• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# Copyright (C) 2016 and later: Unicode, Inc. and others.
2# License & terms of use: http://www.unicode.org/copyright.html
3# Copyright (c) 2002-2016  International Business Machines Corporation and
4# others. All Rights Reserved.
5#
6#  file:  line_cj.txt
7#
8#         Line Breaking Rules
9#         Implement default line breaking as defined by
10#         Unicode Standard Annex #14 (https://www.unicode.org/reports/tr14/)
11#         for Unicode 14.0, with the following modification:
12#
13#         Boundaries between hyphens and following letters are suppressed when
14#         there is a boundary preceding the hyphen. See rule 20.9
15#
16#         This corresponds to CSS line-break=strict (BCP47 -u-lb-strict).
17#         It sets characters of class CJ to behave like NS.
18#         It allows breaking before 201C and after 201D, for zh_Hans, zh_Hant, and ja.
19
20#
21#  Character Classes defined by TR 14.
22#
23
24!!chain;
25!!quoted_literals_only;
26
27$AI = [:LineBreak =  Ambiguous:];
28$AL = [:LineBreak =  Alphabetic:];
29$BA = [:LineBreak =  Break_After:];
30$HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
31$BB = [:LineBreak =  Break_Before:];
32$BK = [:LineBreak =  Mandatory_Break:];
33$B2 = [:LineBreak =  Break_Both:];
34$CB = [:LineBreak =  Contingent_Break:];
35$CJ = [:LineBreak =  Conditional_Japanese_Starter:];
36$CL = [[:LineBreak =  Close_Punctuation:] \u201d];
37# $CM = [:LineBreak =  Combining_Mark:];
38$CP = [:LineBreak =  Close_Parenthesis:];
39$CR = [:LineBreak =  Carriage_Return:];
40$EB = [:LineBreak =  EB:];
41$EM = [:LineBreak =  EM:];
42$EX = [:LineBreak =  Exclamation:];
43$GL = [:LineBreak =  Glue:];
44$HL = [:LineBreak =  Hebrew_Letter:];
45$HY = [:LineBreak =  Hyphen:];
46$H2 = [:LineBreak =  H2:];
47$H3 = [:LineBreak =  H3:];
48$ID = [:LineBreak =  Ideographic:];
49$IN = [:LineBreak =  Inseperable:];
50$IS = [:LineBreak =  Infix_Numeric:];
51$JL = [:LineBreak =  JL:];
52$JV = [:LineBreak =  JV:];
53$JT = [:LineBreak =  JT:];
54$LF = [:LineBreak =  Line_Feed:];
55$NL = [:LineBreak =  Next_Line:];
56# NS includes CJ for CSS strict line breaking.
57$NS = [[:LineBreak =  Nonstarter:] $CJ];
58$NU = [:LineBreak =  Numeric:];
59$OP = [[:LineBreak =  Open_Punctuation:] \u201c];
60$PO = [:LineBreak =  Postfix_Numeric:];
61$PR = [:LineBreak =  Prefix_Numeric:];
62$QU = [[:LineBreak =  Quotation:] - [\u201c\u201d]];
63$RI = [:LineBreak =  Regional_Indicator:];
64$SA = [:LineBreak =  Complex_Context:];
65$SG = [:LineBreak =  Surrogate:];
66$SP = [:LineBreak =  Space:];
67$SY = [:LineBreak =  Break_Symbols:];
68$WJ = [:LineBreak =  Word_Joiner:];
69$XX = [:LineBreak =  Unknown:];
70$ZW = [:LineBreak =  ZWSpace:];
71$ZWJ = [:LineBreak = ZWJ:];
72
73# OP30 and CP30 are variants of OP and CP that appear in-line in rule LB30 from UAX 14,
74# without a formal name. Because ICU rules require multiple uses of the expressions,
75# give them a single definition with a name
76
77$OP30 = [$OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
78$CP30 = [$CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
79
80$ExtPictUnassigned = [\p{Extended_Pictographic} & \p{Cn}];
81
82# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
83#         list it in the numerous rules that use CM.
84# By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
85
86$CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
87$CMX = [[$CM] - [$ZWJ]];
88
89#   Dictionary character set, for triggering language-based break engines. Currently
90#   limited to LineBreak=Complex_Context (SA).
91
92$dictionary = [$SA];
93
94#
95#  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
96#                               SA  (Dictionary chars, excluding Mn and Mc)
97#                               SG  (Unpaired Surrogates)
98#                               XX  (Unknown, unassigned)
99#                         as $AL  (Alphabetic)
100#
101$ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
102
103
104## -------------------------------------------------
105
106#
107# CAN_CM  is the set of characters that may combine with CM combining chars.
108#         Note that Linebreak UAX 14's concept of a combining char and the rules
109#         for what they can combine with are _very_ different from the rest of Unicode.
110#
111#         Note that $CM itself is left out of this set.  If CM is needed as a base
112#         it must be listed separately in the rule.
113#
114$CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
115$CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
116
117#
118# AL_FOLLOW  set of chars that can unconditionally follow an AL
119#            Needed in rules where stand-alone $CM s are treated as AL.
120#
121$AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP30 $QU $BA $HY $NS $IN $NU $PR $PO $ALPlus];
122
123
124#
125#  Rule LB 4, 5    Mandatory (Hard) breaks.
126#
127$LB4Breaks    = [$BK $CR $LF $NL];
128$LB4NonBreaks = [^$BK $CR $LF $NL $CM];
129$CR $LF {100};
130
131#
132#  LB 6    Do not break before hard line breaks.
133#
134$LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
135$CAN_CM $CM*    $LB4Breaks {100};
136^$CM+           $LB4Breaks {100};
137
138# LB 7         x SP
139#              x ZW
140$LB4NonBreaks [$SP $ZW];
141$CAN_CM $CM*  [$SP $ZW];
142^$CM+         [$SP $ZW];
143
144#
145# LB 8         Break after zero width space
146#              ZW SP* ÷
147#
148$LB8Breaks    = [$LB4Breaks $ZW];
149$LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
150$ZW $SP* / [^$SP $ZW $LB4Breaks];
151
152# LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
153#
154$ZWJ [^$CM];
155
156# LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
157#                                $CM not covered by the above needs to behave like $AL
158#                                See definition of $CAN_CM.
159
160$CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
161^$CM+;
162
163#
164# LB 11  Do not break before or after WORD JOINER & related characters.
165#
166$CAN_CM $CM*  $WJ;
167$LB8NonBreaks $WJ;
168^$CM+         $WJ;
169
170$WJ $CM* .;
171
172#
173# LB 12  Do not break after NBSP and related characters.
174#         GL  x
175#
176$GL $CM* .;
177
178#
179# LB 12a  Do not break before NBSP and related characters ...
180#            [^SP BA HY] x GL
181#
182[[$LB8NonBreaks] - [$SP $BA $HY]] $CM* $GL;
183^$CM+ $GL;
184
185
186
187
188# LB 13   Don't break before ']' or '!' or '/', even after spaces.
189#
190$LB8NonBreaks $CL;
191$CAN_CM $CM*  $CL;
192^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
193
194$LB8NonBreaks $CP;
195$CAN_CM $CM*  $CP;
196^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
197
198$LB8NonBreaks $EX;
199$CAN_CM $CM*  $EX;
200^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
201
202$LB8NonBreaks $SY;
203$CAN_CM $CM*  $SY;
204^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
205
206
207#
208# LB 14  Do not break after OP, even after spaces
209#        Note subtle interaction with "SP IS /" rules in LB14a.
210#        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
211#        which is the desired behavior.
212#
213$OP $CM* $SP* .;
214
215$OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
216                                   # by rule 8, CM following a SP is stand-alone.
217
218
219# LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
220#        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
221#        See issue ICU-20303
222
223
224$CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL $IN];
225$SP $IS           / [^ $CanFollowIS $NU $CM];
226$SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
227
228#
229# LB 14b Do not break before numeric separators (IS), even after spaces.
230
231[$LB8NonBreaks - $SP] $IS;
232$SP $IS $CM* [$CanFollowIS {eof}];
233$SP $IS $CM* $ZWJ [^$CM $NU];
234
235$CAN_CM $CM*  $IS;
236^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
237
238
239# LB 15
240$QU $CM* $SP* $OP;
241
242# LB 16
243($CL | $CP) $CM* $SP* $NS;
244
245# LB 17
246$B2 $CM* $SP* $B2;
247
248#
249# LB 18  Break after spaces.
250#
251$LB18NonBreaks = [$LB8NonBreaks - [$SP]];
252$LB18Breaks    = [$LB8Breaks $SP];
253
254
255# LB 19
256#         x QU
257$LB18NonBreaks $CM* $QU;
258^$CM+               $QU;
259
260#         QU  x
261$QU $CM* .;
262
263# LB 20
264#        <break>  $CB
265#        $CB   <break>
266#
267$LB20NonBreaks = [$LB18NonBreaks - $CB];
268
269# LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
270#             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
271#             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
272#
273^($HY | $HH) $CM* $ALPlus;
274
275# LB 21        x   (BA | HY | NS)
276#           BB x
277#
278$LB20NonBreaks $CM* ($BA | $HY | $NS);
279
280
281^$CM+ ($BA | $HY | $NS);
282
283$BB $CM* [^$CB];                                  #  $BB  x
284$BB $CM* $LB20NonBreaks;
285
286# LB 21a Don't break after Hebrew + Hyphen
287#   HL (HY | BA) x
288#
289$HL $CM* ($HY | $BA) $CM* [^$CB]?;
290
291# LB 21b (forward) Don't break between SY and HL
292# (break between HL and SY already disallowed by LB 13 above)
293$SY $CM* $HL;
294
295# LB 22  Do not break before ellipses
296#
297$LB20NonBreaks $CM*    $IN;
298^$CM+ $IN;
299
300
301# LB 23
302#
303($ALPlus | $HL) $CM* $NU;
304^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
305$NU $CM* ($ALPlus | $HL);
306
307# LB 23a
308#
309$PR $CM* ($ID | $EB | $EM);
310($ID | $EB | $EM) $CM*  $PO;
311
312
313#
314# LB 24
315#
316($PR | $PO) $CM* ($ALPlus | $HL);
317($ALPlus | $HL) $CM* ($PR | $PO);
318^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
319
320#
321# LB 25   Numbers.
322#
323(($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
324    ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
325
326# LB 26  Do not break a Korean syllable
327#
328$JL $CM* ($JL | $JV | $H2 | $H3);
329($JV | $H2) $CM* ($JV | $JT);
330($JT | $H3) $CM* $JT;
331
332# LB 27  Treat korean Syllable Block the same as ID  (don't break it)
333($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
334$PR $CM* ($JL | $JV | $JT | $H2 | $H3);
335
336
337# LB 28   Do not break between alphabetics
338#
339($ALPlus | $HL) $CM* ($ALPlus | $HL);
340^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
341
342# LB 29
343$IS $CM* ($ALPlus | $HL);
344
345# LB 30
346($ALPlus | $HL | $NU) $CM* $OP30;
347^$CM+ $OP30;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
348$CP30 $CM* ($ALPlus | $HL | $NU);
349
350# LB 30a  Do not break between regional indicators. Break after pairs of them.
351#         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
352$RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
353$RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
354$RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $ZWJ {eof}];
355# note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
356#       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
357#       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
358
359# LB30b Do not break between an emoji base (or potential emoji) and an emoji modifier.
360$EB $CM* $EM;
361$ExtPictUnassigned $CM* $EM;
362
363# LB 31 Break everywhere else.
364#       Match a single code point if no other rule applies.
365.;
366