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