• 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 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 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 = [$CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
79
80# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
81#         list it in the numerous rules that use CM.
82# By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
83
84$CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
85$CMX = [[$CM] - [$ZWJ]];
86
87#   Dictionary character set, for triggering language-based break engines. Currently
88#   limited to LineBreak=Complex_Context (SA).
89
90$dictionary = [$SA];
91
92#
93#  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
94#                               SA  (Dictionary chars, excluding Mn and Mc)
95#                               SG  (Unpaired Surrogates)
96#                               XX  (Unknown, unassigned)
97#                         as $AL  (Alphabetic)
98#
99$ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
100
101
102## -------------------------------------------------
103
104#
105# CAN_CM  is the set of characters that may combine with CM combining chars.
106#         Note that Linebreak UAX 14's concept of a combining char and the rules
107#         for what they can combine with are _very_ different from the rest of Unicode.
108#
109#         Note that $CM itself is left out of this set.  If CM is needed as a base
110#         it must be listed separately in the rule.
111#
112$CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
113$CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
114
115#
116# AL_FOLLOW  set of chars that can unconditionally follow an AL
117#            Needed in rules where stand-alone $CM s are treated as AL.
118#
119$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 $ALPlus];
120
121
122#
123#  Rule LB 4, 5    Mandatory (Hard) breaks.
124#
125$LB4Breaks    = [$BK $CR $LF $NL];
126$LB4NonBreaks = [^$BK $CR $LF $NL $CM];
127$CR $LF {100};
128
129#
130#  LB 6    Do not break before hard line breaks.
131#
132$LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
133$CAN_CM $CM*    $LB4Breaks {100};
134^$CM+           $LB4Breaks {100};
135
136# LB 7         x SP
137#              x ZW
138$LB4NonBreaks [$SP $ZW];
139$CAN_CM $CM*  [$SP $ZW];
140^$CM+         [$SP $ZW];
141
142#
143# LB 8         Break after zero width space
144#              ZW SP* ÷
145#
146$LB8Breaks    = [$LB4Breaks $ZW];
147$LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
148$ZW $SP* / [^$SP $ZW $LB4Breaks];
149
150# LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
151#
152$ZWJ [^$CM];
153
154# LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
155#                                $CM not covered by the above needs to behave like $AL
156#                                See definition of $CAN_CM.
157
158$CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
159^$CM+;
160
161#
162# LB 11  Do not break before or after WORD JOINER & related characters.
163#
164$CAN_CM $CM*  $WJ;
165$LB8NonBreaks $WJ;
166^$CM+         $WJ;
167
168$WJ $CM* .;
169
170#
171# LB 12  Do not break after NBSP and related characters.
172#         GL  x
173#
174$GL $CM* .;
175
176#
177# LB 12a  Do not break before NBSP and related characters ...
178#            [^SP BA HY] x GL
179#
180[[$LB8NonBreaks] - [$SP $BA $HY]] $CM* $GL;
181^$CM+ $GL;
182
183
184
185
186# LB 13   Don't break before ']' or '!' or or '/', even after spaces.
187#
188$LB8NonBreaks $CL;
189$CAN_CM $CM*  $CL;
190^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
191
192$LB8NonBreaks $CP;
193$CAN_CM $CM*  $CP;
194^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
195
196$LB8NonBreaks $EX;
197$CAN_CM $CM*  $EX;
198^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
199
200$LB8NonBreaks $SY;
201$CAN_CM $CM*  $SY;
202^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
203
204
205#
206# LB 14  Do not break after OP, even after spaces
207#        Note subtle interaction with "SP IS /" rules in LB14a.
208#        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
209#        which is the desired behavior.
210#
211$OP $CM* $SP* .;
212
213$OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
214                                   # by rule 8, CM following a SP is stand-alone.
215
216
217# LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
218#        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
219#        See issue ICU-20303
220
221
222$CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL $IN];
223$SP $IS           / [^ $CanFollowIS $NU $CM];
224$SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
225
226#
227# LB 14b Do not break before numeric separators (IS), even after spaces.
228
229[$LB8NonBreaks - $SP] $IS;
230$SP $IS $CM* [$CanFollowIS {eof}];
231$SP $IS $CM* $ZWJ [^$CM $NU];
232
233$CAN_CM $CM*  $IS;
234^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
235
236
237# LB 15
238$QU $CM* $SP* $OP;
239
240# LB 16
241($CL | $CP) $CM* $SP* $NS;
242
243# LB 17
244$B2 $CM* $SP* $B2;
245
246#
247# LB 18  Break after spaces.
248#
249$LB18NonBreaks = [$LB8NonBreaks - [$SP]];
250$LB18Breaks    = [$LB8Breaks $SP];
251
252
253# LB 19
254#         x QU
255$LB18NonBreaks $CM* $QU;
256^$CM+               $QU;
257
258#         QU  x
259$QU $CM* .;
260
261# LB 20
262#        <break>  $CB
263#        $CB   <break>
264#
265$LB20NonBreaks = [$LB18NonBreaks - $CB];
266
267# LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
268#             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
269#             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
270#
271^($HY | $HH) $CM* $ALPlus;
272
273# LB 21        x   (BA | HY | NS)
274#           BB x
275#
276$LB20NonBreaks $CM* ($BA | $HY | $NS);
277
278
279^$CM+ ($BA | $HY | $NS);
280
281$BB $CM* [^$CB];                                  #  $BB  x
282$BB $CM* $LB20NonBreaks;
283
284# LB 21a Don't break after Hebrew + Hyphen
285#   HL (HY | BA) x
286#
287$HL $CM* ($HY | $BA) $CM* [^$CB]?;
288
289# LB 21b (forward) Don't break between SY and HL
290# (break between HL and SY already disallowed by LB 13 above)
291$SY $CM* $HL;
292
293# LB 22  Do not break before ellipses
294#
295$LB20NonBreaks $CM*    $IN;
296^$CM+ $IN;
297
298
299# LB 23
300#
301($ALPlus | $HL) $CM* $NU;
302^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
303$NU $CM* ($ALPlus | $HL);
304
305# LB 23a
306#
307$PR $CM* ($ID | $EB | $EM);
308($ID | $EB | $EM) $CM*  $PO;
309
310
311#
312# LB 24
313#
314($PR | $PO) $CM* ($ALPlus | $HL);
315($ALPlus | $HL) $CM* ($PR | $PO);
316^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
317
318#
319# LB 25   Numbers.
320#
321(($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
322    ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
323
324# LB 26  Do not break a Korean syllable
325#
326$JL $CM* ($JL | $JV | $H2 | $H3);
327($JV | $H2) $CM* ($JV | $JT);
328($JT | $H3) $CM* $JT;
329
330# LB 27  Treat korean Syllable Block the same as ID  (don't break it)
331($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
332($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
333$PR $CM* ($JL | $JV | $JT | $H2 | $H3);
334
335
336# LB 28   Do not break between alphabetics
337#
338($ALPlus | $HL) $CM* ($ALPlus | $HL);
339^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
340
341# LB 29
342$IS $CM* ($ALPlus | $HL);
343
344# LB 30
345($ALPlus | $HL | $NU) $CM* $OP30;
346^$CM+ $OP30;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
347$CP30 $CM* ($ALPlus | $HL | $NU);
348
349# LB 30a  Do not break between regional indicators. Break after pairs of them.
350#         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
351$RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
352$RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
353$RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $ZWJ {eof}];
354# note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
355#       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
356#       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
357
358# LB 30b Do not break between an Emoji Base and an Emoji Modifier
359$EB $CM* $EM;
360
361# LB 31 Break everywhere else.
362#       Match a single code point if no other rule applies.
363.;
364