icuSources/test/testdata/break_rules/line_normal_cj.txt

   1 #
   2 # Copyright (C) 2016 and later: Unicode, Inc. and others.
   3 # License & terms of use: http://www.unicode.org/copyright.html
   4 # Copyright (c) 2016  International Business Machines Corporation and others. All Rights Reserved.
   5 #
   6 #  file:  line_normal_cj.txt
   7 #
   8 # Reference Line Break rules for intltest rbbi/RBBIMonkeyTest.
   9 # Rules derived from Unicode Standard Annex #14 Revision 44 for Unicode 13.0
  10 #
  11 # Note: Rule syntax and the monkey test itself are still a work in progress.
  12 #       They are expected to change with review and the addition of support for rule tailoring.
  13 #
  14 #         Line Breaking Rules
  15 #         Implement default line breaking as defined by
  16 #         Unicode Standard Annex #14
  17 #         http://www.unicode.org/reports/tr14/, tailored as noted below.
  18 #
  19 #         This tailors the line break behavior to correspond to CSS
  20 #         line-break=normal (BCP47 -u-lb-normal) as defined for Chinese & Japanese.
  21 #         It sets characters of class CJ to behave like ID.
  22 #         In addition, it allows breaks:
  23 #         * before hyphens 2010 & 2013 (both BA) and 301C, 30A0 (both NS)
  24 #         It allows breaking before 201C and after 201D, for zh_Hans, zh_Hant, and ja.
  25
  26 type = line;
  27 locale = ja@lb=normal;
  28
  29 AI = [:LineBreak =  Ambiguous:];
  30 AL = [:LineBreak =  Alphabetic:];
  31 BAX = [\u2010 \u2013];
  32 BA = [[:LineBreak =  Break_After:] - BAX];
  33 HH = [\u2010];      # \u2010 is HYPHEN, default line break is BA.
  34 BB = [:LineBreak =  Break_Before:];
  35 BK = [:LineBreak =  Mandatory_Break:];
  36 B2 = [:LineBreak =  Break_Both:];
  37 CB = [:LineBreak =  Contingent_Break:];
  38 CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  39 CL = [[:LineBreak =  Close_Punctuation:] \u201d];
  40 CMS = [:LineBreak =  Combining_Mark:];
  41 CP = [:LineBreak =  Close_Parenthesis:];
  42 CR = [:LineBreak =  Carriage_Return:];
  43 EB = [[:LineBreak =  EB:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
  44 EM = [:LineBreak =  EM:];
  45 EX = [:LineBreak =  Exclamation:];
  46 GL = [:LineBreak =  Glue:];
  47 HL = [:LineBreak =  Hebrew_Letter:];
  48 HY = [:LineBreak =  Hyphen:];
  49 H2 = [:LineBreak =  H2:];
  50 H3 = [:LineBreak =  H3:];
  51 ID = [[:LineBreak =  Ideographic:] CJ];  # CSS Normal tailoring: CJ resolves to ID
  52 IN = [:LineBreak =  Inseperable:];
  53 IS = [:LineBreak =  Infix_Numeric:];
  54 JL = [:LineBreak =  JL:];
  55 JV = [:LineBreak =  JV:];
  56 JT = [:LineBreak =  JT:];
  57 LF = [:LineBreak =  Line_Feed:];
  58 NL = [:LineBreak =  Next_Line:];
  59 NSX = [\u301C \u30A0];
  60 NS = [[:LineBreak =  Nonstarter:] - NSX];
  61 NU = [:LineBreak =  Numeric:];
  62 OP = [[:LineBreak =  Open_Punctuation:] \u201c];
  63 PO = [:LineBreak =  Postfix_Numeric:];
  64 PR = [:LineBreak =  Prefix_Numeric:];
  65 QU = [[:LineBreak =  Quotation:] - [\u201c\u201d]];
  66 RI = [:LineBreak =  Regional_Indicator:];
  67 SA = [:LineBreak =  Complex_Context:];
  68 SG = [:LineBreak =  Surrogate:];
  69 SP = [:LineBreak =  Space:];
  70 SY = [:LineBreak =  Break_Symbols:];
  71 WJ = [:LineBreak =  Word_Joiner:];
  72 XX = [:LineBreak =  Unknown:];
  73 ZW = [:LineBreak =  ZWSpace:];
  74 ZWJ = [:LineBreak =  ZWJ:];
  75
  76 # OP30 and CP30 are variants of OP and CP that appear in rule LB30 from UAX 14.
  77 # Limitations of this monkey test rule parser require that these definitions be pulled out
  78 # rather than appearing in-line in LB 30.
  79
  80 OP30 = [OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
  81 CP30 = [CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
  82
  83 # LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
  84 AL = [AL AI SG XX ];
  85 dictionary = SA;
  86
  87 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  88 #         list it in the numerous rules that use CM.
  89 CM = [CMS ZWJ];
  90
  91 LB4:        BK ÷;
  92 LB5:        CR LF;
  93 LB5.1:      CR ÷;
  94 LB5.2:      LF ÷;
  95 LB5.3:      NL ÷;
  96
  97 LB6:        . (BK | CR | LF | NL);
  98 LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
  99
 100 # LB8   break after ZW SP*.
 101 # Precedes LB7 because both rules will match the sequences like ZW SP,
 102 # and LB8 must take precedence.
 103
 104 LB8:        ZW SP* ÷ [^ZW SP BK CR LF NL];
 105
 106 # Numbers. Equivalent to Tailoring example 8 from UAX 14.
 107 # Moved up, before LB14, because it can match longer sequences which must take precedence.
 108 LB25:        ((PR | PO)CM*)? ((OP | HY)CM*)? (IS CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?;
 109
 110 # Rules LB14 - LB17.
 111 # Moved up, before LB7, because they can match a longer sequence that would also match LB7.
 112 # For example, the sequence "OP CM SP AL" matches LB14
 113 # while the prefix of it, "OP CM SP" matches LB7.1
 114 LB14:        OP CM* SP* .;
 115
 116 # LB 14a Break before an IS that begins a number and follows a space.
 117 LB14a:       SP ÷ IS CM* NU;
 118
 119 # LB14b       × IS
 120 LB14b.1:     [^SP] CM* IS;
 121 LB14b.2:     SP IS;
 122
 123 LB15:        QU CM* SP* OP;
 124
 125 # Do not break between closing punctuation and $NS, even with intervening spaces
 126 # But DO allow a break between closing punctuation and $NSX, don't include it here
 127 LB16:        (CL | CP)CM* SP* NS;
 128 LB17:        B2 CM* SP* B2;
 129
 130 # LB7 Do not break before spaces or zero width space.
 131
 132 LB7.1:      [^ZW SP] CM* [SP ZW];
 133 LB7.2:      [ZW SP] [SP ZW];
 134
 135 # LB8a
 136 #      ZWJ x
 137 #      Don't match a CM on the right - let other rules pick up CM sequences, where
 138 #      the ZWJ behaves as just another generic CM.
 139 LB8a:       ZWJ [^CM];
 140
 141
 142 # LB9:  X CM -> X
 143 # LB10: Unattached CM -> AL
 144
 145 #LB11:       × WJ;
 146 #            WJ ×
 147
 148 LB11.1:      [^SP] CM* WJ;
 149 LB11.2:      SP WJ;
 150 LB11.3:      WJ CM* [^CM];
 151
 152 LB12:        GL CM* [^CM];
 153
 154 LB12a:       [^SP BA BAX HY] CM* GL;
 155
 156 # LB 13 Do not break before ‘]’ or ‘!’ or ‘/’, even after spaces.
 157 LB13.1: [^SP] CM* [CL CP EX SY];
 158 LB13.2: SP [CL CP EX SY];
 159
 160 # LB 14-17 are moved above LB 7.
 161
 162 LB18:        SP ÷;
 163
 164 LB19:        . CM* QU;
 165 LB19.1:      QU CM* [^CM];
 166
 167 # LB 20   Break before and after CB.
 168 #         Interaction with LB8a:  ZWJ x . is tricky because CM includes ZWJ.
 169 #                                 ZWJ acts like a CM to the left, combining with CB.
 170 #                                 ZWJ acts independently to the right, no break after by LB8a.
 171 LB20.1:      . CM* ZWJ CB;
 172 LB20.2:      . CM* ÷ CB;
 173
 174 LB20.3:      CB CM* ZWJ [^CM];
 175 LB20.4:      CB CM* ÷;
 176
 177 # LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
 178 LB20.09:     ^(HY | HH) CM* AL;
 179
 180 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
 181 #       not picking up the continuing match after the BA from 21a.
 182 # TODO: For CJ tailorings (with BAX) does this rule want to include BAX? If so,
 183 #       should "HL BAX" not break when followed by a CB? Thats what the current
 184 #       rules do, which is why "[^CM CB]?" includes the ?.
 185 LB21a:       HL CM* (HY | BA | BAX) CM* [^CM CB]?;
 186
 187 # DO allow breaks here before $BAXcm and $NSXcm, so don't include them
 188 LB21.1:      . CM* [BA HY NS];
 189 LB21.2:      BB CM* [^CM CB];
 190
 191 LB21b:       SY CM* HL;
 192
 193 LB22:        . CM* IN;
 194
 195 LB23.1:      (AL | HL | CM) CM* NU;
 196 LB23.2:      NU CM* (AL | HL);
 197
 198 LB23a.1:     PR CM* (ID | EB | EM);
 199 LB23a.2:     (ID | EB | EM) CM* PO;
 200
 201 LB24.2:      (PR | PO) CM* (AL | HL);
 202 LB24.3:      (AL | HL | CM) CM* (PR | PO);
 203
 204 LB26.1:      JL CM* (JL | JV | H2 | H3);
 205 LB26.2:      (JV | H2) CM* (JV | JT);
 206 LB26.3:      (JT | H3) CM* JT;
 207
 208 LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
 209 LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
 210 LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
 211
 212 # LB28 Do not break between Alphabetics.
 213 #      Unattached (leading) CM treated as AL.
 214 LB28:        (AL | HL | CM)CM* (AL | HL);
 215
 216 LB29:        IS CM* (AL | HL);
 217
 218 # LB30  is adjusted for unattached leading CM being treated as AL.
 219 LB30.1:      (AL | CM | HL | NU) CM* OP30;
 220 LB30.2:      CP30 CM* (AL | HL | NU);
 221
 222 # LB30a  keep pairs of RI together.
 223 LB30a.1:     RI CM* RI         ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS IN CM];
 224 LB30a.2:     RI CM* RI CM* CMS ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS IN CM];
 225 LB30a.3:     RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS IN ZWJ]?;
 226
 227 # LB30b Do not break between Emoji Base and Emoji Modifier
 228 LB30b:       EB CM* EM;
 229
 230 # LB31 Break Everywhere Else.
 231 #      Include combining marks
 232 LB31.1:        . CM* ZWJ [^CM];
 233 LB31.2:        . CM* ÷;