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$Katakana = [[:Script = KATAKANA:] [:name = KATAKANA-HIRAGANA PROLONGED SOUND MARK:] 20 [:name = HALFWIDTH KATAKANA-HIRAGANA PROLONGED SOUND MARK:] 21 [:name = HALFWIDTH KATAKANA VOICED SOUND MARK:] 22 [:name = HALFWIDTH KATAKANA SEMI-VOICED SOUND MARK:]]; 23 24 25$dash = \u002d; 26 27$ALetter = [[:Alphabetic:] [:name= HEBREW PUNCTUATION GERESH:] 28 [:P:] [:S:] [:LineBreak = Numeric:] 29 - $dash 30 - $Katakana 31 - [:Script = Thai:] 32 - [:Script = Lao:] 33 - [:Script = Hiragana:]]; 34 35$TheZWSP = \u200b; 36 37# 38# Character Class Definitions. 39# The names are those from TR29. 40# 41$CR = \u000d; 42$LF = \u000a; 43$Control = [[[:Zl:] [:Zp:] [:Cc:] [:Cf:]] - $TheZWSP]; 44$Extend = [[:Grapheme_Extend = TRUE:]]; 45 46 47 48 49#################################################################################### 50# 51# Word Break Rules. Definitions and Rules specific to word break begin Here. 52# 53#################################################################################### 54 55$Format = [[:Cf:] - $TheZWSP]; 56 57 58 59# Rule 3: Treat a grapheme cluster as if it were a single character. 60# Hangul Syllables are easier to deal with here than they are in Grapheme Clusters 61# because we don't need to find the boundaries between adjacent syllables - 62# they won't be word boundaries. 63# 64 65 66# 67# "Extended" definitions. Grapheme Cluster + Format Chars, treated like the base char. 68# 69$ALetterEx = $ALetter $Extend*; 70$KatakanaEx = $Katakana $Extend*; 71$FormatEx = $Format $Extend*; 72 73# 74# Numbers. Rules 8, 11, 12 form the TR. 75# 76 77# 78# Words. Alpha-numerics. Rule 5, 6, 7, 9, 10 79# - must include at least one letter. 80# - may include both letters and numbers. 81# - may include MideLetter, MidNumber punctuation. 82# 83$LetterSequence = $ALetterEx ($FormatEx* $ALetterEx)*; # rules #6, #7 84$LetterSequence {200}; 85 86$ALetterEx* $dash+ {200}; 87$ALetterEx* ($dash $LetterSequence)+ $dash* {200}; 88 89# 90# Do not break between Katakana. Rule #13. 91# 92$KatakanaEx ($FormatEx* $KatakanaEx)* {300}; 93[:Hiragana:] $Extend* {300}; 94 95# 96# Ideographic Characters. Stand by themselves as words. 97# Separated from the "Everything Else" rule, below, only so that they 98# can be tagged with a return value. TODO: is this what we want? 99# 100# [:IDEOGRAPHIC:] $Extend* {400}; 101 102# 103# Everything Else, with no tag. 104# Non-Control chars combine with $Extend (combining) chars. 105# Controls are do not. 106# 107[^$Control [:Ideographic:]] $Extend*; 108$CR $LF; 109 110# 111# Reverse Rules. Back up over any of the chars that can group together. 112# (Reverse rules do not need to be exact; they can back up too far, 113# but must back up at least enough, and must stop on a boundary.) 114# 115 116# NonStarters are the set of all characters that can appear at the 2nd - nth position of 117# a word. (They may also be the first.) The reverse rule skips over these, until it 118# reaches something that can only be the start (and probably only) char in a "word". 119# A space or punctuation meets the test. 120# 121$NonStarters = [$ALetter $Katakana $Extend $Format]; 122 123#!.*; 124! ($NonStarters* | \n \r) .; 125 126