icuSources/test/testdata/break_rules/line.txt

   1 #
   2 # Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved.
   3
   4 # file: line.txt
   5 #
   6 # Reference Line Break rules for intltest rbbi/RBBIMonkeyTest
   7 #
   8 # Note: Rule syntax and the monkey test itself are still a work in progress.
   9 #       They are expected to change with review and the addition of support for rule tailoring.
  10
  11
  12 type = line;
  13 locale = en;
  14
  15
  16 AI = [[:LineBreak =  Ambiguous:] - [\u2640\u2642]];
  17 AL = [[:LineBreak =  Alphabetic:]-[\u2764]];
  18 BA = [:LineBreak =  Break_After:];
  19 BB = [:LineBreak =  Break_Before:];
  20 BK = [:LineBreak =  Mandatory_Break:];
  21 B2 = [:LineBreak =  Break_Both:];
  22 CB = [:LineBreak =  Contingent_Break:];
  23 CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  24 CL = [:LineBreak =  Close_Punctuation:];
  25 CM = [[:LineBreak =  Combining_Mark:] \u200D];
  26 CP = [:LineBreak =  Close_Parenthesis:];
  27 CR = [:LineBreak =  Carriage_Return:];
  28
  29 EB = [\u261D\u26F9\u270A-\u270D\U0001F385\U0001F3C3-\U0001F3C4\U0001F3CA-\U0001F3CB\U0001F442-\U0001F443\U0001F446-\U0001F450\U0001F466-\U0001F469\U0001F46E\U0001F470-\U0001F478\U0001F47C\U0001F481-\U0001F483\U0001F485-\U0001F487\U0001F4AA\U0001F575\U0001F590\U0001F595-\U0001F596\U0001F645-\U0001F647\U0001F64B-\U0001F64F\U0001F6A3\U0001F6B4-\U0001F6B6\U0001F6C0\U0001F918];
  30 EM = [\U0001F3FB-\U0001F3FF];
  31
  32 EX = [:LineBreak =  Exclamation:];
  33 GL = [:LineBreak =  Glue:];
  34 HL = [:LineBreak =  Hebrew_Letter:];
  35 HY = [:LineBreak =  Hyphen:];
  36 H2 = [:LineBreak =  H2:];
  37 H3 = [:LineBreak =  H3:];
  38 ID = [[:LineBreak =  Ideographic:][\u2640\u2642\u2764]];
  39 IN = [:LineBreak =  Inseperable:];
  40 IS = [:LineBreak =  Infix_Numeric:];
  41 JL = [:LineBreak =  JL:];
  42 JV = [:LineBreak =  JV:];
  43 JT = [:LineBreak =  JT:];
  44 LF = [:LineBreak =  Line_Feed:];
  45 NL = [:LineBreak =  Next_Line:];
  46 NS = [[:LineBreak =  Nonstarter:] CJ];
  47 NU = [:LineBreak =  Numeric:];
  48 OP = [:LineBreak =  Open_Punctuation:];
  49 PO = [:LineBreak =  Postfix_Numeric:];
  50 PR = [:LineBreak =  Prefix_Numeric:];
  51 QU = [:LineBreak =  Quotation:];
  52 RI = [:LineBreak =  Regional_Indicator:];
  53 SA = [:LineBreak =  Complex_Context:];
  54 SG = [:LineBreak =  Surrogate:];
  55 SP = [:LineBreak =  Space:];
  56 SY = [:LineBreak =  Break_Symbols:];
  57 WJ = [:LineBreak =  Word_Joiner:];
  58 XX = [:LineBreak =  Unknown:];
  59 ZW = [:LineBreak =  ZWSpace:];
  60 ZJ = [\u200D];
  61
  62 # TODO: adjustment to sets needed only until Unicode properties are updated for Emoji.
  63 ID   = [ID - EB];
  64 AL   = [AL - EM];
  65
  66 dictionary = [:LineBreak = Complex_Context:];
  67
  68 # Redfine AL. LB1.  TODO: refine according to latest UAX.
  69 AL = [ AL AI SA SG XX ];
  70
  71 LB4:        BK ÷;
  72 LB5:        CR LF;
  73 LB5.1:      CR ÷;
  74 LB5.2:      LF ÷;
  75 LB5.3:      NL ÷;
  76
  77 LB6:        . (BK | CR | LF | NL);
  78 LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
  79
  80 # Rules LB14 - LB17.
  81 # Moved before LB7, because they can match a longer sequence that would also match LB7,
  82 # for example, the sequence "OP CM SP AL" matches LB14 while the prefix of it,
  83 #                           "while only the prefix "OP CM SP" matches LB7.1
  84 LB14:        OP CM* SP* .;
  85 LB15:        QU CM* SP* OP;
  86 LB16:        (CL | CP)CM* SP* NS;
  87 LB17:        B2 CM* SP* B2;
  88
  89 LB7.1:      [^ZW SP] CM* [SP ZW];
  90 LB7.2:      [ZW SP] [SP ZW];
  91
  92 # LB8, ICU differs from UAX-14,
  93 #    ICU:    ZW ÷;
  94 #    UAX 14: ZW SP* ÷;
  95 LB8:        ZW ÷;
  96
  97 # LB8a, from Emoji proposal L2/16-011R3
  98 #      ZWJ x ID
  99 LB8a:       ZJ (ID | EB | EM);
 100
 101
 102 # LB9:  X CM -> X
 103 # LB10: Unattached CM -> AL
 104
 105 #LB11:       × WJ;
 106 #            WJ ×
 107
 108 LB11.1:      [^BK CR LF NL SP ZW] CM* WJ;
 109 LB11.2:      SP WJ;
 110 LB11.3:      WJ CM* [^CM];
 111
 112 LB12:        GL CM* [^CM];
 113
 114 LB12a:       [^SP BA HY] CM* GL;
 115
 116 # LB 13 ICU Tailoring, matches tailoring exmaple 8 from UAX 14.
 117 #
 118 #   LB13.1   [^SP] CM* [CL CP EX IS SY]    # original UAX 14 rule.
 119 #   LB13.2   SP    CM* [CL CP EX IS SY]
 120
 121 LB13.1: [^NU SP] CM* [CL CP IS SY];
 122 LB13.2: [^SP] CM* EX;
 123 LB13.2: SP [CL CP EX IS SY];
 124
 125
 126 # LB 14-17 are moved above LB 7.
 127
 128 LB18:        SP ÷;
 129
 130 LB19:        . CM* QU;
 131 LB19.1:      QU CM* [^CM];
 132
 133 # LB 20   Break before and after CB.
 134 #         Interaction with LB8a:  ZJ x ID is tricky because CM includes ZJ.
 135 #                                 ZJ acts like a CM to the left, combining with CB.
 136 #                                 ZJ acts independently to the right, no break from ID by LB8a.
 137 LB20:        . CM* ÷ CB;
 138 LB20.1a:     CB CM* ZJ (ID | EB | EM);
 139 LB20.1b:      CB CM* ÷;
 140
 141 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
 142 #       not picking up the continuing match after the BA from 21a.
 143 LB21a:       HL CM* (HY | BA) CM* [^CM CB];
 144
 145 LB21.1:      . CM* [BA HY NS];
 146 LB21.2:      BB CM* [^CM CB];
 147
 148 LB21b:       SY CM* HL;
 149
 150 LB22.1:        (AL | HL | CM) CM* IN;   # The CM is from LB10, treat an unattached CM as AL.
 151 LB22.2:       EX CM* IN;
 152 LB22.3:       (ID | EB | EM) CM* IN;
 153 LB22.4:       IN CM* IN;
 154 LB22.5:       NU CM* IN;
 155
 156 LB23.1:      (ID | EB | EM) CM* PO;
 157 LB23.2:      (AL | HL | CM) CM* NU;
 158 LB23.3:      NU CM* (AL | HL);
 159
 160 LB24.1:      PR CM* (ID | EB | EM);
 161 LB24.2:      PR CM* (AL | HL);
 162 LB24.3:      PO CM* (AL | HL);
 163 LB24.4:      (AL | HL | CM) CM* (PR | PO);
 164
 165 # Numbers. Equivalent to Tailoring example 8 from UAX 14.
 166 LB25:        ((PR | PO)CM*)? ((OP | HY)CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?;
 167
 168 LB26.1:      JL CM* (JL | JV | H2 | H3);
 169 LB26.2:      (JV | H2) CM* (JV | JT);
 170 LB26.3:      (JT | H3) CM* JT;
 171
 172 LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
 173 LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
 174 LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
 175
 176 # LB28 Do not break between Alphabetics.
 177 #      Unattached (leading) CM treated as AL.
 178 LB28:        (AL | HL | CM)CM* (AL | HL);
 179
 180 LB29:        IS CM* (AL | HL);
 181
 182 # LB30  is adjusted for unattached leading CM being treated as AL.
 183 LB30.1:      (AL | CM | HL | NU) CM* OP;
 184 LB30.2:      CP CM* (AL | HL | NU);
 185
 186 # LB31  keep pairs of RI together.
 187 LB30a.1:     RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS];
 188 LB30a.2:     RI CM* RI CM* ZJ (ID | EB | EM);
 189 LB30a.3:     RI CM* RI CM* ÷;
 190
 191 # LB30b Do not break between Emoji Base and Emoji Modifier
 192 LB30b:       EB CM* EM;
 193
 194 # LB31 Break Everywhere Else.
 195 #      Include combining marks
 196 LB31.1:        . CM* ZJ (ID | EB | EM);
 197 LB31.2:        . CM* ÷;