searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-11-22 21:04:27 +01:00
Code Issues Releases Wiki Activity

Merge #3866

Browse Source 
3866: Update charabia v0.8.0 r=dureuill a=ManyTheFish

# Pull Request

Update Charabia:
- enhance Japanese segmentation
- enhance Latin Tokenization
  - words containing `_` are now properly segmented into several words
  - brackets `{([])}` are no more considered as context separators so word separated by brackets are now considered near together for the proximity ranking rule
- fixes #3815
- fixes #3778
- fixes [product#151](https://github.com/meilisearch/product/discussions/151)

> Important note: now the float numbers are segmented around the `.` so `3.22` is segmented as [`3`, `.`, `22`] but the middle dot isn't considered as a hard separator, which means that if we search `3.22` we find documents containing `3.22`

Co-authored-by: ManyTheFish <many@meilisearch.com>
This commit is contained in:
meili-bors[bot] 2023-06-29 15:24:36 +00:00 committed by GitHub
commit 661d1f90dc
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
14 changed files with 252 additions and 187 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

222
Cargo.lock generated
View File

@ -152,7 +152,7 @@ dependencies = [
"pin-project-lite", "pin-project-lite",
"tokio-rustls 0.23.4", "tokio-rustls 0.23.4",
"tokio-util", "tokio-util",
"webpki-roots", "webpki-roots 0.22.6",
] ]
[[package]] [[package]]
@ -705,24 +705,27 @@ dependencies = [
[[package]] [[package]]
name = "charabia" name = "charabia"
version = "0.7.2" version = "0.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "413155d93157bff9130895c3bd83970ac7f35659ca57226a96aa35cf1e8e102c" checksum = "bb49850f555eb71aa6fc6d4d79420e81f4d89fa56e0e9c0f6d19aace2f56c554"
dependencies = [ dependencies = [
"aho-corasick",
"cow-utils", "cow-utils",
"csv", "csv",
"deunicode", "deunicode",
"either",
"finl_unicode", "finl_unicode",
"fst", "fst",
"irg-kvariants", "irg-kvariants",
"jieba-rs", "jieba-rs",
"lindera", "lindera-core",
"lindera-dictionary",
"lindera-tokenizer",
"once_cell", "once_cell",
"pinyin", "pinyin",
"serde", "serde",
"slice-group-by", "slice-group-by",
"unicode-normalization", "unicode-normalization",
"unicode-segmentation",
"wana_kana", "wana_kana",
"whatlang", "whatlang",
] ]
@ -2135,15 +2138,6 @@ dependencies = [
"simple_asn1", "simple_asn1",
] ]
[[package]]
name = "kanaria"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c0f9d9652540055ac4fded998a73aca97d965899077ab1212587437da44196ff"
dependencies = [
"bitflags",
]
[[package]] [[package]]
name = "language-tags" name = "language-tags"
version = "0.3.2" version = "0.3.2"
@ -2211,38 +2205,11 @@ dependencies = [
"vcpkg", "vcpkg",
] ]
[[package]]
name = "lindera"
version = "0.23.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "72be283281bec2768687b1784be03a678609b51f2f90f6f9d9b4f07953e6dd25"
dependencies = [
"anyhow",
"bincode",
"byteorder",
"encoding",
"kanaria",
"lindera-cc-cedict-builder",
"lindera-core",
"lindera-dictionary",
"lindera-filter",
"lindera-ipadic-builder",
"lindera-ko-dic-builder",
"lindera-unidic-builder",
"regex",
"serde",
"serde_json",
"thiserror",
"unicode-blocks",
"unicode-normalization",
"yada",
]
[[package]] [[package]]
name = "lindera-cc-cedict-builder" name = "lindera-cc-cedict-builder"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "10fbafd37adab44ccc2668a40fba2dbc4e665cb3c36018c15dfe2e2b830e28ce" checksum = "4c6bf79b29a90bcd22036e494d6cc9ac3abe9ab604b21f3258ba6dc1ce501801"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"bincode", "bincode",
@ -2259,9 +2226,9 @@ dependencies = [
[[package]] [[package]]
name = "lindera-compress" name = "lindera-compress"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ed9196bf5995503f6878a090dfee6114ba86430c72f67ef3624246b564869937" checksum = "8f2e99e67736352bbb6ed1c273643975822505067ca32194b0981040bc50527a"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"flate2", "flate2",
@ -2270,9 +2237,9 @@ dependencies = [
[[package]] [[package]]
name = "lindera-core" name = "lindera-core"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e5f0baa9932f682e9c5b388897330f155d3c40de80016e60125897fde5e0e246" checksum = "7c3935e966409156f22cb4b334b21b0dce84b7aa1cad62214b466489d249c8e5"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"bincode", "bincode",
@ -2287,9 +2254,9 @@ dependencies = [
[[package]] [[package]]
name = "lindera-decompress" name = "lindera-decompress"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a6e63fa6ef0bc3ce2c26d372aa6185b7a316194494a84f81678f5da2893bf4a2" checksum = "7476406abb63c49d7f59c88b9b868ee8d2981495ea7e2c3ad129902f9916b3c6"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"flate2", "flate2",
@ -2298,63 +2265,50 @@ dependencies = [
[[package]] [[package]]
name = "lindera-dictionary" name = "lindera-dictionary"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fd765c36166016de87a1f447ea971573e4c63e334836c46ad0020f0408c88bfc" checksum = "808b7d2b3cabc25a4022526d484a4cfd1d5924dc76a26e0379707698841acef2"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"bincode", "bincode",
"byteorder", "byteorder",
"lindera-cc-cedict-builder",
"lindera-core", "lindera-core",
"lindera-ipadic",
"lindera-ko-dic",
"serde",
]
[[package]]
name = "lindera-filter"
version = "0.23.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a5345e37fb9521ab3cee19283bed135d46b3521dc1fd13a49fa0992379056203"
dependencies = [
"anyhow",
"bincode",
"byteorder",
"kanaria",
"lindera-core",
"lindera-dictionary",
"once_cell",
"regex",
"serde",
"serde_json",
"unicode-blocks",
"unicode-normalization",
"unicode-segmentation",
"yada",
]
[[package]]
name = "lindera-ipadic"
version = "0.23.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "60eeb356295f784e7db4cfd2c6772f2bd059e565a7744e246642a07bc333a88a"
dependencies = [
"bincode",
"byteorder",
"encoding",
"flate2",
"lindera-core",
"lindera-decompress",
"lindera-ipadic-builder", "lindera-ipadic-builder",
"once_cell", "lindera-ipadic-neologd-builder",
"tar", "lindera-ko-dic",
"lindera-ko-dic-builder",
"lindera-unidic",
"lindera-unidic-builder",
"serde",
] ]
[[package]] [[package]]
name = "lindera-ipadic-builder" name = "lindera-ipadic-builder"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0a16a2a88db9d956f5086bc976deb9951ca2dbbfef41a002df0a7bfb2c845aab" checksum = "31f373a280958c930e5ee4a1e4db3a0ee0542afaf02d3b5cacb8cab4e298648e"
dependencies = [
"anyhow",
"bincode",
"byteorder",
"csv",
"encoding_rs",
"encoding_rs_io",
"env_logger",
"glob",
"lindera-core",
"lindera-decompress",
"log",
"serde",
"yada",
]
[[package]]
name = "lindera-ipadic-neologd-builder"
version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "92eff98e9ed1a7a412b91709c2343457a04ef02fa0c27c27e3a5892f5591eae9"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"bincode", "bincode",
@ -2364,7 +2318,6 @@ dependencies = [
"encoding_rs_io", "encoding_rs_io",
"env_logger", "env_logger",
"glob", "glob",
"lindera-compress",
"lindera-core", "lindera-core",
"lindera-decompress", "lindera-decompress",
"log", "log",
@ -2374,9 +2327,9 @@ dependencies = [
[[package]] [[package]]
name = "lindera-ko-dic" name = "lindera-ko-dic"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "abb479b170a841b8cfbe602d772e30849ffe0562b219190a378368968b8c8f66" checksum = "74c6d5bf7d8092bd6d10de7a5d74b70ea7cf234586235b0d6cdb903b05a6c9e2"
dependencies = [ dependencies = [
"bincode", "bincode",
"byteorder", "byteorder",
@ -2391,9 +2344,9 @@ dependencies = [
[[package]] [[package]]
name = "lindera-ko-dic-builder" name = "lindera-ko-dic-builder"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9b9b58213552560717c48e7833444a20d2d7fe26a6e565f7ce0cbbf85784c7cf" checksum = "f0a4add6d3c1e41ec9e2690d33e287d0223fb59a30ccee4980c23f31368cae1e"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"bincode", "bincode",
@ -2410,10 +2363,42 @@ dependencies = [
] ]
[[package]] [[package]]
name = "lindera-unidic-builder" name = "lindera-tokenizer"
version = "0.23.0" version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6858147cdaf4a7b564c08a247449d3aca38e9b4812499651af08afbf85324596" checksum = "cb6a8acbd068019d1cdac7316f0dcb87f8e33ede2b13aa237f45114f9750afb8"
dependencies = [
"bincode",
"byteorder",
"lindera-core",
"lindera-dictionary",
"once_cell",
"serde",
"serde_json",
]
[[package]]
name = "lindera-unidic"
version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "14abf0613d350b30d3b0406a33b1de8fa8d829f26516909421702174785991c8"
dependencies = [
"bincode",
"byteorder",
"encoding",
"lindera-core",
"lindera-decompress",
"lindera-unidic-builder",
"once_cell",
"ureq",
"zip",
]
[[package]]
name = "lindera-unidic-builder"
version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e204ed53d9bd63227d1e6a6c1f122ca039e00a8634ac32e7fb0281eeec8615c4"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"bincode", "bincode",
@ -2422,6 +2407,7 @@ dependencies = [
"encoding", "encoding",
"env_logger", "env_logger",
"glob", "glob",
"lindera-compress",
"lindera-core", "lindera-core",
"lindera-decompress", "lindera-decompress",
"log", "log",
@ -3428,7 +3414,7 @@ dependencies = [
"wasm-bindgen", "wasm-bindgen",
"wasm-bindgen-futures", "wasm-bindgen-futures",
"web-sys", "web-sys",
"webpki-roots", "webpki-roots 0.22.6",
"winreg", "winreg",
] ]
@ -4211,12 +4197,6 @@ version = "0.3.13"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "92888ba5573ff080736b3648696b70cafad7d250551175acbaa4e0385b3e1460" checksum = "92888ba5573ff080736b3648696b70cafad7d250551175acbaa4e0385b3e1460"
[[package]]
name = "unicode-blocks"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "943e3f1f50cc455d072e0801ccb71ff893b0c88060b1169f92e35fb5bb881cc6"
[[package]] [[package]]
name = "unicode-ident" name = "unicode-ident"
version = "1.0.9" version = "1.0.9"
@ -4250,6 +4230,21 @@ version = "0.7.1"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a156c684c91ea7d62626509bce3cb4e1d9ed5c4d978f7b4352658f96a4c26b4a" checksum = "a156c684c91ea7d62626509bce3cb4e1d9ed5c4d978f7b4352658f96a4c26b4a"
[[package]]
name = "ureq"
version = "2.7.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0b11c96ac7ee530603dcdf68ed1557050f374ce55a5a07193ebf8cbc9f8927e9"
dependencies = [
"base64 0.21.2",
"log",
"once_cell",
"rustls 0.21.1",
"rustls-webpki",
"url",
"webpki-roots 0.23.1",
]
[[package]] [[package]]
name = "url" name = "url"
version = "2.3.1" version = "2.3.1"
@ -4458,6 +4453,15 @@ dependencies = [
"webpki", "webpki",
] ]
[[package]]
name = "webpki-roots"
version = "0.23.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b03058f88386e5ff5310d9111d53f48b17d732b401aeb83a8d5190f2ac459338"
dependencies = [
"rustls-webpki",
]
[[package]] [[package]]
name = "whatlang" name = "whatlang"
version = "0.16.2" version = "0.16.2"

8
meilisearch/src/search.rs
View File

@ -814,10 +814,10 @@ fn extract_field(
} }
} }
fn format_fields<A: AsRef<[u8]>>( fn format_fields<'a>(
document: &Document, document: &Document,
field_ids_map: &FieldsIdsMap, field_ids_map: &FieldsIdsMap,
builder: &MatcherBuilder<'_, A>, builder: &'a MatcherBuilder<'a>,
formatted_options: &BTreeMap<FieldId, FormatOptions>, formatted_options: &BTreeMap<FieldId, FormatOptions>,
compute_matches: bool, compute_matches: bool,
displayable_ids: &BTreeSet<FieldId>, displayable_ids: &BTreeSet<FieldId>,
@ -862,9 +862,9 @@ fn format_fields<A: AsRef<[u8]>>(
Ok((matches_position, document)) Ok((matches_position, document))
} }
fn format_value<A: AsRef<[u8]>>( fn format_value<'a>(
value: Value, value: Value,
builder: &MatcherBuilder<'_, A>, builder: &'a MatcherBuilder<'a>,
format_options: Option<FormatOptions>, format_options: Option<FormatOptions>,
infos: &mut Vec<MatchBounds>, infos: &mut Vec<MatchBounds>,
compute_matches: bool, compute_matches: bool,

2
milli/Cargo.toml
View File

@ -17,7 +17,7 @@ bincode = "1.3.3"
bstr = "1.4.0" bstr = "1.4.0"
bytemuck = { version = "1.13.1", features = ["extern_crate_alloc"] } bytemuck = { version = "1.13.1", features = ["extern_crate_alloc"] }
byteorder = "1.4.3" byteorder = "1.4.3"
charabia = { version = "0.7.2", default-features = false } charabia = { version = "0.8.1", default-features = false }
concat-arrays = "0.1.2" concat-arrays = "0.1.2"
crossbeam-channel = "0.5.8" crossbeam-channel = "0.5.8"
deserr = "0.5.0" deserr = "0.5.0"

3
milli/src/search/new/matches/matching_words.rs
View File

@ -256,7 +256,8 @@ pub(crate) mod tests {
let temp_index = temp_index_with_documents(); let temp_index = temp_index_with_documents();
let rtxn = temp_index.read_txn().unwrap(); let rtxn = temp_index.read_txn().unwrap();
let mut ctx = SearchContext::new(&temp_index, &rtxn); let mut ctx = SearchContext::new(&temp_index, &rtxn);
let tokenizer = TokenizerBuilder::new().build(); let mut builder = TokenizerBuilder::default();
let tokenizer = builder.build();
let tokens = tokenizer.tokenize("split this world"); let tokens = tokenizer.tokenize("split this world");
let query_terms = located_query_terms_from_tokens(&mut ctx, tokens, None).unwrap(); let query_terms = located_query_terms_from_tokens(&mut ctx, tokens, None).unwrap();
let matching_words = MatchingWords::new(ctx, query_terms); let matching_words = MatchingWords::new(ctx, query_terms);

22
milli/src/search/new/matches/mod.rs
View File

@ -12,16 +12,16 @@ const DEFAULT_HIGHLIGHT_PREFIX: &str = "<em>";
const DEFAULT_HIGHLIGHT_SUFFIX: &str = "</em>"; const DEFAULT_HIGHLIGHT_SUFFIX: &str = "</em>";
/// Structure used to build a Matcher allowing to customize formating tags. /// Structure used to build a Matcher allowing to customize formating tags.
pub struct MatcherBuilder<'a, A> { pub struct MatcherBuilder<'m> {
matching_words: MatchingWords, matching_words: MatchingWords,
tokenizer: Tokenizer<'a, 'a, A>, tokenizer: Tokenizer<'m>,
crop_marker: Option<String>, crop_marker: Option<String>,
highlight_prefix: Option<String>, highlight_prefix: Option<String>,
highlight_suffix: Option<String>, highlight_suffix: Option<String>,
} }
impl<'a, A> MatcherBuilder<'a, A> { impl<'m> MatcherBuilder<'m> {
pub fn new(matching_words: MatchingWords, tokenizer: Tokenizer<'a, 'a, A>) -> Self { pub fn new(matching_words: MatchingWords, tokenizer: Tokenizer<'m>) -> Self {
Self { Self {
matching_words, matching_words,
tokenizer, tokenizer,
@ -46,7 +46,7 @@ impl<'a, A> MatcherBuilder<'a, A> {
self self
} }
pub fn build<'t, 'm>(&'m self, text: &'t str) -> Matcher<'t, 'm, A> { pub fn build<'t>(&'m self, text: &'t str) -> Matcher<'t, 'm> {
let crop_marker = match &self.crop_marker { let crop_marker = match &self.crop_marker {
Some(marker) => marker.as_str(), Some(marker) => marker.as_str(),
None => DEFAULT_CROP_MARKER, None => DEFAULT_CROP_MARKER,
@ -103,17 +103,17 @@ pub struct MatchBounds {
/// Structure used to analize a string, compute words that match, /// Structure used to analize a string, compute words that match,
/// and format the source string, returning a highlighted and cropped sub-string. /// and format the source string, returning a highlighted and cropped sub-string.
pub struct Matcher<'t, 'm, A> { pub struct Matcher<'t, 'm> {
text: &'t str, text: &'t str,
matching_words: &'m MatchingWords, matching_words: &'m MatchingWords,
tokenizer: &'m Tokenizer<'m, 'm, A>, tokenizer: &'m Tokenizer<'m>,
crop_marker: &'m str, crop_marker: &'m str,
highlight_prefix: &'m str, highlight_prefix: &'m str,
highlight_suffix: &'m str, highlight_suffix: &'m str,
matches: Option<(Vec<Token<'t>>, Vec<Match>)>, matches: Option<(Vec<Token<'t>>, Vec<Match>)>,
} }
impl<'t, A: AsRef<[u8]>> Matcher<'t, '_, A> { impl<'t> Matcher<'t, '_> {
/// Iterates over tokens and save any of them that matches the query. /// Iterates over tokens and save any of them that matches the query.
fn compute_matches(&mut self) -> &mut Self { fn compute_matches(&mut self) -> &mut Self {
/// some words are counted as matches only if they are close together and in the good order, /// some words are counted as matches only if they are close together and in the good order,
@ -503,7 +503,7 @@ mod tests {
use crate::index::tests::TempIndex; use crate::index::tests::TempIndex;
use crate::{execute_search, SearchContext}; use crate::{execute_search, SearchContext};
impl<'a> MatcherBuilder<'a, &[u8]> { impl<'a> MatcherBuilder<'a> {
fn new_test(rtxn: &'a heed::RoTxn, index: &'a TempIndex, query: &str) -> Self { fn new_test(rtxn: &'a heed::RoTxn, index: &'a TempIndex, query: &str) -> Self {
let mut ctx = SearchContext::new(index, rtxn); let mut ctx = SearchContext::new(index, rtxn);
let crate::search::PartialSearchResult { located_query_terms, .. } = execute_search( let crate::search::PartialSearchResult { located_query_terms, .. } = execute_search(
@ -530,7 +530,7 @@ mod tests {
None => MatchingWords::default(), None => MatchingWords::default(),
}; };
MatcherBuilder::new(matching_words, TokenizerBuilder::new().build()) MatcherBuilder::new(matching_words, TokenizerBuilder::default().into_tokenizer())
} }
} }
@ -690,7 +690,7 @@ mod tests {
// should crop the phrase instead of croping around the match. // should crop the phrase instead of croping around the match.
insta::assert_snapshot!( insta::assert_snapshot!(
matcher.format(format_options), matcher.format(format_options),
@"… Split The World is a book written by Emily Henry…" @"…Split The World is a book written by Emily Henry…"
); );
// Text containing some matches. // Text containing some matches.

5
milli/src/search/new/query_term/parse_query.rs
View File

@ -7,7 +7,7 @@ use crate::{Result, SearchContext, MAX_WORD_LENGTH};
/// Convert the tokenised search query into a list of located query terms. /// Convert the tokenised search query into a list of located query terms.
pub fn located_query_terms_from_tokens( pub fn located_query_terms_from_tokens(
ctx: &mut SearchContext, ctx: &mut SearchContext,
query: NormalizedTokenIter<&[u8]>, query: NormalizedTokenIter,
words_limit: Option<usize>, words_limit: Option<usize>,
) -> Result<Vec<LocatedQueryTerm>> { ) -> Result<Vec<LocatedQueryTerm>> {
let nbr_typos = number_of_typos_allowed(ctx)?; let nbr_typos = number_of_typos_allowed(ctx)?;
@ -303,7 +303,8 @@ mod tests {
#[test] #[test]
fn start_with_hard_separator() -> Result<()> { fn start_with_hard_separator() -> Result<()> {
let tokenizer = TokenizerBuilder::new().build(); let mut builder = TokenizerBuilder::default();
let tokenizer = builder.build();
let tokens = tokenizer.tokenize("."); let tokens = tokenizer.tokenize(".");
let index = temp_index_with_documents(); let index = temp_index_with_documents();
let rtxn = index.read_txn()?; let rtxn = index.read_txn()?;

4
milli/src/update/index_documents/extract/extract_docid_word_positions.rs
View File

@ -128,10 +128,10 @@ pub fn extract_docid_word_positions<R: io::Read + io::Seek>(
.map(|reader| (documents_ids, reader, script_language_docids)) .map(|reader| (documents_ids, reader, script_language_docids))
} }
fn extract_tokens_from_document<T: AsRef<[u8]>>( fn extract_tokens_from_document(
obkv: &KvReader<FieldId>, obkv: &KvReader<FieldId>,
searchable_fields: &Option<HashSet<FieldId>>, searchable_fields: &Option<HashSet<FieldId>>,
tokenizer: &Tokenizer<T>, tokenizer: &Tokenizer,
max_positions_per_attributes: u32, max_positions_per_attributes: u32,
buffers: &mut Buffers, buffers: &mut Buffers,
script_language_word_count: &mut HashMap<Script, Vec<(Language, usize)>>, script_language_word_count: &mut HashMap<Script, Vec<(Language, usize)>>,

11
milli/src/update/index_documents/snapshots/mod.rs/simple_documents_replace/initial/word_docids.snap
View File

@ -1,18 +1,21 @@
--- ---
source: milli/src/update/index_documents/mod.rs source: milli/src/update/index_documents/mod.rs
--- ---
0 [1, ]
1 [2, ] 1 [2, ]
10.0 [1, ] 10 [1, ]
12 [0, ] 12 [0, ]
1344 [3, ] 1344 [3, ]
2 [0, ] 2 [0, ]
23 [5, ] 23 [5, ]
25.99 [2, ] 25 [2, ]
3.5 [0, ] 3 [0, ]
35 [5, ] 35 [5, ]
4 [4, ] 4 [4, ]
42 [0, 5, ] 42 [0, 5, ]
456 [1, ] 456 [1, ]
5 [0, ]
99 [2, ]
adams [5, ] adams [5, ]
adventure [1, ] adventure [1, ]
alice [2, ] alice [2, ]
@ -29,7 +32,7 @@ galaxy [5, ]
guide [5, ] guide [5, ]
half [4, ] half [4, ]
harry [4, ] harry [4, ]
hitchhiker' [5, ] hitchhiker [5, ]
hobbit [3, ] hobbit [3, ]
in [2, ] in [2, ]
j [3, 4, ] j [3, 4, ]

11
milli/src/update/index_documents/snapshots/mod.rs/simple_documents_replace/updated/word_docids.snap
View File

@ -1,19 +1,22 @@
--- ---
source: milli/src/update/index_documents/mod.rs source: milli/src/update/index_documents/mod.rs
--- ---
0 [1, 7, ]
1 [2, ] 1 [2, ]
10.0 [1, 7, ] 10 [1, 7, ]
12 [0, 8, ] 12 [0, 8, ]
1344 [3, ] 1344 [3, ]
1813 [8, ] 1813 [8, ]
2 [0, 8, ] 2 [0, 8, ]
23 [5, ] 23 [5, ]
25.99 [2, ] 25 [2, ]
3.5 [0, 8, ] 3 [0, 8, ]
35 [5, ] 35 [5, ]
4 [4, 6, ] 4 [4, 6, ]
42 [0, 5, 8, ] 42 [0, 5, 8, ]
456 [1, 7, ] 456 [1, 7, ]
5 [0, 8, ]
99 [2, ]
adams [5, ] adams [5, ]
adventure [1, 7, ] adventure [1, 7, ]
alice [2, ] alice [2, ]
@ -31,7 +34,7 @@ galaxy [5, ]
guide [5, ] guide [5, ]
half [4, 6, ] half [4, 6, ]
harry [4, 6, ] harry [4, 6, ]
hitchhiker' [5, ] hitchhiker [5, ]
hobbit [3, ] hobbit [3, ]
in [2, ] in [2, ]
j [3, 4, 6, 8, ] j [3, 4, 6, 8, ]

12
milli/src/update/settings.rs
View File

@ -1,7 +1,7 @@
use std::collections::{BTreeSet, HashMap, HashSet}; use std::collections::{BTreeSet, HashMap, HashSet};
use std::result::Result as StdResult; use std::result::Result as StdResult;
use charabia::{Tokenizer, TokenizerBuilder}; use charabia::{Normalize, Tokenizer, TokenizerBuilder};
use deserr::{DeserializeError, Deserr}; use deserr::{DeserializeError, Deserr};
use itertools::Itertools; use itertools::Itertools;
use serde::{Deserialize, Deserializer, Serialize, Serializer}; use serde::{Deserialize, Deserializer, Serialize, Serializer};
@ -423,6 +423,12 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
match self.stop_words { match self.stop_words {
Setting::Set(ref stop_words) => { Setting::Set(ref stop_words) => {
let current = self.index.stop_words(self.wtxn)?; let current = self.index.stop_words(self.wtxn)?;
// Apply an unlossy normalization on stop_words
let stop_words = stop_words
.iter()
.map(|w| w.as_str().normalize(&Default::default()).into_owned());
// since we can't compare a BTreeSet with an FST we are going to convert the // since we can't compare a BTreeSet with an FST we are going to convert the
// BTreeSet to an FST and then compare bytes per bytes the two FSTs. // BTreeSet to an FST and then compare bytes per bytes the two FSTs.
let fst = fst::Set::from_iter(stop_words)?; let fst = fst::Set::from_iter(stop_words)?;
@ -446,7 +452,7 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
fn update_synonyms(&mut self) -> Result<bool> { fn update_synonyms(&mut self) -> Result<bool> {
match self.synonyms { match self.synonyms {
Setting::Set(ref synonyms) => { Setting::Set(ref synonyms) => {
fn normalize(tokenizer: &Tokenizer<&[u8]>, text: &str) -> Vec<String> { fn normalize(tokenizer: &Tokenizer, text: &str) -> Vec<String> {
tokenizer tokenizer
.tokenize(text) .tokenize(text)
.filter_map(|token| { .filter_map(|token| {
@ -647,7 +653,7 @@ impl<'a, 't, 'u, 'i> Settings<'a, 't, 'u, 'i> {
fn update_exact_words(&mut self) -> Result<()> { fn update_exact_words(&mut self) -> Result<()> {
match self.exact_words { match self.exact_words {
Setting::Set(ref mut words) => { Setting::Set(ref mut words) => {
fn normalize(tokenizer: &Tokenizer<&[u8]>, text: &str) -> String { fn normalize(tokenizer: &Tokenizer, text: &str) -> String {
tokenizer.tokenize(text).map(|token| token.lemma().to_string()).collect() tokenizer.tokenize(text).map(|token| token.lemma().to_string()).collect()
} }

43
milli/src/update/snapshots/delete_documents.rs/filtered_placeholder_search_should_not_return_deleted_documents/always_hard/word_docids.snap
View File

@ -1,27 +1,28 @@
--- ---
source: milli/src/update/delete_documents.rs source: milli/src/update/delete_documents.rs
--- ---
1_36 [3, ] 1 [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21, ]
1_37 [4, ] 2 [21, ]
1_38 [5, ] 36 [3, ]
1_39 [6, ] 37 [4, ]
1_40 [7, ] 38 [5, ]
1_41 [8, ] 39 [6, ]
1_42 [9, ] 40 [7, ]
1_43 [10, ] 41 [8, ]
1_44 [11, ] 42 [9, ]
1_45 [12, ] 43 [10, ]
1_46 [13, ] 44 [11, ]
1_47 [14, ] 45 [12, ]
1_5 [1, ] 46 [13, ]
1_52 [15, ] 47 [14, ]
1_57 [16, ] 5 [1, ]
1_58 [17, ] 52 [15, ]
1_68 [18, ] 57 [16, ]
1_69 [19, ] 58 [17, ]
1_7 [2, ] 68 [18, ]
1_71 [21, ] 69 [19, ]
2.2 [21, ] 7 [2, ]
71 [21, ]
abstract [2, 6, 10, 13, 14, 15, 16, 17, ] abstract [2, 6, 10, 13, 14, 15, 16, 17, ]
aquarium [5, ] aquarium [5, ]
art [4, 5, 8, 9, 10, 12, 17, ] art [4, 5, 8, 9, 10, 12, 17, ]

21
milli/src/update/snapshots/delete_documents.rs/filtered_placeholder_search_should_not_return_deleted_documents/always_hard/word_pair_proximity_docids.snap
View File

@ -1,4 +1,25 @@
--- ---
source: milli/src/update/delete_documents.rs source: milli/src/update/delete_documents.rs
--- ---
1 1 36 [3, ]
1 1 37 [4, ]
1 1 38 [5, ]
1 1 39 [6, ]
1 1 40 [7, ]
1 1 41 [8, ]
1 1 42 [9, ]
1 1 43 [10, ]
1 1 44 [11, ]
1 1 45 [12, ]
1 1 46 [13, ]
1 1 47 [14, ]
1 1 5 [1, ]
1 1 52 [15, ]
1 1 57 [16, ]
1 1 58 [17, ]
1 1 68 [18, ]
1 1 69 [19, ]
1 1 7 [2, ]
1 1 71 [21, ]
1 2 2 [21, ]

50
milli/src/update/snapshots/delete_documents.rs/filtered_placeholder_search_should_not_return_deleted_documents/always_soft/word_docids.snap
View File

@ -1,31 +1,31 @@
--- ---
source: milli/src/update/delete_documents.rs source: milli/src/update/delete_documents.rs
--- ---
1.2 [20, 22, ] 1 [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, ]
1_36 [3, ] 2 [20, 21, 22, ]
1_37 [4, ] 36 [3, ]
1_38 [5, ] 37 [4, ]
1_39 [6, ] 38 [5, ]
1_4 [0, ] 39 [6, ]
1_40 [7, ] 4 [0, ]
1_41 [8, ] 40 [7, ]
1_42 [9, ] 41 [8, ]
1_43 [10, ] 42 [9, ]
1_44 [11, ] 43 [10, ]
1_45 [12, ] 44 [11, ]
1_46 [13, ] 45 [12, ]
1_47 [14, ] 46 [13, ]
1_5 [1, ] 47 [14, ]
1_52 [15, ] 5 [1, ]
1_57 [16, ] 52 [15, ]
1_58 [17, ] 57 [16, ]
1_68 [18, ] 58 [17, ]
1_69 [19, ] 68 [18, ]
1_7 [2, ] 69 [19, ]
1_70 [20, ] 7 [2, ]
1_71 [21, ] 70 [20, ]
1_72 [22, ] 71 [21, ]
2.2 [21, ] 72 [22, ]
abstract [2, 6, 10, 13, 14, 15, 16, 17, ] abstract [2, 6, 10, 13, 14, 15, 16, 17, ]
aquarium [5, ] aquarium [5, ]
art [4, 5, 8, 9, 10, 12, 17, ] art [4, 5, 8, 9, 10, 12, 17, ]

25
milli/src/update/snapshots/delete_documents.rs/filtered_placeholder_search_should_not_return_deleted_documents/always_soft/word_pair_proximity_docids.snap
View File

@ -1,4 +1,29 @@
--- ---
source: milli/src/update/delete_documents.rs source: milli/src/update/delete_documents.rs
--- ---
1 1 2 [20, 22, ]
1 1 36 [3, ]
1 1 37 [4, ]
1 1 38 [5, ]
1 1 39 [6, ]
1 1 4 [0, ]
1 1 40 [7, ]
1 1 41 [8, ]
1 1 42 [9, ]
1 1 43 [10, ]
1 1 44 [11, ]
1 1 45 [12, ]
1 1 46 [13, ]
1 1 47 [14, ]
1 1 5 [1, ]
1 1 52 [15, ]
1 1 57 [16, ]
1 1 58 [17, ]
1 1 68 [18, ]
1 1 69 [19, ]
1 1 7 [2, ]
1 1 70 [20, ]
1 1 71 [21, ]
1 1 72 [22, ]
1 2 2 [21, ]