icuSources/data/brkitr/rules/line_normal_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_normal_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 tailors the line break behavior to correspond to CSS
  17 #         line-break=normal (BCP47 -u-lb-normal) 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 #         It allows breaking before 201C and after 201D, for zh_Hans, zh_Hant, and ja.
  22
  23 #
  24 #  Character Classes defined by TR 14.
  25 #
  26
  27 !!chain;
  28 !!quoted_literals_only;
  29
  30 $AI = [:LineBreak =  Ambiguous:];
  31 $AL = [:LineBreak =  Alphabetic:];
  32 $BAX = [\u2010 \u2013];
  33 $BA = [[:LineBreak =  Break_After:] - $BAX];
  34 $HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
  35 $BB = [:LineBreak =  Break_Before:];
  36 $BK = [:LineBreak =  Mandatory_Break:];
  37 $B2 = [:LineBreak =  Break_Both:];
  38 $CB = [:LineBreak =  Contingent_Break:];
  39 $CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  40 $CL = [[:LineBreak =  Close_Punctuation:] \u201d];
  41 # $CM = [:LineBreak =  Combining_Mark:];
  42 $CP = [:LineBreak =  Close_Parenthesis:];
  43 $CR = [:LineBreak =  Carriage_Return:];
  44 $EB = [:LineBreak =  EB:];
  45 $EM = [:LineBreak =  EM:];
  46 $EX = [:LineBreak =  Exclamation:];
  47 $GL = [:LineBreak =  Glue:];
  48 $HL = [:LineBreak =  Hebrew_Letter:];
  49 $HY = [:LineBreak =  Hyphen:];
  50 $H2 = [:LineBreak =  H2:];
  51 $H3 = [:LineBreak =  H3:];
  52 # CSS Normal tailoring: CJ resolves to ID
  53 $ID = [[:LineBreak =  Ideographic:] $CJ];
  54 $IN = [:LineBreak =  Inseperable:];
  55 $IS = [:LineBreak =  Infix_Numeric:];
  56 $JL = [:LineBreak =  JL:];
  57 $JV = [:LineBreak =  JV:];
  58 $JT = [:LineBreak =  JT:];
  59 $LF = [:LineBreak =  Line_Feed:];
  60 $NL = [:LineBreak =  Next_Line:];
  61 $NSX = [\u301C \u30A0];
  62 $NS = [[:LineBreak =  Nonstarter:] - $NSX];
  63 $NU = [:LineBreak =  Numeric:];
  64 $OP = [[:LineBreak =  Open_Punctuation:] \u201c];
  65 $PO = [:LineBreak =  Postfix_Numeric:];
  66 $PR = [:LineBreak =  Prefix_Numeric:];
  67 $QU = [[:LineBreak =  Quotation:] - [\u201c\u201d]];
  68 $RI = [:LineBreak =  Regional_Indicator:];
  69 $SA = [:LineBreak =  Complex_Context:];
  70 $SG = [:LineBreak =  Surrogate:];
  71 $SP = [:LineBreak =  Space:];
  72 $SY = [:LineBreak =  Break_Symbols:];
  73 $WJ = [:LineBreak =  Word_Joiner:];
  74 $XX = [:LineBreak =  Unknown:];
  75 $ZW = [:LineBreak =  ZWSpace:];
  76 $ZWJ = [:LineBreak = ZWJ:];
  77
  78 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  79 #         list it in the numerous rules that use CM.
  80 # By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
  81
  82 $CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
  83 $CMX = [[$CM] - [$ZWJ]];
  84
  85 #   Dictionary character set, for triggering language-based break engines. Currently
  86 #   limited to LineBreak=Complex_Context (SA).
  87
  88 $dictionary = [$SA];
  89
  90 #
  91 #  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
  92 #                               SA  (Dictionary chars, excluding Mn and Mc)
  93 #                               SG  (Unpaired Surrogates)
  94 #                               XX  (Unknown, unassigned)
  95 #                         as $AL  (Alphabetic)
  96 #
  97 $ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
  98
  99
 100 ## -------------------------------------------------
 101
 102 #
 103 # CAN_CM  is the set of characters that may combine with CM combining chars.
 104 #         Note that Linebreak UAX 14's concept of a combining char and the rules
 105 #         for what they can combine with are _very_ different from the rest of Unicode.
 106 #
 107 #         Note that $CM itself is left out of this set.  If CM is needed as a base
 108 #         it must be listed separately in the rule.
 109 #
 110 $CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
 111 $CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
 112
 113 #
 114 # AL_FOLLOW  set of chars that can unconditionally follow an AL
 115 #            Needed in rules where stand-alone $CM s are treated as AL.
 116 #
 117 $AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP $QU $BA $HY $NS $IN $NU $PR $PO $ALPlus];
 118
 119
 120 #
 121 #  Rule LB 4, 5    Mandatory (Hard) breaks.
 122 #
 123 $LB4Breaks    = [$BK $CR $LF $NL];
 124 $LB4NonBreaks = [^$BK $CR $LF $NL $CM];
 125 $CR $LF {100};
 126
 127 #
 128 #  LB 6    Do not break before hard line breaks.
 129 #
 130 $LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
 131 $CAN_CM $CM*    $LB4Breaks {100};
 132 ^$CM+           $LB4Breaks {100};
 133
 134 # LB 7         x SP
 135 #              x ZW
 136 $LB4NonBreaks [$SP $ZW];
 137 $CAN_CM $CM*  [$SP $ZW];
 138 ^$CM+         [$SP $ZW];
 139
 140 #
 141 # LB 8         Break after zero width space
 142 #              ZW SP* ÷
 143 #
 144 $LB8Breaks    = [$LB4Breaks $ZW];
 145 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 146 $ZW $SP* / [^$SP $ZW $LB4Breaks];
 147
 148 # LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
 149 #
 150 $ZWJ [^$CM];
 151
 152 # LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
 153 #                                $CM not covered by the above needs to behave like $AL
 154 #                                See definition of $CAN_CM.
 155
 156 $CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
 157 ^$CM+;
 158
 159 #
 160 # LB 11  Do not break before or after WORD JOINER & related characters.
 161 #
 162 $CAN_CM $CM*  $WJ;
 163 $LB8NonBreaks $WJ;
 164 ^$CM+         $WJ;
 165
 166 $WJ $CM* .;
 167
 168 #
 169 # LB 12  Do not break after NBSP and related characters.
 170 #         GL  x
 171 #
 172 $GL $CM* .;
 173
 174 #
 175 # LB 12a  Do not break before NBSP and related characters ...
 176 #            [^SP BA HY] x GL
 177 #
 178 [[$LB8NonBreaks] - [$SP $BA $BAX $HY]] $CM* $GL;
 179 ^$CM+ $GL;
 180
 181
 182
 183
 184 # LB 13   Don't break before ']' or '!' or or '/', even after spaces.
 185 #
 186 $LB8NonBreaks $CL;
 187 $CAN_CM $CM*  $CL;
 188 ^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
 189
 190 $LB8NonBreaks $CP;
 191 $CAN_CM $CM*  $CP;
 192 ^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
 193
 194 $LB8NonBreaks $EX;
 195 $CAN_CM $CM*  $EX;
 196 ^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
 197
 198 $LB8NonBreaks $SY;
 199 $CAN_CM $CM*  $SY;
 200 ^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
 201
 202
 203 #
 204 # LB 14  Do not break after OP, even after spaces
 205 #        Note subtle interaction with "SP IS /" rules in LB14a.
 206 #        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
 207 #        which is the desired behavior.
 208 #
 209 $OP $CM* $SP* .;
 210
 211 $OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
 212                                    # by rule 8, CM following a SP is stand-alone.
 213
 214
 215 # LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
 216 #        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
 217 #        See issue ICU-20303
 218
 219
 220 $CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL];
 221 $SP $IS           / [^ $CanFollowIS $NU $CM];
 222 $SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
 223
 224 #
 225 # LB 14b Do not break before numeric separators (IS), even after spaces.
 226
 227 [$LB8NonBreaks - $SP] $IS;
 228 $SP $IS $CM* [$CanFollowIS {eof}];
 229 $SP $IS $CM* $ZWJ [^$CM $NU];
 230
 231 $CAN_CM $CM*  $IS;
 232 ^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
 233
 234
 235 # LB 15
 236 $QU $CM* $SP* $OP;
 237
 238 # LB 16
 239 # Do not break between closing punctuation and $NS, even with intervening spaces
 240 # But DO allow a break between closing punctuation and $NSX, don't include it here
 241 ($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 # DO allow breaks here before $BAX and $NSX, so don't include them
 277 $LB20NonBreaks $CM* ($BA | $HY | $NS);
 278
 279
 280 ^$CM+ ($BA | $HY | $NS);
 281
 282 $BB $CM* [^$CB];                                  #  $BB  x
 283 $BB $CM* $LB20NonBreaks;
 284
 285 # LB 21a Don't break after Hebrew + Hyphen
 286 #   HL (HY | BA) x
 287 #
 288 $HL $CM* ($HY | $BA | $BAX) $CM* [^$CB]?;
 289
 290 # LB 21b (forward) Don't break between SY and HL
 291 # (break between HL and SY already disallowed by LB 13 above)
 292 $SY $CM* $HL;
 293
 294 # LB 22
 295 ($ALPlus | $HL) $CM* $IN;
 296 ^$CM+    $IN;     #  by rule 10, any otherwise unattached CM behaves as AL
 297 $EX $CM*    $IN;
 298 ($ID | $EB | $EM) $CM*  $IN;
 299 $IN $CM*    $IN;
 300 $NU $CM*    $IN;
 301
 302
 303 # $LB 23
 304 #
 305 ($ALPlus | $HL) $CM* $NU;
 306 ^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
 307 $NU $CM* ($ALPlus | $HL);
 308
 309 # LB 23a
 310 #
 311 $PR $CM* ($ID | $EB | $EM);
 312 ($ID | $EB | $EM) $CM*  $PO;
 313
 314
 315 #
 316 # LB 24
 317 #
 318 ($PR | $PO) $CM* ($ALPlus | $HL);
 319 ($ALPlus | $HL) $CM* ($PR | $PO);
 320 ^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
 321
 322 #
 323 # LB 25   Numbers.
 324 #
 325 (($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
 326     ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
 327
 328 # LB 26  Do not break a Korean syllable
 329 #
 330 $JL $CM* ($JL | $JV | $H2 | $H3);
 331 ($JV | $H2) $CM* ($JV | $JT);
 332 ($JT | $H3) $CM* $JT;
 333
 334 # LB 27  Treat korean Syllable Block the same as ID  (don't break it)
 335 ($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
 336 ($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
 337 $PR $CM* ($JL | $JV | $JT | $H2 | $H3);
 338
 339
 340 # LB 28   Do not break between alphabetics
 341 #
 342 ($ALPlus | $HL) $CM* ($ALPlus | $HL);
 343 ^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
 344
 345 # LB 29
 346 $IS $CM* ($ALPlus | $HL);
 347
 348 # LB 30
 349 ($ALPlus | $HL | $NU) $CM* $OP;
 350 ^$CM+ $OP;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
 351 $CP $CM* ($ALPlus | $HL | $NU);
 352
 353 # LB 30a  Do not break between regional indicators. Break after pairs of them.
 354 #         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
 355 $RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 356 $RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 357 $RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $ZWJ {eof}];
 358 # note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
 359 #       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
 360 #       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
 361
 362 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 363 $EB $CM* $EM;
 364
 365 # LB 31 Break everywhere else.
 366 #       Match a single code point if no other rule applies.
 367 .;