Commit Graph

998 Commits

Author SHA1 Message Date
8b14090927 fix min-word-len-for-typo not reset properly 2022-04-19 15:20:16 +02:00
ea4bb9402f Merge #483
483: Enhance matching words r=Kerollmops a=ManyTheFish

# Summary

Enhance milli word-matcher making it handle match computing and cropping.

# Implementation

## Computing best matches for cropping

Before we were considering that the first match of the attribute was the best one, this was accurate when only one word was searched but was missing the target when more than one word was searched.

Now we are searching for the best matches interval to crop around, the chosen interval is the one:
1) that have the highest count of unique matches
> for example, if we have a query `split the world`, then the interval `the split the split the` has 5 matches but only 2 unique matches (1 for `split` and 1 for `the`) where the interval `split of the world` has 3 matches and 3 unique matches. So the interval `split of the world` is considered better.
2) that have the minimum distance between matches
> for example, if we have a query `split the world`, then the interval `split of the world` has a distance of 3 (2 between `split` and `the`, and 1 between `the` and `world`) where the interval `split the world` has a distance of 2. So the interval `split the world` is considered better.
3) that have the highest count of ordered matches
> for example, if we have a query `split the world`, then the interval `the world split` has 2 ordered words where the interval `split the world` has 3. So the interval `split the world` is considered better.

## Cropping around the best matches interval

Before we were cropping around the interval without checking the context.

Now we are cropping around words in the same context as matching words.
This means that we will keep words that are farther from the matching words but are in the same phrase, than words that are nearer but separated by a dot.

> For instance, for the matching word `Split` the text:
`Natalie risk her future. Split The World is a book written by Emily Henry. I never read it.`
will be cropped like:
`…. Split The World is a book written by Emily Henry. …`
and  not like:
`Natalie risk her future. Split The World is a book …`


Co-authored-by: ManyTheFish <many@meilisearch.com>
2022-04-19 11:42:32 +00:00
f1115e274f Use Copy impl of FormatOption instead of clonning 2022-04-19 10:35:50 +02:00
00f78d6b5a Apply code suggestions
Co-authored-by: Clément Renault <clement@meilisearch.com>
2022-04-14 11:14:08 +02:00
399fba16bb only flatten an object if it's nested 2022-04-14 11:14:08 +02:00
ee64f4a936 Use smartstring to store the external id in our hashmap
We need to store all the external id (primary key) in a hashmap
associated to their internal id during.
The smartstring remove heap allocation / memory usage and should
improve the cache locality.
2022-04-13 21:22:07 +02:00
dda28d7415 exclude excluded canditates from search result candidates 2022-04-13 12:10:35 +02:00
bbb6728d2f add distinct attributes to cli 2022-04-13 12:10:35 +02:00
5809d3ae0d Add first benchmarks on formatting 2022-04-12 16:31:58 +02:00
827cedcd15 Add format option structure 2022-04-12 13:42:14 +02:00
011f8210ed Make compute_matches more rust idiomatic 2022-04-12 10:19:02 +02:00
a16de5de84 Symplify format and remove intermediate function 2022-04-08 11:20:41 +02:00
a769e09dfa Make token_crop_bounds more rust idiomatic 2022-04-07 20:15:14 +02:00
c8ed1675a7 Add some documentation 2022-04-07 17:32:13 +02:00
b1905dfa24 Make split_best_frequency returns references instead of owned data 2022-04-07 17:05:44 +02:00
4f3ce6d9cd nested fields 2022-04-07 16:58:46 +02:00
b799f3326b rename merge_nothing to merge_ignore_values 2022-04-05 18:44:35 +02:00
fa7d3a37c0 Make some cleaning and add comments 2022-04-05 17:48:56 +02:00
3bb1e35ada Fix match count 2022-04-05 17:48:45 +02:00
56e0edd621 Put crop markers direclty around words 2022-04-05 17:41:32 +02:00
a93cd8c61c Fix prefix highlight with special chars 2022-04-05 17:41:32 +02:00
b3f0f39106 Make some cleaning 2022-04-05 17:41:32 +02:00
6dc345bc53 Test and Fix prefix highlight 2022-04-05 17:41:32 +02:00
bd30ee97b8 Keep separators at start of the croped string 2022-04-05 17:41:32 +02:00
29c5f76d7f Use new matcher in http-ui 2022-04-05 17:41:32 +02:00
734d0899d3 Publish Matcher 2022-04-05 17:41:32 +02:00
4428cb5909 Add some tests and fix some corner cases 2022-04-05 17:41:32 +02:00
844f546a8b Add matches algorithm V1 2022-04-05 17:41:32 +02:00
3be1790803 Add crop algorithm with naive match algorithm 2022-04-05 17:41:32 +02:00
d96e72e5dc Create formater with some tests 2022-04-05 17:41:32 +02:00
201fea0fda limit extract_word_docids memory usage 2022-04-05 14:14:15 +02:00
5cfd3d8407 add exact attributes documentation 2022-04-05 14:10:22 +02:00
b85cd4983e remove field_id_from_position 2022-04-05 09:50:34 +02:00
ab185a59b5 fix infos 2022-04-05 09:46:56 +02:00
1810927dbd rephrase exact_attributes doc 2022-04-04 21:04:49 +02:00
b7694c34f5 remove println 2022-04-04 21:00:07 +02:00
6cabd47c32 fix typo in comment 2022-04-04 20:59:20 +02:00
6b2c2509b2 fix bug in exact search 2022-04-04 20:54:03 +02:00
56b4f5dce2 add exact prefix to query_docids 2022-04-04 20:54:03 +02:00
21ae4143b1 add exact_word_prefix to Context 2022-04-04 20:54:03 +02:00
e8f06f6c06 extract exact_word_prefix_docids 2022-04-04 20:54:03 +02:00
6dd2e4ffbd introduce exact_word_prefix database in index 2022-04-04 20:54:03 +02:00
ba0bb29cd8 refactor WordPrefixDocids to take dbs instead of indexes 2022-04-04 20:54:02 +02:00
c4c6e35352 query exact_word_docids in resolve_query_tree 2022-04-04 20:54:02 +02:00
8d46a5b0b5 extract exact word docids 2022-04-04 20:54:02 +02:00
0a77be4ec0 introduce exact_word_docids db 2022-04-04 20:54:02 +02:00
5f9f82757d refactor spawn_extraction_task 2022-04-04 20:54:02 +02:00
f82d4b36eb introduce exact attribute setting 2022-04-04 20:54:02 +02:00
c882d8daf0 add test for exact words 2022-04-04 20:54:01 +02:00
7e9d56a9e7 disable typos on exact words 2022-04-04 20:54:01 +02:00