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