icuSources/test/testdata/break_rules/line_normal.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.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 languages other than
  21 #         Chinese & Japanese.
  22 #         It sets characters of class CJ to behave like ID.
  23
  24
  25 type = line;
  26 locale = en@lb=normal;
  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 CMS = [:LineBreak =  Combining_Mark:];
  39 CP = [:LineBreak =  Close_Parenthesis:];
  40 CR = [:LineBreak =  Carriage_Return:];
  41 EB = [[:LineBreak =  EB:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
  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 ID = [[:LineBreak =  Ideographic:] CJ];  # CSS Normal tailoring: CJ resolves to ID
  50 IN = [:LineBreak =  Inseperable:];
  51 IS = [:LineBreak =  Infix_Numeric:];
  52 JL = [:LineBreak =  JL:];
  53 JV = [:LineBreak =  JV:];
  54 JT = [:LineBreak =  JT:];
  55 LF = [:LineBreak =  Line_Feed:];
  56 NL = [:LineBreak =  Next_Line:];
  57 NS = [:LineBreak =  Nonstarter:];
  58 NU = [:LineBreak =  Numeric:];
  59 OP = [:LineBreak =  Open_Punctuation:];
  60 PO = [:LineBreak =  Postfix_Numeric:];
  61 PR = [:LineBreak =  Prefix_Numeric:];
  62 QU = [:LineBreak =  Quotation:];
  63 RI = [:LineBreak =  Regional_Indicator:];
  64 SA = [:LineBreak =  Complex_Context:];
  65 SG = [:LineBreak =  Surrogate:];
  66 SP = [:LineBreak =  Space:];
  67 SY = [:LineBreak =  Break_Symbols:];
  68 WJ = [:LineBreak =  Word_Joiner:];
  69 XX = [:LineBreak =  Unknown:];
  70 ZW = [:LineBreak =  ZWSpace:];
  71 ZWJ = [:LineBreak =  ZWJ:];
  72
  73 # OP30 and CP30 are variants of OP and CP that appear in rule LB30 from UAX 14.
  74 # Limitations of this monkey test rule parser require that these definitions be pulled out
  75 # rather than appearing in-line in LB 30.
  76
  77 OP30 = [OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
  78 CP30 = [CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
  79
  80 # LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
  81 AL = [AL AI SG XX ];
  82 dictionary = SA;
  83
  84 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  85 #         list it in the numerous rules that use CM.
  86 CM = [CMS ZWJ];
  87
  88 LB4:        BK ÷;
  89 LB5:        CR LF;
  90 LB5.1:      CR ÷;
  91 LB5.2:      LF ÷;
  92 LB5.3:      NL ÷;
  93
  94 LB6:        . (BK | CR | LF | NL);
  95 LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
  96
  97 # LB8   break after ZW SP*.
  98 # Precedes LB7 because both rules will match the sequences like ZW SP,
  99 # and LB8 must take precedence.
 100
 101 LB8:        ZW SP* ÷ [^ZW SP BK CR LF NL];
 102
 103 # Numbers. Equivalent to Tailoring example 8 from UAX 14.
 104 # Moved up, before LB14, because it can match longer sequences which must take precedence.
 105 LB25:        ((PR | PO)CM*)? ((OP | HY)CM*)? (IS CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?;
 106
 107 # Rules LB14 - LB17.
 108 # Moved up, before LB7, because they can match a longer sequence that would also match LB7.
 109 # For example, the sequence "OP CM SP AL" matches LB14
 110 # while the prefix of it, "OP CM SP" matches LB7.1
 111 LB14:        OP CM* SP* .;
 112
 113 # LB 14a Break before an IS that begins a number and follows a space.
 114 LB14a:       SP ÷ IS CM* NU;
 115
 116 # LB14b       × IS
 117 LB14b.1:     [^SP] CM* IS;
 118 LB14b.2:     SP IS;
 119
 120 LB15:        QU CM* SP* OP;
 121 LB16:        (CL | CP)CM* SP* NS;
 122 LB17:        B2 CM* SP* B2;
 123
 124
 125 # LB7 Do not break before spaces or zero width space.
 126
 127 LB7.1:      [^ZW SP] CM* [SP ZW];
 128 LB7.2:      [ZW SP] [SP ZW];
 129
 130 # LB8a
 131 #      ZWJ x
 132 #      Don't match a CM on the right - let other rules pick up CM sequences, where
 133 #      the ZWJ behaves as just another generic CM.
 134 LB8a:       ZWJ [^CM];
 135
 136
 137 # LB9:  X CM -> X
 138 # LB10: Unattached CM -> AL
 139
 140 #LB11:       × WJ;
 141 #            WJ ×
 142
 143 LB11.1:      [^SP] CM* WJ;
 144 LB11.2:      SP WJ;
 145 LB11.3:      WJ CM* [^CM];
 146
 147 LB12:        GL CM* [^CM];
 148
 149 LB12a:       [^SP BA HY] CM* GL;
 150
 151 # LB 13 Do not break before ‘]’ or ‘!’ or ‘/’, even after spaces.
 152 LB13.1: [^SP] CM* [CL CP EX SY];
 153 LB13.2: SP [CL CP EX SY];
 154
 155 # LB 14-17 are moved above LB 7.
 156
 157 LB18:        SP ÷;
 158
 159 LB19:        . CM* QU;
 160 LB19.1:      QU CM* [^CM];
 161
 162 # LB 20   Break before and after CB.
 163 #         Interaction with LB8a:  ZWJ x . is tricky because CM includes ZWJ.
 164 #                                 ZWJ acts like a CM to the left, combining with CB.
 165 #                                 ZWJ acts independently to the right, no break after by LB8a.
 166 LB20.1:      . CM* ZWJ CB;
 167 LB20.2:      . CM* ÷ CB;
 168
 169 LB20.3:      CB CM* ZWJ [^CM];
 170 LB20.4:      CB CM* ÷;
 171
 172 # LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
 173 LB20.09:     ^(HY | HH) CM* AL;
 174
 175 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
 176 #       not picking up the continuing match after the BA from 21a.
 177 LB21a:       HL CM* (HY | BA) CM* [^CM CB];
 178
 179 LB21.1:      . CM* [BA HY NS];
 180 LB21.2:      BB CM* [^CM CB];
 181
 182 LB21b:       SY CM* HL;
 183
 184 LB22:        . CM* IN;
 185
 186 LB23.1:      (AL | HL | CM) CM* NU;
 187 LB23.2:      NU CM* (AL | HL);
 188
 189 LB23a.1:     PR CM* (ID | EB | EM);
 190 LB23a.2:     (ID | EB | EM) CM* PO;
 191
 192 LB24.2:      (PR | PO) CM* (AL | HL);
 193 LB24.3:      (AL | HL | CM) CM* (PR | PO);
 194
 195
 196 LB26.1:      JL CM* (JL | JV | H2 | H3);
 197 LB26.2:      (JV | H2) CM* (JV | JT);
 198 LB26.3:      (JT | H3) CM* JT;
 199
 200 LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
 201 LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
 202 LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
 203
 204 # LB28 Do not break between Alphabetics.
 205 #      Unattached (leading) CM treated as AL.
 206 LB28:        (AL | HL | CM)CM* (AL | HL);
 207
 208 LB29:        IS CM* (AL | HL);
 209
 210 # LB30  is adjusted for unattached leading CM being treated as AL.
 211 LB30.1:      (AL | CM | HL | NU) CM* OP30;
 212 LB30.2:      CP30 CM* (AL | HL | NU);
 213
 214 # LB30a  keep pairs of RI together.
 215 LB30a.1:     RI CM* RI         ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS IN CM];
 216 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];
 217 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]?;
 218
 219 # LB30b Do not break between Emoji Base and Emoji Modifier
 220 LB30b:       EB CM* EM;
 221
 222 # LB31 Break Everywhere Else.
 223 #      Include combining marks
 224 LB31.1:        . CM* ZWJ [^CM];
 225 LB31.2:        . CM* ÷;