• 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#
4# Copyright (c) 2002-2016  International Business Machines Corporation and
5# others. All Rights Reserved.
6#
7#  file:  line_loose.txt
8#
9#         Line Breaking Rules
10#         Implement default line breaking as defined by
11#         Unicode Standard Annex #14 (https://www.unicode.org/reports/tr14/)
12#         for Unicode 14.0, with the following modification:
13#
14#         Boundaries between hyphens and following letters are suppressed when
15#         there is a boundary preceding the hyphen. See rule 20.9
16#
17#         This tailors the line break behavior to correspond to CSS
18#         line-break=loose (BCP47 -u-lb-loose) as defined for languages other than
19#         Chinese & Japanese.
20#         It sets characters of class CJ to behave like ID.
21#         In addition, it allows breaks:
22#         * before iteration marks 3005, 303B, 309D, 309E, 30FD, 30FE (all NS)
23#         * between characters of LineBreak class IN
24
25#
26#  Character Classes defined by TR 14.
27#
28
29!!chain;
30!!quoted_literals_only;
31
32$AI = [:LineBreak =  Ambiguous:];
33$AL = [:LineBreak =  Alphabetic:];
34$BA = [:LineBreak =  Break_After:];
35$HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
36$BB = [:LineBreak =  Break_Before:];
37$BK = [:LineBreak =  Mandatory_Break:];
38$B2 = [:LineBreak =  Break_Both:];
39$CB = [:LineBreak =  Contingent_Break:];
40$CJ = [:LineBreak =  Conditional_Japanese_Starter:];
41$CL = [:LineBreak =  Close_Punctuation:];
42# $CM = [:LineBreak =  Combining_Mark:];
43$CP = [:LineBreak =  Close_Parenthesis:];
44$CR = [:LineBreak =  Carriage_Return:];
45$EB = [:LineBreak =  EB:];
46$EM = [:LineBreak =  EM:];
47$EX = [:LineBreak =  Exclamation:];
48$GL = [:LineBreak =  Glue:];
49$HL = [:LineBreak =  Hebrew_Letter:];
50$HY = [:LineBreak =  Hyphen:];
51$H2 = [:LineBreak =  H2:];
52$H3 = [:LineBreak =  H3:];
53# CSS Loose tailoring: CJ resolves to ID
54$ID = [[:LineBreak =  Ideographic:] $CJ];
55$IN = [:LineBreak =  Inseperable:];
56$IS = [:LineBreak =  Infix_Numeric:];
57$JL = [:LineBreak =  JL:];
58$JV = [:LineBreak =  JV:];
59$JT = [:LineBreak =  JT:];
60$LF = [:LineBreak =  Line_Feed:];
61$NL = [:LineBreak =  Next_Line:];
62$NSX = [\u3005 \u303B \u309D \u309E \u30FD \u30FE];
63$NS = [[:LineBreak =  Nonstarter:] - $NSX];
64$NU = [:LineBreak =  Numeric:];
65$OP = [:LineBreak =  Open_Punctuation:];
66$PO = [:LineBreak =  Postfix_Numeric:];
67$PR = [:LineBreak =  Prefix_Numeric:];
68$QU = [:LineBreak =  Quotation:];
69$RI = [:LineBreak =  Regional_Indicator:];
70$SA = [:LineBreak =  Complex_Context:];
71$SG = [:LineBreak =  Surrogate:];
72$SP = [:LineBreak =  Space:];
73$SY = [:LineBreak =  Break_Symbols:];
74$WJ = [:LineBreak =  Word_Joiner:];
75$XX = [:LineBreak =  Unknown:];
76$ZW = [:LineBreak =  ZWSpace:];
77$ZWJ = [:LineBreak = ZWJ:];
78
79# OP30 and CP30 are variants of OP and CP that appear in-line in rule LB30 from UAX 14,
80# without a formal name. Because ICU rules require multiple uses of the expressions,
81# give them a single definition with a name
82
83$OP30 = [$OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
84$CP30 = [$CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
85
86$ExtPictUnassigned = [\p{Extended_Pictographic} & \p{Cn}];
87
88# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
89#         list it in the numerous rules that use CM.
90# By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
91
92$CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
93$CMX = [[$CM] - [$ZWJ]];
94
95#   Dictionary character set, for triggering language-based break engines. Currently
96#   limited to LineBreak=Complex_Context (SA).
97
98$dictionary = [$SA];
99
100#
101#  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
102#                               SA  (Dictionary chars, excluding Mn and Mc)
103#                               SG  (Unpaired Surrogates)
104#                               XX  (Unknown, unassigned)
105#                         as $AL  (Alphabetic)
106#
107$ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
108
109
110## -------------------------------------------------
111
112#
113# CAN_CM  is the set of characters that may combine with CM combining chars.
114#         Note that Linebreak UAX 14's concept of a combining char and the rules
115#         for what they can combine with are _very_ different from the rest of Unicode.
116#
117#         Note that $CM itself is left out of this set.  If CM is needed as a base
118#         it must be listed separately in the rule.
119#
120$CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
121$CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
122
123#
124# AL_FOLLOW  set of chars that can unconditionally follow an AL
125#            Needed in rules where stand-alone $CM s are treated as AL.
126#
127$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 $ALPlus];
128
129
130#
131#  Rule LB 4, 5    Mandatory (Hard) breaks.
132#
133$LB4Breaks    = [$BK $CR $LF $NL];
134$LB4NonBreaks = [^$BK $CR $LF $NL $CM];
135$CR $LF {100};
136
137#
138#  LB 6    Do not break before hard line breaks.
139#
140$LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
141$CAN_CM $CM*    $LB4Breaks {100};
142^$CM+           $LB4Breaks {100};
143
144# LB 7         x SP
145#              x ZW
146$LB4NonBreaks [$SP $ZW];
147$CAN_CM $CM*  [$SP $ZW];
148^$CM+         [$SP $ZW];
149
150#
151# LB 8         Break after zero width space
152#              ZW SP* ÷
153#
154$LB8Breaks    = [$LB4Breaks $ZW];
155$LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
156$ZW $SP* / [^$SP $ZW $LB4Breaks];
157
158# LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
159#
160$ZWJ [^$CM];
161
162# LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
163#                                $CM not covered by the above needs to behave like $AL
164#                                See definition of $CAN_CM.
165
166$CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
167^$CM+;
168
169#
170# LB 11  Do not break before or after WORD JOINER & related characters.
171#
172$CAN_CM $CM*  $WJ;
173$LB8NonBreaks $WJ;
174^$CM+         $WJ;
175
176$WJ $CM* .;
177
178#
179# LB 12  Do not break after NBSP and related characters.
180#         GL  x
181#
182$GL $CM* .;
183
184#
185# LB 12a  Do not break before NBSP and related characters ...
186#            [^SP BA HY] x GL
187#
188[[$LB8NonBreaks] - [$SP $BA $HY]] $CM* $GL;
189^$CM+ $GL;
190
191
192
193
194# LB 13   Don't break before ']' or '!' or '/', even after spaces.
195#
196$LB8NonBreaks $CL;
197$CAN_CM $CM*  $CL;
198^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
199
200$LB8NonBreaks $CP;
201$CAN_CM $CM*  $CP;
202^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
203
204$LB8NonBreaks $EX;
205$CAN_CM $CM*  $EX;
206^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
207
208$LB8NonBreaks $SY;
209$CAN_CM $CM*  $SY;
210^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
211
212
213#
214# LB 14  Do not break after OP, even after spaces
215#        Note subtle interaction with "SP IS /" rules in LB14a.
216#        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
217#        which is the desired behavior.
218#
219$OP $CM* $SP* .;
220
221$OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
222                                   # by rule 8, CM following a SP is stand-alone.
223
224
225# LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
226#        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
227#        See issue ICU-20303
228
229
230$CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL $IN];
231$SP $IS           / [^ $CanFollowIS $NU $CM];
232$SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
233
234#
235# LB 14b Do not break before numeric separators (IS), even after spaces.
236
237[$LB8NonBreaks - $SP] $IS;
238$SP $IS $CM* [$CanFollowIS {eof}];
239$SP $IS $CM* $ZWJ [^$CM $NU];
240
241$CAN_CM $CM*  $IS;
242^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
243
244
245# LB 15
246$QU $CM* $SP* $OP;
247
248# LB 16
249# Do not break between closing punctuation and $NS, even with intervening spaces
250# But DO allow a break between closing punctuation and $NSX, don't include it here
251($CL | $CP) $CM* $SP* $NS;
252
253# LB 17
254$B2 $CM* $SP* $B2;
255
256#
257# LB 18  Break after spaces.
258#
259$LB18NonBreaks = [$LB8NonBreaks - [$SP]];
260$LB18Breaks    = [$LB8Breaks $SP];
261
262
263# LB 19
264#         x QU
265$LB18NonBreaks $CM* $QU;
266^$CM+               $QU;
267
268#         QU  x
269$QU $CM* .;
270
271# LB 20
272#        <break>  $CB
273#        $CB   <break>
274#
275$LB20NonBreaks = [$LB18NonBreaks - $CB];
276
277# LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
278#             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
279#             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
280#
281^($HY | $HH) $CM* $ALPlus;
282
283# LB 21        x   (BA | HY | NS)
284#           BB x
285#
286# DO allow breaks here before NSX, so don't include it
287$LB20NonBreaks $CM* ($BA | $HY | $NS);
288
289
290^$CM+ ($BA | $HY | $NS);
291
292$BB $CM* [^$CB];                                  #  $BB  x
293$BB $CM* $LB20NonBreaks;
294
295# LB 21a Don't break after Hebrew + Hyphen
296#   HL (HY | BA) x
297#
298$HL $CM* ($HY | $BA) $CM* [^$CB]?;
299
300# LB 21b (forward) Don't break between SY and HL
301# (break between HL and SY already disallowed by LB 13 above)
302$SY $CM* $HL;
303
304
305# LB 22  Do not break before ellipses
306#
307[$LB20NonBreaks - $IN] $CM*    $IN;    # line_loose tailoring
308^$CM+ $IN;
309
310
311# LB 23
312#
313($ALPlus | $HL) $CM* $NU;
314^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
315$NU $CM* ($ALPlus | $HL);
316
317# LB 23a
318#
319$PR $CM* ($ID | $EB | $EM);
320($ID | $EB | $EM) $CM*  $PO;
321
322
323#
324# LB 24
325#
326($PR | $PO) $CM* ($ALPlus | $HL);
327($ALPlus | $HL) $CM* ($PR | $PO);
328^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
329
330#
331# LB 25   Numbers.
332#
333(($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
334    ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
335
336# LB 26  Do not break a Korean syllable
337#
338$JL $CM* ($JL | $JV | $H2 | $H3);
339($JV | $H2) $CM* ($JV | $JT);
340($JT | $H3) $CM* $JT;
341
342# LB 27  Treat korean Syllable Block the same as ID  (don't break it)
343($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
344$PR $CM* ($JL | $JV | $JT | $H2 | $H3);
345
346
347# LB 28   Do not break between alphabetics
348#
349($ALPlus | $HL) $CM* ($ALPlus | $HL);
350^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
351
352# LB 29
353$IS $CM* ($ALPlus | $HL);
354
355# LB 30
356($ALPlus | $HL | $NU) $CM* $OP30;
357^$CM+ $OP30;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
358$CP30 $CM* ($ALPlus | $HL | $NU);
359
360# LB 30a  Do not break between regional indicators. Break after pairs of them.
361#         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
362$RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
363$RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
364$RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $ZWJ {eof}];
365# note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
366#       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
367#       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
368
369# LB30b Do not break between an emoji base (or potential emoji) and an emoji modifier.
370$EB $CM* $EM;
371$ExtPictUnassigned $CM* $EM;
372
373# LB 31 Break everywhere else.
374#       Match a single code point if no other rule applies.
375.;
376