ICU update to 54.1 - step 5

source/data/brkitr/word_ja.txt

Index: source/data/brkitr/word_ja.txt
diff --git a/source/data/brkitr/word.txt b/source/data/brkitr/word_ja.txt
similarity index 88%
copy from source/data/brkitr/word.txt
copy to source/data/brkitr/word_ja.txt
index f89a2fe74526fc0e521f942b5f50957377a1a040..fb77507c8bf9185a7312fe38c261ea87f7b7d803 100644
--- a/source/data/brkitr/word.txt
+++ b/source/data/brkitr/word_ja.txt
@@ -2,7 +2,7 @@
 # Copyright (C) 2002-2013, International Business Machines Corporation 
 # and others. All Rights Reserved.
 #
-# file:  word.txt
+# file:  word_ja.txt
 #
 # ICU Word Break Rules
 #      See Unicode Standard Annex #29.
@@ -35,10 +35,16 @@ $Hebrew_Letter      = [\p{Word_Break = Hebrew_Letter}];
 $ALetter            = [\p{Word_Break = ALetter}];
 $Single_Quote       = [\p{Word_Break = Single_Quote}];
 $Double_Quote       = [\p{Word_Break = Double_Quote}];
-$MidNumLet          = [\p{Word_Break = MidNumLet}];
+# Remove two full stop characters from $MidNumLet and add them to $MidNum
+# to break a hostname into its components at the cost of breaking
+# 'e.g.' and 'i.e.' as well.
+# $MidNumLet is used in rules 6/7 (rules of our interest) and rules 11/12.
+# Because it's OR'd with $MidNum in rules 11/12, rules 11/12 are not affected
+# while rules 6/7 are reverted to the old behavior we want.
+$MidNumLet    = [[\p{Word_Break = MidNumLet}] - [\u002E \uFF0E]];
 $MidLetter          = [\p{Word_Break = MidLetter}];
-$MidNum             = [\p{Word_Break = MidNum}];
-$Numeric            = [\p{Word_Break = Numeric}];
+$MidNum       = [\p{Word_Break = MidNum}[\u002E \uFF0E]];
+$Numeric      = [\p{Word_Break = Numeric}[\uff10-\uff19]]; #includes fullwidth digits
 $ExtendNumLet       = [\p{Word_Break = ExtendNumLet}];
 
 $Han                = [:Han:];
@@ -50,15 +56,13 @@ $Hiragana           = [:Hiragana:];
 #   5.0 or later as the definition of Complex_Context was corrected to include all
 #   characters requiring dictionary break.
 
-$Control        = [\p{Grapheme_Cluster_Break = Control}]; 
+$Control        = [\p{Grapheme_Cluster_Break = Control}];
 $HangulSyllable = [\uac00-\ud7a3];
 $ComplexContext = [:LineBreak = Complex_Context:];
 $KanaKanji      = [$Han $Hiragana $Katakana];
-$dictionaryCJK  = [$KanaKanji $HangulSyllable];
-$dictionary     = [$ComplexContext $dictionaryCJK];
+$dictionary     = [$ComplexContext];
 
-# leave CJK scripts out of ALetterPlus
-$ALetterPlus  = [$ALetter-$dictionaryCJK [$ComplexContext-$Extend-$Control]];
+$ALetterPlus  = [$ALetter [$ComplexContext-$Extend-$Control]];
 
 
 #
@@ -78,7 +82,7 @@ $NumericEx            = $Numeric            ($Extend |  $Format)*;
 $ExtendNumLetEx       = $ExtendNumLet       ($Extend |  $Format)*;
 $Regional_IndicatorEx = $Regional_Indicator ($Extend |  $Format)*;
 
-$Ideographic    = [\p{Ideographic}];
+$Ideographic    = [\p{Ideographic} [\u3005 \u3007 \u303B]];
 $HiraganaEx     = $Hiragana     ($Extend |  $Format)*;
 $IdeographicEx  = $Ideographic  ($Extend |  $Format)*;
 
@@ -142,6 +146,8 @@ $NumericEx ($MidNumEx | $MidNumLetEx | $Single_QuoteEx) $NumericEx {100};
 # from 300 to 400.
 # See also TestRuleStatus in intltest/rbbiapts.cpp
 $KatakanaEx  $KatakanaEx {400};
+$HiraganaEx  $HiraganaEx {400};
+$IdeographicEx  $IdeographicEx {400};
 
 # rule 13a/b
 
@@ -160,11 +166,6 @@ $ExtendNumLetEx  $KatakanaEx     {400};    #  (13b)
 
 $Regional_IndicatorEx $Regional_IndicatorEx;
 
-# special handling for CJK characters: chain for later dictionary segmentation
-$HangulSyllable $HangulSyllable {200};
-$KanaKanji $KanaKanji {400}; # different rule status if both kana and kanji found 
-
-
 ## -------------------------------------------------
 
 !!reverse;
@@ -179,6 +180,7 @@ $BackMidNumEx             = ($Format | $Extend)* $MidNum;
 $BackMidLetterEx          = ($Format | $Extend)* $MidLetter;
 $BackKatakanaEx           = ($Format | $Extend)* $Katakana;
 $BackHiraganaEx           = ($Format | $Extend)* $Hiragana;
+$BackIdeographicEx        = ($Format | $Extend)* $Ideographic;
 $BackExtendNumLetEx       = ($Format | $Extend)* $ExtendNumLet;
 $BackRegional_IndicatorEx = ($Format | $Extend)* $Regional_Indicator;
 
@@ -221,6 +223,8 @@ $BackNumericEx ($BackMidNumEx | $BackMidNumLetEx | $BackSingle_QuoteEx) $BackNum
 # rule 13
 
 $BackKatakanaEx $BackKatakanaEx;
+$BackHiraganaEx $BackHiraganaEx;
+$BackIdeographicEx $BackIdeographicEx;
 
 # rules 13 a/b
 #
@@ -231,10 +235,6 @@ $BackExtendNumLetEx ($BackALetterEx | $BackHebrew_LetterEx | $BackNumericEx | $B
 
 $BackRegional_IndicatorEx $BackRegional_IndicatorEx;
 
-# special handling for CJK characters: chain for later dictionary segmentation
-$HangulSyllable $HangulSyllable;
-$KanaKanji $KanaKanji; #different rule status if both kanji and kana found
-
 ## -------------------------------------------------
 
 !!safe_reverse;