icuSources/data/brkitr/rules/line.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.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
  19 #
  20 #  Character Classes defined by TR 14.
  21 #
  22
  23 !!chain;
  24 !!quoted_literals_only;
  25
  26 $AI = [:LineBreak =  Ambiguous:];
  27 $AL = [:LineBreak =  Alphabetic:];
  28 $BA = [:LineBreak =  Break_After:];
  29 $HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
  30 $BB = [:LineBreak =  Break_Before:];
  31 $BK = [:LineBreak =  Mandatory_Break:];
  32 $B2 = [:LineBreak =  Break_Both:];
  33 $CB = [:LineBreak =  Contingent_Break:];
  34 $CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  35 $CL = [:LineBreak =  Close_Punctuation:];
  36 # $CM = [:LineBreak =  Combining_Mark:];
  37 $CP = [:LineBreak =  Close_Parenthesis:];
  38 $CR = [:LineBreak =  Carriage_Return:];
  39 $EB = [:LineBreak =  EB:];
  40 $EM = [:LineBreak =  EM:];
  41 $EX = [:LineBreak =  Exclamation:];
  42 $GL = [:LineBreak =  Glue:];
  43 $HL = [:LineBreak =  Hebrew_Letter:];
  44 $HY = [:LineBreak =  Hyphen:];
  45 $H2 = [:LineBreak =  H2:];
  46 $H3 = [:LineBreak =  H3:];
  47 $ID = [:LineBreak =  Ideographic:];
  48 $IN = [:LineBreak =  Inseperable:];
  49 $IS = [:LineBreak =  Infix_Numeric:];
  50 $JL = [:LineBreak =  JL:];
  51 $JV = [:LineBreak =  JV:];
  52 $JT = [:LineBreak =  JT:];
  53 $LF = [:LineBreak =  Line_Feed:];
  54 $NL = [:LineBreak =  Next_Line:];
  55 # NS includes CJ for CSS strict line breaking.
  56 $NS = [[:LineBreak =  Nonstarter:] $CJ];
  57 $NU = [:LineBreak =  Numeric:];
  58 $OP = [:LineBreak =  Open_Punctuation:];
  59 $PO = [:LineBreak =  Postfix_Numeric:];
  60 $PR = [:LineBreak =  Prefix_Numeric:];
  61 $QU = [:LineBreak =  Quotation:];
  62 $RI = [:LineBreak =  Regional_Indicator:];
  63 $SA = [:LineBreak =  Complex_Context:];
  64 $SG = [:LineBreak =  Surrogate:];
  65 $SP = [:LineBreak =  Space:];
  66 $SY = [:LineBreak =  Break_Symbols:];
  67 $WJ = [:LineBreak =  Word_Joiner:];
  68 $XX = [:LineBreak =  Unknown:];
  69 $ZW = [:LineBreak =  ZWSpace:];
  70 $ZWJ = [:LineBreak = ZWJ:];
  71
  72 # The following classes are no longer needed for ICU rules but may
  73 # still be needed for WebKit rules in lineClasses.txt
  74 #  $EmojiNRK = [[\p{Emoji}] - [$RI \u002a\u00230-9©®™〰〽]];
  75 #  $Extended_Pict = [:ExtPict:];
  76
  77 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  78 #         list it in the numerous rules that use CM.
  79 # By LB1, SA characters with general categor of Mn or Mc also resolve to CM.
  80
  81 $CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]];
  82 $CMX = [[$CM] - [$ZWJ]];
  83
  84 #   Dictionary character set, for triggering language-based break engines. Currently
  85 #   limited to LineBreak=Complex_Context (SA).
  86
  87 $dictionary = [$SA];
  88
  89 #
  90 #  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
  91 #                               SA  (Dictionary chars, excluding Mn and Mc)
  92 #                               SG  (Unpaired Surrogates)
  93 #                               XX  (Unknown, unassigned)
  94 #                         as $AL  (Alphabetic)
  95 #
  96 $ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
  97
  98
  99 ## -------------------------------------------------
 100
 101 #
 102 # CAN_CM  is the set of characters that may combine with CM combining chars.
 103 #         Note that Linebreak UAX 14's concept of a combining char and the rules
 104 #         for what they can combine with are _very_ different from the rest of Unicode.
 105 #
 106 #         Note that $CM itself is left out of this set.  If CM is needed as a base
 107 #         it must be listed separately in the rule.
 108 #
 109 $CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
 110 $CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
 111
 112 #
 113 # AL_FOLLOW  set of chars that can unconditionally follow an AL
 114 #            Needed in rules where stand-alone $CM s are treated as AL.
 115 #
 116 $AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP $QU $BA $HY $NS $IN $NU $PR $PO $ALPlus];
 117
 118
 119 #
 120 #  Rule LB 4, 5    Mandatory (Hard) breaks.
 121 #
 122 $LB4Breaks    = [$BK $CR $LF $NL];
 123 $LB4NonBreaks = [^$BK $CR $LF $NL $CM];
 124 $CR $LF {100};
 125
 126 #
 127 #  LB 6    Do not break before hard line breaks.
 128 #
 129 $LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
 130 $CAN_CM $CM*    $LB4Breaks {100};
 131 ^$CM+           $LB4Breaks {100};
 132
 133 # LB 7         x SP
 134 #              x ZW
 135 $LB4NonBreaks [$SP $ZW];
 136 $CAN_CM $CM*  [$SP $ZW];
 137 ^$CM+         [$SP $ZW];
 138
 139 #
 140 # LB 8         Break after zero width space
 141 #              ZW SP* ÷
 142 #
 143 $LB8Breaks    = [$LB4Breaks $ZW];
 144 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 145 $ZW $SP* / [^$SP $ZW $LB4Breaks];
 146
 147 # LB 8a        ZWJ x            Do not break Emoji ZWJ sequences.
 148 #
 149 $ZWJ [^$CM];
 150
 151 # LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL
 152 #                                $CM not covered by the above needs to behave like $AL
 153 #                                See definition of $CAN_CM.
 154
 155 $CAN_CM $CM+;                   #  Stick together any combining sequences that don't match other rules.
 156 ^$CM+;
 157
 158 #
 159 # LB 11  Do not break before or after WORD JOINER & related characters.
 160 #
 161 $CAN_CM $CM*  $WJ;
 162 $LB8NonBreaks $WJ;
 163 ^$CM+         $WJ;
 164
 165 $WJ $CM* .;
 166
 167 #
 168 # LB 12  Do not break after NBSP and related characters.
 169 #         GL  x
 170 #
 171 $GL $CM* .;
 172
 173 #
 174 # LB 12a  Do not break before NBSP and related characters ...
 175 #            [^SP BA HY] x GL
 176 #
 177 [[$LB8NonBreaks] - [$SP $BA $HY]] $CM* $GL;
 178 ^$CM+ $GL;
 179
 180
 181
 182
 183 # LB 13   Don't break before ']' or '!' or or '/', even after spaces.
 184 #
 185 $LB8NonBreaks $CL;
 186 $CAN_CM $CM*  $CL;
 187 ^$CM+         $CL;              # by rule 10, stand-alone CM behaves as AL
 188
 189 $LB8NonBreaks $CP;
 190 $CAN_CM $CM*  $CP;
 191 ^$CM+         $CP;              # by rule 10, stand-alone CM behaves as AL
 192
 193 $LB8NonBreaks $EX;
 194 $CAN_CM $CM*  $EX;
 195 ^$CM+         $EX;              # by rule 10, stand-alone CM behaves as AL
 196
 197 $LB8NonBreaks $SY;
 198 $CAN_CM $CM*  $SY;
 199 ^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
 200
 201
 202 #
 203 # LB 14  Do not break after OP, even after spaces
 204 #        Note subtle interaction with "SP IS /" rules in LB14a.
 205 #        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
 206 #        which is the desired behavior.
 207 #
 208 $OP $CM* $SP* .;
 209
 210 $OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
 211                                    # by rule 8, CM following a SP is stand-alone.
 212
 213
 214 # LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
 215 #        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
 216 #        See issue ICU-20303
 217
 218
 219 $CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL];
 220 $SP $IS           / [^ $CanFollowIS $NU $CM];
 221 $SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
 222
 223 #
 224 # LB 14b Do not break before numeric separators (IS), even after spaces.
 225
 226 [$LB8NonBreaks - $SP] $IS;
 227 $SP $IS $CM* [$CanFollowIS {eof}];
 228 $SP $IS $CM* $ZWJ [^$CM $NU];
 229
 230 $CAN_CM $CM*  $IS;
 231 ^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
 232
 233
 234 # LB 15
 235 $QU $CM* $SP* $OP;
 236
 237 # LB 16
 238 ($CL | $CP) $CM* $SP* $NS;
 239
 240 # LB 17
 241 $B2 $CM* $SP* $B2;
 242
 243 #
 244 # LB 18  Break after spaces.
 245 #
 246 $LB18NonBreaks = [$LB8NonBreaks - [$SP]];
 247 $LB18Breaks    = [$LB8Breaks $SP];
 248
 249
 250 # LB 19
 251 #         x QU
 252 $LB18NonBreaks $CM* $QU;
 253 ^$CM+               $QU;
 254
 255 #         QU  x
 256 $QU $CM* .;
 257
 258 # LB 20
 259 #        <break>  $CB
 260 #        $CB   <break>
 261 #
 262 $LB20NonBreaks = [$LB18NonBreaks - $CB];
 263
 264 # LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
 265 #             Originally added as a Finnish tailoring, now promoted to default ICU behavior.
 266 #             Note: this is not default UAX-14 behaviour. See issue ICU-8151.
 267 #
 268 ^($HY | $HH) $CM* $ALPlus;
 269
 270 # LB 21        x   (BA | HY | NS)
 271 #           BB x
 272 #
 273 $LB20NonBreaks $CM* ($BA | $HY | $NS);
 274
 275
 276 ^$CM+ ($BA | $HY | $NS);
 277
 278 $BB $CM* [^$CB];                                  #  $BB  x
 279 $BB $CM* $LB20NonBreaks;
 280
 281 # LB 21a Don't break after Hebrew + Hyphen
 282 #   HL (HY | BA) x
 283 #
 284 $HL $CM* ($HY | $BA) $CM* [^$CB]?;
 285
 286 # LB 21b (forward) Don't break between SY and HL
 287 # (break between HL and SY already disallowed by LB 13 above)
 288 $SY $CM* $HL;
 289
 290 # LB 22
 291 ($ALPlus | $HL) $CM* $IN;
 292 ^$CM+    $IN;     #  by rule 10, any otherwise unattached CM behaves as AL
 293 $EX $CM*    $IN;
 294 ($ID | $EB | $EM) $CM*  $IN;
 295 $IN $CM*    $IN;
 296 $NU $CM*    $IN;
 297
 298
 299 # $LB 23
 300 #
 301 ($ALPlus | $HL) $CM* $NU;
 302 ^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
 303 $NU $CM* ($ALPlus | $HL);
 304
 305 # LB 23a
 306 #
 307 $PR $CM* ($ID | $EB | $EM);
 308 ($ID | $EB | $EM) $CM*  $PO;
 309
 310
 311 #
 312 # LB 24
 313 #
 314 ($PR | $PO) $CM* ($ALPlus | $HL);
 315 ($ALPlus | $HL) $CM* ($PR | $PO);
 316 ^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
 317
 318 #
 319 # LB 25   Numbers.
 320 #
 321 (($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
 322     ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
 323
 324 # LB 26  Do not break a Korean syllable
 325 #
 326 $JL $CM* ($JL | $JV | $H2 | $H3);
 327 ($JV | $H2) $CM* ($JV | $JT);
 328 ($JT | $H3) $CM* $JT;
 329
 330 # LB 27  Treat korean Syllable Block the same as ID  (don't break it)
 331 ($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
 332 ($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
 333 $PR $CM* ($JL | $JV | $JT | $H2 | $H3);
 334
 335
 336 # LB 28   Do not break between alphabetics
 337 #
 338 ($ALPlus | $HL) $CM* ($ALPlus | $HL);
 339 ^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
 340
 341 # LB 29
 342 $IS $CM* ($ALPlus | $HL);
 343
 344 # LB 30
 345 ($ALPlus | $HL | $NU) $CM* $OP;
 346 ^$CM+ $OP;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
 347 $CP $CM* ($ALPlus | $HL | $NU);
 348
 349 # LB 30a  Do not break between regional indicators. Break after pairs of them.
 350 #         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
 351 $RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 352 $RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 353 $RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $ZWJ {eof}];
 354 # note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
 355 #       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
 356 #       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
 357
 358 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 359 $EB $CM* $EM;
 360
 361 # LB 31 Break everywhere else.
 362 #       Match a single code point if no other rule applies.
 363 .;