icuSources/data/brkitr/rules/line_loose.txt

   1 # Copyright (C) 2016 and later: Unicode, Inc. and others.
   2 # License & terms of use: http://www.unicode.org/copyright.html
   3 #
   4 # Copyright (c) 2002-2016  International Business Machines Corporation and
   5 # others. All Rights Reserved.
   6 #
   7 #  file:  line_loose.txt
   8 #
   9 #         Line Breaking Rules
  10 #         Implement default line breaking as defined by
  11 #         Unicode Standard Annex #14 Revision 42 for Unicode 12.0
  12 #         http://www.unicode.org/reports/tr14/, with the following modification:
  13 #
  14 #         Boundaries between hyphens and following letters are suppressed when
  15 #         there is a boundary preceding the hyphen. See rule 20.9
  16 #
  17 #         This tailors the line break behavior to correspond to CSS
  18 #         line-break=loose (BCP47 -u-lb-loose) as defined for languages other than
  19 #         Chinese & Japanese.
  20 #         It sets characters of class CJ to behave like ID.
  21 #         In addition, it allows breaks:
  22 #         * before iteration marks 3005, 303B, 309D, 309E, 30FD, 30FE (all NS)
  23 #         * between characters of LineBreak class IN
  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 $BA = [:LineBreak =  Break_After:];
  35 $HH = [\u2010];     # \u2010 is HYPHEN, default line break is BA.
  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:];
  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 Loose 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 = [\u3005 \u303B \u309D \u309E \u30FD \u30FE];
  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 $CMX = [[$CM] - [$ZWJ]];
  85
  86 #   Dictionary character set, for triggering language-based break engines. Currently
  87 #   limited to LineBreak=Complex_Context (SA).
  88
  89 $dictionary = [$SA];
  90
  91 #
  92 #  Rule LB1.  By default, treat AI  (characters with ambiguous east Asian width),
  93 #                               SA  (Dictionary chars, excluding Mn and Mc)
  94 #                               SG  (Unpaired Surrogates)
  95 #                               XX  (Unknown, unassigned)
  96 #                         as $AL  (Alphabetic)
  97 #
  98 $ALPlus = [$AL $AI $SG $XX [$SA-[[:Mn:][:Mc:]]]];
  99
 100
 101 ## -------------------------------------------------
 102
 103 #
 104 # CAN_CM  is the set of characters that may combine with CM combining chars.
 105 #         Note that Linebreak UAX 14's concept of a combining char and the rules
 106 #         for what they can combine with are _very_ different from the rest of Unicode.
 107 #
 108 #         Note that $CM itself is left out of this set.  If CM is needed as a base
 109 #         it must be listed separately in the rule.
 110 #
 111 $CAN_CM  = [^$SP $BK $CR $LF $NL $ZW $CM];       # Bases that can   take CMs
 112 $CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM];       # Bases that can't take CMs
 113
 114 #
 115 # AL_FOLLOW  set of chars that can unconditionally follow an AL
 116 #            Needed in rules where stand-alone $CM s are treated as AL.
 117 #
 118 $AL_FOLLOW      = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP $QU $BA $HY $NS $IN $NU $PR $PO $ALPlus];
 119
 120
 121 #
 122 #  Rule LB 4, 5    Mandatory (Hard) breaks.
 123 #
 124 $LB4Breaks    = [$BK $CR $LF $NL];
 125 $LB4NonBreaks = [^$BK $CR $LF $NL $CM];
 126 $CR $LF {100};
 127
 128 #
 129 #  LB 6    Do not break before hard line breaks.
 130 #
 131 $LB4NonBreaks?  $LB4Breaks {100};    # LB 5  do not break before hard breaks.
 132 $CAN_CM $CM*    $LB4Breaks {100};
 133 ^$CM+           $LB4Breaks {100};
 134
 135 # LB 7         x SP
 136 #              x ZW
 137 $LB4NonBreaks [$SP $ZW];
 138 $CAN_CM $CM*  [$SP $ZW];
 139 ^$CM+         [$SP $ZW];
 140
 141 #
 142 # LB 8         Break after zero width space
 143 #              ZW SP* ÷
 144 #
 145 $LB8Breaks    = [$LB4Breaks $ZW];
 146 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 147 $ZW $SP* / [^$SP $ZW $LB4Breaks];
 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 $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 $SY;
 200 $CAN_CM $CM*  $SY;
 201 ^$CM+         $SY;              # by rule 10, stand-alone CM behaves as AL
 202
 203
 204 #
 205 # LB 14  Do not break after OP, even after spaces
 206 #        Note subtle interaction with "SP IS /" rules in LB14a.
 207 #        This rule consumes the SP, chaining happens on the IS, effectivley overriding the  SP IS rules,
 208 #        which is the desired behavior.
 209 #
 210 $OP $CM* $SP* .;
 211
 212 $OP $CM* $SP+ $CM+ $AL_FOLLOW?;    # by rule 10, stand-alone CM behaves as AL
 213                                    # by rule 8, CM following a SP is stand-alone.
 214
 215
 216 # LB 14a Force a break before start of a number with a leading decimal pt, e.g. " .23"
 217 #        Note: would be simpler to express as "$SP / $IS $CM* $NU;", but ICU rules have limitations.
 218 #        See issue ICU-20303
 219
 220
 221 $CanFollowIS = [$BK $CR $LF $NL $SP $ZW $WJ $GL $CL $CP $EX $IS $SY $QU $BA $HY $NS $ALPlus $HL];
 222 $SP $IS           / [^ $CanFollowIS $NU $CM];
 223 $SP $IS $CM* $CMX / [^ $CanFollowIS $NU $CM];
 224
 225 #
 226 # LB 14b Do not break before numeric separators (IS), even after spaces.
 227
 228 [$LB8NonBreaks - $SP] $IS;
 229 $SP $IS $CM* [$CanFollowIS {eof}];
 230 $SP $IS $CM* $ZWJ [^$CM $NU];
 231
 232 $CAN_CM $CM*  $IS;
 233 ^$CM+         $IS;              # by rule 10, stand-alone CM behaves as AL
 234
 235
 236 # LB 15
 237 $QU $CM* $SP* $OP;
 238
 239 # LB 16
 240 # Do not break between closing punctuation and $NS, even with intervening spaces
 241 # But DO allow a break between closing punctuation and $NSX, don't include it here
 242 ($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 # DO allow breaks here before NSX, so don't include it
 278 $LB20NonBreaks $CM* ($BA | $HY | $NS);
 279
 280
 281 ^$CM+ ($BA | $HY | $NS);
 282
 283 $BB $CM* [^$CB];                                  #  $BB  x
 284 $BB $CM* $LB20NonBreaks;
 285
 286 # LB 21a Don't break after Hebrew + Hyphen
 287 #   HL (HY | BA) x
 288 #
 289 $HL $CM* ($HY | $BA) $CM* [^$CB]?;
 290
 291 # LB 21b (forward) Don't break between SY and HL
 292 # (break between HL and SY already disallowed by LB 13 above)
 293 $SY $CM* $HL;
 294
 295 # LB 22
 296 ($ALPlus | $HL) $CM* $IN;
 297 ^$CM+    $IN;     #  by rule 10, any otherwise unattached CM behaves as AL
 298 $EX $CM*    $IN;
 299 ($ID | $EB | $EM) $CM*  $IN;
 300 # $IN $CM*    $IN;  # delete this rule for CSS loose
 301 $NU $CM*    $IN;
 302
 303
 304 # $LB 23
 305 #
 306 ($ALPlus | $HL) $CM* $NU;
 307 ^$CM+  $NU;       # Rule 10, any otherwise unattached CM behaves as AL
 308 $NU $CM* ($ALPlus | $HL);
 309
 310 # LB 23a
 311 #
 312 $PR $CM* ($ID | $EB | $EM);
 313 ($ID | $EB | $EM) $CM*  $PO;
 314
 315
 316 #
 317 # LB 24
 318 #
 319 ($PR | $PO) $CM* ($ALPlus | $HL);
 320 ($ALPlus | $HL) $CM* ($PR | $PO);
 321 ^$CM+ ($PR | $PO);       # Rule 10, any otherwise unattached CM behaves as AL
 322
 323 #
 324 # LB 25   Numbers.
 325 #
 326 (($PR | $PO) $CM*)? (($OP | $HY) $CM*)? ($IS $CM*)? $NU ($CM* ($NU | $SY | $IS))*
 327     ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?;
 328
 329 # LB 26  Do not break a Korean syllable
 330 #
 331 $JL $CM* ($JL | $JV | $H2 | $H3);
 332 ($JV | $H2) $CM* ($JV | $JT);
 333 ($JT | $H3) $CM* $JT;
 334
 335 # LB 27  Treat korean Syllable Block the same as ID  (don't break it)
 336 ($JL | $JV | $JT | $H2 | $H3) $CM* $IN;
 337 ($JL | $JV | $JT | $H2 | $H3) $CM* $PO;
 338 $PR $CM* ($JL | $JV | $JT | $H2 | $H3);
 339
 340
 341 # LB 28   Do not break between alphabetics
 342 #
 343 ($ALPlus | $HL) $CM* ($ALPlus | $HL);
 344 ^$CM+ ($ALPlus | $HL);      # The $CM+ is from rule 10, an unattached CM is treated as AL
 345
 346 # LB 29
 347 $IS $CM* ($ALPlus | $HL);
 348
 349 # LB 30
 350 ($ALPlus | $HL | $NU) $CM* $OP;
 351 ^$CM+ $OP;         # The $CM+ is from rule 10, an unattached CM is treated as AL.
 352 $CP $CM* ($ALPlus | $HL | $NU);
 353
 354 # LB 30a  Do not break between regional indicators. Break after pairs of them.
 355 #         Tricky interaction with LB8a: ZWJ x .   together with ZWJ acting like a CM.
 356 $RI $CM* $RI                 / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 357 $RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM]];
 358 $RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $ZWJ {eof}];
 359 # note: the preceding rule includes {eof} rather than having the last [set] term qualified with '?'
 360 #       because of the chain-out behavior difference. The rule must chain out only from the [set characters],
 361 #       not from the preceding $RI or $CM, which it would be able to do if the set were optional.
 362
 363 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 364 $EB $CM* $EM;
 365
 366 # LB 31 Break everywhere else.
 367 #       Match a single code point if no other rule applies.
 368 .;