cldr-segments-full 43.0.0 → 44.0.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (17) hide show
  1. package/LICENSE +32 -39
  2. package/bower.json +2 -2
  3. package/package.json +2 -2
  4. package/segments/de/suppressions.json +1 -1
  5. package/segments/el/suppressions.json +1 -1
  6. package/segments/en/suppressions.json +1 -1
  7. package/segments/es/suppressions.json +1 -1
  8. package/segments/fi/suppressions.json +1 -1
  9. package/segments/fr/suppressions.json +1 -1
  10. package/segments/it/suppressions.json +1 -1
  11. package/segments/ja/suppressions.json +1 -1
  12. package/segments/pt/suppressions.json +1 -1
  13. package/segments/ru/suppressions.json +1 -1
  14. package/segments/sv/suppressions.json +1 -1
  15. package/segments/und/suppressions.json +86 -29
  16. package/segments/zh/suppressions.json +1 -1
  17. package/segments/zh-Hant/suppressions.json +1 -1
package/LICENSE CHANGED
@@ -1,46 +1,39 @@
1
- UNICODE, INC. LICENSE AGREEMENT - DATA FILES AND SOFTWARE
1
+ UNICODE LICENSE V3
2
2
 
3
- See Terms of Use <https://www.unicode.org/copyright.html>
4
- for definitions of Unicode Inc.’s Data Files and Software.
3
+ COPYRIGHT AND PERMISSION NOTICE
5
4
 
6
- NOTICE TO USER: Carefully read the following legal agreement.
7
- BY DOWNLOADING, INSTALLING, COPYING OR OTHERWISE USING UNICODE INC.'S
8
- DATA FILES ("DATA FILES"), AND/OR SOFTWARE ("SOFTWARE"),
9
- YOU UNEQUIVOCALLY ACCEPT, AND AGREE TO BE BOUND BY, ALL OF THE
10
- TERMS AND CONDITIONS OF THIS AGREEMENT.
11
- IF YOU DO NOT AGREE, DO NOT DOWNLOAD, INSTALL, COPY, DISTRIBUTE OR USE
12
- THE DATA FILES OR SOFTWARE.
5
+ Copyright © 2004-2023 Unicode, Inc.
13
6
 
14
- COPYRIGHT AND PERMISSION NOTICE
7
+ NOTICE TO USER: Carefully read the following legal agreement. BY
8
+ DOWNLOADING, INSTALLING, COPYING OR OTHERWISE USING DATA FILES, AND/OR
9
+ SOFTWARE, YOU UNEQUIVOCALLY ACCEPT, AND AGREE TO BE BOUND BY, ALL OF THE
10
+ TERMS AND CONDITIONS OF THIS AGREEMENT. IF YOU DO NOT AGREE, DO NOT
11
+ DOWNLOAD, INSTALL, COPY, DISTRIBUTE OR USE THE DATA FILES OR SOFTWARE.
15
12
 
16
- Copyright © 1991-2023 Unicode, Inc. All rights reserved.
17
- Distributed under the Terms of Use in https://www.unicode.org/copyright.html.
13
+ Permission is hereby granted, free of charge, to any person obtaining a
14
+ copy of data files and any associated documentation (the "Data Files") or
15
+ software and any associated documentation (the "Software") to deal in the
16
+ Data Files or Software without restriction, including without limitation
17
+ the rights to use, copy, modify, merge, publish, distribute, and/or sell
18
+ copies of the Data Files or Software, and to permit persons to whom the
19
+ Data Files or Software are furnished to do so, provided that either (a)
20
+ this copyright and permission notice appear with all copies of the Data
21
+ Files or Software, or (b) this copyright and permission notice appear in
22
+ associated Documentation.
18
23
 
19
- Permission is hereby granted, free of charge, to any person obtaining
20
- a copy of the Unicode data files and any associated documentation
21
- (the "Data Files") or Unicode software and any associated documentation
22
- (the "Software") to deal in the Data Files or Software
23
- without restriction, including without limitation the rights to use,
24
- copy, modify, merge, publish, distribute, and/or sell copies of
25
- the Data Files or Software, and to permit persons to whom the Data Files
26
- or Software are furnished to do so, provided that either
27
- (a) this copyright and permission notice appear with all copies
28
- of the Data Files or Software, or
29
- (b) this copyright and permission notice appear in associated
30
- Documentation.
24
+ THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY
25
+ KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
26
+ MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF
27
+ THIRD PARTY RIGHTS.
31
28
 
32
- THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF
33
- ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
34
- WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
35
- NONINFRINGEMENT OF THIRD PARTY RIGHTS.
36
- IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN THIS
37
- NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL
38
- DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
39
- DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
40
- TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
41
- PERFORMANCE OF THE DATA FILES OR SOFTWARE.
29
+ IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN THIS NOTICE
30
+ BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL DAMAGES,
31
+ OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
32
+ WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
33
+ ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THE DATA
34
+ FILES OR SOFTWARE.
42
35
 
43
- Except as contained in this notice, the name of a copyright holder
44
- shall not be used in advertising or otherwise to promote the sale,
45
- use or other dealings in these Data Files or Software without prior
46
- written authorization of the copyright holder.
36
+ Except as contained in this notice, the name of a copyright holder shall
37
+ not be used in advertising or otherwise to promote the sale, use or other
38
+ dealings in these Data Files or Software without prior written
39
+ authorization of the copyright holder.
package/bower.json CHANGED
@@ -1,8 +1,8 @@
1
1
  {
2
2
  "name": "cldr-segments-full",
3
- "version": "43.0.0",
3
+ "version": "44.0.0",
4
4
  "dependencies": {
5
- "cldr-core": "43.0.0"
5
+ "cldr-core": "44.0.0"
6
6
  },
7
7
  "main": "segments/**/*.json",
8
8
  "ignore": [
package/package.json CHANGED
@@ -1,8 +1,8 @@
1
1
  {
2
2
  "name": "cldr-segments-full",
3
- "version": "43.0.0",
3
+ "version": "44.0.0",
4
4
  "peerDependencies": {
5
- "cldr-core": "43.0.0"
5
+ "cldr-core": "44.0.0"
6
6
  },
7
7
  "description": "Text Segmentation data (all locales)",
8
8
  "homepage": "https://cldr.unicode.org",
package/segments/de/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "de"
8
8
  },
package/segments/el/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "el"
8
8
  },
package/segments/en/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "en"
8
8
  },
package/segments/es/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "es"
8
8
  },
package/segments/fi/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "fi"
8
8
  },
package/segments/fr/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "fr"
8
8
  },
package/segments/it/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "it"
8
8
  },
package/segments/ja/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "ja"
8
8
  },
package/segments/pt/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "pt"
8
8
  },
package/segments/ru/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "ru"
8
8
  },
package/segments/sv/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "sv"
8
8
  },
package/segments/und/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "root"
8
8
  },
@@ -49,16 +49,19 @@
49
49
  "$LVT": "\\\\p{Grapheme_Cluster_Break=LVT}"
50
50
  },
51
51
  {
52
- "$Virama": "[\\\\p{Gujr}\\\\p{sc=Telu}\\\\p{sc=Mlym}\\\\p{sc=Orya}\\\\p{sc=Beng}\\\\p{sc=Deva}&\\\\p{Indic_Syllabic_Category=Virama}]"
52
+ "$ConjunctLinkingScripts": "[\\\\p{Gujr}\\\\p{sc=Telu}\\\\p{sc=Mlym}\\\\p{sc=Orya}\\\\p{sc=Beng}\\\\p{sc=Deva}]"
53
53
  },
54
54
  {
55
- "$LinkingConsonant": "[\\\\p{Gujr}\\\\p{sc=Telu}\\\\p{sc=Mlym}\\\\p{sc=Orya}\\\\p{sc=Beng}\\\\p{sc=Deva}&\\\\p{Indic_Syllabic_Category=Consonant}]"
55
+ "$ConjunctLinker": "\\\\p{Indic_Conjunct_Break=Linker}"
56
+ },
57
+ {
58
+ "$LinkingConsonant": "\\\\p{Indic_Conjunct_Break=Consonant}"
56
59
  },
57
60
  {
58
61
  "$ExtPict": "\\\\p{Extended_Pictographic}"
59
62
  },
60
63
  {
61
- "$ExtCccZwj": "[[$Extend-\\\\p{ccc=0}] $ZWJ]"
64
+ "$ExtCccZwj": "[\\\\p{Indic_Conjunct_Break=Linker}\\\\p{Indic_Conjunct_Break=Extend}]"
62
65
  }
63
66
  ],
64
67
  "segmentRules": {
@@ -71,7 +74,7 @@
71
74
  "9": " × ($Extend | $ZWJ) ",
72
75
  "9.1": " × $SpacingMark ",
73
76
  "9.2": " $Prepend × ",
74
- "9.3": " $LinkingConsonant $ExtCccZwj* $Virama $ExtCccZwj* × $LinkingConsonant ",
77
+ "9.3": " $LinkingConsonant $ExtCccZwj* $ConjunctLinker $ExtCccZwj* × $LinkingConsonant ",
75
78
  "11": " $ExtPict $Extend* $ZWJ × $ExtPict ",
76
79
  "12": " ^ ($RI $RI)* $RI × $RI ",
77
80
  "13": " [^$RI] ($RI $RI)* $RI × $RI "
@@ -82,9 +85,18 @@
82
85
  {
83
86
  "$AI": "\\\\p{Line_Break=Ambiguous}"
84
87
  },
88
+ {
89
+ "$AK": "\\\\p{Line_Break=Aksara}"
90
+ },
85
91
  {
86
92
  "$AL": "\\\\p{Line_Break=Alphabetic}"
87
93
  },
94
+ {
95
+ "$AP": "\\\\p{Line_Break=Aksara_Prebase}"
96
+ },
97
+ {
98
+ "$AS": "\\\\p{Line_Break=Aksara_Start}"
99
+ },
88
100
  {
89
101
  "$B2": "\\\\p{Line_Break=Break_Both}"
90
102
  },
@@ -106,9 +118,6 @@
106
118
  {
107
119
  "$CP": "\\\\p{Line_Break=CP}"
108
120
  },
109
- {
110
- "$CP30": "[$CP - [\\\\p{ea=F}\\\\p{ea=W}\\\\p{ea=H}]]"
111
- },
112
121
  {
113
122
  "$CM1": "\\\\p{Line_Break=Combining_Mark}"
114
123
  },
@@ -166,9 +175,6 @@
166
175
  {
167
176
  "$OP": "\\\\p{Line_Break=Open_Punctuation}"
168
177
  },
169
- {
170
- "$OP30": "[$OP - [\\\\p{ea=F}\\\\p{ea=W}\\\\p{ea=H}]]"
171
- },
172
178
  {
173
179
  "$PO": "\\\\p{Line_Break=Postfix_Numeric}"
174
180
  },
@@ -190,6 +196,12 @@
190
196
  {
191
197
  "$SY": "\\\\p{Line_Break=Break_Symbols}"
192
198
  },
199
+ {
200
+ "$VF": "\\\\p{Line_Break=Virama_Final}"
201
+ },
202
+ {
203
+ "$VI": "\\\\p{Line_Break=Virama}"
204
+ },
193
205
  {
194
206
  "$WJ": "\\\\p{Line_Break=Word_Joiner}"
195
207
  },
@@ -217,9 +229,30 @@
217
229
  {
218
230
  "$ZWJ": "\\\\p{Line_Break=ZWJ}"
219
231
  },
232
+ {
233
+ "$QU_Pi": "[$QU & \\\\p{gc=Pi}]"
234
+ },
235
+ {
236
+ "$QU_Pf": "[$QU & \\\\p{gc=Pf}]"
237
+ },
238
+ {
239
+ "$DottedCircle": "◌"
240
+ },
241
+ {
242
+ "$CP30": "[$CP-[\\\\p{ea=F}\\\\p{ea=W}\\\\p{ea=H}]]"
243
+ },
244
+ {
245
+ "$OP30": "[$OP-[\\\\p{ea=F}\\\\p{ea=W}\\\\p{ea=H}]]"
246
+ },
220
247
  {
221
248
  "$ExtPictUnassigned": "[\\\\p{Extended_Pictographic}&\\\\p{gc=Cn}]"
222
249
  },
250
+ {
251
+ "$sot": "^"
252
+ },
253
+ {
254
+ "$eot": "(?!.)"
255
+ },
223
256
  {
224
257
  "$CM": "[$CM1 $ZWJ]"
225
258
  },
@@ -248,14 +281,20 @@
248
281
  "$Spec4_": "[^ $NU $CM]"
249
282
  },
250
283
  {
251
- "$Spec5_": "[$BK $CB $CR $LF $NL $SP $ZW]"
284
+ "$AI": "($AI $X)"
252
285
  },
253
286
  {
254
- "$AI": "($AI $X)"
287
+ "$AK": "($AK $X)"
255
288
  },
256
289
  {
257
290
  "$AL": "($AL $X)"
258
291
  },
292
+ {
293
+ "$AP": "($AP $X)"
294
+ },
295
+ {
296
+ "$AS": "($AS $X)"
297
+ },
259
298
  {
260
299
  "$B2": "($B2 $X)"
261
300
  },
@@ -274,9 +313,6 @@
274
313
  {
275
314
  "$CP": "($CP $X)"
276
315
  },
277
- {
278
- "$CP30": "($CP30 $X)"
279
- },
280
316
  {
281
317
  "$CM": "($CM $X)"
282
318
  },
@@ -325,9 +361,6 @@
325
361
  {
326
362
  "$OP": "($OP $X)"
327
363
  },
328
- {
329
- "$OP30": "($OP30 $X)"
330
- },
331
364
  {
332
365
  "$PO": "($PO $X)"
333
366
  },
@@ -346,6 +379,12 @@
346
379
  {
347
380
  "$SY": "($SY $X)"
348
381
  },
382
+ {
383
+ "$VF": "($VF $X)"
384
+ },
385
+ {
386
+ "$VI": "($VI $X)"
387
+ },
349
388
  {
350
389
  "$WJ": "($WJ $X)"
351
390
  },
@@ -364,6 +403,21 @@
364
403
  {
365
404
  "$ZWJ": "($ZWJ $X)"
366
405
  },
406
+ {
407
+ "$QU_Pi": "($QU_Pi $X)"
408
+ },
409
+ {
410
+ "$QU_Pf": "($QU_Pf $X)"
411
+ },
412
+ {
413
+ "$DottedCircle": "($DottedCircle $X)"
414
+ },
415
+ {
416
+ "$CP30": "($CP30 $X)"
417
+ },
418
+ {
419
+ "$OP30": "($OP30 $X)"
420
+ },
367
421
  {
368
422
  "$AL": "($AL | ^ $CM | (?<=$Spec1_) $CM)"
369
423
  }
@@ -391,7 +445,8 @@
391
445
  "13.03": " $Spec4_ $CM+ × ($CL | $CP | $IS | $SY) ",
392
446
  "13.04": " ^ $CM+ × ($CL | $CP | $IS | $SY) ",
393
447
  "14": " $OP $SP* × ",
394
- "15": " $QU $SP* × $OP ",
448
+ "15.11": " ( $sot | $BK | $CR | $LF | $NL | $OP | $QU | $GL | $SP | $ZW ) $QU_Pi $SP* × ",
449
+ "15.21": " × $QU_Pf ( $SP | $GL | $WJ | $CL | $QU | $CP | $EX | $IS | $SY | $BK | $CR | $LF | $NL | $ZW | $eot ) ",
395
450
  "16": " ($CL | $CP) $SP* × $NS ",
396
451
  "17": " $B2 $SP* × $B2 ",
397
452
  "18": " $SP ÷ ",
@@ -399,14 +454,13 @@
399
454
  "19.02": " $QU × ",
400
455
  "20.01": " ÷ $CB ",
401
456
  "20.02": " $CB ÷ ",
402
- "20.09": " $Spec5_ $HY × $AL ",
403
457
  "21.01": " × $BA ",
404
458
  "21.02": " × $HY ",
405
459
  "21.03": " × $NS ",
406
460
  "21.04": " $BB × ",
407
461
  "21.1": " $HL ($HY | $BA) × ",
408
462
  "21.2": " $SY × $HL ",
409
- "22.01": " × $IN ",
463
+ "22": " × $IN ",
410
464
  "23.02": " ($AL | $HL) × $NU ",
411
465
  "23.03": " $NU × ($AL | $HL) ",
412
466
  "23.12": " $PR × ($ID | $EB | $EM) ",
@@ -421,14 +475,17 @@
421
475
  "26.01": " $JL × $JL | $JV | $H2 | $H3 ",
422
476
  "26.02": " $JV | $H2 × $JV | $JT ",
423
477
  "26.03": " $JT | $H3 × $JT ",
424
- "27.01": " $JL | $JV | $JT | $H2 | $H3 × $IN ",
425
- "27.02": " $JL | $JV | $JT | $H2 | $H3 × $PO ",
426
- "27.03": " $PR × $JL | $JV | $JT | $H2 | $H3 ",
478
+ "27.01": " $JL | $JV | $JT | $H2 | $H3 × $PO ",
479
+ "27.02": " $PR × $JL | $JV | $JT | $H2 | $H3 ",
427
480
  "28": " ($AL | $HL) × ($AL | $HL) ",
481
+ "28.11": " $AP × ($AK | $DottedCircle | $AS) ",
482
+ "28.12": " ($AK | $DottedCircle | $AS) × ($VF | $VI) ",
483
+ "28.13": " ($AK | $DottedCircle | $AS) $VI × ($AK | $DottedCircle) ",
484
+ "28.14": " ($AK | $DottedCircle | $AS) × ($AK | $DottedCircle | $AS) $VF ",
428
485
  "29": " $IS × ($AL | $HL) ",
429
486
  "30.01": " ($AL | $HL | $NU) × $OP30 ",
430
487
  "30.02": " $CP30 × ($AL | $HL | $NU) ",
431
- "30.11": " ^ ($RI $RI)* $RI × $RI ",
488
+ "30.11": " $sot ($RI $RI)* $RI × $RI ",
432
489
  "30.12": " [^$RI] ($RI $RI)* $RI × $RI ",
433
490
  "30.13": " $RI ÷ $RI ",
434
491
  "30.21": " $EB × $EM ",
@@ -551,16 +608,16 @@
551
608
  "$Extend": "\\\\p{Word_Break=Extend}"
552
609
  },
553
610
  {
554
- "$Format": "\\\\p{Word_Break=Format}"
611
+ "$Format": "[\\\\p{Word_Break=Format}]"
555
612
  },
556
613
  {
557
614
  "$Katakana": "\\\\p{Word_Break=Katakana}"
558
615
  },
559
616
  {
560
- "$ALetter": "[\\\\p{Word_Break=ALetter}@]"
617
+ "$ALetter": "\\\\p{Word_Break=ALetter}"
561
618
  },
562
619
  {
563
- "$MidLetter": "[[\\\\p{Word_Break=MidLetter}] - [\\u003A \\uFE55 \\uFF1A]]"
620
+ "$MidLetter": "\\\\p{Word_Break=MidLetter}"
564
621
  },
565
622
  {
566
623
  "$MidNum": "\\\\p{Word_Break=MidNum}"
package/segments/zh/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "zh"
8
8
  },
package/segments/zh-Hant/suppressions.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "segments": {
3
3
  "identity": {
4
4
  "version": {
5
- "_cldrVersion": "43"
5
+ "_cldrVersion": "44"
6
6
  },
7
7
  "language": "zh",
8
8
  "script": "Hant"