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 40 for Unicode 11.0
  11 #         http://www.unicode.org/reports/tr14/
  12 #
  13 #         tailored as noted in 2nd paragraph below.
  14 #
  15 #         TODO:  Rule LB 8 remains as it was in Unicode 5.2
  16 #         This is only because of a limitation of ICU break engine implementation,
  17 #         not because the older behavior is desirable.
  18 #
  19 #         This tailors the line break behavior to correspond to CSS
  20 #         line-break=normal (BCP47 -u-lb-normal) as defined for Chinese & Japanese.
  21 #         It sets characters of class CJ to behave like ID.
  22 #         In addition, it allows breaks:
  23 #         * before hyphens 2010 & 2013 (both BA) and 301C, 30A0 (both NS)
  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 $BAX = [\u2010 \u2013];
  35 $BA = [[:LineBreak =  Break_After:] - $BAX];
  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:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
  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 Normal 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 = [\u301C \u30A0];
  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 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  80 #         list it in the numerous rules that use CM.
  81 # By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
  82
  83 $CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
  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 #              TODO:  ZW SP* <break>
 143 #              An engine change is required to write the reverse rule for this.
 144 #              For now, leave the Unicode 5.2 rule, ZW <break>
 145 #
 146 $LB8Breaks    = [$LB4Breaks $ZW];
 147 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 148
 149 # LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
 150 #
 151 $ZWJ [^$CM];
 152
 153 # LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
 154 #                                $CM not covered by the above needs to behave like $AL
 155 #                                See definition of $CAN_CM.
 156
 157 $CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
 158 ^$CM+;
 159
 160 #
 161 # LB 11  Do not break before or after WORD JOINER & related characters.
 162 #
 163 $CAN_CM $CM*  $WJ;
 164 $LB8NonBreaks $WJ;
 165 ^$CM+         $WJ;
 166
 167 $WJ $CM* .;
 168
 169 #
 170 # LB 12  Do not break after NBSP and related characters.
 171 #         GL  x
 172 #
 173 $GL $CM* .;
 174
 175 #
 176 # LB 12a  Do not break before NBSP and related characters ...
 177 #            [^SP BA HY] x GL
 178 #
 179 [[$LB8NonBreaks] - [$SP $BA $BAX $HY]] $CM* $GL;
 180 ^$CM+ $GL;
 181
 182
 183
 184 #
 185 # LB 13   Don't break before ']' or '!' or ';' or '/', even after spaces.
 186 #
 187 $LB8NonBreaks $CL;
 188 $CAN_CM $CM*  $CL;
 189 ^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
 190
 191 $LB8NonBreaks $CP;
 192 $CAN_CM $CM*  $CP;
 193 ^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
 194
 195 $LB8NonBreaks $EX;
 196 $CAN_CM $CM*  $EX;
 197 ^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
 198
 199 $LB8NonBreaks $IS;
 200 $CAN_CM $CM*  $IS;
 201 ^$CM+         $IS;              # 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 #
 211 $OP $CM* $SP* .;
 212
 213 $OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
 214                                    # by rule 8, CM following a SP is stand-alone.
 215
 216 # LB 15
 217 $QU $CM* $SP* $OP;
 218
 219 # LB 16
 220 # Do not break between closing punctuation and $NS, even with intervening spaces
 221 # But DO allow a break between closing punctuation and $NSX, don't include it here
 222 ($CL | $CP) $CM* $SP* $NS;
 223
 224 # LB 17
 225 $B2 $CM* $SP* $B2;
 226
 227 #
 228 # LB 18  Break after spaces.
 229 #
 230 $LB18NonBreaks = [$LB8NonBreaks - [$SP]];
 231 $LB18Breaks    = [$LB8Breaks $SP];
 232
 233
 234 # LB 19
 235 #         x QU
 236 $LB18NonBreaks $CM* $QU;
 237 ^$CM+               $QU;
 238
 239 #         QU  x
 240 $QU $CM* .;
 241
 242
 243 # LB 20
 244 #        <break>  $CB
 245 #        $CB   <break>
 246
 247 $LB20NonBreaks = [$LB18NonBreaks - $CB];
 248
 249 # LB 21        x   (BA | HY | NS)
 250 #           BB x
 251 #
 252 # DO allow breaks here before $BAXcm and $NSXcm, so don't include them
 253 $LB20NonBreaks $CM* ($BA | $HY | $NS);
 254 ^$CM+ ($BA | $HY | $NS);
 255
 256 $BB $CM* [^$CB];                                  #  $BB  x
 257 $BB $CM* $LB20NonBreaks;
 258
 259 # LB 21a Don't break after Hebrew + Hyphen
 260 #   HL (HY | BA) x
 261 #
 262 $HL $CM* ($HY | $BA | $BAX) $CM* [^$CB]?;
 263
 264 # LB 21b (forward) Don't break between SY and HL
 265 # (break between HL and SY already disallowed by LB 13 above)
 266 $SY $CM* $HL;
 267
 268 # LB 22
 269 ($ALPlus | $HL) $CM* $IN;
 270 ^$CM+    $IN;     #  by rule 10, any otherwise unattached CM behaves as AL
 271 $EX $CM*    $IN;
 272 ($ID | $EB | $EM) $CM*  $IN;
 273 $IN $CM*    $IN;
 274 $NU $CM*    $IN;
 275
 276
 277 # $LB 23
 278 #
 279 ($ALPlus | $HL) $CM* $NU;
 280 ^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
 281 $NU $CM* ($ALPlus | $HL);
 282
 283 # LB 23a
 284 #
 285 $PR $CM* ($ID | $EB | $EM);
 286 ($ID | $EB | $EM) $CM*  $PO;
 287
 288
 289 #
 290 # LB 24
 291 #
 292 ($PR | $PO) $CM* ($ALPlus | $HL);
 293 ($ALPlus | $HL) $CM* ($PR | $PO);
 294 ^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
 295
 296 #
 297 # LB 25   Numbers.
 298 #
 299 (($PR | $PO) $CM*)? (($OP | $HY) $CM*)? $NU ($CM* ($NU | $SY | $IS))*
 300     ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
 301
 302 # LB 26  Do not break a Korean syllable
 303 #
 304 $JL $CM* ($JL | $JV | $H2 | $H3);
 305 ($JV | $H2) $CM* ($JV | $JT);
 306 ($JT | $H3) $CM* $JT;
 307
 308 # LB 27  Treat korean Syllable Block the same as ID  (don't break it)
 309 ($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
 310 ($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
 311 $PR $CM* ($JL | $JV | $JT | $H2 | $H3);
 312
 313
 314 # LB 28   Do not break between alphabetics
 315 #
 316 ($ALPlus | $HL) $CM* ($ALPlus | $HL);
 317 ^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
 318
 319 # LB 29
 320 $IS $CM* ($ALPlus | $HL);
 321
 322 # LB 30
 323 ($ALPlus | $HL | $NU) $CM* $OP;
 324 ^$CM+ $OP;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
 325 $CP $CM* ($ALPlus | $HL | $NU);
 326
 327 # LB 30a  Do not break between regional indicators. Break after pairs of them.
 328 #         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
 329 $RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 330 $RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 331 $RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $ZWJ {eof}];
 332 # note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
 333 #       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
 334 #       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
 335
 336 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 337 $EB $CM* $EM;
 338
 339 # LB 31 Break everywhere else.
 340 #       Match a single code point if no other rule applies.
 341 .;