icuSources/data/brkitr/rules/line_cj.txt

   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 42 for Unicode 12.0
  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 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  74 #         list it in the numerous rules that use CM.
  75 # By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
  76
  77 $CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
  78 $CMX = [[$CM] - [$ZWJ]];
  79
  80 #   Dictionary character set, for triggering language-based break engines. Currently
  81 #   limited to LineBreak=Complex_Context (SA).
  82
  83 $dictionary = [$SA];
  84
  85 #
  86 #  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
  87 #                               SA  (Dictionary chars, excluding Mn and Mc)
  88 #                               SG  (Unpaired Surrogates)
  89 #                               XX  (Unknown, unassigned)
  90 #                         as $AL  (Alphabetic)
  91 #
  92 $ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
  93
  94
  95 ## -------------------------------------------------
  96
  97 #
  98 # CAN_CM  is the set of characters that may combine with CM combining chars.
  99 #         Note that Linebreak UAX 14's concept of a combining char and the rules
 100 #         for what they can combine with are _very_ different from the rest of Unicode.
 101 #
 102 #         Note that $CM itself is left out of this set.  If CM is needed as a base
 103 #         it must be listed separately in the rule.
 104 #
 105 $CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
 106 $CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
 107
 108 #
 109 # AL_FOLLOW  set of chars that can unconditionally follow an AL
 110 #            Needed in rules where stand-alone $CM s are treated as AL.
 111 #
 112 $AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP $QU $BA $HY $NS $IN $NU $PR $PO $ALPlus];
 113
 114
 115 #
 116 #  Rule LB 4, 5    Mandatory (Hard) breaks.
 117 #
 118 $LB4Breaks    = [$BK $CR $LF $NL];
 119 $LB4NonBreaks = [^$BK $CR $LF $NL $CM];
 120 $CR $LF {100};
 121
 122 #
 123 #  LB 6    Do not break before hard line breaks.
 124 #
 125 $LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
 126 $CAN_CM $CM*    $LB4Breaks {100};
 127 ^$CM+           $LB4Breaks {100};
 128
 129 # LB 7         x SP
 130 #              x ZW
 131 $LB4NonBreaks [$SP $ZW];
 132 $CAN_CM $CM*  [$SP $ZW];
 133 ^$CM+         [$SP $ZW];
 134
 135 #
 136 # LB 8         Break after zero width space
 137 #              ZW SP* ÷
 138 #
 139 $LB8Breaks    = [$LB4Breaks $ZW];
 140 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 141 $ZW $SP* / [^$SP $ZW $LB4Breaks];
 142
 143 # LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
 144 #
 145 $ZWJ [^$CM];
 146
 147 # LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
 148 #                                $CM not covered by the above needs to behave like $AL
 149 #                                See definition of $CAN_CM.
 150
 151 $CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
 152 ^$CM+;
 153
 154 #
 155 # LB 11  Do not break before or after WORD JOINER & related characters.
 156 #
 157 $CAN_CM $CM*  $WJ;
 158 $LB8NonBreaks $WJ;
 159 ^$CM+         $WJ;
 160
 161 $WJ $CM* .;
 162
 163 #
 164 # LB 12  Do not break after NBSP and related characters.
 165 #         GL  x
 166 #
 167 $GL $CM* .;
 168
 169 #
 170 # LB 12a  Do not break before NBSP and related characters ...
 171 #            [^SP BA HY] x GL
 172 #
 173 [[$LB8NonBreaks] - [$SP $BA $HY]] $CM* $GL;
 174 ^$CM+ $GL;
 175
 176
 177
 178
 179 # LB 13   Don't break before ']' or '!' or or '/', even after spaces.
 180 #
 181 $LB8NonBreaks $CL;
 182 $CAN_CM $CM*  $CL;
 183 ^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
 184
 185 $LB8NonBreaks $CP;
 186 $CAN_CM $CM*  $CP;
 187 ^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
 188
 189 $LB8NonBreaks $EX;
 190 $CAN_CM $CM*  $EX;
 191 ^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
 192
 193 $LB8NonBreaks $SY;
 194 $CAN_CM $CM*  $SY;
 195 ^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
 196
 197
 198 #
 199 # LB 14  Do not break after OP, even after spaces
 200 #        Note subtle interaction with "SP IS /" rules in LB14a.
 201 #        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
 202 #        which is the desired behavior.
 203 #
 204 $OP $CM* $SP* .;
 205
 206 $OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
 207                                    # by rule 8, CM following a SP is stand-alone.
 208
 209
 210 # LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
 211 #        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
 212 #        See issue ICU-20303
 213
 214
 215 $CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL];
 216 $SP $IS           / [^ $CanFollowIS $NU $CM];
 217 $SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
 218
 219 #
 220 # LB 14b Do not break before numeric separators (IS), even after spaces.
 221
 222 [$LB8NonBreaks - $SP] $IS;
 223 $SP $IS $CM* [$CanFollowIS {eof}];
 224 $SP $IS $CM* $ZWJ [^$CM $NU];
 225
 226 $CAN_CM $CM*  $IS;
 227 ^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
 228
 229
 230 # LB 15
 231 $QU $CM* $SP* $OP;
 232
 233 # LB 16
 234 ($CL | $CP) $CM* $SP* $NS;
 235
 236 # LB 17
 237 $B2 $CM* $SP* $B2;
 238
 239 #
 240 # LB 18  Break after spaces.
 241 #
 242 $LB18NonBreaks = [$LB8NonBreaks - [$SP]];
 243 $LB18Breaks    = [$LB8Breaks $SP];
 244
 245
 246 # LB 19
 247 #         x QU
 248 $LB18NonBreaks $CM* $QU;
 249 ^$CM+               $QU;
 250
 251 #         QU  x
 252 $QU $CM* .;
 253
 254 # LB 20
 255 #        <break>  $CB
 256 #        $CB   <break>
 257 #
 258 $LB20NonBreaks = [$LB18NonBreaks - $CB];
 259
 260 # LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
 261 #             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
 262 #             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
 263 #
 264 ^($HY | $HH) $CM* $ALPlus;
 265
 266 # LB 21        x   (BA | HY | NS)
 267 #           BB x
 268 #
 269 $LB20NonBreaks $CM* ($BA | $HY | $NS);
 270
 271
 272 ^$CM+ ($BA | $HY | $NS);
 273
 274 $BB $CM* [^$CB];                                  #  $BB  x
 275 $BB $CM* $LB20NonBreaks;
 276
 277 # LB 21a Don't break after Hebrew + Hyphen
 278 #   HL (HY | BA) x
 279 #
 280 $HL $CM* ($HY | $BA) $CM* [^$CB]?;
 281
 282 # LB 21b (forward) Don't break between SY and HL
 283 # (break between HL and SY already disallowed by LB 13 above)
 284 $SY $CM* $HL;
 285
 286 # LB 22
 287 ($ALPlus | $HL) $CM* $IN;
 288 ^$CM+    $IN;     #  by rule 10, any otherwise unattached CM behaves as AL
 289 $EX $CM*    $IN;
 290 ($ID | $EB | $EM) $CM*  $IN;
 291 $IN $CM*    $IN;
 292 $NU $CM*    $IN;
 293
 294
 295 # $LB 23
 296 #
 297 ($ALPlus | $HL) $CM* $NU;
 298 ^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
 299 $NU $CM* ($ALPlus | $HL);
 300
 301 # LB 23a
 302 #
 303 $PR $CM* ($ID | $EB | $EM);
 304 ($ID | $EB | $EM) $CM*  $PO;
 305
 306
 307 #
 308 # LB 24
 309 #
 310 ($PR | $PO) $CM* ($ALPlus | $HL);
 311 ($ALPlus | $HL) $CM* ($PR | $PO);
 312 ^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
 313
 314 #
 315 # LB 25   Numbers.
 316 #
 317 (($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
 318     ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
 319
 320 # LB 26  Do not break a Korean syllable
 321 #
 322 $JL $CM* ($JL | $JV | $H2 | $H3);
 323 ($JV | $H2) $CM* ($JV | $JT);
 324 ($JT | $H3) $CM* $JT;
 325
 326 # LB 27  Treat korean Syllable Block the same as ID  (don't break it)
 327 ($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
 328 ($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
 329 $PR $CM* ($JL | $JV | $JT | $H2 | $H3);
 330
 331
 332 # LB 28   Do not break between alphabetics
 333 #
 334 ($ALPlus | $HL) $CM* ($ALPlus | $HL);
 335 ^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
 336
 337 # LB 29
 338 $IS $CM* ($ALPlus | $HL);
 339
 340 # LB 30
 341 ($ALPlus | $HL | $NU) $CM* $OP;
 342 ^$CM+ $OP;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
 343 $CP $CM* ($ALPlus | $HL | $NU);
 344
 345 # LB 30a  Do not break between regional indicators. Break after pairs of them.
 346 #         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
 347 $RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 348 $RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 349 $RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $ZWJ {eof}];
 350 # note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
 351 #       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
 352 #       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
 353
 354 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 355 $EB $CM* $EM;
 356
 357 # LB 31 Break everywhere else.
 358 #       Match a single code point if no other rule applies.
 359 .;