1# 2# Copyright (C) 2002-2003, International Business Machines Corporation and others. 3# All Rights Reserved. 4# 5# file: count_word.txt 6# 7# ICU Word Break Rules 8# See Unicode Standard Annex #29. 9# These rules are based on Version 4.0.0, dated 2003-04-17 10# 11 12 13 14#################################################################################### 15# 16# Character class definitions from TR 29 17# 18#################################################################################### 19$Katakana = [[:Script = KATAKANA:] [:name = KATAKANA-HIRAGANA PROLONGED SOUND MARK:] 20 [:name = HALFWIDTH KATAKANA-HIRAGANA PROLONGED SOUND MARK:] 21 [:name = HALFWIDTH KATAKANA VOICED SOUND MARK:] 22 [:name = HALFWIDTH KATAKANA SEMI-VOICED SOUND MARK:]]; 23 24$ALetter = [[:Alphabetic:] [:name= HEBREW PUNCTUATION GERESH:] [:name = HYPHEN-MINUS:] 25 - $Katakana 26 - [:Script = Thai:] 27 - [:Script = Lao:] 28 - [:Script = Hiragana:]]; 29 30$MidLetter = [[:name = APOSTROPHE:] [:name = MIDDLE DOT:] [:name = HEBREW PUNCTUATION GERSHAYIM:] 31 [:name = RIGHT SINGLE QUOTATION MARK:] [:name = HYPHENATION POINT:] [:name = COLON:]]; 32 33$MidNumLet = [:name = FULL STOP:]; 34 35$MidNum = [[:LineBreak = Infix_Numeric:] - $MidNumLet]; 36$Numeric = [:LineBreak = Numeric:]; 37 38 39$TheZWSP = \u200b; 40 41# 42# Character Class Definitions. 43# The names are those from TR29. 44# 45$CR = \u000d; 46$LF = \u000a; 47$Control = [[[:Zl:] [:Zp:] [:Cc:] [:Cf:]] - $TheZWSP]; 48$Extend = [[:Grapheme_Extend = TRUE:]]; 49 50 51 52 53#################################################################################### 54# 55# Word Break Rules. Definitions and Rules specific to word break begin Here. 56# 57#################################################################################### 58 59$Format = [[:Cf:] - $TheZWSP]; 60 61 62 63# Rule 3: Treat a grapheme cluster as if it were a single character. 64# Hangul Syllables are easier to deal with here than they are in Grapheme Clusters 65# because we don't need to find the boundaries between adjacent syllables - 66# they won't be word boundaries. 67# 68 69 70# 71# "Extended" definitions. Grapheme Cluster + Format Chars, treated like the base char. 72# 73$ALetterEx = $ALetter $Extend*; 74$NumericEx = $Numeric $Extend*; 75$MidNumEx = $MidNum $Extend*; 76$MidNumLetEx = $MidNumLet $Extend*; 77$MidLetterEx = $MidLetter $Extend*; 78$KatakanaEx = $Katakana $Extend*; 79$FormatEx = $Format $Extend*; 80 81$word_pad=[[:P:][:S:][:Z:][:C:]]; 82 83# 84# Numbers. Rules 8, 11, 12 form the TR. 85# 86$NumberSequence = $NumericEx ($FormatEx* ($MidNumEx | $MidNumLetEx)? $FormatEx* $NumericEx)*; 87$NumberSequence $word_pad* {100}; 88 89# 90# Words. Alpha-numerics. Rule 5, 6, 7, 9, 10 91# - must include at least one letter. 92# - may include both letters and numbers. 93# - may include MideLetter, MidNumber punctuation. 94# 95$LetterSequence = $ALetterEx ($FormatEx* ($MidLetterEx | $MidNumLetEx)? $FormatEx* $ALetterEx)*; # rules #6, #7 96($NumberSequence $FormatEx*)? $LetterSequence ($FormatEx* ($NumberSequence | $LetterSequence))* $word_pad* {200}; 97 98# 99# Do not break between Katakana. Rule #13. 100# 101$KatakanaEx ($FormatEx* $KatakanaEx)* {300}; 102[:Hiragana:] $Extend* {300}; 103 104# 105# Ideographic Characters. Stand by themselves as words. 106# Separated from the "Everything Else" rule, below, only so that they 107# can be tagged with a return value. TODO: is this what we want? 108# 109# [:IDEOGRAPHIC:] $Extend* $word_pad* {400}; 110 111# 112# Everything Else, with no tag. 113# Non-Control chars combine with $Extend (combining) chars. 114# Controls are do not. 115# 116[^$Control [:Ideographic:]] $Extend* $word_pad*; 117$CR $LF; 118 119# 120# Reverse Rules. Back up over any of the chars that can group together. 121# (Reverse rules do not need to be exact; they can back up too far, 122# but must back up at least enough, and must stop on a boundary.) 123# 124 125# NonStarters are the set of all characters that can appear at the 2nd - nth position of 126# a word. (They may also be the first.) The reverse rule skips over these, until it 127# reaches something that can only be the start (and probably only) char in a "word". 128# A space or punctuation meets the test. 129# 130$NonStarters = [$Numeric $ALetter $Katakana $MidLetter $MidNum $MidNumLet $Extend $Format]; 131 132#!.*; 133! ($NonStarters* | \n \r) .; 134 135