icuSources/test/testdata/break_rules/line_loose_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_loose_cj.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 #         Line Breaking Rules
  15 #         Implement default line breaking as defined by
  16 #         Unicode Standard Annex #14 Revision 34 for Unicode 8.0
  17 #         http://www.unicode.org/reports/tr14/
  18 #         tailored as noted in 2nd paragraph below..
  19 #
  20 #         This tailors the line break behavior to correspond to CSS
  21 #         line-break=loose (BCP47 -u-lb-loose) as defined for Chinese & Japanese.
  22 #         It sets characters of class CJ to behave like ID.
  23 #         In addition, it allows breaks:
  24 #         * before hyphens 2010 & 2013 (both BA) and 301C, 30A0 (both NS)
  25 #         * before iteration marks 3005, 303B, 309D, 309E, 30FD, 30FE (all NS)
  26 #         * between characters of LineBreak class IN such as 2026
  27 #         * before some centered punct 203C, 2047, 2048, 2049, 30FB, FF1A, FF1B,
  28 #           FF65 (all NS) and FF01, FF1F (both EX).
  29 #         * before suffix characters with LineBreak class PO and EastAsianWidth A,F,W;
  30 #           this includes: 00B0 2030 2032 2033 2035 2103 2109 FE6A FF05 FFE0
  31 #         * after prefix characters with LineBreak class PR and EastAsianWidth A,F,W;
  32 #           this includes: 00A4 00B1 20AC 2116 FE69 FF04 FFE1 FFE5 FFE6
  33 #         It allows breaking before 201C and after 201D, for zh_Hans, zh_Hant, and ja.
  34
  35
  36 type = line;
  37 locale = ja@lb=loose;
  38
  39
  40 AI = [:LineBreak =  Ambiguous:];
  41 AL = [:LineBreak =  Alphabetic:];
  42 BAX = [\u2010 \u2013];
  43 BA = [[:LineBreak =  Break_After:] - BAX];
  44 HH = [\u2010];      # \u2010 is HYPHEN, default line break is BA.
  45 BB = [:LineBreak =  Break_Before:];
  46 BK = [:LineBreak =  Mandatory_Break:];
  47 B2 = [:LineBreak =  Break_Both:];
  48 CB = [:LineBreak =  Contingent_Break:];
  49 CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  50 CL = [[:LineBreak =  Close_Punctuation:] \u201d];
  51 CMS = [:LineBreak =  Combining_Mark:];
  52 CP = [:LineBreak =  Close_Parenthesis:];
  53 CR = [:LineBreak =  Carriage_Return:];
  54 EB = [[:LineBreak =  EB:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
  55 EM = [:LineBreak =  EM:];
  56 EXX = [\uFF01 \uFF1F];
  57 EX = [[:LineBreak =  Exclamation:] - EXX];
  58 GL = [:LineBreak =  Glue:];
  59 HL = [:LineBreak =  Hebrew_Letter:];
  60 HY = [:LineBreak =  Hyphen:];
  61 H2 = [:LineBreak =  H2:];
  62 H3 = [:LineBreak =  H3:];
  63 ID = [[:LineBreak =  Ideographic:] CJ];  # CSS Loose tailoring: CJ resolves to ID
  64 IN = [:LineBreak =  Inseperable:];
  65 IS = [:LineBreak =  Infix_Numeric:];
  66 JL = [:LineBreak =  JL:];
  67 JV = [:LineBreak =  JV:];
  68 JT = [:LineBreak =  JT:];
  69 LF = [:LineBreak =  Line_Feed:];
  70 NL = [:LineBreak =  Next_Line:];
  71 NSX = [\u301C \u30A0 \u3005 \u303B \u309D \u309E \u30FD \u30FE \u203C \u2047 \u2048 \u2049 \u30FB \uFF1A \uFF1B \uFF65];
  72 NS = [[:LineBreak =  Nonstarter:] - NSX];
  73 NU = [:LineBreak =  Numeric:];
  74 OP = [[:LineBreak =  Open_Punctuation:] \u201c];
  75 POX = [\u00B0 \u2030 \u2032 \u2033 \u2035 \u2103 \u2109 \uFE6A \uFF05 \uFFE0];
  76 PO = [[:LineBreak =  Postfix_Numeric:] - POX];
  77 PRX = [\u00A4 \u00B1 \u20AC \u2116 \uFE69 \uFF04 \uFFE1 \uFFE5 \uFFE6];
  78 PR = [[:LineBreak =  Prefix_Numeric:] - PRX];
  79 QU = [[:LineBreak =  Quotation:] - [\u201c\u201d]];
  80 RI = [:LineBreak =  Regional_Indicator:];
  81 SA = [:LineBreak =  Complex_Context:];
  82 SG = [:LineBreak =  Surrogate:];
  83 SP = [:LineBreak =  Space:];
  84 SY = [:LineBreak =  Break_Symbols:];
  85 WJ = [:LineBreak =  Word_Joiner:];
  86 XX = [:LineBreak =  Unknown:];
  87 ZW = [:LineBreak =  ZWSpace:];
  88 ZWJ = [:LineBreak =  ZWJ:];
  89
  90 # LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
  91 AL = [AL AI SG XX ];
  92 dictionary = SA;
  93
  94 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
  95 #         list it in the numerous rules that use CM.
  96 CM = [CMS ZWJ];
  97
  98 LB4:        BK ÷;
  99 LB5:        CR LF;
 100 LB5.1:      CR ÷;
 101 LB5.2:      LF ÷;
 102 LB5.3:      NL ÷;
 103
 104 LB6:        . (BK | CR | LF | NL);
 105 LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
 106
 107 # LB8   break after ZW SP*.
 108 # Precedes LB7 because both rules will match the sequences like ZW SP,
 109 # and LB8 must take precedence.
 110
 111 LB8:        ZW SP* ÷ [^ZW SP BK CR LF NL];
 112
 113 # Numbers. Equivalent to Tailoring example 8 from UAX 14.
 114 # Moved up, before LB14, because it can match longer sequences which must take precedence.
 115 #          Loose_cj tailoring: do not include $PRX at the beginning or $POX at the end.
 116 LB25:        ((PR | PO | POX)CM*)? ((OP | HY)CM*)? (IS CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PRX | PO))?;
 117
 118 # Rules LB14 - LB17.
 119 # Moved up, before LB7, because they can match a longer sequence that would also match LB7.
 120 # For example, the sequence "OP CM SP AL" matches LB14
 121 # while the prefix of it, "OP CM SP" matches LB7.1
 122 LB14:        OP CM* SP* .;
 123
 124 # LB 14a Break before an IS that begins a number and follows a space.
 125 LB14a:       SP ÷ IS CM* NU;
 126
 127 # LB14b       × IS
 128 LB14b.1:     [^SP] CM* IS;
 129 LB14b.2:     SP IS;
 130
 131 LB15:        QU CM* SP* OP;
 132 LB16:        (CL | CP)CM* SP* NS;
 133 LB17:        B2 CM* SP* B2;
 134
 135
 136 # LB7 Do not break before spaces or zero width space.
 137
 138 LB7.1:      [^ZW SP] CM* [SP ZW];
 139 LB7.2:      [ZW SP] [SP ZW];
 140
 141 # LB8a
 142 #      ZWJ x
 143 #      Don't match a CM on the right - let other rules pick up CM sequences, where
 144 #      the ZWJ behaves as just another generic CM.
 145 LB8a:       ZWJ [^CM];
 146
 147
 148 # LB9:  X CM -> X
 149 # LB10: Unattached CM -> AL
 150
 151 #LB11:       × WJ;
 152 #            WJ ×
 153
 154 LB11.1:      [^SP] CM* WJ;
 155 LB11.2:      SP WJ;
 156 LB11.3:      WJ CM* [^CM];
 157
 158 LB12:        GL CM* [^CM];
 159
 160 LB12a:       [^SP BA BAX HY] CM* GL;
 161
 162 # LB 13 Do not break before ‘]’ or ‘!’ or ‘/’, even after spaces.
 163 LB13.1: [^SP] CM* [CL CP EX SY];
 164 LB13.2: SP [CL CP EX SY];
 165
 166
 167 # LB 14-17 are moved above LB 7.
 168
 169 LB18:        SP ÷;
 170
 171 LB19:        . CM* QU;
 172 LB19.1:      QU CM* [^CM];
 173
 174 # LB 20   Break before and after CB.
 175 #         Interaction with LB8a:  ZWJ x . is tricky because CM includes ZWJ.
 176 #                                 ZWJ acts like a CM to the left, combining with CB.
 177 #                                 ZWJ acts independently to the right, no break after by LB8a.
 178 LB20.1:      . CM* ZWJ CB;
 179 LB20.2:      . CM* ÷ CB;
 180
 181 LB20.3:      CB CM* ZWJ [^CM];
 182 LB20.4:      CB CM* ÷;
 183
 184 # LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
 185 LB20.09:     ^(HY | HH) CM* AL;
 186
 187 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
 188 #       not picking up the continuing match after the BA from 21a.
 189 # LB 21a Don't break after Hebrew + Hyphen
 190 #   HL (HY | BA) x
 191
 192 LB21a:       HL CM* (HY | BA | BAX) CM* [^CM CB]?;
 193
 194 LB21.1:      . CM* [BA HY NS];
 195 LB21.2:      BB CM* [^CM CB];
 196
 197 LB21b:       SY CM* HL;
 198
 199 LB22.1:        (AL | HL | CM) CM* IN;   # The CM is from LB10, treat an unattached CM as AL.
 200 LB22.2:       EX CM* IN;
 201 LB22.3:       (ID | EB | EM) CM* IN;
 202 # LB22.4:       IN CM* IN;  # delete this rule for CSS loose.
 203 LB22.5:       NU CM* IN;
 204
 205 LB23.1:      (AL | HL | CM) CM* NU;
 206 LB23.2:      NU CM* (AL | HL);
 207
 208 LB23a.1:     PR CM* (ID | EB | EM);
 209 LB23a.2:     (ID | EB | EM) CM* PO;
 210
 211 LB24.2:      (PR | PO | POX) CM* (AL | HL);
 212 LB24.3:      (AL | HL | CM) CM* (PR | PO | POX);
 213
 214
 215 LB26.1:      JL CM* (JL | JV | H2 | H3);
 216 LB26.2:      (JV | H2) CM* (JV | JT);
 217 LB26.3:      (JT | H3) CM* JT;
 218
 219 LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
 220 LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
 221 LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
 222
 223 # LB28 Do not break between Alphabetics.
 224 #      Unattached (leading) CM treated as AL.
 225 LB28:        (AL | HL | CM)CM* (AL | HL);
 226
 227 LB29:        IS CM* (AL | HL);
 228
 229 # LB30  is adjusted for unattached leading CM being treated as AL.
 230 LB30.1:      (AL | CM | HL | NU) CM* OP;
 231 LB30.2:      CP CM* (AL | HL | NU);
 232
 233 # LB30a  keep pairs of RI together.
 234 LB30a.1:     RI CM* RI         ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS CM];
 235 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];
 236 LB30a.3:     RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS ZWJ]?;
 237
 238 # LB30b Do not break between Emoji Base and Emoji Modifier
 239 LB30b:       EB CM* EM;
 240
 241 # LB31 Break Everywhere Else.
 242 #      Include combining marks
 243 LB31.1:        . CM* ZWJ [^CM];
 244 LB31.2:        . CM* ÷;