patches/brkitr.patch

Index: source/data/brkitr/word.txt
===================================================================
--- source/data/brkitr/word.txt	(revision 259715)
+++ source/data/brkitr/word.txt	(working copy)
@@ -35,10 +35,16 @@
 $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:];
Index: source/data/brkitr/line.txt
===================================================================
--- source/data/brkitr/line.txt	(revision 259715)
+++ source/data/brkitr/line.txt	(working copy)
@@ -12,9 +12,8 @@
 #         This is only because of a limitation of ICU break engine implementation,
 #         not because the older behavior is desirable.
 
-#
-#  Character Classes defined by TR 14.
-#
+# CHROME: 1. Use line_ja.txt to apply small kana rules in all locales.
+#         2. Adjust CL, OP, and IS to handle 'comma-variants' consistently.
 
 !!chain;
 !!LBCMNoChain;
@@ -57,14 +56,14 @@
 #
 
 $AI = [:LineBreak =  Ambiguous:];
-$AL = [:LineBreak =  Alphabetic:];
+$AL = [[:LineBreak =  Alphabetic:] - [\u23B4\u23B5]];
 $BA = [:LineBreak =  Break_After:];
 $BB = [:LineBreak =  Break_Before:];
 $BK = [:LineBreak =  Mandatory_Break:];
 $B2 = [:LineBreak =  Break_Both:];
 $CB = [:LineBreak =  Contingent_Break:];
 $CJ = [:LineBreak =  Conditional_Japanese_Starter:];
-$CL = [:LineBreak =  Close_Punctuation:];
+$CL = [[:LineBreak =  Close_Punctuation:] [\uFE51\uFE10\u23B5]];
 $CM = [:LineBreak =  Combining_Mark:];
 $CP = [:LineBreak =  Close_Parenthesis:];
 $CR = [:LineBreak =  Carriage_Return:];
@@ -74,16 +73,16 @@
 $HY = [:LineBreak =  Hyphen:];
 $H2 = [:LineBreak =  H2:];
 $H3 = [:LineBreak =  H3:];
-$ID = [:LineBreak =  Ideographic:];
+$ID = [[[:LineBreak =  Ideographic:] $CJ] - [\uFE51]];
 $IN = [:LineBreak =  Inseperable:];
-$IS = [:LineBreak =  Infix_Numeric:];
+$IS = [[:LineBreak =  Infix_Numeric:] - [\uFE10]];
 $JL = [:LineBreak =  JL:];
 $JV = [:LineBreak =  JV:];
 $JT = [:LineBreak =  JT:];
 $LF = [:LineBreak =  Line_Feed:];
 $NL = [:LineBreak =  Next_Line:];
-$NS = [[:LineBreak =  Nonstarter:] $CJ];
+$NS = [:LineBreak =  Nonstarter:];
 $NU = [:LineBreak =  Numeric:];
-$OP = [:LineBreak =  Open_Punctuation:];
+$OP = [[:LineBreak =  Open_Punctuation:] \u23B4];
 $PO = [:LineBreak =  Postfix_Numeric:];
 $PR = [:LineBreak =  Prefix_Numeric:];