Shikhar Bhushan
72cef12db1
Add none compression ( #1208 )
2021-11-16 10:50:42 +09:00
Paul Masurel
8802d125f8
Prepare commit is public again ( #1202 )
...
- Simplified some of the prepare commit & segment updater code using
async.
- Made PrepareCommit public again.
2021-11-12 23:25:39 +09:00
Paul Masurel
7234bef0eb
Issue/1198 ( #1201 )
...
* Unit test reproducing #1198
* Fixing unit test to handle the error from add_document.
* Bump project version
2021-11-11 16:42:19 +09:00
azerowall
fcff91559b
Fix the deserialization error of FieldEntry when the 'options' field appears before the 'type' field ( #1199 )
...
Co-authored-by: quel <azerowall>
2021-11-10 18:39:58 +09:00
Paul Masurel
b75d4e59d1
Remove the broken panic on drop unit test. ( #1200 )
2021-11-10 18:39:37 +09:00
Paul Masurel
c6b5ab1dbe
Replacing the panic check in the RAM Directory on lack of flush.
2021-11-09 11:04:31 +09:00
PSeitz
7dc0dc1c9b
extend proptests with adding case ( #1191 )
...
This extends the proptest to cover a case where up to a 100 documents are added to an index.
2021-11-01 09:27:10 +09:00
François Massot
0462754673
Optimize block wand for one and several TermScorer. ( #1190 )
...
* Added optimisation using block wand for single TermScorer.
A proptest was also added.
* Fix block wand algorithm by taking the last doc id of scores until the pivot scorer (included).
* In block wand, when block max score is lower than the threshold, advance the scorer with best score.
* Fix wrong condition in block_wand_single_scorer and add debug_assert to have an equality check on doc to break the loop.
2021-11-01 09:18:05 +09:00
PSeitz
dbaf4f3623
Merge pull request #1187 from PSeitz/sort_issue
...
check searcher num docs in proptest
2021-10-29 16:19:24 +08:00
Pascal Seitz
4808648322
check searcher num docs in proptest
2021-10-29 14:38:30 +08:00
Paul Masurel
54afb9b34a
Made PrepareCommit private
2021-10-29 14:13:14 +09:00
Dan Cecile
6317982876
Make indexer::prepared_commit public ( #1184 )
...
* Make indexer::prepared_commit public
* Add PreparedCommit to lib
2021-10-26 12:21:24 +09:00
PSeitz
e2fbbc08ca
Merge pull request #1182 from PSeitz/remove_directory_generic
...
use Box<dyn Directory> as parameter to open/create an Index
2021-10-25 12:49:55 +08:00
Pascal Seitz
99cd25beae
use <T: Into<Box<dyn Directory>>> as parameter to open/create an Index
...
This is done in order to support Box<dyn Directory> additionally to generic implementations of the trait Directory.
Remove boxing in ManagedDirectory.
2021-10-25 12:34:40 +08:00
Kanji Yomoda
737ecc7015
Fix outdated comment for IndexWriter::new ( #1183 )
2021-10-25 10:59:18 +09:00
PSeitz
3a78402496
update links ( #1176 )
2021-10-18 20:45:40 +09:00
Paul Masurel
d18ac136c0
Search simplified ( #1175 )
2021-10-18 12:52:43 +09:00
Paul Masurel
27acfa4dea
Removing dead file ( #1170 )
2021-10-07 14:15:21 +09:00
Paul Masurel
02cffa4dea
Code simplification. ( #1169 )
...
Code simplification and Clippy
2021-10-07 14:11:44 +09:00
Paul Masurel
b52abbc771
Bugfix transposition_cost_one in FuzzyQuery ( #1167 )
2021-10-07 09:38:39 +09:00
Paul Masurel
894c61867f
Fix test compilation ( #1168 )
2021-10-06 17:50:10 +09:00
PSeitz
352e0cc58d
Adde demux operation ( #1150 )
...
* add merge for DeleteBitSet, allow custom DeleteBitSet on merge
* forward delete bitsets on merge, add tests
* add demux operation and tests
2021-10-06 16:05:16 +09:00
Paul Masurel
ffe4446d90
Minor lint comments ( #1166 )
2021-10-06 11:27:48 +09:00
Paul Masurel
0855649986
Leaning more on the alive (vs delete) semantics. ( #1164 )
2021-10-05 18:53:29 +09:00
Pascal Seitz
aa0396fe27
fix variable names
2021-10-01 13:48:51 +08:00
Pascal Seitz
8d8315f8d0
prealloc vec in postinglist
2021-09-29 09:02:38 +08:00
Pascal Seitz
078c0a2e2e
reserve vec
2021-09-29 08:45:04 +08:00
Pascal Seitz
f21e8dd875
use only segment ordinal in docidmapping
2021-09-29 08:44:56 +08:00
Tomoko Uchida
74e36c7e97
Add unit tests for tokenizers and filters ( #1156 )
...
* add unit test for SimpleTokenizer
* add unit tests for tokenizers and filters.
2021-09-27 10:22:01 +09:00
PSeitz
0ce49c9dd4
use lz4_flex 0.9.0 ( #1160 )
2021-09-27 10:12:20 +09:00
PSeitz
fe8e58e078
Merge pull request #1154 from PSeitz/delete_bitset
...
add DeleteBitSet iterator
2021-09-24 09:37:39 +02:00
Pascal Seitz
22bcc83d10
fix padding in initialization
2021-09-24 14:43:04 +08:00
Pascal Seitz
5ee5037934
create and use ReadSerializedBitSet
2021-09-24 12:53:33 +08:00
Pascal Seitz
c217bfed1e
cargo fmt
2021-09-23 21:02:19 +08:00
Pascal Seitz
c27ccd3e24
improve naming
2021-09-23 21:02:09 +08:00
Paul Masurel
367f5da782
Fixed comment to the index accessor
2021-09-23 21:53:48 +09:00
Mestery
b256df6599
add index accessor for index writer ( #1159 )
...
* add index accessor for index writer
* Update src/indexer/index_writer.rs
Co-authored-by: Paul Masurel <paul@quickwit.io >
2021-09-23 21:49:20 +09:00
Pascal Seitz
d7a6a409a1
renames
2021-09-23 20:33:11 +08:00
Pascal Seitz
a1f5cead96
AliveBitSet instead of DeleteBitSet
2021-09-23 20:03:57 +08:00
Pascal Seitz
4583fa270b
fixes
2021-09-23 10:39:53 +08:00
Pascal Seitz
93cbd52bf0
move code to biset, add inline, add benchmark
2021-09-18 17:35:22 +08:00
Pascal Seitz
c22177a005
add iterator
2021-09-17 15:29:27 +08:00
Pascal Seitz
4da71273e1
add de/serialization for bitset
...
remove len footgun
2021-09-17 10:28:12 +08:00
Pascal Seitz
4ae1d87632
add DeleteBitSet iterator
2021-09-15 23:10:04 +08:00
PSeitz
3bc177e69d
fix #1151 ( #1152 )
...
* fix #1151
Fixes a off by one error in the stats for the index fast field in the multi value fast field.
When retrieving the data range for a docid, `get(doc)..get(docid+1)` is requested. On creation
the num_vals statistic was set to doc instead of docid + 1. In the multivaluelinearinterpol fast
field the last value was therefore not serialized (and would return 0 instead in most cases).
So the last document get(lastdoc)..get(lastdoc + 1) would return the invalid range `value..0`.
This PR adds a proptest to cover this scenario. A combination of a large number values, since multilinear
interpolation is only active for more than 5_000 values, and a merge is required.
2021-09-10 23:00:37 +09:00
Kanji Yomoda
9d87b89718
Fix incorrect comment for Index::create_in_dir ( #1148 )
...
* Fix incorrect comment for Index::create_in_dir
2021-09-03 10:37:16 +09:00
Tomoko Uchida
dd81e38e53
Add WhitespaceTokenizer ( #1147 )
...
* Add WhitespaceTokenizer.
2021-08-29 18:20:49 +09:00
sigaloid
096ce7488e
Resolve some clippys, format ( #1144 )
...
* cargo +nightly clippy --fix -Z unstable-options
2021-08-26 08:46:00 +09:00
Pascal Seitz
e0b83eb291
cargo fmt
2021-08-21 18:52:10 +01:00
PSeitz
13401f46ea
add wildcard mention
2021-08-21 18:10:33 +01:00