Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix bug in segmenter involving short strings of complex scripts #3341

Merged
merged 4 commits into from
Apr 18, 2023
Merged

Fix bug in segmenter involving short strings of complex scripts #3341

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
7670d16
Start debugging Segmenter bug
sffc Apr 17, 2023
dc54147
Fix the bug (in rule segmenter UTF-8 only)
sffc Apr 17, 2023
27b94f0
Apply fixes to Line and make UTF-16 more consistent
sffc Apr 17, 2023
61e5730
Better test cases
sffc Apr 17, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Better test cases
  • Loading branch information
@sffc
sffc committed Apr 17, 2023
commit 61e57303a49fe0e2318f2b2dc417f79bee3ba70e
12 changes: 10 additions & 2 deletions components/segmenter/src/line.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1080,7 +1080,11 @@ where
iter.result_cache = iter.result_cache.iter().skip(1).map(|r| r - i).collect();
return iter.get_current_position();
}
debug_assert!(i < first_pos, "we should always arrive at first_pos: near index {:?}", iter.get_current_position());
debug_assert!(
i < first_pos,
"we should always arrive at first_pos: near index {:?}",
iter.get_current_position()
);
i += T::get_current_position_character_len(iter);
iter.advance_iter();
if iter.is_eof() {
Expand Down Expand Up @@ -1189,7 +1193,11 @@ impl<'l, 's> LineBreakType<'l, 's> for LineBreakTypeUtf16 {
.collect();
return iterator.get_current_position();
}
debug_assert!(i < first_pos, "we should always arrive at first_pos: near index {:?}", iterator.get_current_position());
debug_assert!(
i < first_pos,
"we should always arrive at first_pos: near index {:?}",
iterator.get_current_position()
);
i += 1;
iterator.advance_iter();
if iterator.is_eof() {
Expand Down
12 changes: 10 additions & 2 deletions components/segmenter/src/word.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -468,7 +468,11 @@ where
iter.result_cache = iter.result_cache.iter().skip(1).map(|r| r - i).collect();
return iter.get_current_position();
}
debug_assert!(i < first_pos, "we should always arrive at first_pos: near index {:?}", iter.get_current_position());
debug_assert!(
i < first_pos,
"we should always arrive at first_pos: near index {:?}",
iter.get_current_position()
);
i += T::get_current_position_character_len(iter);
iter.advance_iter();
if iter.is_eof() {
Expand Down Expand Up @@ -530,7 +534,11 @@ impl<'l, 's> RuleBreakType<'l, 's> for WordBreakTypeUtf16 {
iter.result_cache = iter.result_cache.iter().skip(1).map(|r| r - i).collect();
return iter.get_current_position();
}
debug_assert!(i < first_pos, "we should always arrive at first_pos: near index {:?}", iter.get_current_position());
debug_assert!(
i < first_pos,
"we should always arrive at first_pos: near index {:?}",
iter.get_current_position()
);
i += 1;
iter.advance_iter();
if iter.is_eof() {
Expand Down
49 changes: 43 additions & 6 deletions components/segmenter/tests/complex_word.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,22 +107,59 @@ fn word_break_mixed_han() {
}

#[test]
fn word_break_th_wikipedia() {
let text = "แพนด้าแดง (อังกฤษ: Red panda, Shining cat; จีน: 小熊貓; พินอิน: Xiǎo xióngmāo) สัตว์เลี้ยงลูกด้วยนมชนิดหนึ่ง มีชื่อวิทยาศาสตร์ว่า Ailurus fulgens";
fn word_line_th_wikipedia_auto() {
use icu_segmenter::LineSegmenter;

let text = "แพนด้าแดง (อังกฤษ: Red panda, Shining cat; จีน: 小熊貓; พินอิน: Xiǎo xióngmāo) สัตว์เลี้ยงลูกด้วยนมชนิดหนึ่ง มีชื่อวิทยาศาสตร์ว่า Ailurus fulgens";
assert_eq!(text.len(), 297);
let utf16: Vec<u16> = text.encode_utf16().collect();
assert_eq!(utf16.len(), 142);

let segmenter_line_auto =
let segmenter_word_auto =
WordSegmenter::try_new_auto_unstable(&icu_testdata::unstable()).expect("Data exists");
let segmenter_line_auto =
LineSegmenter::try_new_auto_unstable(&icu_testdata::unstable()).expect("Data exists");

let breakpoints = segmenter_line_auto.segment_str(text).collect::<Vec<_>>();

let breakpoints_word_utf8 = segmenter_word_auto.segment_str(text).collect::<Vec<_>>();
assert_eq!(
breakpoints,
breakpoints_word_utf8,
[
0, 9, 18, 27, 28, 29, 38, 47, 48, 49, 52, 53, 58, 59, 60, 67, 68, 71, 72, 73, 82, 83,
84, 90, 93, 94, 95, 104, 113, 114, 115, 120, 121, 131, 132, 133, 148, 166, 175, 187,
193, 205, 220, 221, 227, 239, 272, 281, 282, 289, 290, 297
]
);

let breakpoints_line_utf8 = segmenter_line_auto.segment_str(text).collect::<Vec<_>>();
assert_eq!(
breakpoints_line_utf8,
[
0, 9, 18, 27, 28, 38, 47, 49, 53, 60, 68, 73, 82, 84, 87, 90, 95, 104, 113, 115, 121,
133, 148, 166, 175, 187, 193, 205, 220, 221, 227, 239, 272, 281, 282, 290, 297
]
);

let breakpoints_word_utf16 = segmenter_word_auto
.segment_utf16(&utf16)
.collect::<Vec<_>>();
assert_eq!(
breakpoints_word_utf16,
[
0, 3, 6, 9, 10, 11, 14, 17, 18, 19, 22, 23, 28, 29, 30, 37, 38, 41, 42, 43, 46, 47, 48,
50, 51, 52, 53, 56, 59, 60, 61, 65, 66, 74, 75, 76, 81, 87, 90, 94, 96, 100, 105, 106,
108, 112, 123, 126, 127, 134, 135, 142
]
);

let breakpoints_word_utf16 = segmenter_word_auto
.segment_utf16(&utf16)
.collect::<Vec<_>>();
assert_eq!(
breakpoints_word_utf16,
[
0, 3, 6, 9, 10, 11, 14, 17, 18, 19, 22, 23, 28, 29, 30, 37, 38, 41, 42, 43, 46, 47, 48,
50, 51, 52, 53, 56, 59, 60, 61, 65, 66, 74, 75, 76, 81, 87, 90, 94, 96, 100, 105, 106,
108, 112, 123, 126, 127, 134, 135, 142
]
);
}