searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-12-27 15:10:05 +01:00
Code Issues Releases Wiki Activity

fix: Change the tokenizer to mesure cjk chars positions

Browse Source
This commit is contained in:
Clément Renault 2019-02-22 23:06:42 +01:00
parent 10414791a2
commit a799470997
No known key found for this signature in database
GPG Key ID: 0151CDAB43460DAE
1 changed files with 9 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
src/tokenizer/mod.rs
View File

@ -124,8 +124,6 @@ impl<'a> Iterator for Tokenizer<'a> {
(c >= '\u{4e00}' && c <= '\u{9fff}') ||
(c >= '\u{f900}' && c <= '\u{faff}')
{
let char_len = c.len_utf8();
match start_word {
Some(start_word) => {
let (prefix, tail) = self.inner.split_at(i);
@ -147,7 +145,7 @@ impl<'a> Iterator for Tokenizer<'a> {
return Some(token)
},
None => {
let (prefix, tail) = self.inner.split_at(i + char_len);
let (prefix, tail) = self.inner.split_at(i + c.len_utf8());
let (spaces, word) = prefix.split_at(i);
self.inner = tail;
@ -163,7 +161,7 @@ impl<'a> Iterator for Tokenizer<'a> {
if tail.chars().next().and_then(detect_separator).is_none() {
self.word_index += 1;
}
self.char_index += char_len;
self.char_index += 1;
return Some(token)
}
@ -252,18 +250,18 @@ mod tests {
let mut tokenizer = Tokenizer::new("\u{2ec4}lolilol\u{2ec7}");
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ec4}", word_index: 0, char_index: 0 }));
assert_eq!(tokenizer.next(), Some(Token { word: "lolilol", word_index: 1, char_index: 3 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ec7}", word_index: 2, char_index: 10 }));
assert_eq!(tokenizer.next(), Some(Token { word: "lolilol", word_index: 1, char_index: 1 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ec7}", word_index: 2, char_index: 8 }));
assert_eq!(tokenizer.next(), None);
let mut tokenizer = Tokenizer::new("\u{2ec4}\u{2ed3}\u{2ef2} lolilol - hello \u{2ec7}");
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ec4}", word_index: 0, char_index: 0 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ed3}", word_index: 1, char_index: 3 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ef2}", word_index: 2, char_index: 6 }));
assert_eq!(tokenizer.next(), Some(Token { word: "lolilol", word_index: 3, char_index: 10 }));
assert_eq!(tokenizer.next(), Some(Token { word: "hello", word_index: 11, char_index: 20 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ec7}", word_index: 12, char_index: 29 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ed3}", word_index: 1, char_index: 1 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ef2}", word_index: 2, char_index: 2 }));
assert_eq!(tokenizer.next(), Some(Token { word: "lolilol", word_index: 3, char_index: 4 }));
assert_eq!(tokenizer.next(), Some(Token { word: "hello", word_index: 11, char_index: 14 }));
assert_eq!(tokenizer.next(), Some(Token { word: "\u{2ec7}", word_index: 12, char_index: 23 }));
assert_eq!(tokenizer.next(), None);
}
}