiValueColumn<'a> {
self.max_value
}
fn num_vals(&self) -> u64 {
fn num_vals(&self) -> u32 {
self.num_vals
}
}
pub(crate) struct RemappedDocIdMultiValueIndexColumn<'a> {
doc_id_mapping: &'a SegmentDocIdMapping,
multi_value_length_readers: Vec<&'a MultiValueIndex>,
min_value: u64,
max_value: u64,
num_vals: u32,
}
impl<'a> RemappedDocIdMultiValueIndexColumn<'a> {
pub(crate) fn new(
segment_and_ff_readers: &'a [(&'a SegmentReader, &'a MultiValueIndex)],
doc_id_mapping: &'a SegmentDocIdMapping,
) -> Self {
// We go through a complete first pass to compute the minimum and the
// maximum value and initialize our Column.
let mut num_vals = 0;
let min_value = 0;
let mut max_value = 0;
let mut multi_value_length_readers = Vec::with_capacity(segment_and_ff_readers.len());
for segment_and_ff_reader in segment_and_ff_readers {
let segment_reader = segment_and_ff_reader.0;
let multi_value_length_reader = segment_and_ff_reader.1;
if !segment_reader.has_deletes() {
max_value += multi_value_length_reader.total_num_vals() as u64;
} else {
for doc in segment_reader.doc_ids_alive() {
max_value += multi_value_length_reader.num_vals_for_doc(doc) as u64;
}
}
num_vals += segment_reader.num_docs();
multi_value_length_readers.push(multi_value_length_reader);
}
// The value range is always get_val(doc)..get_val(doc + 1)
num_vals += 1;
Self {
doc_id_mapping,
multi_value_length_readers,
min_value,
max_value,
num_vals,
}
}
}
impl<'a> Column for RemappedDocIdMultiValueIndexColumn<'a> {
fn get_val(&self, _pos: u32) -> u64 {
unimplemented!()
}
fn iter(&self) -> Box<dyn Iterator<Item = u64> + '_> {
let mut offset = 0;
Box::new(
std::iter::once(0).chain(self.doc_id_mapping.iter_old_doc_addrs().map(
move |old_doc_addr| {
let ff_reader =
&self.multi_value_length_readers[old_doc_addr.segment_ord as usize];
offset += ff_reader.num_vals_for_doc(old_doc_addr.doc_id);
offset as u64
},
)),
)
}
fn min_value(&self) -> u64 {
self.min_value
}
fn max_value(&self) -> u64 {
self.max_value
}
fn num_vals(&self) -> u32 {
self.num_vals
}
}

8
src/lib.rs
View File

@@ -277,6 +277,8 @@ pub mod fastfield;
pub mod fieldnorm;
pub mod positions;
pub mod postings;
/// Module containing the different query implementations.
pub mod query;
pub mod schema;
pub mod space_usage;
@@ -311,7 +313,7 @@ pub use crate::postings::Postings;
pub use crate::schema::{DateOptions, DatePrecision, Document, Term};
/// Index format version.
const INDEX_FORMAT_VERSION: u32 = 4;
const INDEX_FORMAT_VERSION: u32 = 5;
/// Structure version for the index.
#[derive(Clone, PartialEq, Eq, Serialize, Deserialize)]
@@ -819,7 +821,7 @@ pub mod tests {
fn test_indexedfield_not_in_documents() -> crate::Result<()> {
let mut schema_builder = Schema::builder();
let text_field = schema_builder.add_text_field("text", TEXT);
let absent_field = schema_builder.add_text_field("text", TEXT);
let absent_field = schema_builder.add_text_field("absent_text", TEXT);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_for_tests()?;
@@ -1001,7 +1003,7 @@ pub mod tests {
let fast_field_signed = schema_builder.add_i64_field("signed", FAST);
let fast_field_float = schema_builder.add_f64_field("float", FAST);
let text_field = schema_builder.add_text_field("text", TEXT);
let stored_int_field = schema_builder.add_u64_field("text", STORED);
let stored_int_field = schema_builder.add_u64_field("stored_int", STORED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);

2
src/postings/compression/mod.rs
View File

@@ -106,7 +106,7 @@ impl BlockDecoder {
pub trait VIntEncoder {
/// Compresses an array of `u32` integers,
/// using [delta-encoding](https://en.wikipedia.org/wiki/Delta_ encoding)
/// using [delta-encoding](https://en.wikipedia.org/wiki/Delta_encoding)
/// and variable bytes encoding.
///
/// The method takes an array of ints to compress, and returns

6
src/postings/json_postings_writer.rs
View File

@@ -3,7 +3,7 @@ use std::io;
use crate::fastfield::MultiValuedFastFieldWriter;
use crate::indexer::doc_id_mapping::DocIdMapping;
use crate::postings::postings_writer::SpecializedPostingsWriter;
use crate::postings::recorder::{BufferLender, NothingRecorder, Recorder};
use crate::postings::recorder::{BufferLender, DocIdRecorder, Recorder};
use crate::postings::stacker::Addr;
use crate::postings::{
FieldSerializer, IndexingContext, IndexingPosition, PostingsWriter, UnorderedTermId,
@@ -16,7 +16,7 @@ use crate::{DocId, Term};
#[derive(Default)]
pub(crate) struct JsonPostingsWriter<Rec: Recorder> {
str_posting_writer: SpecializedPostingsWriter<Rec>,
non_str_posting_writer: SpecializedPostingsWriter<NothingRecorder>,
non_str_posting_writer: SpecializedPostingsWriter<DocIdRecorder>,
}
impl<Rec: Recorder> From<JsonPostingsWriter<Rec>> for Box<dyn PostingsWriter> {
@@ -77,7 +77,7 @@ impl<Rec: Recorder> PostingsWriter for JsonPostingsWriter<Rec> {
serializer,
)?;
} else {
SpecializedPostingsWriter::<NothingRecorder>::serialize_one_term(
SpecializedPostingsWriter::<DocIdRecorder>::serialize_one_term(
term,
*addr,
doc_id_map,

13
src/postings/per_field_postings_writer.rs
View File

@@ -1,6 +1,6 @@
use crate::postings::json_postings_writer::JsonPostingsWriter;
use crate::postings::postings_writer::SpecializedPostingsWriter;
use crate::postings::recorder::{NothingRecorder, TermFrequencyRecorder, TfAndPositionRecorder};
use crate::postings::recorder::{DocIdRecorder, TermFrequencyRecorder, TfAndPositionRecorder};
use crate::postings::PostingsWriter;
use crate::schema::{Field, FieldEntry, FieldType, IndexRecordOption, Schema};
@@ -34,7 +34,7 @@ fn posting_writer_from_field_entry(field_entry: &FieldEntry) -> Box<dyn Postings
.get_indexing_options()
.map(|indexing_options| match indexing_options.index_option() {
IndexRecordOption::Basic => {
SpecializedPostingsWriter::<NothingRecorder>::default().into()
SpecializedPostingsWriter::<DocIdRecorder>::default().into()
}
IndexRecordOption::WithFreqs => {
SpecializedPostingsWriter::<TermFrequencyRecorder>::default().into()
@@ -43,19 +43,20 @@ fn posting_writer_from_field_entry(field_entry: &FieldEntry) -> Box<dyn Postings
SpecializedPostingsWriter::<TfAndPositionRecorder>::default().into()
}
})
.unwrap_or_else(|| SpecializedPostingsWriter::<NothingRecorder>::default().into()),
.unwrap_or_else(|| SpecializedPostingsWriter::<DocIdRecorder>::default().into()),
FieldType::U64(_)
| FieldType::I64(_)
| FieldType::F64(_)
| FieldType::Bool(_)
| FieldType::Date(_)
| FieldType::Bytes(_)
| FieldType::Facet(_) => Box::new(SpecializedPostingsWriter::<NothingRecorder>::default()),
| FieldType::IpAddr(_)
| FieldType::Facet(_) => Box::new(SpecializedPostingsWriter::<DocIdRecorder>::default()),
FieldType::JsonObject(ref json_object_options) => {
if let Some(text_indexing_option) = json_object_options.get_text_indexing_options() {
match text_indexing_option.index_option() {
IndexRecordOption::Basic => {
JsonPostingsWriter::<NothingRecorder>::default().into()
JsonPostingsWriter::<DocIdRecorder>::default().into()
}
IndexRecordOption::WithFreqs => {
JsonPostingsWriter::<TermFrequencyRecorder>::default().into()
@@ -65,7 +66,7 @@ fn posting_writer_from_field_entry(field_entry: &FieldEntry) -> Box<dyn Postings
}
}
} else {
JsonPostingsWriter::<NothingRecorder>::default().into()
JsonPostingsWriter::<DocIdRecorder>::default().into()
}
}
}

19
src/postings/postings_writer.rs
View File

@@ -3,7 +3,7 @@ use std::io;
use std::marker::PhantomData;
use std::ops::Range;
use fnv::FnvHashMap;
use rustc_hash::FxHashMap;
use super::stacker::Addr;
use crate::fastfield::MultiValuedFastFieldWriter;
@@ -56,12 +56,12 @@ pub(crate) fn serialize_postings(
doc_id_map: Option<&DocIdMapping>,
schema: &Schema,
serializer: &mut InvertedIndexSerializer,
) -> crate::Result<HashMap<Field, FnvHashMap<UnorderedTermId, TermOrdinal>>> {
) -> crate::Result<HashMap<Field, FxHashMap<UnorderedTermId, TermOrdinal>>> {
let mut term_offsets: Vec<(Term<&[u8]>, Addr, UnorderedTermId)> =
Vec::with_capacity(ctx.term_index.len());
term_offsets.extend(ctx.term_index.iter());
term_offsets.sort_unstable_by_key(|(k, _, _)| k.clone());
let mut unordered_term_mappings: HashMap<Field, FnvHashMap<UnorderedTermId, TermOrdinal>> =
let mut unordered_term_mappings: HashMap<Field, FxHashMap<UnorderedTermId, TermOrdinal>> =
HashMap::new();
let field_offsets = make_field_partition(&term_offsets);
@@ -74,7 +74,7 @@ pub(crate) fn serialize_postings(
let unordered_term_ids = term_offsets[byte_offsets.clone()]
.iter()
.map(|&(_, _, bucket)| bucket);
let mapping: FnvHashMap<UnorderedTermId, TermOrdinal> = unordered_term_ids
let mapping: FxHashMap<UnorderedTermId, TermOrdinal> = unordered_term_ids
.enumerate()
.map(|(term_ord, unord_term_id)| {
(unord_term_id as UnorderedTermId, term_ord as TermOrdinal)
@@ -89,6 +89,7 @@ pub(crate) fn serialize_postings(
| FieldType::Bool(_) => {}
FieldType::Bytes(_) => {}
FieldType::JsonObject(_) => {}
FieldType::IpAddr(_) => {}
}
let postings_writer = per_field_postings_writers.get_for_field(field);
@@ -152,9 +153,9 @@ pub(crate) trait PostingsWriter: Send + Sync {
indexing_position: &mut IndexingPosition,
mut term_id_fast_field_writer_opt: Option<&mut MultiValuedFastFieldWriter>,
) {
let end_of_path_idx = term_buffer.as_slice().len();
let end_of_path_idx = term_buffer.len_bytes();
let mut num_tokens = 0;
let mut end_position = 0;
let mut end_position = indexing_position.end_position;
token_stream.process(&mut |token: &Token| {
// We skip all tokens with a len greater than u16.
if token.text.len() > MAX_TOKEN_LEN {
@@ -166,10 +167,10 @@ pub(crate) trait PostingsWriter: Send + Sync {
);
return;
}
term_buffer.truncate(end_of_path_idx);
term_buffer.truncate_value_bytes(end_of_path_idx);
term_buffer.append_bytes(token.text.as_bytes());
let start_position = indexing_position.end_position + token.position as u32;
end_position = start_position + token.position_length as u32;
end_position = end_position.max(start_position + token.position_length as u32);
let unordered_term_id = self.subscribe(doc_id, start_position, term_buffer, ctx);
if let Some(term_id_fast_field_writer) = term_id_fast_field_writer_opt.as_mut() {
term_id_fast_field_writer.add_val(unordered_term_id);
@@ -180,7 +181,7 @@ pub(crate) trait PostingsWriter: Send + Sync {
indexing_position.end_position = end_position + POSITION_GAP;
indexing_position.num_tokens += num_tokens;
term_buffer.truncate(end_of_path_idx);
term_buffer.truncate_value_bytes(end_of_path_idx);
}
fn total_num_tokens(&self) -> u64;

14
src/postings/recorder.rs
View File

@@ -47,11 +47,11 @@ impl<'a> Iterator for VInt32Reader<'a> {
}
}
/// Recorder is in charge of recording relevant information about
/// `Recorder` is in charge of recording relevant information about
/// the presence of a term in a document.
///
/// Depending on the `TextIndexingOptions` associated with the
/// field, the recorder may records
/// Depending on the [`TextOptions`](crate::schema::TextOptions) associated
/// with the field, the recorder may record:
/// * the document frequency
/// * the document id
/// * the term frequency
@@ -83,21 +83,21 @@ pub(crate) trait Recorder: Copy + Default + Send + Sync + 'static {
/// Only records the doc ids
#[derive(Clone, Copy)]
pub struct NothingRecorder {
pub struct DocIdRecorder {
stack: ExpUnrolledLinkedList,
current_doc: DocId,
}
impl Default for NothingRecorder {
impl Default for DocIdRecorder {
fn default() -> Self {
NothingRecorder {
DocIdRecorder {
stack: ExpUnrolledLinkedList::new(),
current_doc: u32::MAX,
}
}
}
impl Recorder for NothingRecorder {
impl Recorder for DocIdRecorder {
fn current_doc(&self) -> DocId {
self.current_doc
}

4
src/postings/stacker/term_hashmap.rs
View File

@@ -98,7 +98,7 @@ impl<'a> Iterator for Iter<'a> {
/// # Panics if n == 0
fn compute_previous_power_of_two(n: usize) -> usize {
assert!(n > 0);
let msb = (63u32 - n.leading_zeros()) as u8;
let msb = (63u32 - (n as u64).leading_zeros()) as u8;
1 << msb
}
@@ -199,7 +199,7 @@ impl TermHashMap {
/// `update` create a new entry for a given key if it does not exists
/// or updates the existing entry.
///
/// The actual logic for this update is define in the the `updater`
/// The actual logic for this update is define in the `updater`
/// argument.
///
/// If the key is not present, `updater` will receive `None` and

12
src/query/all_query.rs
View File

@@ -1,8 +1,8 @@
use crate::core::{Searcher, SegmentReader};
use crate::core::SegmentReader;
use crate::docset::{DocSet, TERMINATED};
use crate::query::boost_query::BoostScorer;
use crate::query::explanation::does_not_match;
use crate::query::{Explanation, Query, Scorer, Weight};
use crate::query::{EnableScoring, Explanation, Query, Scorer, Weight};
use crate::{DocId, Score};
/// Query that matches all of the documents.
@@ -12,7 +12,7 @@ use crate::{DocId, Score};
pub struct AllQuery;
impl Query for AllQuery {
fn weight(&self, _: &Searcher, _: bool) -> crate::Result<Box<dyn Weight>> {
fn weight(&self, _: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
Ok(Box::new(AllWeight))
}
}
@@ -72,7 +72,7 @@ impl Scorer for AllScorer {
mod tests {
use super::AllQuery;
use crate::docset::TERMINATED;
use crate::query::Query;
use crate::query::{EnableScoring, Query};
use crate::schema::{Schema, TEXT};
use crate::Index;
@@ -95,7 +95,7 @@ mod tests {
let index = create_test_index()?;
let reader = index.reader()?;
let searcher = reader.searcher();
let weight = AllQuery.weight(&searcher, false)?;
let weight = AllQuery.weight(EnableScoring::Disabled(&index.schema()))?;
{
let reader = searcher.segment_reader(0);
let mut scorer = weight.scorer(reader, 1.0)?;
@@ -118,7 +118,7 @@ mod tests {
let index = create_test_index()?;
let reader = index.reader()?;
let searcher = reader.searcher();
let weight = AllQuery.weight(&searcher, false)?;
let weight = AllQuery.weight(EnableScoring::Disabled(searcher.schema()))?;
let reader = searcher.segment_reader(0);
{
let mut scorer = weight.scorer(reader, 2.0)?;

2
src/query/automaton_weight.rs
View File

@@ -33,7 +33,7 @@ where
&'a self,
term_dict: &'a TermDictionary,
) -> io::Result<TermStreamer<'a, &'a A>> {
let automaton: &A = &*self.automaton;
let automaton: &A = &self.automaton;
let term_stream_builder = term_dict.search(automaton);
term_stream_builder.into_stream()
}

5
src/query/bitset/mod.rs
View File

@@ -86,10 +86,7 @@ impl DocSet for BitSetDocSet {
self.doc
}
/// Returns half of the `max_doc`
/// This is quite a terrible heuristic,
/// but we don't have access to any better
/// value.
/// Returns the number of values set in the underlying bitset.
fn size_hint(&self) -> u32 {
self.docs.len() as u32
}

6
src/query/boolean_query/block_wand.rs
View File

@@ -144,7 +144,7 @@ fn advance_all_scorers_on_pivot(term_scorers: &mut Vec<TermScorerWithMaxScore>,
/// Implements the WAND (Weak AND) algorithm for dynamic pruning
/// described in the paper "Faster Top-k Document Retrieval Using Block-Max Indexes".
/// Link: http://engineering.nyu.edu/~suel/papers/bmw.pdf
/// Link: <http://engineering.nyu.edu/~suel/papers/bmw.pdf>
pub fn block_wand(
mut scorers: Vec<TermScorer>,
mut threshold: Score,
@@ -212,12 +212,12 @@ pub fn block_wand(
}
/// Specialized version of [`block_wand`] for a single scorer.
/// In this case, the algorithm is simple and readable and faster (~ x3)
/// In this case, the algorithm is simple, readable and faster (~ x3)
/// than the generic algorithm.
/// The algorithm behaves as follows:
/// - While we don't hit the end of the docset:
/// - While the block max score is under the `threshold`, go to the next block.
/// - On a block, advance until the end and execute `callback`` when the doc score is greater or
/// - On a block, advance until the end and execute `callback` when the doc score is greater or
/// equal to the `threshold`.
pub fn block_wand_single_scorer(
mut scorer: TermScorer,

11
src/query/boolean_query/boolean_query.rs
View File

@@ -1,7 +1,6 @@
use super::boolean_weight::BooleanWeight;
use crate::query::{Occur, Query, SumWithCoordsCombiner, TermQuery, Weight};
use crate::query::{EnableScoring, Occur, Query, SumWithCoordsCombiner, TermQuery, Weight};
use crate::schema::{IndexRecordOption, Term};
use crate::Searcher;
/// The boolean query returns a set of documents
/// that matches the Boolean combination of constituent subqueries.
@@ -143,17 +142,15 @@ impl From<Vec<(Occur, Box<dyn Query>)>> for BooleanQuery {
}
impl Query for BooleanQuery {
fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
let sub_weights = self
.subqueries
.iter()
.map(|&(ref occur, ref subquery)| {
Ok((*occur, subquery.weight(searcher, scoring_enabled)?))
})
.map(|&(ref occur, ref subquery)| Ok((*occur, subquery.weight(enable_scoring)?)))
.collect::<crate::Result<_>>()?;
Ok(Box::new(BooleanWeight::new(
sub_weights,
scoring_enabled,
enable_scoring.is_scoring_enabled(),
Box::new(SumWithCoordsCombiner::default),
)))
}

24
src/query/boolean_query/boolean_weight.rs
View File

@@ -5,7 +5,7 @@ use crate::postings::FreqReadingOption;
use crate::query::explanation::does_not_match;
use crate::query::score_combiner::{DoNothingCombiner, ScoreCombiner};
use crate::query::term_query::TermScorer;
use crate::query::weight::{for_each_pruning_scorer, for_each_scorer};
use crate::query::weight::{for_each_docset, for_each_pruning_scorer, for_each_scorer};
use crate::query::{
intersect_scorers, EmptyScorer, Exclude, Explanation, Occur, RequiredOptionalScorer, Scorer,
Union, Weight,
@@ -174,9 +174,9 @@ impl<TScoreCombiner: ScoreCombiner + Sync> Weight for BooleanWeight<TScoreCombin
into_box_scorer(specialized_scorer, &self.score_combiner_fn)
})
} else {
self.complex_scorer(reader, boost, &DoNothingCombiner::default)
self.complex_scorer(reader, boost, DoNothingCombiner::default)
.map(|specialized_scorer| {
into_box_scorer(specialized_scorer, &DoNothingCombiner::default)
into_box_scorer(specialized_scorer, DoNothingCombiner::default)
})
}
}
@@ -219,6 +219,24 @@ impl<TScoreCombiner: ScoreCombiner + Sync> Weight for BooleanWeight<TScoreCombin
Ok(())
}
fn for_each_no_score(
&self,
reader: &SegmentReader,
callback: &mut dyn FnMut(DocId),
) -> crate::Result<()> {
let scorer = self.complex_scorer(reader, 1.0, || DoNothingCombiner)?;
match scorer {
SpecializedScorer::TermUnion(term_scorers) => {
let mut union_scorer = Union::build(term_scorers, &self.score_combiner_fn);
for_each_docset(&mut union_scorer, callback);
}
SpecializedScorer::Other(mut scorer) => {
for_each_docset(scorer.as_mut(), callback);
}
}
Ok(())
}
/// Calls `callback` with all of the `(doc, score)` for which score
/// is exceeding a given threshold.
///

17
src/query/boolean_query/mod.rs
View File

@@ -15,7 +15,8 @@ mod tests {
use crate::query::score_combiner::SumWithCoordsCombiner;
use crate::query::term_query::TermScorer;
use crate::query::{
Intersection, Occur, Query, QueryParser, RequiredOptionalScorer, Scorer, TermQuery,
EnableScoring, Intersection, Occur, Query, QueryParser, RequiredOptionalScorer, Scorer,
TermQuery,
};
use crate::schema::*;
use crate::{assert_nearly_equals, DocAddress, DocId, Index, Score};
@@ -54,7 +55,7 @@ mod tests {
let query_parser = QueryParser::for_index(&index, vec![text_field]);
let query = query_parser.parse_query("+a")?;
let searcher = index.reader()?.searcher();
let weight = query.weight(&searcher, true)?;
let weight = query.weight(EnableScoring::Enabled(&searcher))?;
let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
assert!(scorer.is::<TermScorer>());
Ok(())
@@ -67,13 +68,13 @@ mod tests {
let searcher = index.reader()?.searcher();
{
let query = query_parser.parse_query("+a +b +c")?;
let weight = query.weight(&searcher, true)?;
let weight = query.weight(EnableScoring::Enabled(&searcher))?;
let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
assert!(scorer.is::<Intersection<TermScorer>>());
}
{
let query = query_parser.parse_query("+a +(b c)")?;
let weight = query.weight(&searcher, true)?;
let weight = query.weight(EnableScoring::Enabled(&searcher))?;
let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
assert!(scorer.is::<Intersection<Box<dyn Scorer>>>());
}
@@ -87,7 +88,7 @@ mod tests {
let searcher = index.reader()?.searcher();
{
let query = query_parser.parse_query("+a b")?;
let weight = query.weight(&searcher, true)?;
let weight = query.weight(EnableScoring::Enabled(&searcher))?;
let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
assert!(scorer.is::<RequiredOptionalScorer<
Box<dyn Scorer>,
@@ -97,7 +98,7 @@ mod tests {
}
{
let query = query_parser.parse_query("+a b")?;
let weight = query.weight(&searcher, false)?;
let weight = query.weight(EnableScoring::Disabled(searcher.schema()))?;
let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
assert!(scorer.is::<TermScorer>());
}
@@ -241,7 +242,9 @@ mod tests {
let searcher = reader.searcher();
let boolean_query =
BooleanQuery::new(vec![(Occur::Should, term_a), (Occur::Should, term_b)]);
let boolean_weight = boolean_query.weight(&searcher, true).unwrap();
let boolean_weight = boolean_query
.weight(EnableScoring::Enabled(&searcher))
.unwrap();
{
let mut boolean_scorer = boolean_weight.scorer(searcher.segment_reader(0u32), 1.0)?;
assert_eq!(boolean_scorer.doc(), 0u32);

10
src/query/boost_query.rs
View File

@@ -2,8 +2,8 @@ use std::fmt;
use crate::fastfield::AliveBitSet;
use crate::query::explanation::does_not_match;
use crate::query::{Explanation, Query, Scorer, Weight};
use crate::{DocId, DocSet, Score, Searcher, SegmentReader, Term};
use crate::query::{EnableScoring, Explanation, Query, Scorer, Weight};
use crate::{DocId, DocSet, Score, SegmentReader, Term};
/// `BoostQuery` is a wrapper over a query used to boost its score.
///
@@ -38,9 +38,9 @@ impl fmt::Debug for BoostQuery {
}
impl Query for BoostQuery {
fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
let weight_without_boost = self.query.weight(searcher, scoring_enabled)?;
let boosted_weight = if scoring_enabled {
fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
let weight_without_boost = self.query.weight(enable_scoring)?;
let boosted_weight = if enable_scoring.is_scoring_enabled() {
Box::new(BoostWeight::new(weight_without_boost, self.boost))
} else {
weight_without_boost

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