icuSources/test/testdata/break_rules/line_loose_cj.txt

   1 #
   2 # Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved.
   3 #
   4 #  file:  line_loose_cj.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 #         Line Breaking Rules
  12 #         Implement default line breaking as defined by
  13 #         Unicode Standard Annex #14 Revision 34 for Unicode 8.0
  14 #         http://www.unicode.org/reports/tr14/
  15 #         tailored as noted in 2nd paragraph below..
  16 #
  17 #         This tailors the line break behavior to correspond to CSS
  18 #         line-break=loose (BCP47 -u-lb-loose) as defined for Chinese & Japanese.
  19 #         It sets characters of class CJ to behave like ID.
  20 #         In addition, it allows breaks:
  21 #         * before hyphens 2010 & 2013 (both BA) and 301C, 30A0 (both NS)
  22 #         * before iteration marks 3005, 303B, 309D, 309E, 30FD, 30FE (all NS)
  23 #         * between characters of LineBreak class IN such as 2026
  24 #         * before some centered punct 203C, 2047, 2048, 2049, 30FB, FF1A, FF1B,
  25 #           FF65 (all NS) and FF01, FF1F (both EX).
  26 #         * before suffix characters with LineBreak class PO and EastAsianWidth A,F,W;
  27 #           this includes: 00B0 2030 2032 2033 2035 2103 2109 FE6A FF05 FFE0
  28 #         * after prefix characters with LineBreak class PR and EastAsianWidth A,F,W;
  29 #           this includes: 00A4 00B1 20AC 2116 FE69 FF04 FFE1 FFE5 FFE6
  30
  31
  32 type = line;
  33 locale = ja@lb=loose;
  34
  35
  36 AI = [:LineBreak =  Ambiguous:];
  37 AL = [[:LineBreak =  Alphabetic:]-[\u2764]];
  38 BAX = [\u2010 \u2013];
  39 BA = [[:LineBreak =  Break_After:] - BAX];
  40 BB = [:LineBreak =  Break_Before:];
  41 BK = [:LineBreak =  Mandatory_Break:];
  42 B2 = [:LineBreak =  Break_Both:];
  43 CB = [:LineBreak =  Contingent_Break:];
  44 CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  45 CL = [:LineBreak =  Close_Punctuation:];
  46 CM = [:LineBreak =  Combining_Mark:];
  47 CP = [:LineBreak =  Close_Parenthesis:];
  48 CR = [:LineBreak =  Carriage_Return:];
  49
  50 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];
  51 EM = [\U0001F3FB-\U0001F3FF];
  52
  53 EXX = [\uFF01 \uFF1F];
  54 EX = [[:LineBreak =  Exclamation:] - EXX];
  55 GL = [:LineBreak =  Glue:];
  56 HL = [:LineBreak =  Hebrew_Letter:];
  57 HY = [:LineBreak =  Hyphen:];
  58 H2 = [:LineBreak =  H2:];
  59 H3 = [:LineBreak =  H3:];
  60 ID = [[:LineBreak =  Ideographic:][\u2764]CJ];
  61 IN = [:LineBreak =  Inseperable:];
  62 IS = [:LineBreak =  Infix_Numeric:];
  63 JL = [:LineBreak =  JL:];
  64 JV = [:LineBreak =  JV:];
  65 JT = [:LineBreak =  JT:];
  66 LF = [:LineBreak =  Line_Feed:];
  67 NL = [:LineBreak =  Next_Line:];
  68 NSX = [\u301C \u30A0 \u3005 \u303B \u309D \u309E \u30FD \u30FE \u203C \u2047 \u2048 \u2049 \u30FB \uFF1A \uFF1B \uFF65];
  69 NS = [[:LineBreak =  Nonstarter:] - NSX];
  70 NU = [:LineBreak =  Numeric:];
  71 OP = [:LineBreak =  Open_Punctuation:];
  72 POX = [\u00B0 \u2030 \u2032 \u2033 \u2035 \u2103 \u2109 \uFE6A \uFF05 \uFFE0];
  73 PO = [[:LineBreak =  Postfix_Numeric:] - POX];
  74 PRX = [\u00A4 \u00B1 \u20AC \u2116 \uFE69 \uFF04 \uFFE1 \uFFE5 \uFFE6];
  75 PR = [[:LineBreak =  Prefix_Numeric:] - PRX];
  76 QU = [:LineBreak =  Quotation:];
  77 RI = [:LineBreak =  Regional_Indicator:];
  78 SA = [:LineBreak =  Complex_Context:];
  79 SG = [:LineBreak =  Surrogate:];
  80 SP = [:LineBreak =  Space:];
  81 SY = [:LineBreak =  Break_Symbols:];
  82 WJ = [:LineBreak =  Word_Joiner:];
  83 XX = [:LineBreak =  Unknown:];
  84 ZW = [:LineBreak =  ZWSpace:];
  85 ZJ = [\u200D];
  86
  87 # TODO: adjustment to sets needed only until Unicode properties are updated for Emoji.
  88 ID   = [ID - EB];
  89 AL   = [AL - EM];
  90
  91 dictionary = [:LineBreak = Complex_Context:];
  92
  93 # Redfine AL. LB1.  TODO: refine according to latest UAX.
  94 AL = [ AL AI SA SG XX ];
  95
  96 LB4:        BK ÷;
  97 LB5:        CR LF;
  98 LB5.1:      CR ÷;
  99 LB5.2:      LF ÷;
 100 LB5.3:      NL ÷;
 101
 102 LB6:        . (BK | CR | LF | NL);
 103 LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
 104
 105 # Rules LB14 - LB17.
 106 # Moved before LB7, because they can match a longer sequence that would also match LB7,
 107 # for example, the sequence "OP CM SP AL" matches LB14 while the prefix of it,
 108 #                           "while only the prefix "OP CM SP" matches LB7.1
 109 LB14:        OP CM* SP* .;
 110 LB15:        QU CM* SP* OP;
 111 LB16:        (CL | CP)CM* SP* NS;
 112 LB17:        B2 CM* SP* B2;
 113
 114 LB7.1:      [^ZW SP] CM* [SP ZW];
 115 LB7.2:      [ZW SP] [SP ZW];
 116
 117 # LB8, ICU differs from UAX-14,
 118 #    ICU:    ZW ÷;
 119 #    UAX 14: ZW SP* ÷;
 120 LB8:        ZW ÷;
 121
 122 # LB8a, from Emoji proposal L2/16-011R3
 123 #      ZWJ x ID
 124 LB8a:       ZJ (ID | EB | EM);
 125
 126
 127 # LB9:  X CM -> X
 128 # LB10: Unattached CM -> AL
 129
 130 #LB11:       × WJ;
 131 #            WJ ×
 132
 133 LB11.1:      [^BK CR LF NL SP ZW] CM* WJ;
 134 LB11.2:      SP WJ;
 135 LB11.3:      WJ CM* [^CM];
 136
 137 LB12:        GL CM* [^CM];
 138
 139 LB12a:       [^SP BA BAX HY] CM* GL;
 140
 141 # LB 13 ICU Tailoring, matches tailoring exmaple 8 from UAX 14.
 142 #
 143 #   LB13.1   [^SP] CM* [CL CP EX IS SY]    # original UAX 14 rule.
 144 #   LB13.2   SP    CM* [CL CP EX IS SY]
 145
 146 LB13.1: [^NU SP] CM* [CL CP IS SY];
 147 LB13.2: [^SP] CM* EX;
 148 LB13.2: SP [CL CP EX IS SY];
 149
 150
 151 # LB 14-17 are moved above LB 7.
 152
 153 LB18:        SP ÷;
 154
 155 LB19:        . CM* QU;
 156 LB19.1:      QU CM* [^CM];
 157
 158 # LB 20   Break before and after CB.
 159 #         Interaction with LB8a:  ZJ x ID is tricky because CM includes ZJ.
 160 #                                 ZJ acts like a CM to the left, combining with CB.
 161 #                                 ZJ acts independently to the right, no break from ID by LB8a.
 162 LB20:        . CM* ÷ CB;
 163 LB20.1a:     CB CM* ZJ (ID | EB | EM);
 164 LB20.1b:      CB CM* ÷;
 165
 166 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
 167 #       not picking up the continuing match after the BA from 21a.
 168 # LB 21a Don't break after Hebrew + Hyphen
 169 #   HL (HY | BA) x
 170
 171 LB21a:       HL CM* (HY | BA | BAX) CM* [^CM CB]?;
 172
 173 LB21.1:      . CM* [BA HY NS];
 174 LB21.2:      BB CM* [^CM CB];
 175
 176 LB21b:       SY CM* HL;
 177
 178 LB22.1:        (AL | HL | CM) CM* IN;   # The CM is from LB10, treat an unattached CM as AL.
 179 LB22.2:       EX CM* IN;
 180 LB22.3:       (ID | EB | EM) CM* IN;
 181 # LB22.4:       IN CM* IN;  # delete this rule for CSS loose.
 182 LB22.5:       NU CM* IN;
 183
 184 LB23.1:      (ID | EB | EM) CM* PO;
 185 LB23.2:      (AL | HL | CM) CM* NU;
 186 LB23.3:      NU CM* (AL | HL);
 187
 188 LB24.1:      PR CM* (ID | EB | EM);
 189 LB24.2:      PR CM* (AL | HL);
 190 LB24.3:      (PO | POX) CM* (AL | HL);
 191 LB24.4:      (AL | HL | CM) CM* (PR | PO | POX);
 192
 193 # Numbers. Equivalent to Tailoring example 8 from UAx 14.
 194 #          Loose_cj tailoring: do not include $PRX at the beginning or $POX at the end.
 195 LB25:        ((PR | PO | POX)CM*)? ((OP | HY)CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PRX | PO))?;
 196
 197 LB26.1:      JL CM* (JL | JV | H2 | H3);
 198 LB26.2:      (JV | H2) CM* (JV | JT);
 199 LB26.3:      (JT | H3) CM* JT;
 200
 201 LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
 202 LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
 203 LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
 204
 205 # LB28 Do not break between Alphabetics.
 206 #      Unattached (leading) CM treated as AL.
 207 LB28:        (AL | HL | CM)CM* (AL | HL);
 208
 209 LB29:        IS CM* (AL | HL);
 210
 211 # LB30  is adjusted for unattached leading CM being treated as AL.
 212 LB30.1:      (AL | CM | HL | NU) CM* OP;
 213 LB30.2:      CP CM* (AL | HL | NU);
 214
 215 # LB31  keep pairs of RI together.
 216 LB30a.1:     RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS];
 217 LB30a.2:     RI CM* RI CM* ZJ (ID | EB | EM);
 218 LB30a.3:     RI CM* RI CM* ÷;
 219
 220 # LB30b Do not break between Emoji Base and Emoji Modifier
 221 LB30b:       EB CM* EM;
 222
 223 # LB31 Break Everywhere Else.
 224 #      Include combining marks
 225 LB31.1:        . CM* ZJ (ID | EB | EM);
 226 LB31.2:        . CM* ÷;