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_fi.txt 7# 8# Line Breaking Rules 9# Implement default line breaking as defined by 10# Unicode Standard Annex #14 Revision 37 for Unicode 9.0 11# http://www.unicode.org/reports/tr14/ 12# tailored as noted in 3rd paragraph below. 13# 14# Includes extensions to the handling of emoji ZWJ sequences from 15# https://goo.gl/cluFCn 16# 17# TODO: Rule LB 8 remains as it was in Unicode 5.2 18# This is only because of a limitation of ICU break engine implementation, 19# not because the older behavior is desirable. 20# 21# This tailors the line break behavior for Finnish, and to correspond to CSS 22# line-break=normal (BCP47 -u-lb-normal) as defined for languages other than 23# Chinese & Japanese. 24# It sets characters of class CJ to behave like ID. 25 26# 27# Character Classes defined by TR 14. 28# 29 30!!chain; 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:]\U0001F3C2\U0001F3C7\U0001F3CC\U0001F46A-\U0001F46D\U0001F46F\U0001F574\U0001F6CC]; 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$H3 = [:LineBreak = H3:]; 53# CSS Normal tailoring: CJ resolves to ID 54$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$NS = [:LineBreak = Nonstarter:]; 63$NU = [:LineBreak = Numeric:]; 64$OP = [:LineBreak = Open_Punctuation:]; 65$PO = [:LineBreak = Postfix_Numeric:]; 66$PR = [:LineBreak = Prefix_Numeric:]; 67$QU = [:LineBreak = Quotation:]; 68$RI = [:LineBreak = Regional_Indicator:]; 69$SA = [:LineBreak = Complex_Context:]; 70$SG = [:LineBreak = Surrogate:]; 71$SP = [:LineBreak = Space:]; 72$SY = [:LineBreak = Break_Symbols:]; 73$WJ = [:LineBreak = Word_Joiner:]; 74$XX = [:LineBreak = Unknown:]; 75$ZW = [:LineBreak = ZWSpace:]; 76$ZWJ = [:LineBreak = ZWJ:]; 77 78$EmojiNRK = [[\p{Emoji}] - [$RI \u002a\u00230-9©®™〰〽]]; 79# Data for Extended Pictographic scraped from CLDR common/properties/ExtendedPictographic.txt, r12773 80$Extended_Pict = [\U0001F774-\U0001F77F\u2700-\u2701\u2703-\u2704\u270E\u2710-\u2711\u2765-\u2767\U0001F030-\U0001F093\U0001F094-\U0001F09F\U0001F10D-\U0001F10F\U0001F12F\U0001F16C-\U0001F16F\U0001F1AD-\U0001F1E5\U0001F203-\U0001F20F\U0001F23C-\U0001F23F\U0001F249-\U0001F24F\U0001F252-\U0001F2FF\U0001F7D5-\U0001F7FF\U0001F000-\U0001F003\U0001F005-\U0001F02B\U0001F02C-\U0001F02F\U0001F322-\U0001F323\U0001F394-\U0001F395\U0001F398\U0001F39C-\U0001F39D\U0001F3F1-\U0001F3F2\U0001F3F6\U0001F4FE\U0001F53E-\U0001F548\U0001F54F\U0001F568-\U0001F56E\U0001F571-\U0001F572\U0001F57B-\U0001F586\U0001F588-\U0001F589\U0001F58E-\U0001F58F\U0001F591-\U0001F594\U0001F597-\U0001F5A3\U0001F5A6-\U0001F5A7\U0001F5A9-\U0001F5B0\U0001F5B3-\U0001F5BB\U0001F5BD-\U0001F5C1\U0001F5C5-\U0001F5D0\U0001F5D4-\U0001F5DB\U0001F5DF-\U0001F5E0\U0001F5E2\U0001F5E4-\U0001F5E7\U0001F5E9-\U0001F5EE\U0001F5F0-\U0001F5F2\U0001F5F4-\U0001F5F9\u2605\u2607-\u260D\u260F-\u2610\u2612\u2616-\u2617\u2619-\u261C\u261E-\u261F\u2621\u2624-\u2625\u2627-\u2629\u262B-\u262D\u2630-\u2637\u263B-\u2647\u2654-\u265F\u2661-\u2662\u2664\u2667\u2669-\u267A\u267C-\u267E\u2680-\u2691\u2695\u2698\u269A\u269D-\u269F\u26A2-\u26A9\u26AC-\u26AF\u26B2-\u26BC\u26BF-\u26C3\u26C6-\u26C7\u26C9-\u26CD\u26D0\u26D2\u26D5-\u26E8\u26EB-\u26EF\u26F6\u26FB-\u26FC\u26FE-\u26FF\u2388\U0001FA00-\U0001FFFD\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F0AF-\U0001F0B0\U0001F0C0\U0001F0D0\U0001F0F6-\U0001F0FF\U0001F80C-\U0001F80F\U0001F848-\U0001F84F\U0001F85A-\U0001F85F\U0001F888-\U0001F88F\U0001F8AE-\U0001F8FF\U0001F900-\U0001F90F\U0001F91F\U0001F928-\U0001F92F\U0001F931-\U0001F932\U0001F93F\U0001F94C-\U0001F94F\U0001F95F-\U0001F97F\U0001F992-\U0001F9BF\U0001F9C1-\U0001F9FF\U0001F6C6-\U0001F6CA\U0001F6E6-\U0001F6E8\U0001F6EA\U0001F6F1-\U0001F6F2\U0001F6D3-\U0001F6DF\U0001F6ED-\U0001F6EF\U0001F6F7-\U0001F6FF]; 81 82# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly 83# list it in the numerous rules that use CM. 84# By LB1, SA characters with general categor of Mn or Mc also resolve to CM. 85 86$CM = [[:LineBreak = Combining_Mark:] $ZWJ [$SA & [[:Mn:][:Mc:]]]]; 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!!forward; 106 107# 108# CAN_CM is the set of characters that may combine with CM combining chars. 109# Note that Linebreak UAX 14's concept of a combining char and the rules 110# for what they can combine with are _very_ different from the rest of Unicode. 111# 112# Note that $CM itself is left out of this set. If CM is needed as a base 113# it must be listed separately in the rule. 114# 115$CAN_CM = [^$SP $BK $CR $LF $NL $ZW $CM]; # Bases that can take CMs 116$CANT_CM = [ $SP $BK $CR $LF $NL $ZW $CM]; # Bases that can't take CMs 117 118# 119# AL_FOLLOW set of chars that can unconditionally follow an AL 120# Needed in rules where stand-alone $CM s are treated as AL. 121# 122$AL_FOLLOW = [$BK $CR $LF $NL $ZW $SP $CL $CP $EX $HL $IS $SY $WJ $GL $OP $QU $BA $HH $HY $NS $IN $NU $PR $PO $ALPlus]; 123 124 125# 126# Rule LB 4, 5 Mandatory (Hard) breaks. 127# 128$LB4Breaks = [$BK $CR $LF $NL]; 129$LB4NonBreaks = [^$BK $CR $LF $NL $CM]; 130$CR $LF {100}; 131 132# 133# LB 6 Do not break before hard line breaks. 134# 135$LB4NonBreaks? $LB4Breaks {100}; # LB 5 do not break before hard breaks. 136$CAN_CM $CM* $LB4Breaks {100}; 137^$CM+ $LB4Breaks {100}; 138 139# LB 7 x SP 140# x ZW 141$LB4NonBreaks [$SP $ZW]; 142$CAN_CM $CM* [$SP $ZW]; 143^$CM+ [$SP $ZW]; 144 145# 146# LB 8 Break after zero width space 147# TODO: ZW SP* <break> 148# An engine change is required to write the reverse rule for this. 149# For now, leave the Unicode 5.2 rule, ZW <break> 150# 151$LB8Breaks = [$LB4Breaks $ZW]; 152$LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]]; 153 154# LB 8a ZWJ x (ID | Extended_Pict | EmojiNRK) Emoji ZWJ sequences. 155# 156$ZWJ ($ID | $Extended_Pict | $EmojiNRK); 157 158# LB 9 Combining marks. X $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL 159# $CM not covered by the above needs to behave like $AL 160# See definition of $CAN_CM. 161 162$CAN_CM $CM+; # Stick together any combining sequences that don't match other rules. 163^$CM+; 164 165# 166# LB 11 Do not break before or after WORD JOINER & related characters. 167# 168$CAN_CM $CM* $WJ; 169$LB8NonBreaks $WJ; 170^$CM+ $WJ; 171 172$WJ $CM* .; 173 174# 175# LB 12 Do not break after NBSP and related characters. 176# GL x 177# 178$GL $CM* .; 179 180# 181# LB 12a Do not break before NBSP and related characters ... 182# [^SP BA HY] x GL 183# 184[[$LB8NonBreaks] - [$SP $BA $HH $HY]] $CM* $GL; 185^$CM+ $GL; 186 187 188 189# 190# LB 13 Don't break before ']' or '!' or ';' or '/', even after spaces. 191# 192$LB8NonBreaks $CL; 193$CAN_CM $CM* $CL; 194^$CM+ $CL; # by rule 10, stand-alone CM behaves as AL 195 196$LB8NonBreaks $CP; 197$CAN_CM $CM* $CP; 198^$CM+ $CP; # by rule 10, stand-alone CM behaves as AL 199 200$LB8NonBreaks $EX; 201$CAN_CM $CM* $EX; 202^$CM+ $EX; # by rule 10, stand-alone CM behaves as AL 203 204$LB8NonBreaks $IS; 205$CAN_CM $CM* $IS; 206^$CM+ $IS; # by rule 10, stand-alone CM behaves as AL 207 208$LB8NonBreaks $SY; 209$CAN_CM $CM* $SY; 210^$CM+ $SY; # by rule 10, stand-alone CM behaves as AL 211 212 213# 214# LB 14 Do not break after OP, even after spaces 215# 216$OP $CM* $SP* .; 217 218$OP $CM* $SP+ $CM+ $AL_FOLLOW?; # by rule 10, stand-alone CM behaves as AL 219 # by rule 8, CM following a SP is stand-alone. 220 221# LB 15 222$QU $CM* $SP* $OP; 223 224# LB 16 225($CL | $CP) $CM* $SP* $NS; 226 227# LB 17 228$B2 $CM* $SP* $B2; 229 230# 231# LB 18 Break after spaces. 232# 233$LB18NonBreaks = [$LB8NonBreaks - [$SP]]; 234$LB18Breaks = [$LB8Breaks $SP]; 235 236 237# LB 19 238# x QU 239$LB18NonBreaks $CM* $QU; 240^$CM+ $QU; 241 242# QU x 243$QU $CM* .; 244 245 246# LB 20 247# <break> $CB 248# $CB <break> 249 250$LB20NonBreaks = [$LB18NonBreaks - $CB]; 251 252# LB 20.09 added rule for Finnish tailoring 253# LB 21 x (BA | HY | NS) 254# BB x 255# 256$LB20NonBreaks $CM* ($BA | $HH | $HY | $NS) $CM* / $AL; 257$LB20NonBreaks $CM* ($BA | $HH | $HY | $NS); 258($HY | $HH) $AL; 259^$CM+ ($BA | $HY | $HH | $NS); 260 261$BB $CM* [^$CB]; # $BB x 262$BB $CM* $LB20NonBreaks; 263 264# LB 21a Don't break after Hebrew + Hyphen 265# HL (HY | BA) x 266# 267$HL $CM* ($HY | $BA | $HH) $CM* [^$CB]?; 268 269# LB 21b (forward) Don't break between SY and HL 270# (break between HL and SY already disallowed by LB 13 above) 271$SY $CM* $HL; 272 273# LB 22 274($ALPlus | $HL) $CM* $IN; 275^$CM+ $IN; # by rule 10, any otherwise unattached CM behaves as AL 276$EX $CM* $IN; 277($ID | $EB | $EM) $CM* $IN; 278$IN $CM* $IN; 279$NU $CM* $IN; 280 281 282# $LB 23 283# 284($ALPlus | $HL) $CM* $NU; 285^$CM+ $NU; # Rule 10, any otherwise unattached CM behaves as AL 286$NU $CM* ($ALPlus | $HL); 287 288# LB 23a 289# 290$PR $CM* ($ID | $EB | $EM); 291($ID | $EB | $EM) $CM* $PO; 292 293 294# 295# LB 24 296# 297($PR | $PO) $CM* ($ALPlus | $HL); 298($ALPlus | $HL) $CM* ($PR | $PO); 299^$CM+ ($PR | $PO); # Rule 10, any otherwise unattached CM behaves as AL 300 301# 302# LB 25 Numbers. 303# 304(($PR | $PO) $CM*)? (($OP | $HY) $CM*)? $NU ($CM* ($NU | $SY | $IS))* 305 ($CM* ($CL | $CP))? ($CM* ($PR | $PO))?; 306 307# LB 26 Do not break a Korean syllable 308# 309$JL $CM* ($JL | $JV | $H2 | $H3); 310($JV | $H2) $CM* ($JV | $JT); 311($JT | $H3) $CM* $JT; 312 313# LB 27 Treat korean Syllable Block the same as ID (don't break it) 314($JL | $JV | $JT | $H2 | $H3) $CM* $IN; 315($JL | $JV | $JT | $H2 | $H3) $CM* $PO; 316$PR $CM* ($JL | $JV | $JT | $H2 | $H3); 317 318 319# LB 28 Do not break between alphabetics 320# 321($ALPlus | $HL) $CM* ($ALPlus | $HL); 322^$CM+ ($ALPlus | $HL); # The $CM+ is from rule 10, an unattached CM is treated as AL 323 324# LB 29 325$IS $CM* ($ALPlus | $HL); 326 327# LB 30 328($ALPlus | $HL | $NU) $CM* $OP; 329^$CM+ $OP; # The $CM+ is from rule 10, an unattached CM is treated as AL. 330$CP $CM* ($ALPlus | $HL | $NU); 331 332# LB 30a Do not break between regional indicators. Break after pairs of them. 333# Tricky interaction with LB8a: ZWJ x ID 334$RI $CM* $RI / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM] {eof}]; 335$RI $CM* $RI $CM* $ZWJ / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM $ID $Extended_Pict $EmojiNRK] {eof}]; 336$RI $CM* $RI $CM* [$CM-$ZWJ] / [[^$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS $CM] {eof}]; 337 338$RI $CM* $RI $CM* [$BK $CR $LF $NL $SP $ZW $WJ $CL $CP $EX $IS $SY $GL $QU $BA $HY $NS {eof}]; 339$RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK); 340 341# LB 30b Do not break between an Emoji Base and an Emoji Modifier 342$EB $CM* $EM; 343 344# 345# Reverse Rules. 346# 347## ------------------------------------------------- 348 349!!reverse; 350 351# LB 9 Combining Marks. 352# Stick together any combining sequences that don't match other rules. 353 354^$CM+ $CAN_CM?; 355 356# 357# Sequences of the form (shown forwards) 358# [CANT_CM] <break> [CM] [whatever] 359# The CM needs to behave as an AL 360# 361$AL_FOLLOW $CM+ / ( 362 [$BK $CR $LF $NL $ZW {eof}] | 363 $SP+ $CM+ $SP | 364 $SP+ $CM* ([^$OP $CM $SP] | [$AL {eof}])); # if LB 14 will match, need to surpress this break. 365 # LB14 says OP SP* x . 366 # becomes OP SP* x AL 367 # becomes OP SP* x CM+ AL_FOLLOW 368 # 369 # Further note: the $AL in [$AL {eof}] is only to work around 370 # a rule compiler bug which complains about 371 # empty sets otherwise. 372 373 374# LB 4, 5, 6 375 376$LB4Breaks [$LB4NonBreaks-$CM]; 377$LB4Breaks $CM+ $CAN_CM; 378$LF $CR; 379 380 381# LB 7 x SP 382# x ZW 383[$SP $ZW] [$LB4NonBreaks-$CM]; 384[$SP $ZW] $CM+ $CAN_CM; 385 386# LB 8 ZW SP* <break> 387# TODO: to implement this, we need more than one look-ahead hard break in play at a time. 388# Requires an engine enhancement. 389# / $SP* $ZW 390 391# LB 8a ZWJ x (ID | Extended_Pict | EmojiNRK) 392# 393($ID | $Extended_Pict | $EmojiNRK) $ZWJ $CM* $CAN_CM?; 394 395 396# LB 9,10 Combining marks. 397# X $CM needs to behave like X, where X is not $SP or controls. 398# $CM not covered by the above needs to behave like $AL 399# Stick together any combining sequences that don't match other rules. 400^$CM+ $CAN_CM; 401 402 403# LB 11 404# 405$WJ $CM* $CAN_CM; 406$WJ [$LB8NonBreaks-$CM]; 407 408 $CANT_CM $CM* $WJ; 409$CAN_CM $CM* $WJ; 410 411# LB 12a 412# [^SP BA HY] x GL 413# 414$GL $CM* [$LB8NonBreaks-[$CM $SP $BA $HH $HY]]; 415 416# LB 12 417# GL x 418# 419$CANT_CM $CM* $GL; 420$CAN_CM $CM* $GL; 421 422 423# LB 13 424$CL $CM+ $CAN_CM; 425$CP $CM+ $CAN_CM; 426$EX $CM+ $CAN_CM; 427$IS $CM+ $CAN_CM; 428$SY $CM+ $CAN_CM; 429 430$CL [$LB8NonBreaks-$CM]; 431$CP [$LB8NonBreaks-$CM]; 432$EX [$LB8NonBreaks-$CM]; 433$IS [$LB8NonBreaks-$CM]; 434$SY [$LB8NonBreaks-$CM]; 435 436 437# LB 14 OP SP* x 438# 439. $SP* $CM* $OP; 440$AL_FOLLOW? $CM+ $SP+ $CM* $OP; # by LB 10, behaves like $AL_FOLLOW? $AL $SP+ $CM* $OP 441 442 443# LB 15 444$OP $SP* $CM* $QU; 445 446# LB 16 447$NS $SP* $CM* ($CL | $CP); 448 449# LB 17 450$B2 $SP* $CM* $B2; 451 452# LB 18 break after spaces 453# Nothing explicit needed here. 454 455 456# 457# LB 19 458# 459$QU $CM* $CAN_CM; # . x QU 460$QU $LB18NonBreaks; 461 462 463$CAN_CM $CM* $QU; # QU x . 464 $CANT_CM $CM* $QU; 465 466# 467# LB 20 Break before and after CB. 468# nothing needed here. 469# 470 471# LB 20.09 added rule for Finnish tailoring 472$AL ($HY | $HH) / $SP; 473 474# LB 21 475($BA | $HH | $HY | $NS) $CM* [$LB20NonBreaks-$CM]; # . x (BA | HY | NS) 476 477[$LB20NonBreaks-$CM] $CM* $BB; # BB x . 478[^$CB] $CM* $BB; # 479 480# LB21a 481[^$CB] $CM* ($HY | $BA | $HH) $CM* $HL; 482 483# LB21b (reverse) 484$HL $CM* $SY; 485 486# LB 22 487$IN $CM* ($ALPlus | $HL); 488$IN $CM* $EX; 489$IN $CM* ($ID | $EB | $EM); 490$IN $CM* $IN; 491$IN $CM* $NU; 492 493# LB 23 494$NU $CM* ($ALPlus | $HL); 495($ALPlus | $HL) $CM* $NU; 496 497# LB23a 498($ID | $EB | $EM) $CM* $PR; 499$PO $CM* ($ID | $EB | $EM); 500 501# LB 24 502($ALPlus | $HL) $CM* ($PR | $PO); 503($PR | $PO) $CM* ($ALPlus | $HL); 504 505 506# LB 25 507($CM* ($PR | $PO))? ($CM* ($CL | $CP))? ($CM* ($NU | $IS | $SY))* $CM* $NU ($CM* ($OP | $HY))? ($CM* ($PR | $PO))?; 508 509# LB 26 510($H3 | $H2 | $JV | $JL) $CM* $JL; 511($JT | $JV) $CM* ($H2 | $JV); 512$JT $CM* ($H3 | $JT); 513 514# LB 27 515$IN $CM* ($H3 | $H2 | $JT | $JV | $JL); 516$PO $CM* ($H3 | $H2 | $JT | $JV | $JL); 517 ($H3 | $H2 | $JT | $JV | $JL) $CM* $PR; 518 519# LB 28 520($ALPlus | $HL) $CM* ($ALPlus | $HL); 521 522 523# LB 29 524($ALPlus | $HL) $CM* $IS; 525 526# LB 30 527$OP $CM* ($ALPlus | $HL | $NU); 528($ALPlus | $HL | $NU) $CM* $CP; 529 530# LB 30a 531# Pairs of Regional Indicators. 532# The following two rules are nearly identical. The first matches only sequences with an odd number of adjacent RIs, 533# the second with an even number. Stripping away the cruft they look like 534# [^RI] RI / (RI RI)+ ^RI; 535# [^RI] RI RI / (RI RI)+ ^RI; 536# 537[{bof} $NS $HY $BA $QU $CL $CP $EX $IS $SY $WJ $GL $ZW $SP $BK $CR $LF $NL $ZWJ] $CM* $RI / ($CM* $RI $CM* $RI)+ $CM* [{eof}[^$RI $CM]]; 538[{bof} $NS $HY $BA $QU $CL $CP $EX $IS $SY $WJ $GL $ZW $SP $BK $CR $LF $NL $ZWJ] $CM* $RI $CM* $RI / ($CM* $RI $CM* $RI)+ $CM* [{eof}[^$RI $CM]]; 539 540# In general, adjacent RIs stay together. The hard-break rules, above, overide this, forcing in the boundaries between pairs. 541$RI $CM* $RI; 542 543# WJ, GL, QU, etc. are classes with rules like "WJ x " which includes "WJ x RI". 544$RI $CM* ([$WJ $GL $QU $BB] | (($HY | $BA)$CM* $HL)); 545 546 547# LB 30b Do not break between an Emoji Base and an Emoji Modifier 548$EM $CM* $EB; 549 550 551## ------------------------------------------------- 552 553!!safe_reverse; 554 555# LB 9 556^$CM+ [^$CM $BK $CR $LF $NL $ZW $SP]; 557^$CM+ $SP / .; 558 559# LB 14 560$SP+ $CM* $OP; 561 562# LB 15 563$SP+ $CM* $QU; 564 565# LB 16 566$SP+ $CM* ($CL | $CP); 567 568# LB 17 569$SP+ $CM* $B2; 570 571# LB 21 572$CM* ($HY | $BA | $HH) $CM* $HL; 573 574# LB 25 575($CM* ($IS | $SY))+ $CM* $NU; 576($CL | $CP) $CM* ($NU | $IS | $SY); 577 578# LB 30 579($CM* $RI)+; 580 581# For dictionary-based break 582$dictionary $dictionary; 583 584## ------------------------------------------------- 585 586!!safe_forward; 587 588# Skip forward over all character classes that are involved in 589# rules containing patterns with possibly more than one char 590# of context. 591# 592# It might be slightly more efficient to have specific rules 593# instead of one generic one, but only if we could 594# turn off rule chaining. We don't want to move more 595# than necessary. 596# 597^[$CM $OP $QU $CL $CP $B2 $PR $HY $BA $SP $RI $ZWJ $dictionary]+ [^$CM $OP $QU $CL $CP $B2 $PR $HY $BA $RI $ZWJ $dictionary]; 598$dictionary $dictionary; 599 600