Skip to content

Add support for Grapheme_Cluster_Break and Sentence_Break #96

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
+4,635 −42
Open

Add support for Grapheme_Cluster_Break and Sentence_Break #96

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
1,513 changes: 1,513 additions & 0 deletions data/17.0.0-grapheme-cluster-break.txt
View file
Open in desktop

Large diffs are not rendered by default.

3,019 changes: 3,019 additions & 0 deletions data/17.0.0-sentence-break.txt
View file
Open in desktop

Large diffs are not rendered by default.

121 changes: 84 additions & 37 deletions data/resources.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -176,8 +176,12 @@ const resources = [
'special-casing': 'https://unicode.org/Public/4.1.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/4.1.0/ucd/BidiMirroring.txt',
'line-break': 'https://unicode.org/Public/4.1.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/4.1.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/4.1.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/4.1.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '5.0.0',
Expand All @@ -199,8 +203,12 @@ const resources = [
'special-casing': 'https://unicode.org/Public/5.0.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/5.0.0/ucd/BidiMirroring.txt',
'line-break': 'https://unicode.org/Public/5.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/5.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/5.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/5.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '5.1.0',
Expand All @@ -222,8 +230,12 @@ const resources = [
'special-casing': 'https://unicode.org/Public/5.1.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/5.1.0/ucd/BidiMirroring.txt',
'line-break': 'https://unicode.org/Public/5.1.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/5.1.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/5.1.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/5.1.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '5.2.0',
Expand All @@ -245,8 +257,12 @@ const resources = [
'special-casing': 'https://unicode.org/Public/5.2.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/5.2.0/ucd/BidiMirroring.txt',
'line-break': 'https://unicode.org/Public/5.2.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/5.2.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/5.2.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/5.2.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '6.0.0',
Expand All @@ -270,8 +286,12 @@ const resources = [
'special-casing': 'https://unicode.org/Public/6.0.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/6.0.0/ucd/BidiMirroring.txt',
'line-break': 'https://unicode.org/Public/6.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/6.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/6.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/6.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '6.1.0',
Expand All @@ -296,8 +316,12 @@ const resources = [
'special-casing': 'https://unicode.org/Public/6.1.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/6.1.0/ucd/BidiMirroring.txt',
'line-break': 'https://unicode.org/Public/6.1.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/6.1.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/6.1.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/6.1.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '6.2.0',
Expand All @@ -322,8 +346,12 @@ const resources = [
'special-casing': 'https://unicode.org/Public/6.2.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/6.2.0/ucd/BidiMirroring.txt',
'line-break': 'https://unicode.org/Public/6.2.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/6.2.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/6.2.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/6.2.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '6.3.0',
Expand All @@ -349,8 +377,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/6.3.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/6.3.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/6.3.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/6.3.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/6.3.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/6.3.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '7.0.0',
Expand All @@ -376,8 +408,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/7.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/7.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/7.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/7.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/7.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/7.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '8.0.0',
Expand All @@ -403,8 +439,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/8.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/8.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/8.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/8.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/8.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/8.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '9.0.0',
Expand All @@ -430,8 +470,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/9.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/9.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/9.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/9.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/9.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/9.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '10.0.0',
Expand All @@ -457,8 +501,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/10.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/10.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/10.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/10.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/10.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/10.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
},
{
version: '11.0.0',
Expand All @@ -484,8 +532,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/11.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/11.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/11.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/11.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/11.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/11.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/emoji/11.0/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -521,8 +573,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/12.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/12.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/12.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/12.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/12.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/12.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/emoji/12.0/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -558,8 +614,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/12.1.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/12.1.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/12.1.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/12.1.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/12.1.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/12.1.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/emoji/12.1/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -595,8 +655,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/13.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/13.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/13.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/13.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/13.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/13.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/13.0.0/ucd/emoji/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -632,8 +696,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/14.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/14.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/14.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/14.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/14.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/14.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/14.0.0/ucd/emoji/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -669,8 +737,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/15.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/15.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/15.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/15.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/15.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/15.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/15.0.0/ucd/emoji/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -706,8 +778,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/15.1.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/15.1.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/15.1.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/15.1.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/15.1.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/15.1.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/15.1.0/ucd/emoji/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -743,8 +819,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/16.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/16.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/16.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/16.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/16.0.0/ucd/auxiliary/WordBreakProperty.txt',
'sentence-break':
'https://unicode.org/Public/16.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/16.0.0/ucd/emoji/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down Expand Up @@ -780,45 +860,12 @@ const resources = [
'bidi-mirroring': 'https://unicode.org/Public/17.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/17.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/17.0.0/ucd/LineBreak.txt',
'grapheme-cluster-break':
'https://unicode.org/Public/17.0.0/ucd/auxiliary/GraphemeBreakProperty.txt',
'word-break':
'https://unicode.org/Public/17.0.0/ucd/auxiliary/WordBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/17.0.0/ucd/emoji/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
'emoji-sequences':
'https://unicode.org/Public/emoji/17.0/emoji-sequences.txt',
// Emoji_ZWJ_Sequence
'emoji-zwj-sequences':
'https://unicode.org/Public/emoji/17.0/emoji-zwj-sequences.txt',
// Emoji_Test (not an official property)
'emoji-test': 'https://unicode.org/Public/emoji/17.0/emoji-test.txt',
},
{
version: '17.0.0',
main: 'https://unicode.org/Public/17.0.0/ucd/UnicodeData.txt',
scripts: 'https://unicode.org/Public/17.0.0/ucd/Scripts.txt',
'script-extensions':
'https://unicode.org/Public/17.0.0/ucd/ScriptExtensions.txt',
blocks: 'https://unicode.org/Public/17.0.0/ucd/Blocks.txt',
properties: 'https://unicode.org/Public/17.0.0/ucd/PropList.txt',
'name-aliases': 'https://unicode.org/Public/17.0.0/ucd/NameAliases.txt',
'derived-binary-properties':
'https://unicode.org/Public/17.0.0/ucd/extracted/DerivedBinaryProperties.txt',
'derived-core-properties':
'https://unicode.org/Public/17.0.0/ucd/DerivedCoreProperties.txt',
'derived-general-category':
'https://unicode.org/Public/17.0.0/ucd/extracted/DerivedGeneralCategory.txt',
'derived-normalization-properties':
'https://unicode.org/Public/17.0.0/ucd/DerivedNormalizationProps.txt',
'composition-exclusions':
'https://unicode.org/Public/17.0.0/ucd/CompositionExclusions.txt',
'case-folding': 'https://unicode.org/Public/17.0.0/ucd/CaseFolding.txt',
'special-casing': 'https://unicode.org/Public/17.0.0/ucd/SpecialCasing.txt',
'bidi-mirroring': 'https://unicode.org/Public/17.0.0/ucd/BidiMirroring.txt',
'bidi-brackets': 'https://unicode.org/Public/17.0.0/ucd/BidiBrackets.txt',
'line-break': 'https://unicode.org/Public/17.0.0/ucd/LineBreak.txt',
'word-break':
'https://unicode.org/Public/17.0.0/ucd/auxiliary/WordBreakProperty.txt',
Comment on lines -785 to -821
Copy link
Author

@lionel-rowe lionel-rowe Nov 15, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The data for version: '17.0.0' was duplicated, presumably in error (?), so I deleted the duplicate lines.

'sentence-break':
'https://unicode.org/Public/17.0.0/ucd/auxiliary/SentenceBreakProperty.txt',
// Emoji, Emoji_Presentation, Emoji_Modifier, Emoji_Modifier_Base, Emoji_Component, Extended_Pictographic
emoji: 'https://unicode.org/Public/17.0.0/ucd/emoji/emoji-data.txt',
// Emoji_Keycap_Sequence, Emoji_Flag_Sequence, Emoji_Modifier_Sequence
Expand Down
16 changes: 14 additions & 2 deletions index.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ parsers.parseLineBreak = require('./scripts/parse-line-break.js');
parsers.parseScriptExtensions = require('./scripts/parse-script-extensions.js');
parsers.parseSpecialCasing = require('./scripts/parse-special-casing.js');
parsers.parseSimpleCaseMapping = require('./scripts/parse-simple-case-mapping.js');
parsers.parseWordBreak = require('./scripts/parse-word-break.js');
parsers.parseGraphemeWordSentenceBreak = require('./scripts/parse-grapheme-word-sentence-break.js');
parsers.parseEmoji = require('./scripts/parse-emoji.js');
parsers.parseEmojiSequences = require('./scripts/parse-emoji-sequences.js');
parsers.parseNames = require('./scripts/parse-names.js');
Expand Down Expand Up @@ -126,12 +126,24 @@ const generateData = function(version) {
'map': parsers.parseLineBreak(version),
'type': 'Line_Break'
}));
console.log('Parsing Unicode v%s `Grapheme_Cluster_Break`…', version);
extend(dirMap, utils.writeFiles({
'version': version,
'map': parsers.parseGraphemeWordSentenceBreak(version, 'grapheme-cluster-break'),
'type': 'Grapheme_Cluster_Break'
}));
console.log('Parsing Unicode v%s `Word_Break`…', version);
extend(dirMap, utils.writeFiles({
'version': version,
'map': parsers.parseWordBreak(version),
'map': parsers.parseGraphemeWordSentenceBreak(version, 'word-break'),
'type': 'Word_Break'
}));
console.log('Parsing Unicode v%s `Sentence_Break`…', version);
extend(dirMap, utils.writeFiles({
'version': version,
'map': parsers.parseGraphemeWordSentenceBreak(version, 'sentence-break'),
'type': 'Sentence_Break'
}));
console.log('Parsing Unicode v%s binary emoji properties…', version);
extend(dirMap, utils.writeFiles({
'version': version,
Expand Down
2 changes: 2 additions & 0 deletions scripts/download.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,9 @@ const TYPES = [
'bidi-mirroring',
'bidi-brackets',
'line-break',
'grapheme-cluster-break',
'word-break',
'sentence-break',
'emoji',
'emoji-sequences',
'emoji-test',
Expand Down
6 changes: 3 additions & 3 deletions scripts/parse-word-break.js → ...pts/parse-grapheme-word-sentence-break.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,8 @@
const utils = require('./utils.js');
const regenerate = require('regenerate');

const parseWordBreak = function(version) {
const source = utils.readDataFile(version, 'word-break');
const parseGraphemeWordSentenceBreak = function(version, kind) {
const source = utils.readDataFile(version, kind);
if (!source) {
return;
}
Expand Down Expand Up @@ -39,4 +39,4 @@ const parseWordBreak = function(version) {
return map;
};

module.exports = parseWordBreak;
module.exports = parseGraphemeWordSentenceBreak;