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