• 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_normal_cj.txt
7#
8#         Line Breaking Rules
9#         Implement default line breaking as defined by
10#         Unicode Standard Annex #14 Revision 44 for Unicode 13.0
11#         http://www.unicode.org/reports/tr14/, 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=normal (BCP47 -u-lb-normal) as defined for Chinese & Japanese.
18#         It sets characters of class CJ to behave like ID.
19#         In addition, it allows breaks:
20#         * before hyphens 2010 & 2013 (both BA) and 301C, 30A0 (both NS)
21#         It allows breaking before 201C and after 201D, for zh_Hans, zh_Hant, and ja.
22
23#
24#  Character Classes defined by TR 14.
25#
26
27!!chain;
28!!quoted_literals_only;
29
30$AI = [:LineBreak =  Ambiguous:];
31$AL = [:LineBreak =  Alphabetic:];
32$BAX = [\u2010 \u2013];
33$BA = [[:LineBreak =  Break_After:] - $BAX];
34$HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
35$BB = [:LineBreak =  Break_Before:];
36$BK = [:LineBreak =  Mandatory_Break:];
37$B2 = [:LineBreak =  Break_Both:];
38$CB = [:LineBreak =  Contingent_Break:];
39$CJ = [:LineBreak =  Conditional_Japanese_Starter:];
40$CL = [[:LineBreak =  Close_Punctuation:] \u201d];
41# $CM = [:LineBreak =  Combining_Mark:];
42$CP = [:LineBreak =  Close_Parenthesis:];
43$CR = [:LineBreak =  Carriage_Return:];
44$EB = [:LineBreak =  EB:];
45$EM = [:LineBreak =  EM:];
46$EX = [:LineBreak =  Exclamation:];
47$GL = [:LineBreak =  Glue:];
48$HL = [:LineBreak =  Hebrew_Letter:];
49$HY = [:LineBreak =  Hyphen:];
50$H2 = [:LineBreak =  H2:];
51$H3 = [:LineBreak =  H3:];
52# CSS Normal tailoring: CJ resolves to ID
53$ID = [[:LineBreak =  Ideographic:] $CJ];
54$IN = [:LineBreak =  Inseperable:];
55$IS = [:LineBreak =  Infix_Numeric:];
56$JL = [:LineBreak =  JL:];
57$JV = [:LineBreak =  JV:];
58$JT = [:LineBreak =  JT:];
59$LF = [:LineBreak =  Line_Feed:];
60$NL = [:LineBreak =  Next_Line:];
61$NSX = [\u301C \u30A0];
62$NS = [[:LineBreak =  Nonstarter:] - $NSX];
63$NU = [:LineBreak =  Numeric:];
64$OP = [[:LineBreak =  Open_Punctuation:] \u201c];
65$PO = [:LineBreak =  Postfix_Numeric:];
66$PR = [:LineBreak =  Prefix_Numeric:];
67$QU = [[:LineBreak =  Quotation:] - [\u201c\u201d]];
68$RI = [:LineBreak =  Regional_Indicator:];
69$SA = [:LineBreak =  Complex_Context:];
70$SG = [:LineBreak =  Surrogate:];
71$SP = [:LineBreak =  Space:];
72$SY = [:LineBreak =  Break_Symbols:];
73$WJ = [:LineBreak =  Word_Joiner:];
74$XX = [:LineBreak =  Unknown:];
75$ZW = [:LineBreak =  ZWSpace:];
76$ZWJ = [:LineBreak = ZWJ:];
77
78# OP30 and CP30 are variants of OP and CP that appear in-line in rule LB30 from UAX 14,
79# without a formal name. Because ICU rules require multiple uses of the expressions,
80# give them a single definition with a name
81
82$OP30 = [$OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
83$CP30 = [$CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
84
85# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
86#         list it in the numerous rules that use CM.
87# By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
88
89$CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
90$CMX = [[$CM] - [$ZWJ]];
91
92#   Dictionary character set, for triggering language-based break engines. Currently
93#   limited to LineBreak=Complex_Context (SA).
94
95$dictionary = [$SA];
96
97#
98#  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
99#                               SA  (Dictionary chars, excluding Mn and Mc)
100#                               SG  (Unpaired Surrogates)
101#                               XX  (Unknown, unassigned)
102#                         as $AL  (Alphabetic)
103#
104$ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
105
106
107## -------------------------------------------------
108
109#
110# CAN_CM  is the set of characters that may combine with CM combining chars.
111#         Note that Linebreak UAX 14's concept of a combining char and the rules
112#         for what they can combine with are _very_ different from the rest of Unicode.
113#
114#         Note that $CM itself is left out of this set.  If CM is needed as a base
115#         it must be listed separately in the rule.
116#
117$CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
118$CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
119
120#
121# AL_FOLLOW  set of chars that can unconditionally follow an AL
122#            Needed in rules where stand-alone $CM s are treated as AL.
123#
124$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 $ALPlus];
125
126
127#
128#  Rule LB 4, 5    Mandatory (Hard) breaks.
129#
130$LB4Breaks    = [$BK $CR $LF $NL];
131$LB4NonBreaks = [^$BK $CR $LF $NL $CM];
132$CR $LF {100};
133
134#
135#  LB 6    Do not break before hard line breaks.
136#
137$LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
138$CAN_CM $CM*    $LB4Breaks {100};
139^$CM+           $LB4Breaks {100};
140
141# LB 7         x SP
142#              x ZW
143$LB4NonBreaks [$SP $ZW];
144$CAN_CM $CM*  [$SP $ZW];
145^$CM+         [$SP $ZW];
146
147#
148# LB 8         Break after zero width space
149#              ZW SP* ÷
150#
151$LB8Breaks    = [$LB4Breaks $ZW];
152$LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
153$ZW $SP* / [^$SP $ZW $LB4Breaks];
154
155# LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
156#
157$ZWJ [^$CM];
158
159# LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
160#                                $CM not covered by the above needs to behave like $AL
161#                                See definition of $CAN_CM.
162
163$CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
164^$CM+;
165
166#
167# LB 11  Do not break before or after WORD JOINER & related characters.
168#
169$CAN_CM $CM*  $WJ;
170$LB8NonBreaks $WJ;
171^$CM+         $WJ;
172
173$WJ $CM* .;
174
175#
176# LB 12  Do not break after NBSP and related characters.
177#         GL  x
178#
179$GL $CM* .;
180
181#
182# LB 12a  Do not break before NBSP and related characters ...
183#            [^SP BA HY] x GL
184#
185[[$LB8NonBreaks] - [$SP $BA $BAX $HY]] $CM* $GL;
186^$CM+ $GL;
187
188
189
190
191# LB 13   Don't break before ']' or '!' or or '/', even after spaces.
192#
193$LB8NonBreaks $CL;
194$CAN_CM $CM*  $CL;
195^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
196
197$LB8NonBreaks $CP;
198$CAN_CM $CM*  $CP;
199^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
200
201$LB8NonBreaks $EX;
202$CAN_CM $CM*  $EX;
203^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
204
205$LB8NonBreaks $SY;
206$CAN_CM $CM*  $SY;
207^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
208
209
210#
211# LB 14  Do not break after OP, even after spaces
212#        Note subtle interaction with "SP IS /" rules in LB14a.
213#        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
214#        which is the desired behavior.
215#
216$OP $CM* $SP* .;
217
218$OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
219                                   # by rule 8, CM following a SP is stand-alone.
220
221
222# LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
223#        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
224#        See issue ICU-20303
225
226
227$CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL $IN];
228$SP $IS           / [^ $CanFollowIS $NU $CM];
229$SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
230
231#
232# LB 14b Do not break before numeric separators (IS), even after spaces.
233
234[$LB8NonBreaks - $SP] $IS;
235$SP $IS $CM* [$CanFollowIS {eof}];
236$SP $IS $CM* $ZWJ [^$CM $NU];
237
238$CAN_CM $CM*  $IS;
239^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
240
241
242# LB 15
243$QU $CM* $SP* $OP;
244
245# LB 16
246# Do not break between closing punctuation and $NS, even with intervening spaces
247# But DO allow a break between closing punctuation and $NSX, don't include it here
248($CL | $CP) $CM* $SP* $NS;
249
250# LB 17
251$B2 $CM* $SP* $B2;
252
253#
254# LB 18  Break after spaces.
255#
256$LB18NonBreaks = [$LB8NonBreaks - [$SP]];
257$LB18Breaks    = [$LB8Breaks $SP];
258
259
260# LB 19
261#         x QU
262$LB18NonBreaks $CM* $QU;
263^$CM+               $QU;
264
265#         QU  x
266$QU $CM* .;
267
268# LB 20
269#        <break>  $CB
270#        $CB   <break>
271#
272$LB20NonBreaks = [$LB18NonBreaks - $CB];
273
274# LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
275#             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
276#             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
277#
278^($HY | $HH) $CM* $ALPlus;
279
280# LB 21        x   (BA | HY | NS)
281#           BB x
282#
283# DO allow breaks here before $BAX and $NSX, so don't include them
284$LB20NonBreaks $CM* ($BA | $HY | $NS);
285
286
287^$CM+ ($BA | $HY | $NS);
288
289$BB $CM* [^$CB];                                  #  $BB  x
290$BB $CM* $LB20NonBreaks;
291
292# LB 21a Don't break after Hebrew + Hyphen
293#   HL (HY | BA) x
294#
295$HL $CM* ($HY | $BA | $BAX) $CM* [^$CB]?;
296
297# LB 21b (forward) Don't break between SY and HL
298# (break between HL and SY already disallowed by LB 13 above)
299$SY $CM* $HL;
300
301# LB 22  Do not break before ellipses
302#
303$LB20NonBreaks $CM*    $IN;
304^$CM+ $IN;
305
306
307# LB 23
308#
309($ALPlus | $HL) $CM* $NU;
310^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
311$NU $CM* ($ALPlus | $HL);
312
313# LB 23a
314#
315$PR $CM* ($ID | $EB | $EM);
316($ID | $EB | $EM) $CM*  $PO;
317
318
319#
320# LB 24
321#
322($PR | $PO) $CM* ($ALPlus | $HL);
323($ALPlus | $HL) $CM* ($PR | $PO);
324^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
325
326#
327# LB 25   Numbers.
328#
329(($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
330    ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
331
332# LB 26  Do not break a Korean syllable
333#
334$JL $CM* ($JL | $JV | $H2 | $H3);
335($JV | $H2) $CM* ($JV | $JT);
336($JT | $H3) $CM* $JT;
337
338# LB 27  Treat korean Syllable Block the same as ID  (don't break it)
339($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
340($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
341$PR $CM* ($JL | $JV | $JT | $H2 | $H3);
342
343
344# LB 28   Do not break between alphabetics
345#
346($ALPlus | $HL) $CM* ($ALPlus | $HL);
347^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
348
349# LB 29
350$IS $CM* ($ALPlus | $HL);
351
352# LB 30
353($ALPlus | $HL | $NU) $CM* $OP30;
354^$CM+ $OP30;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
355$CP30 $CM* ($ALPlus | $HL | $NU);
356
357# LB 30a  Do not break between regional indicators. Break after pairs of them.
358#         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
359$RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
360$RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
361$RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $ZWJ {eof}];
362# note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
363#       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
364#       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
365
366# LB 30b Do not break between an Emoji Base and an Emoji Modifier
367$EB $CM* $EM;
368
369# LB 31 Break everywhere else.
370#       Match a single code point if no other rule applies.
371.;
372