icuSources/data/brkitr/rules/line_loose_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_loose_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=loose (BCP47 -u-lb-loose) 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 #         * before iteration marks 3005, 303B, 309D, 309E, 30FD, 30FE (all NS)
  22 #         * between characters of LineBreak class IN such as 2026
  23 #         * before some centered punct 203C, 2047, 2048, 2049, 30FB, FF1A, FF1B,
  24 #           FF65 (all NS) and FF01, FF1F (both EX).
  25 #         * before suffix characters with LineBreak class PO and EastAsianWidth A,F,W;
  26 #           this includes: 00B0 2030 2032 2033 2035 2103 2109 FE6A FF05 FFE0
  27 #         * after prefix characters with LineBreak class PR and EastAsianWidth A,F,W;
  28 #           this includes: 00A4 00B1 20AC 2116 FE69 FF04 FFE1 FFE5 FFE6
  29 #         It allows breaking before 201C and after 201D, for zh_Hans, zh_Hant, and ja.
  30
  31
  32 #
  33 #  Character Classes defined by TR 14.
  34 #
  35
  36 !!chain;
  37 !!quoted_literals_only;
  38
  39 $AI = [:LineBreak =  Ambiguous:];
  40 $AL = [:LineBreak =  Alphabetic:];
  41 $BAX = [\u2010 \u2013];
  42 $BA = [[:LineBreak =  Break_After:] - $BAX];
  43 $HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
  44 $BB = [:LineBreak =  Break_Before:];
  45 $BK = [:LineBreak =  Mandatory_Break:];
  46 $B2 = [:LineBreak =  Break_Both:];
  47 $CB = [:LineBreak =  Contingent_Break:];
  48 $CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  49 $CL = [[:LineBreak =  Close_Punctuation:] \u201d];
  50 # $CM = [:LineBreak =  Combining_Mark:];
  51 $CP = [:LineBreak =  Close_Parenthesis:];
  52 $CR = [:LineBreak =  Carriage_Return:];
  53 $EB = [:LineBreak =  EB:];
  54 $EM = [:LineBreak =  EM:];
  55 $EXX = [\uFF01 \uFF1F];
  56 $EX = [[:LineBreak =  Exclamation:] - $EXX];
  57 $GL = [:LineBreak =  Glue:];
  58 $HL = [:LineBreak =  Hebrew_Letter:];
  59 $HY = [:LineBreak =  Hyphen:];
  60 $H2 = [:LineBreak =  H2:];
  61 $H3 = [:LineBreak =  H3:];
  62 # CSS Loose tailoring: CJ resolves to ID
  63 $ID = [[:LineBreak =  Ideographic:] $CJ];
  64 $IN = [:LineBreak =  Inseperable:];
  65 $IS = [:LineBreak =  Infix_Numeric:];
  66 $JL = [:LineBreak =  JL:];
  67 $JV = [:LineBreak =  JV:];
  68 $JT = [:LineBreak =  JT:];
  69 $LF = [:LineBreak =  Line_Feed:];
  70 $NL = [:LineBreak =  Next_Line:];
  71 $NSX = [\u301C \u30A0 \u3005 \u303B \u309D \u309E \u30FD \u30FE \u203C \u2047 \u2048 \u2049 \u30FB \uFF1A \uFF1B \uFF65];
  72 $NS = [[:LineBreak =  Nonstarter:] - $NSX];
  73 $NU = [:LineBreak =  Numeric:];
  74 $OP = [[:LineBreak =  Open_Punctuation:] \u201c];
  75 $POX = [\u00B0 \u2030 \u2032 \u2033 \u2035 \u2103 \u2109 \uFE6A \uFF05 \uFFE0];
  76 $PO = [[:LineBreak =  Postfix_Numeric:] - $POX];
  77 $PRX = [\u00A4 \u00B1 \u20AC \u2116 \uFE69 \uFF04 \uFFE1 \uFFE5 \uFFE6];
  78 $PR = [[:LineBreak =  Prefix_Numeric:] - $PRX];
  79 $QU = [[:LineBreak =  Quotation:] - [\u201c\u201d]];
  80 $RI = [:LineBreak =  Regional_Indicator:];
  81 $SA = [:LineBreak =  Complex_Context:];
  82 $SG = [:LineBreak =  Surrogate:];
  83 $SP = [:LineBreak =  Space:];
  84 $SY = [:LineBreak =  Break_Symbols:];
  85 $WJ = [:LineBreak =  Word_Joiner:];
  86 $XX = [:LineBreak =  Unknown:];
  87 $ZW = [:LineBreak =  ZWSpace:];
  88 $ZWJ = [:LineBreak = ZWJ:];
  89
  90 # OP30 and CP30 are variants of OP and CP that appear in-line in rule LB30 from UAX 14,
  91 # without a formal name. Because ICU rules require multiple uses of the expressions,
  92 # give them a single definition with a name
  93
  94 $OP30 = [$OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
  95 $CP30 = [$CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
  96
  97 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  98 #         list it in the numerous rules that use CM.
  99 # By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
 100
 101 $CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
 102 $CMX = [[$CM] - [$ZWJ]];
 103
 104 #   Dictionary character set, for triggering language-based break engines. Currently
 105 #   limited to LineBreak=Complex_Context (SA).
 106
 107 $dictionary = [$SA];
 108
 109 #
 110 #  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
 111 #                               SA  (Dictionary chars, excluding Mn and Mc)
 112 #                               SG  (Unpaired Surrogates)
 113 #                               XX  (Unknown, unassigned)
 114 #                         as $AL  (Alphabetic)
 115 #
 116 $ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
 117
 118
 119 ## -------------------------------------------------
 120
 121 #
 122 # CAN_CM  is the set of characters that may combine with CM combining chars.
 123 #         Note that Linebreak UAX 14's concept of a combining char and the rules
 124 #         for what they can combine with are _very_ different from the rest of Unicode.
 125 #
 126 #         Note that $CM itself is left out of this set.  If CM is needed as a base
 127 #         it must be listed separately in the rule.
 128 #
 129 $CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
 130 $CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
 131
 132 #
 133 # AL_FOLLOW  set of chars that can unconditionally follow an AL
 134 #            Needed in rules where stand-alone $CM s are treated as AL.
 135 #
 136 $AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP30 $QU $BA $HY $NS $IN $NU $PR $PO $POX $ALPlus];
 137
 138
 139 #
 140 #  Rule LB 4, 5    Mandatory (Hard) breaks.
 141 #
 142 $LB4Breaks    = [$BK $CR $LF $NL];
 143 $LB4NonBreaks = [^$BK $CR $LF $NL $CM];
 144 $CR $LF {100};
 145
 146 #
 147 #  LB 6    Do not break before hard line breaks.
 148 #
 149 $LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
 150 $CAN_CM $CM*    $LB4Breaks {100};
 151 ^$CM+           $LB4Breaks {100};
 152
 153 # LB 7         x SP
 154 #              x ZW
 155 $LB4NonBreaks [$SP $ZW];
 156 $CAN_CM $CM*  [$SP $ZW];
 157 ^$CM+         [$SP $ZW];
 158
 159 #
 160 # LB 8         Break after zero width space
 161 #              ZW SP* ÷
 162 #
 163 $LB8Breaks    = [$LB4Breaks $ZW];
 164 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 165 $ZW $SP* / [^$SP $ZW $LB4Breaks];
 166
 167 # LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
 168 #
 169 $ZWJ [^$CM];
 170
 171 # LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
 172 #                                $CM not covered by the above needs to behave like $AL
 173 #                                See definition of $CAN_CM.
 174
 175 $CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
 176 ^$CM+;
 177
 178 #
 179 # LB 11  Do not break before or after WORD JOINER & related characters.
 180 #
 181 $CAN_CM $CM*  $WJ;
 182 $LB8NonBreaks $WJ;
 183 ^$CM+         $WJ;
 184
 185 $WJ $CM* .;
 186
 187 #
 188 # LB 12  Do not break after NBSP and related characters.
 189 #         GL  x
 190 #
 191 $GL $CM* .;
 192
 193 #
 194 # LB 12a  Do not break before NBSP and related characters ...
 195 #            [^SP BA HY] x GL
 196 #
 197 [[$LB8NonBreaks] - [$SP $BA $BAX $HY]] $CM* $GL;
 198 ^$CM+ $GL;
 199
 200
 201
 202 # LB 13   Don't break before ']' or '!' or or '/', even after spaces.
 203 #
 204 # Do not include $EXX here
 205 $LB8NonBreaks $CL;
 206 $CAN_CM $CM*  $CL;
 207 ^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
 208
 209 $LB8NonBreaks $CP;
 210 $CAN_CM $CM*  $CP;
 211 ^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
 212
 213 $LB8NonBreaks $EX;
 214 $CAN_CM $CM*  $EX;
 215 ^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
 216
 217 $LB8NonBreaks $SY;
 218 $CAN_CM $CM*  $SY;
 219 ^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
 220
 221
 222 #
 223 # LB 14  Do not break after OP, even after spaces
 224 #        Note subtle interaction with "SP IS /" rules in LB14a.
 225 #        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
 226 #        which is the desired behavior.
 227 #
 228 $OP $CM* $SP* .;
 229
 230 $OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
 231                                    # by rule 8, CM following a SP is stand-alone.
 232
 233
 234 # LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
 235 #        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
 236 #        See issue ICU-20303
 237
 238
 239 $CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL $IN];
 240 $SP $IS           / [^ $CanFollowIS $NU $CM];
 241 $SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
 242
 243 #
 244 # LB 14b Do not break before numeric separators (IS), even after spaces.
 245
 246 [$LB8NonBreaks - $SP] $IS;
 247 $SP $IS $CM* [$CanFollowIS {eof}];
 248 $SP $IS $CM* $ZWJ [^$CM $NU];
 249
 250 $CAN_CM $CM*  $IS;
 251 ^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
 252
 253
 254 # LB 15
 255 $QU $CM* $SP* $OP;
 256
 257 # LB 16
 258 # Do not break between closing punctuation and $NS, even with intervening spaces
 259 # But DO allow a break between closing punctuation and $NSX, don't include it here
 260 ($CL | $CP) $CM* $SP* $NS;
 261
 262 # LB 17
 263 $B2 $CM* $SP* $B2;
 264
 265 #
 266 # LB 18  Break after spaces.
 267 #
 268 $LB18NonBreaks = [$LB8NonBreaks - [$SP]];
 269 $LB18Breaks    = [$LB8Breaks $SP];
 270
 271
 272 # LB 19
 273 #         x QU
 274 $LB18NonBreaks $CM* $QU;
 275 ^$CM+               $QU;
 276
 277 #         QU  x
 278 $QU $CM* .;
 279
 280 # LB 20
 281 #        <break>  $CB
 282 #        $CB   <break>
 283 #
 284 $LB20NonBreaks = [$LB18NonBreaks - $CB];
 285
 286 # LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
 287 #             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
 288 #             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
 289 #
 290 ^($HY | $HH) $CM* $ALPlus;
 291
 292 # LB 21        x   (BA | HY | NS)
 293 #           BB x
 294 #
 295 # DO allow breaks here before $BAX and $NSX, so don't include them
 296 $LB20NonBreaks $CM* ($BA | $HY | $NS);
 297
 298
 299 ^$CM+ ($BA | $HY | $NS);
 300
 301 $BB $CM* [^$CB];                                  #  $BB  x
 302 $BB $CM* $LB20NonBreaks;
 303
 304 # LB 21a Don't break after Hebrew + Hyphen
 305 #   HL (HY | BA) x
 306 #
 307 $HL $CM* ($HY | $BA | $BAX) $CM* [^$CB]?;
 308
 309 # LB 21b (forward) Don't break between SY and HL
 310 # (break between HL and SY already disallowed by LB 13 above)
 311 $SY $CM* $HL;
 312
 313
 314 # LB 22  Do not break before ellipses
 315 #
 316 [$LB20NonBreaks - $IN] $CM*    $IN;    # line_loose tailoring
 317 ^$CM+ $IN;
 318
 319
 320 # LB 23
 321 #
 322 ($ALPlus | $HL) $CM* $NU;
 323 ^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
 324 $NU $CM* ($ALPlus | $HL);
 325
 326 # LB 23a
 327 # Do not include $POX here
 328 #
 329 $PR $CM* ($ID | $EB | $EM);
 330 ($ID | $EB | $EM) $CM*  $PO;
 331
 332
 333 #
 334 # LB 24
 335 #
 336 # Do not include $PRX here
 337 ($PR | $PO | $POX) $CM* ($ALPlus | $HL);
 338 ($ALPlus | $HL) $CM* ($PR | $PO | $POX);     # TODO: should this be ($PR | $PRX | $PO)
 339 ^$CM+ ($PR | $PO | $POX);       # Rule 10, any otherwise unattached CM behaves as AL
 340
 341 #
 342 # LB 25   Numbers.
 343 #
 344 # Here do not include $PRX at the beginning or $POX at the end
 345 (($PR | $PO | $POX) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
 346     ($CM* ($CL | $CP))? ($CM* ($PR | $PRX | $PO))?;
 347
 348 # LB 26  Do not break a Korean syllable
 349 #
 350 $JL $CM* ($JL | $JV | $H2 | $H3);
 351 ($JV | $H2) $CM* ($JV | $JT);
 352 ($JT | $H3) $CM* $JT;
 353
 354 # LB 27  Treat korean Syllable Block the same as ID  (don't break it)
 355 # Do not include $POX or $PRX here
 356 ($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
 357 ($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
 358 $PR $CM* ($JL | $JV | $JT | $H2 | $H3);
 359
 360
 361 # LB 28   Do not break between alphabetics
 362 #
 363 ($ALPlus | $HL) $CM* ($ALPlus | $HL);
 364 ^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
 365
 366 # LB 29
 367 $IS $CM* ($ALPlus | $HL);
 368
 369 # LB 30
 370 ($ALPlus | $HL | $NU) $CM* $OP30;
 371 ^$CM+ $OP30;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
 372 $CP30 $CM* ($ALPlus | $HL | $NU);
 373
 374 # LB 30a  Do not break between regional indicators. Break after pairs of them.
 375 #         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
 376 $RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $CM]];
 377 $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]];
 378 $RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $IN $ZWJ {eof}];
 379 # note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
 380 #       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
 381 #       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
 382
 383 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 384 $EB $CM* $EM;
 385
 386 # LB 31 Break everywhere else.
 387 #       Match a single code point if no other rule applies.
 388 .;