ICU-551.30.tar.gz

[apple/icu.git] / icuSources / data / brkitr / line.txt

diff --git a/icuSources/data/brkitr/line.txt b/icuSources/data/brkitr/line.txt
index e72d188667d1244b74438c891fdcdbc38d21cada..8c207abb8e14a5ce1bb54652ef2f84ed525577ce 100644 (file)
--- a/icuSources/data/brkitr/line.txt
+++ b/icuSources/data/brkitr/line.txt
@@ -1,4 +1,4 @@
-# Copyright (c) 2002-2012  International Business Machines Corporation and
+# Copyright (c) 2002-2015  International Business Machines Corporation and
 # others. All Rights Reserved.
 #
 #  file:  line.txt
@@ -11,6 +11,9 @@
 #         TODO:  Rule LB 8 remains as it was in Unicode 5.2
 #         This is only because of a limitation of ICU break engine implementation,
 #         not because the older behavior is desirable.
+#
+#         This corresponds to CSS line-break=strict (BCP47 -u-lb-strict).
+#         It sets characters of class CJ to behave like NS.
 
 #
 #  Character Classes defined by TR 14.
@@ -97,6 +100,15 @@ $WJ = [:LineBreak =  Word_Joiner:];
 $XX = [:LineBreak =  Unknown:];
 $ZW = [:LineBreak =  ZWSpace:];
 
+# Special character classes for people & body part emoji:
+# Subsets of $CM:
+$ZWJ = \u200D;
+$EmojiVar = \uFE0F;
+# The following are subsets of $ID
+$EmojiForSeqs = [\u2764 \U0001F441 \U0001F466-\U0001F469 \U0001F48B \U0001F5E8];
+$EmojiForMods = [\u261D \u26F9 \u270A-\u270D \U0001F385 \U0001F3C3-\U0001F3C4 \U0001F3C7 \U0001F3CA-\U0001F3CB \U0001F442-\U0001F443 \U0001F446-\U0001F450 \U0001F466-\U0001F469 \U0001F46E-\U0001F478 \U0001F47C \U0001F481-\U0001F483 \U0001F485-\U0001F487 \U0001F4AA \U0001F590 \U0001F595 \U0001F596 \U0001F645-\U0001F647 \U0001F64B-\U0001F64F \U0001F6A3 \U0001F6B4-\U0001F6B6 \U0001F6C0 \U0001F918];
+$EmojiMods = [\U0001F3FB-\U0001F3FF];
+
 #   Dictionary character set, for triggering language-based break engines. Currently
 #   limited to LineBreak=Complex_Context. Note that this set only works in Unicode
 #   5.0 or later as the definition of Complex_Context was corrected to include all
@@ -231,6 +243,10 @@ $LB8Breaks    = [$LB4Breaks $ZW];
 $LB8NonBreaks = [[$LB4NonBreaks] - [$ZW]];
 
 
+# Special forward rule for people & body part emoji:
+# don't break $ZWJ from subsequent $EmojiForSeqs
+$EmojiForSeqs $EmojiVar? $EmojiMods? $ZWJ $EmojiForSeqs;
+
 # LB 9     Combining marks.      X   $CM needs to behave like X, where X is not $SP, $BK $CR $LF $NL 
 #                                $CM not covered by the above needs to behave like $AL   
 #                                See definition of $CAN_CM.
@@ -402,6 +418,10 @@ $CPcm ($ALcm | $HLcm | $NUcm);
 # LB 30a  Do not break between regional indicators.
 $RIcm $RIcm;
 
+# Special forward rule for people & body part emoji:
+# don't break between relevant emoji and $EmojiMods
+$EmojiForMods $EmojiVar? $EmojiMods;
+
 #
 #  Reverse Rules.
 #
@@ -484,6 +504,10 @@ $LF $CR;
 #           Requires an engine enhancement.
 #   / $SP* $ZW
 
+# Special reverse rule for people & body part emoji:
+# don't break $ZWJ from subsequent $EmojiForSeqs
+$EmojiForSeqs $ZWJ $EmojiMods? $EmojiVar? $EmojiForSeqs;
+
 # LB 9,10  Combining marks.
 #    X   $CM needs to behave like X, where X is not $SP or controls.
 #    $CM not covered by the above needs to behave like $AL
@@ -626,6 +650,10 @@ $CM* ($ALPlus | $HL | $NU) $CM* $CP;
 # LB 30a
 $CM* $RI $CM* $RI;
 
+# Special reverse rule for people & body part emoji:
+# don't break between relevant emoji and $EmojiMods
+$EmojiMods $EmojiVar? $EmojiForMods;
+
 ## -------------------------------------------------
 
 !!safe_reverse;