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 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=loose (BCP47 -u-lb-loose) 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 #         * before iteration marks 3005, 303B, 309D, 309E, 30FD, 30FE (all NS)
  25 #         * between characters of LineBreak class IN such as 2026
  26 #         * before some centered punct 203C, 2047, 2048, 2049, 30FB, FF1A, FF1B,
  27 #           FF65 (all NS) and FF01, FF1F (both EX).
  28 #         * before suffix characters with LineBreak class PO and EastAsianWidth A,F,W;
  29 #           this includes: 00B0 2030 2032 2033 2035 2103 2109 FE6A FF05 FFE0
  30 #         * after prefix characters with LineBreak class PR and EastAsianWidth A,F,W;
  31 #           this includes: 00A4 00B1 20AC 2116 FE69 FF04 FFE1 FFE5 FFE6
  32
  33
  34 #
  35 #  Character Classes defined by TR 14.
  36 #
  37
  38 !!chain;
  39 !!quoted_literals_only;
  40
  41 $AI = [:LineBreak =  Ambiguous:];
  42 $AL = [:LineBreak =  Alphabetic:];
  43 $BAX = [\u2010 \u2013];
  44 $BA = [[:LineBreak =  Break_After:] - $BAX];
  45 $BB = [:LineBreak =  Break_Before:];
  46 $BK = [:LineBreak =  Mandatory_Break:];
  47 $B2 = [:LineBreak =  Break_Both:];
  48 $CB = [:LineBreak =  Contingent_Break:];
  49 $CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  50 $CL = [:LineBreak =  Close_Punctuation:];
  51 # $CM = [:LineBreak =  Combining_Mark:];
  52 $CP = [:LineBreak =  Close_Parenthesis:];
  53 $CR = [:LineBreak =  Carriage_Return:];
  54 $EB = [[:LineBreak =  EB:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
  55 $EM = [:LineBreak =  EM:];
  56 $EXX = [\uFF01 \uFF1F];
  57 $EX = [[:LineBreak =  Exclamation:] - $EXX];
  58 $GL = [:LineBreak =  Glue:];
  59 $HL = [:LineBreak =  Hebrew_Letter:];
  60 $HY = [:LineBreak =  Hyphen:];
  61 $H2 = [:LineBreak =  H2:];
  62 $H3 = [:LineBreak =  H3:];
  63 # CSS Loose tailoring: CJ resolves to ID
  64 $ID = [[:LineBreak =  Ideographic:] $CJ];
  65 $IN = [:LineBreak =  Inseperable:];
  66 $IS = [:LineBreak =  Infix_Numeric:];
  67 $JL = [:LineBreak =  JL:];
  68 $JV = [:LineBreak =  JV:];
  69 $JT = [:LineBreak =  JT:];
  70 $LF = [:LineBreak =  Line_Feed:];
  71 $NL = [:LineBreak =  Next_Line:];
  72 $NSX = [\u301C \u30A0 \u3005 \u303B \u309D \u309E \u30FD \u30FE \u203C \u2047 \u2048 \u2049 \u30FB \uFF1A \uFF1B \uFF65];
  73 $NS = [[:LineBreak =  Nonstarter:] - $NSX];
  74 $NU = [:LineBreak =  Numeric:];
  75 $OP = [:LineBreak =  Open_Punctuation:];
  76 $POX = [\u00B0 \u2030 \u2032 \u2033 \u2035 \u2103 \u2109 \uFE6A \uFF05 \uFFE0];
  77 $PO = [[:LineBreak =  Postfix_Numeric:] - $POX];
  78 $PRX = [\u00A4 \u00B1 \u20AC \u2116 \uFE69 \uFF04 \uFFE1 \uFFE5 \uFFE6];
  79 $PR = [[:LineBreak =  Prefix_Numeric:] - $PRX];
  80 $QU = [:LineBreak =  Quotation:];
  81 $RI = [:LineBreak =  Regional_Indicator:];
  82 $SA = [:LineBreak =  Complex_Context:];
  83 $SG = [:LineBreak =  Surrogate:];
  84 $SP = [:LineBreak =  Space:];
  85 $SY = [:LineBreak =  Break_Symbols:];
  86 $WJ = [:LineBreak =  Word_Joiner:];
  87 $XX = [:LineBreak =  Unknown:];
  88 $ZW = [:LineBreak =  ZWSpace:];
  89 $ZWJ = [:LineBreak = ZWJ:];
  90
  91 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  92 #         list it in the numerous rules that use CM.
  93 # By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
  94
  95 $CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
  96
  97 #   Dictionary character set, for triggering language-based break engines. Currently
  98 #   limited to LineBreak=Complex_Context (SA).
  99
 100 $dictionary = [$SA];
 101
 102 #
 103 #  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
 104 #                               SA  (Dictionary chars, excluding Mn and Mc)
 105 #                               SG  (Unpaired Surrogates)
 106 #                               XX  (Unknown, unassigned)
 107 #                         as $AL  (Alphabetic)
 108 #
 109 $ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
 110
 111
 112 ## -------------------------------------------------
 113
 114 #
 115 # CAN_CM  is the set of characters that may combine with CM combining chars.
 116 #         Note that Linebreak UAX 14's concept of a combining char and the rules
 117 #         for what they can combine with are _very_ different from the rest of Unicode.
 118 #
 119 #         Note that $CM itself is left out of this set.  If CM is needed as a base
 120 #         it must be listed separately in the rule.
 121 #
 122 $CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
 123 $CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
 124
 125 #
 126 # AL_FOLLOW  set of chars that can unconditionally follow an AL
 127 #            Needed in rules where stand-alone $CM s are treated as AL.
 128 #
 129 $AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP $QU $BA $HY $NS $IN $NU $PR $PO $POX $ALPlus];
 130
 131
 132 #
 133 #  Rule LB 4, 5    Mandatory (Hard) breaks.
 134 #
 135 $LB4Breaks    = [$BK $CR $LF $NL];
 136 $LB4NonBreaks = [^$BK $CR $LF $NL $CM];
 137 $CR $LF {100};
 138
 139 #
 140 #  LB 6    Do not break before hard line breaks.
 141 #
 142 $LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
 143 $CAN_CM $CM*    $LB4Breaks {100};
 144 ^$CM+           $LB4Breaks {100};
 145
 146 # LB 7         x SP
 147 #              x ZW
 148 $LB4NonBreaks [$SP $ZW];
 149 $CAN_CM $CM*  [$SP $ZW];
 150 ^$CM+         [$SP $ZW];
 151
 152 #
 153 # LB 8         Break after zero width space
 154 #              TODO:  ZW SP* <break>
 155 #              An engine change is required to write the reverse rule for this.
 156 #              For now, leave the Unicode 5.2 rule, ZW <break>
 157 #
 158 $LB8Breaks    = [$LB4Breaks $ZW];
 159 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 160
 161 # LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
 162 #
 163 $ZWJ [^$CM];
 164
 165 # LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
 166 #                                $CM not covered by the above needs to behave like $AL
 167 #                                See definition of $CAN_CM.
 168
 169 $CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
 170 ^$CM+;
 171
 172 #
 173 # LB 11  Do not break before or after WORD JOINER & related characters.
 174 #
 175 $CAN_CM $CM*  $WJ;
 176 $LB8NonBreaks $WJ;
 177 ^$CM+         $WJ;
 178
 179 $WJ $CM* .;
 180
 181 #
 182 # LB 12  Do not break after NBSP and related characters.
 183 #         GL  x
 184 #
 185 $GL $CM* .;
 186
 187 #
 188 # LB 12a  Do not break before NBSP and related characters ...
 189 #            [^SP BA HY] x GL
 190 #
 191 [[$LB8NonBreaks] - [$SP $BA $BAX $HY]] $CM* $GL;
 192 ^$CM+ $GL;
 193
 194
 195 #
 196 # LB 13   Don't break before ']' or '!' or ';' or '/', even after spaces.
 197 #
 198 # Do not include $EXX here
 199 $LB8NonBreaks $CL;
 200 $CAN_CM $CM*  $CL;
 201 ^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
 202
 203 $LB8NonBreaks $CP;
 204 $CAN_CM $CM*  $CP;
 205 ^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
 206
 207 $LB8NonBreaks $EX;
 208 $CAN_CM $CM*  $EX;
 209 ^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
 210
 211 $LB8NonBreaks $IS;
 212 $CAN_CM $CM*  $IS;
 213 ^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
 214
 215 $LB8NonBreaks $SY;
 216 $CAN_CM $CM*  $SY;
 217 ^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
 218
 219
 220 #
 221 # LB 14  Do not break after OP, even after spaces
 222 #
 223 $OP $CM* $SP* .;
 224
 225 $OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
 226                                    # by rule 8, CM following a SP is stand-alone.
 227
 228 # LB 15
 229 $QU $CM* $SP* $OP;
 230
 231 # LB 16
 232 # Do not break between closing punctuation and $NS, even with intervening spaces
 233 # But DO allow a break between closing punctuation and $NSX, don't include it here
 234 ($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
 255 # LB 20
 256 #        <break>  $CB
 257 #        $CB   <break>
 258
 259 $LB20NonBreaks = [$LB18NonBreaks - $CB];
 260
 261 # LB 21        x   (BA | HY | NS)
 262 #           BB x
 263 #
 264 # DO allow breaks here before $BAX and $NSX, so don't include them
 265 $LB20NonBreaks $CM* ($BA | $HY | $NS);
 266 ^$CM+ ($BA | $HY | $NS);
 267
 268 $BB $CM* [^$CB];                                  #  $BB  x
 269 $BB $CM* $LB20NonBreaks;
 270
 271 # LB 21a Don't break after Hebrew + Hyphen
 272 #   HL (HY | BA) x
 273 #
 274 $HL $CM* ($HY | $BA | $BAX) $CM* [^$CB]?;
 275
 276 # LB 21b (forward) Don't break between SY and HL
 277 # (break between HL and SY already disallowed by LB 13 above)
 278 $SY $CM* $HL;
 279
 280 # LB 22
 281 ($ALPlus | $HL) $CM* $IN;
 282 ^$CM+    $IN;     #  by rule 10, any otherwise unattached CM behaves as AL
 283 $EX $CM*    $IN;
 284 ($ID | $EB | $EM) $CM*  $IN;
 285 # $IN $CM*    $IN;  # delete this rule for CSS loose
 286 $NU $CM*    $IN;
 287
 288
 289 # $LB 23
 290 #
 291 ($ALPlus | $HL) $CM* $NU;
 292 ^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
 293 $NU $CM* ($ALPlus | $HL);
 294
 295 # LB 23a
 296 # Do not include $POX here
 297 #
 298 $PR $CM* ($ID | $EB | $EM);
 299 ($ID | $EB | $EM) $CM*  $PO;
 300
 301
 302 #
 303 # LB 24
 304 #
 305 # Do not include $PRX here
 306 ($PR | $PO | $POX) $CM* ($ALPlus | $HL);
 307 ($ALPlus | $HL) $CM* ($PR | $PO | $POX);     # TODO: should this be ($PR | $PRX | $PO)
 308 ^$CM+ ($PR | $PO | $POX);       # Rule 10, any otherwise unattached CM behaves as AL
 309
 310 #
 311 # LB 25   Numbers.
 312 #
 313 # Here do not include $PRX at the beginning or $POX at the end
 314 (($PR | $PO | $POX) $CM*)? (($OP | $HY) $CM*)? $NU ($CM* ($NU | $SY | $IS))*
 315     ($CM* ($CL | $CP))? ($CM* ($PR | $PRX | $PO))?;
 316
 317 # LB 26  Do not break a Korean syllable
 318 #
 319 $JL $CM* ($JL | $JV | $H2 | $H3);
 320 ($JV | $H2) $CM* ($JV | $JT);
 321 ($JT | $H3) $CM* $JT;
 322
 323 # LB 27  Treat korean Syllable Block the same as ID  (don't break it)
 324 # Do not include $POX or $PRX here
 325 ($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
 326 ($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
 327 $PR $CM* ($JL | $JV | $JT | $H2 | $H3);
 328
 329
 330 # LB 28   Do not break between alphabetics
 331 #
 332 ($ALPlus | $HL) $CM* ($ALPlus | $HL);
 333 ^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
 334
 335 # LB 29
 336 $IS $CM* ($ALPlus | $HL);
 337
 338 # LB 30
 339 ($ALPlus | $HL | $NU) $CM* $OP;
 340 ^$CM+ $OP;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
 341 $CP $CM* ($ALPlus | $HL | $NU);
 342
 343 # LB 30a  Do not break between regional indicators. Break after pairs of them.
 344 #         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
 345 $RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 346 $RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 347 $RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $ZWJ {eof}];
 348 # note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
 349 #       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
 350 #       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
 351
 352 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 353 $EB $CM* $EM;
 354
 355 # LB 31 Break everywhere else.
 356 #       Match a single code point if no other rule applies.
 357 .;