icuSources/test/testdata/break_rules/word.txt

   1 #
   2 # Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved.
   3
   4 # file: word.txt
   5 #
   6 # Reference Word 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 = word;      # one of grapheme | word | line | sentence
  13 locale = en;
  14
  15 E_Base = [\u261D\u26F9\u270A-\u270D\U0001F385\U0001F3C2-\U0001F3C4\U0001F3C7\U0001F3CA-\U0001F3CC\U0001F442-\U0001F443\U0001F446-\U0001F450\U0001F466-\U0001F478\U0001F47C\U0001F481-\U0001F483\U0001F485-\U0001F487\U0001F4AA\U0001F574-\U0001F575\U0001F57A\U0001F590\U0001F595-\U0001F596\U0001F645-\U0001F647\U0001F64B-\U0001F64F\U0001F6A3\U0001F6B4-\U0001F6B6\U0001F6C0\U0001F6CC\U0001F918-\U0001F91E\U0001F926\U0001F930\U0001F933-\U0001F939\U0001F93C-\U0001F93E];
  16 E_Modifier = [\U0001F3FB-\U0001F3FF];
  17 ZWJ = [\u200D];
  18 GAZ = [\u2640\u2642\u2764\u2695-\u2696\u2708\U0001F308\U0001F33E\U0001F373\U0001F393\U0001F3A4\U0001F3A8\U0001F3EB\U0001F3ED\U0001F466-\U0001F469\U0001F48B\U0001F4BB-\U0001F4BC\U0001F527\U0001F52C\U0001F5E8\U0001F680\U0001F692];
  19
  20 CR                 = [\p{Word_Break = CR}];
  21 LF                 = [\p{Word_Break = LF}];
  22 Newline            = [\p{Word_Break = Newline}];
  23 Extend             = [[[\p{Word_Break = Extend}][:Block=Tags:]]-ZWJ];
  24 Regional_Indicator = [\p{Word_Break = Regional_Indicator}];
  25 Format             = [[\p{Word_Break = Format}]-[:Block=Tags:]];
  26 Katakana           = [\p{Word_Break = Katakana}];
  27 Hebrew_Letter      = [\p{Word_Break = Hebrew_Letter}];
  28 ALetter            = [\p{Word_Break = ALetter}];
  29 Single_Quote       = [\p{Word_Break = Single_Quote}];
  30 Double_Quote       = [\p{Word_Break = Double_Quote}];
  31 MidNumLet          = [\p{Word_Break = MidNumLet}];
  32 MidLetter          = [\p{Word_Break = MidLetter} - [\:]];
  33 MidNum             = [\p{Word_Break = MidNum}];
  34 Numeric            = [\p{Word_Break = Numeric}];
  35 ExtendNumLet       = [\p{Word_Break = ExtendNumLet}];
  36
  37 #define dicitionary, with the effect being that those characters don't appear in test data.
  38
  39 Han            = [:Han:];
  40 Hiragana       = [:Hiragana:];
  41
  42 Control        = [\p{Grapheme_Cluster_Break = Control}];
  43 HangulSyllable = [\uac00-\ud7a3];
  44 ComplexContext = [:LineBreak = Complex_Context:];
  45 KanaKanji      = [Han Hiragana Katakana];
  46 dictionaryCJK  = [KanaKanji HangulSyllable];
  47 dictionary     = [ComplexContext dictionaryCJK];
  48
  49 # leave CJK scripts out of ALetterPlus
  50 #   Tricky. Redfine a set.
  51 #   For tailorings, if it modifies itself, do at end of sets ????
  52 #   Tweak redefine to mean replace existing definition at its original location.
  53 #   Insert defs without redefine just after last pre-existing def of that name.
  54 #   Maybe drop redefine, add warning for sets defined and not used, should catch typos.
  55
  56 ALetter        = [ALetter - dictionary];
  57
  58 AHLetter       = [ALetter  Hebrew_Letter];
  59 MidNumLetQ     = [MidNumLet  Single_Quote];
  60 ExtFmt         = [Extend Format ZWJ];
  61
  62 WB3:   CR LF;
  63 WB3a:  (Newline | CR | LF) ÷;
  64 WB3b:  . ÷ (Newline | CR | LF);   # actually redundant? No other rule combines.
  65                                   # (but needed with UAX treat-as scheme.)
  66 WB3c:   ZWJ GAZ;
  67
  68 WB5:    AHLetter ExtFmt* AHLetter;
  69
  70 # includes both WB6 and WB7
  71 WB6:    AHLetter ExtFmt* (MidLetter | MidNumLetQ) ExtFmt*  AHLetter;
  72
  73 WB7a:   Hebrew_Letter ExtFmt* Single_Quote;
  74 WB7b:   Hebrew_Letter ExtFmt* Double_Quote ExtFmt* Hebrew_Letter;   # Include WB7c
  75
  76 WB8:    Numeric ExtFmt* Numeric;
  77 WB9:    AHLetter ExtFmt* Numeric;
  78 WB10:   Numeric ExtFmt* AHLetter;
  79
  80 WB11:   Numeric ExtFmt* (MidNum | MidNumLetQ) ExtFmt* Numeric;    # includes WB12
  81 WB13:   Katakana ExtFmt* Katakana;
  82
  83 WB13a:  (AHLetter | Numeric | Katakana | ExtendNumLet) ExtFmt* ExtendNumLet;
  84 WB13b:  ExtendNumLet ExtFmt* (AHLetter | Numeric | Katakana);
  85
  86 # WB rule 13c, pairs of Regional Indicators stay unbroken.
  87 #              Interacts with WB3c.
  88 WB13c.1:  Regional_Indicator ExtFmt* Regional_Indicator ExtFmt* ZWJ GAZ;
  89 WB13c.2:  Regional_Indicator ExtFmt* Regional_Indicator ExtFmt* ÷;
  90
  91 WB13d:  (E_Base | GAZ) ExtFmt* E_Modifier;
  92
  93 # Rule WB 14   Any ÷ Any
  94 #    Interacts with WB3c, do not break between ZWJ and GAZ.
  95 WB14.1: . ExtFmt* ZWJ GAZ;
  96 WB14.2: . ExtFmt* ÷;
  97