RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-27 19:37:34 +00:00
Code Issues Activity Actions

LibUnicode: Fix typos causing text segmentation on mid-word punctuation

Browse source 
For example the words "can't" and "32.3" should not have boundaries
detected on the "'" and "." code points, respectively.

The String test cases fixed here are because "b'ar" is now considered
one word.
This commit is contained in:
Timothy Flynn 2023-02-14 13:56:19 -05:00 • committed by Linus Groh
parent 6e7a6e2d02
commit 5cbf054651
4 changed files with 45 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

42
Userland/Libraries/LibJS/Tests/builtins/Intl/Segmenter/Segmenter.prototype.segment.js
View file

@ -82,7 +82,6 @@ describe("correct behavior", () => {
];
index = 0;
for (const segment of wordSegments) {
console.log(JSON.stringify(segment));
expect(segment.segment).toBe(expectedSegments[index].segment);
expect(segment.index).toBe(expectedSegments[index].index);
expect(segment.input).toBe(string);
@ -103,4 +102,45 @@ describe("correct behavior", () => {
}
expect(index).toBe(1);
});
test("word segmentation of string with mid-word punctuation", () => {
const string = "The quick (“brown”) fox can’t jump 32.3 feet, right?";
const segmenter = new Intl.Segmenter([], { granularity: "word" });
const segments = segmenter.segment(string);
const expectedSegments = [
{ segment: "The", index: 0, isWordLike: true },
{ segment: " ", index: 3, isWordLike: false },
{ segment: "quick", index: 4, isWordLike: true },
{ segment: " ", index: 9, isWordLike: false },
{ segment: "(", index: 10, isWordLike: false },
{ segment: "“", index: 11, isWordLike: false },
{ segment: "brown", index: 12, isWordLike: true },
{ segment: "”", index: 17, isWordLike: false },
{ segment: ")", index: 18, isWordLike: false },
{ segment: " ", index: 19, isWordLike: false },
{ segment: "fox", index: 20, isWordLike: true },
{ segment: " ", index: 23, isWordLike: false },
{ segment: "can’t", index: 24, isWordLike: true },
{ segment: " ", index: 29, isWordLike: false },
{ segment: "jump", index: 30, isWordLike: true },
{ segment: " ", index: 34, isWordLike: false },
{ segment: "32.3", index: 35, isWordLike: true },
{ segment: " ", index: 39, isWordLike: false },
{ segment: "feet", index: 40, isWordLike: true },
{ segment: ",", index: 44, isWordLike: false },
{ segment: " ", index: 45, isWordLike: false },
{ segment: "right", index: 46, isWordLike: true },
{ segment: "?", index: 51, isWordLike: false },
];
let index = 0;
for (const segment of segments) {
expect(segment.segment).toBe(expectedSegments[index].segment);
expect(segment.index).toBe(expectedSegments[index].index);
expect(segment.input).toBe(string);
index++;
}
});
});