• 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_loose_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 tailors the line break behavior to correspond to CSS
17#         line-break=loose (BCP47 -u-lb-loose) as defined for Chinese & Japanese.
18#         It sets characters of class CJ to behave like ID.
19#         In addition, it allows breaks:
20#         * between ID and hyphens 2010 & 2013 (both BA)
21#         * before 301C, 30A0 (both NS)
22#         * before iteration marks 3005, 303B, 309D, 309E, 30FD, 30FE (all NS)
23#         * between characters of LineBreak class IN such as 2026
24#         * before some centered punct 203C, 2047, 2048, 2049, 30FB, FF1A, FF1B,
25#           FF65 (all NS) and FF01, FF1F (both EX).
26#         * before suffix characters with LineBreak class PO and EastAsianWidth A,F,W;
27#           this includes: 00B0 2030 2032 2033 2035 2103 2109 FE6A FF05 FFE0
28#         * after prefix characters with LineBreak class PR and EastAsianWidth A,F,W;
29#           this includes: 00A4 00B1 20AC 2116 FE69 FF04 FFE1 FFE5 FFE6
30#         It allows breaking before 201C and after 201D, for zh_Hans, zh_Hant, and ja.
31
32
33#
34#  Character Classes defined by TR 14.
35#
36
37!!chain;
38!!quoted_literals_only;
39
40$AI = [:LineBreak =  Ambiguous:];
41$AL = [:LineBreak =  Alphabetic:];
42$BAX = [\u2010 \u2013];
43$BA = [[:LineBreak =  Break_After:] - $BAX];
44$HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
45$BB = [:LineBreak =  Break_Before:];
46$BK = [:LineBreak =  Mandatory_Break:];
47$B2 = [:LineBreak =  Break_Both:];
48$CB = [:LineBreak =  Contingent_Break:];
49$CJ = [:LineBreak =  Conditional_Japanese_Starter:];
50$CL = [[:LineBreak =  Close_Punctuation:] \u201d];
51# $CM = [:LineBreak =  Combining_Mark:];
52$CP = [:LineBreak =  Close_Parenthesis:];
53$CR = [:LineBreak =  Carriage_Return:];
54$EB = [:LineBreak =  EB:];
55$EM = [:LineBreak =  EM:];
56$EXX = [\uFF01 \uFF1F];
57$EX = [[:LineBreak =  Exclamation:] - $EXX];
58$GL = [:LineBreak =  Glue:];
59$HL = [:LineBreak =  Hebrew_Letter:];
60$HY = [:LineBreak =  Hyphen:];
61$H2 = [:LineBreak =  H2:];
62$H3 = [:LineBreak =  H3:];
63# CSS Loose tailoring: CJ resolves to ID
64$ID = [[:LineBreak =  Ideographic:] $CJ];
65$IN = [:LineBreak =  Inseperable:];
66$IS = [:LineBreak =  Infix_Numeric:];
67$JL = [:LineBreak =  JL:];
68$JV = [:LineBreak =  JV:];
69$JT = [:LineBreak =  JT:];
70$LF = [:LineBreak =  Line_Feed:];
71$NL = [:LineBreak =  Next_Line:];
72$NSX = [\u301C \u30A0 \u3005 \u303B \u309D \u309E \u30FD \u30FE \u203C \u2047 \u2048 \u2049 \u30FB \uFF1A \uFF1B \uFF65];
73$NS = [[:LineBreak =  Nonstarter:] - $NSX];
74$NU = [:LineBreak =  Numeric:];
75$OP = [[:LineBreak =  Open_Punctuation:] \u201c];
76$POX = [\u00B0 \u2030 \u2032 \u2033 \u2035 \u2103 \u2109 \uFE6A \uFF05 \uFFE0];
77$PO = [[:LineBreak =  Postfix_Numeric:] - $POX];
78$PRX = [\u00A4 \u00B1 \u20AC \u2116 \uFE69 \uFF04 \uFFE1 \uFFE5 \uFFE6];
79$PR = [[:LineBreak =  Prefix_Numeric:] - $PRX];
80$QU = [[:LineBreak =  Quotation:] - [\u201c\u201d]];
81$RI = [:LineBreak =  Regional_Indicator:];
82$SA = [:LineBreak =  Complex_Context:];
83$SG = [:LineBreak =  Surrogate:];
84$SP = [:LineBreak =  Space:];
85$SY = [:LineBreak =  Break_Symbols:];
86$WJ = [:LineBreak =  Word_Joiner:];
87$XX = [:LineBreak =  Unknown:];
88$ZW = [:LineBreak =  ZWSpace:];
89$ZWJ = [:LineBreak = ZWJ:];
90
91# OP30 and CP30 are variants of OP and CP that appear in-line in rule LB30 from UAX 14,
92# without a formal name. Because ICU rules require multiple uses of the expressions,
93# give them a single definition with a name
94
95$OP30 = [$OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
96$CP30 = [$CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
97
98$ExtPictUnassigned = [\p{Extended_Pictographic} & \p{Cn}];
99
100# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
101#         list it in the numerous rules that use CM.
102# By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
103
104$CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
105$CMX = [[$CM] - [$ZWJ]];
106
107#   Dictionary character set, for triggering language-based break engines. Currently
108#   limited to LineBreak=Complex_Context (SA).
109
110$dictionary = [$SA];
111
112#
113#  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
114#                               SA  (Dictionary chars, excluding Mn and Mc)
115#                               SG  (Unpaired Surrogates)
116#                               XX  (Unknown, unassigned)
117#                         as $AL  (Alphabetic)
118#
119$ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
120
121
122## -------------------------------------------------
123
124#
125# CAN_CM  is the set of characters that may combine with CM combining chars.
126#         Note that Linebreak UAX 14's concept of a combining char and the rules
127#         for what they can combine with are _very_ different from the rest of Unicode.
128#
129#         Note that $CM itself is left out of this set.  If CM is needed as a base
130#         it must be listed separately in the rule.
131#
132$CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
133$CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
134
135#
136# AL_FOLLOW  set of chars that can unconditionally follow an AL
137#            Needed in rules where stand-alone $CM s are treated as AL.
138#
139$AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP30 $QU $BA $BAX $HY $NS $IN $NU $PR $PO $POX $ALPlus];
140
141
142#
143#  Rule LB 4, 5    Mandatory (Hard) breaks.
144#
145$LB4Breaks    = [$BK $CR $LF $NL];
146$LB4NonBreaks = [^$BK $CR $LF $NL $CM];
147$CR $LF {100};
148
149#
150#  LB 6    Do not break before hard line breaks.
151#
152$LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
153$CAN_CM $CM*    $LB4Breaks {100};
154^$CM+           $LB4Breaks {100};
155
156# LB 7         x SP
157#              x ZW
158$LB4NonBreaks [$SP $ZW];
159$CAN_CM $CM*  [$SP $ZW];
160^$CM+         [$SP $ZW];
161
162#
163# LB 8         Break after zero width space
164#              ZW SP* ÷
165#
166$LB8Breaks    = [$LB4Breaks $ZW];
167$LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
168$ZW $SP* / [^$SP $ZW $LB4Breaks];
169
170# LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
171#
172$ZWJ [^$CM];
173
174# LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
175#                                $CM not covered by the above needs to behave like $AL
176#                                See definition of $CAN_CM.
177
178$CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
179^$CM+;
180
181#
182# LB 11  Do not break before or after WORD JOINER & related characters.
183#
184$CAN_CM $CM*  $WJ;
185$LB8NonBreaks $WJ;
186^$CM+         $WJ;
187
188$WJ $CM* .;
189
190#
191# LB 12  Do not break after NBSP and related characters.
192#         GL  x
193#
194$GL $CM* .;
195
196#
197# LB 12a  Do not break before NBSP and related characters ...
198#            [^SP BA HY] x GL
199#
200[[$LB8NonBreaks] - [$SP $BA $BAX $HY]] $CM* $GL;
201^$CM+ $GL;
202
203
204
205# LB 13   Don't break before ']' or '!' or '/', even after spaces.
206#
207# Do not include $EXX here
208$LB8NonBreaks $CL;
209$CAN_CM $CM*  $CL;
210^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
211
212$LB8NonBreaks $CP;
213$CAN_CM $CM*  $CP;
214^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
215
216$LB8NonBreaks $EX;
217$CAN_CM $CM*  $EX;
218^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
219
220$LB8NonBreaks $SY;
221$CAN_CM $CM*  $SY;
222^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
223
224
225#
226# LB 14  Do not break after OP, even after spaces
227#        Note subtle interaction with "SP IS /" rules in LB14a.
228#        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
229#        which is the desired behavior.
230#
231$OP $CM* $SP* .;
232
233$OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
234                                   # by rule 8, CM following a SP is stand-alone.
235
236
237# LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
238#        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
239#        See issue ICU-20303
240
241
242$CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $BAX $HY $NS $ALPlus $HL $IN];
243$SP $IS           / [^ $CanFollowIS $NU $CM];
244$SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
245
246#
247# LB 14b Do not break before numeric separators (IS), even after spaces.
248
249[$LB8NonBreaks - $SP] $IS;
250$SP $IS $CM* [$CanFollowIS {eof}];
251$SP $IS $CM* $ZWJ [^$CM $NU];
252
253$CAN_CM $CM*  $IS;
254^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
255
256
257# LB 15
258$QU $CM* $SP* $OP;
259
260# LB 16
261# Do not break between closing punctuation and $NS, even with intervening spaces
262# But DO allow a break between closing punctuation and $NSX, don't include it here
263($CL | $CP) $CM* $SP* $NS;
264
265# LB 17
266$B2 $CM* $SP* $B2;
267
268#
269# LB 18  Break after spaces.
270#
271$LB18NonBreaks = [$LB8NonBreaks - [$SP]];
272$LB18Breaks    = [$LB8Breaks $SP];
273
274
275# LB 19
276#         x QU
277$LB18NonBreaks $CM* $QU;
278^$CM+               $QU;
279
280#         QU  x
281$QU $CM* .;
282
283# LB 20
284#        <break>  $CB
285#        $CB   <break>
286#
287$LB20NonBreaks = [$LB18NonBreaks - $CB];
288
289# LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
290#             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
291#             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
292#
293^($HY | $HH) $CM* $ALPlus;
294
295# LB 21        x   (BA | HY | NS)
296#           BB x
297#
298# DO allow breaks here before $NSX, so don't include it.
299# And DO allow breaks between ID and $BAX, so split out the handling of ID and do not include $BAX for them.
300[$LB20NonBreaks - $ID] $CM* ($BA | $BAX | $HY | $NS);
301$ID $CM* ($BA | $HY | $NS);
302
303
304^$CM+ ($BA | $BAX | $HY | $NS);
305
306$BB $CM* [^$CB];                                  #  $BB  x
307$BB $CM* $LB20NonBreaks;
308
309# LB 21a Don't break after Hebrew + Hyphen
310#   HL (HY | BA) x
311#
312$HL $CM* ($HY | $BA | $BAX) $CM* [^$CB]?;
313
314# LB 21b (forward) Don't break between SY and HL
315# (break between HL and SY already disallowed by LB 13 above)
316$SY $CM* $HL;
317
318
319# LB 22  Do not break before ellipses
320#
321[$LB20NonBreaks - $IN] $CM*    $IN;    # line_loose tailoring
322^$CM+ $IN;
323
324
325# LB 23
326#
327($ALPlus | $HL) $CM* $NU;
328^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
329$NU $CM* ($ALPlus | $HL);
330
331# LB 23a
332# Do not include $POX here
333#
334$PR $CM* ($ID | $EB | $EM);
335($ID | $EB | $EM) $CM*  $PO;
336
337
338#
339# LB 24
340#
341# Do not include $PRX here
342($PR | $PO | $POX) $CM* ($ALPlus | $HL);
343($ALPlus | $HL) $CM* ($PR | $PO | $POX);     # TODO: should this be ($PR | $PRX | $PO)
344^$CM+ ($PR | $PO | $POX);       # Rule 10, any otherwise unattached CM behaves as AL
345
346#
347# LB 25   Numbers.
348#
349# Here do not include $PRX at the beginning or $POX at the end
350(($PR | $PO | $POX) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
351    ($CM* ($CL | $CP))? ($CM* ($PR | $PRX | $PO))?;
352
353# LB 26  Do not break a Korean syllable
354#
355$JL $CM* ($JL | $JV | $H2 | $H3);
356($JV | $H2) $CM* ($JV | $JT);
357($JT | $H3) $CM* $JT;
358
359# LB 27  Treat korean Syllable Block the same as ID  (don't break it)
360# Do not include $POX or $PRX here
361($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
362$PR $CM* ($JL | $JV | $JT | $H2 | $H3);
363
364
365# LB 28   Do not break between alphabetics
366#
367($ALPlus | $HL) $CM* ($ALPlus | $HL);
368^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
369
370# LB 29
371$IS $CM* ($ALPlus | $HL);
372
373# LB 30
374($ALPlus | $HL | $NU) $CM* $OP30;
375^$CM+ $OP30;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
376$CP30 $CM* ($ALPlus | $HL | $NU);
377
378# LB 30a  Do not break between regional indicators. Break after pairs of them.
379#         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
380$RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $BAX $HY $NS $IN $CM]];
381$RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $BAX $HY $NS $IN $CM]];
382$RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $BAX $HY $NS $IN $ZWJ {eof}];
383# note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
384#       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
385#       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
386
387# LB30b Do not break between an emoji base (or potential emoji) and an emoji modifier.
388$EB $CM* $EM;
389$ExtPictUnassigned $CM* $EM;
390
391# LB 31 Break everywhere else.
392#       Match a single code point if no other rule applies.
393.;
394