1 Index: source/data/brkitr/word.txt 2 =================================================================== 3 --- source/data/brkitr/word.txt (revision 259715) 4 +++ source/data/brkitr/word.txt (working copy) 5 @@ -35,10 +35,16 @@ 6 $ALetter = [\p{Word_Break = ALetter}]; 7 $Single_Quote = [\p{Word_Break = Single_Quote}]; 8 $Double_Quote = [\p{Word_Break = Double_Quote}]; 9 -$MidNumLet = [\p{Word_Break = MidNumLet}]; 10 +# Remove two full stop characters from $MidNumLet and add them to $MidNum 11 +# to break a hostname into its components at the cost of breaking 12 +# 'e.g.' and 'i.e.' as well. 13 +# $MidNumLet is used in rules 6/7 (rules of our interest) and rules 11/12. 14 +# Because it's OR'd with $MidNum in rules 11/12, rules 11/12 are not affected 15 +# while rules 6/7 are reverted to the old behavior we want. 16 +$MidNumLet = [[\p{Word_Break = MidNumLet}] - [\u002E \uFF0E]]; 17 $MidLetter = [\p{Word_Break = MidLetter}]; 18 -$MidNum = [\p{Word_Break = MidNum}]; 19 -$Numeric = [\p{Word_Break = Numeric}]; 20 +$MidNum = [\p{Word_Break = MidNum}[\u002E \uFF0E]]; 21 +$Numeric = [\p{Word_Break = Numeric}[\uff10-\uff19]]; #includes fullwidth digits 22 $ExtendNumLet = [\p{Word_Break = ExtendNumLet}]; 23 24 $Han = [:Han:]; 25 Index: source/data/brkitr/line.txt 26 =================================================================== 27 --- source/data/brkitr/line.txt (revision 259715) 28 +++ source/data/brkitr/line.txt (working copy) 29 @@ -12,9 +12,8 @@ 30 # This is only because of a limitation of ICU break engine implementation, 31 # not because the older behavior is desirable. 32 33 -# 34 -# Character Classes defined by TR 14. 35 -# 36 +# CHROME: 1. Use line_ja.txt to apply small kana rules in all locales. 37 +# 2. Adjust CL, OP, and IS to handle 'comma-variants' consistently. 38 39 !!chain; 40 !!LBCMNoChain; 41 @@ -57,14 +56,14 @@ 42 # 43 44 $AI = [:LineBreak = Ambiguous:]; 45 -$AL = [:LineBreak = Alphabetic:]; 46 +$AL = [[:LineBreak = Alphabetic:] - [\u23B4\u23B5]]; 47 $BA = [:LineBreak = Break_After:]; 48 $BB = [:LineBreak = Break_Before:]; 49 $BK = [:LineBreak = Mandatory_Break:]; 50 $B2 = [:LineBreak = Break_Both:]; 51 $CB = [:LineBreak = Contingent_Break:]; 52 $CJ = [:LineBreak = Conditional_Japanese_Starter:]; 53 -$CL = [:LineBreak = Close_Punctuation:]; 54 +$CL = [[:LineBreak = Close_Punctuation:] [\uFE51\uFE10\u23B5]]; 55 $CM = [:LineBreak = Combining_Mark:]; 56 $CP = [:LineBreak = Close_Parenthesis:]; 57 $CR = [:LineBreak = Carriage_Return:]; 58 @@ -74,16 +73,16 @@ 59 $HY = [:LineBreak = Hyphen:]; 60 $H2 = [:LineBreak = H2:]; 61 $H3 = [:LineBreak = H3:]; 62 -$ID = [:LineBreak = Ideographic:]; 63 +$ID = [[[:LineBreak = Ideographic:] $CJ] - [\uFE51]]; 64 $IN = [:LineBreak = Inseperable:]; 65 -$IS = [:LineBreak = Infix_Numeric:]; 66 +$IS = [[:LineBreak = Infix_Numeric:] - [\uFE10]]; 67 $JL = [:LineBreak = JL:]; 68 $JV = [:LineBreak = JV:]; 69 $JT = [:LineBreak = JT:]; 70 $LF = [:LineBreak = Line_Feed:]; 71 $NL = [:LineBreak = Next_Line:]; 72 -$NS = [[:LineBreak = Nonstarter:] $CJ]; 73 +$NS = [:LineBreak = Nonstarter:]; 74 $NU = [:LineBreak = Numeric:]; 75 -$OP = [:LineBreak = Open_Punctuation:]; 76 +$OP = [[:LineBreak = Open_Punctuation:] \u23B4]; 77 $PO = [:LineBreak = Postfix_Numeric:]; 78 $PR = [:LineBreak = Prefix_Numeric:]; 79
