icuSources/test/testdata/break_rules/line.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.txt
   7 #
   8 # Reference Line Break rules for intltest rbbi/RBBIMonkeyTest.
   9 # Rules derived from Unicode Standard Annex #14 Revision 40 for Unicode 11.0
  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
  15 type = line;
  16 locale = en;
  17
  18
  19 AI = [:LineBreak =  Ambiguous:];
  20 AL = [:LineBreak =  Alphabetic:];
  21 BA = [:LineBreak =  Break_After:];
  22 HH = [\u2010];      # \u2010 is HYPHEN, default line break is BA.
  23 BB = [:LineBreak =  Break_Before:];
  24 BK = [:LineBreak =  Mandatory_Break:];
  25 B2 = [:LineBreak =  Break_Both:];
  26 CB = [:LineBreak =  Contingent_Break:];
  27 CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  28 CL = [:LineBreak =  Close_Punctuation:];
  29 CMS = [:LineBreak =  Combining_Mark:];
  30 CP = [:LineBreak =  Close_Parenthesis:];
  31 CR = [:LineBreak =  Carriage_Return:];
  32 EB = [[:LineBreak =  EB:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
  33 EM = [:LineBreak =  EM:];
  34 EX = [:LineBreak =  Exclamation:];
  35 GL = [:LineBreak =  Glue:];
  36 HL = [:LineBreak =  Hebrew_Letter:];
  37 HY = [:LineBreak =  Hyphen:];
  38 H2 = [:LineBreak =  H2:];
  39 H3 = [:LineBreak =  H3:];
  40 ID = [:LineBreak =  Ideographic:];
  41 IN = [:LineBreak =  Inseperable:];
  42 IS = [:LineBreak =  Infix_Numeric:];
  43 JL = [:LineBreak =  JL:];
  44 JV = [:LineBreak =  JV:];
  45 JT = [:LineBreak =  JT:];
  46 LF = [:LineBreak =  Line_Feed:];
  47 NL = [:LineBreak =  Next_Line:];
  48 NS = [[:LineBreak =  Nonstarter:] CJ];   # CSS Strict tailoring: CJ resolves to NS.
  49 NU = [:LineBreak =  Numeric:];
  50 OP = [:LineBreak =  Open_Punctuation:];
  51 PO = [:LineBreak =  Postfix_Numeric:];
  52 PR = [:LineBreak =  Prefix_Numeric:];
  53 QU = [:LineBreak =  Quotation:];
  54 RI = [:LineBreak =  Regional_Indicator:];
  55 SA = [:LineBreak =  Complex_Context:];
  56 SG = [:LineBreak =  Surrogate:];
  57 SP = [:LineBreak =  Space:];
  58 SY = [:LineBreak =  Break_Symbols:];
  59 WJ = [:LineBreak =  Word_Joiner:];
  60 XX = [:LineBreak =  Unknown:];
  61 ZW = [:LineBreak =  ZWSpace:];
  62 ZWJ = [:LineBreak =  ZWJ:];
  63
  64 # LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
  65 AL = [AL AI SG XX ];
  66 dictionary = SA;
  67
  68 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  69 #         list it in the numerous rules that use CM.
  70 CM = [CMS ZWJ];
  71
  72 LB4:        BK ÷;
  73 LB5:        CR LF;
  74 LB5.1:      CR ÷;
  75 LB5.2:      LF ÷;
  76 LB5.3:      NL ÷;
  77
  78 LB6:        . (BK | CR | LF | NL);
  79 LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
  80
  81 # LB8   break after ZW SP*.
  82 # Precedes LB7 because both rules will match the sequences like ZW SP,
  83 # and LB8 must take precedence.
  84
  85 LB8:        ZW SP* ÷ [^ZW SP BK CR LF NL];
  86
  87 # Numbers. Equivalent to Tailoring example 8 from UAX 14.
  88 # Moved up, before LB14, because it can match longer sequences which must take precedence.
  89 LB25:        ((PR | PO)CM*)? ((OP | HY)CM*)? (IS CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?;
  90
  91 # Rules LB14 - LB17.
  92 # Moved up, before LB7, because they can match a longer sequence that would also match LB7.
  93 # For example, the sequence "OP CM SP AL" matches LB14
  94 # while the prefix of it, "OP CM SP" matches LB7.1
  95 LB14:        OP CM* SP* .;
  96
  97 # LB 14a Break before an IS that begins a number and follows a space.
  98 LB14a:       SP ÷ IS CM* NU;
  99
 100 # LB14b       × IS
 101 LB14b.1:     [^SP] CM* IS;
 102 LB14b.2:     SP IS;
 103
 104 LB15:        QU CM* SP* OP;
 105 LB16:        (CL | CP)CM* SP* NS;
 106 LB17:        B2 CM* SP* B2;
 107
 108
 109 # LB7 Do not break before spaces or zero width space.
 110
 111 LB7.1:      [^ZW SP] CM* [SP ZW];
 112 LB7.2:      [ZW SP] [SP ZW];
 113
 114 # LB8a
 115 #      ZWJ x
 116 #      Don't match a CM on the right - let other rules pick up CM sequences, where
 117 #      the ZWJ behaves as just another generic CM.
 118 LB8a:       ZWJ [^CM];
 119
 120
 121 # LB9:  X CM -> X
 122 # LB10: Unattached CM -> AL
 123
 124 #LB11:       × WJ;
 125 #            WJ ×
 126
 127 LB11.1:      [^SP] CM* WJ;
 128 LB11.2:      SP WJ;
 129 LB11.3:      WJ CM* [^CM];
 130
 131 LB12:        GL CM* [^CM];
 132
 133 LB12a:       [^SP BA HY] CM* GL;
 134
 135 # LB 13 Do not break before ‘]’ or ‘!’ or ‘/’, even after spaces.
 136 LB13.1: [^SP] CM* [CL CP EX SY];
 137 LB13.2: SP [CL CP EX SY];
 138
 139 # LB 14-17 are moved above LB 7.
 140
 141 LB18:        SP ÷;
 142
 143 LB19:        . CM* QU;
 144 LB19.1:      QU CM* [^CM];
 145
 146 # LB 20   Break before and after CB.
 147 #         Interaction with LB8a:  ZWJ x . is tricky because CM includes ZWJ.
 148 #                                 ZWJ acts like a CM to the left, combining with CB.
 149 #                                 ZWJ acts independently to the right, no break after by LB8a.
 150 LB20.1:      . CM* ZWJ CB;
 151 LB20.2:      . CM* ÷ CB;
 152
 153 LB20.3:      CB CM* ZWJ [^CM];
 154 LB20.4:      CB CM* ÷;
 155
 156 # LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
 157 LB20.09:     ^(HY | HH) CM* AL;
 158
 159 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
 160 #       not picking up the continuing match after the BA from 21a.
 161 LB21a:       HL CM* (HY | BA) CM* [^CM CB];
 162
 163 LB21.1:      . CM* [BA HY NS];
 164 LB21.2:      BB CM* [^CM CB];
 165
 166 LB21b:       SY CM* HL;
 167
 168 LB22.1:        (AL | HL | CM) CM* IN;   # The CM is from LB10, treat an unattached CM as AL.
 169 LB22.2:       EX CM* IN;
 170 LB22.3:       (ID | EB | EM) CM* IN;
 171 LB22.4:       IN CM* IN;
 172 LB22.5:       NU CM* IN;
 173
 174 LB23.1:      (AL | HL | CM) CM* NU;
 175 LB23.2:      NU CM* (AL | HL);
 176
 177 LB23a.1:     PR CM* (ID | EB | EM);
 178 LB23a.2:     (ID | EB | EM) CM* PO;
 179
 180 LB24.2:      (PR | PO) CM* (AL | HL);
 181 LB24.3:      (AL | HL | CM) CM* (PR | PO);
 182
 183
 184 LB26.1:      JL CM* (JL | JV | H2 | H3);
 185 LB26.2:      (JV | H2) CM* (JV | JT);
 186 LB26.3:      (JT | H3) CM* JT;
 187
 188 LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
 189 LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
 190 LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
 191
 192 # LB28 Do not break between Alphabetics.
 193 #      Unattached (leading) CM treated as AL.
 194 LB28:        (AL | HL | CM)CM* (AL | HL);
 195
 196 LB29:        IS CM* (AL | HL);
 197
 198 # LB30  is adjusted for unattached leading CM being treated as AL.
 199 LB30.1:      (AL | CM | HL | NU) CM* OP;
 200 LB30.2:      CP CM* (AL | HL | NU);
 201
 202 # LB30a  keep pairs of RI together.
 203 LB30a.1:     RI CM* RI         ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS CM];
 204 LB30a.2:     RI CM* RI CM* CMS ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS CM];
 205 LB30a.3:     RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS ZWJ]?;
 206
 207 # LB30b Do not break between Emoji Base and Emoji Modifier
 208 LB30b:       EB CM* EM;
 209
 210 # LB31 Break Everywhere Else.
 211 #      Include combining marks
 212 LB31.1:        . CM* ZWJ [^CM];
 213 LB31.2:        . CM* ÷;