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

#
# Copyright (C) 2002-2015, International Business Machines Corporation 
# and others. All Rights Reserved.
#
# file:  word.txt
#
# ICU Word Break Rules
#      See Unicode Standard Annex #29.
#      These rules are based on UAX #29 Revision 22 for Unicode Version 6.3
#
# Note:  Updates to word.txt will usually need to be merged into
#        word_POSIX.txt also.

##############################################################################
#
#  Character class definitions from TR 29
#
##############################################################################

!!chain;
!!RINoChain;


#
#  Character Class Definitions.
#

$CR                 = [\p{Word_Break = CR}];
$LF                 = [\p{Word_Break = LF}];
$Newline            = [\p{Word_Break = Newline}];
$Extend             = [\p{Word_Break = Extend}];
$RI                 = [\p{Word_Break = Regional_Indicator}];
$Format             = [\p{Word_Break = Format}];
$Katakana           = [\p{Word_Break = Katakana}];
$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}];
$MidLetter          = [\p{Word_Break = MidLetter}];
$MidNum             = [\p{Word_Break = MidNum}];
$Numeric            = [\p{Word_Break = Numeric}];
$ExtendNumLet       = [\p{Word_Break = ExtendNumLet}];

$Han                = [:Han:];
$Hiragana           = [:Hiragana:];

# Special character classes for people & body part emoji:
# Subsets of $Extend:
$ZWJ = \u200D;
$EmojiVar = \uFE0F;
# The following are subsets of \p{Word_Break = Other}
$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
#   characters requiring dictionary break.

$Control        = [\p{Grapheme_Cluster_Break = Control}]; 
$HangulSyllable = [\uac00-\ud7a3];
$ComplexContext = [:LineBreak = Complex_Context:];
$KanaKanji      = [$Han $Hiragana $Katakana];
$dictionaryCJK  = [$KanaKanji $HangulSyllable];
$dictionary     = [$ComplexContext $dictionaryCJK];

# leave CJK scripts out of ALetterPlus
$ALetterPlus  = [$ALetter-$dictionaryCJK [$ComplexContext-$Extend-$Control]];


#
#  Rules 4    Ignore Format and Extend characters, 
#             except when they appear at the beginning of a region of text.
#
# TODO: check if handling of katakana in dictionary makes rules incorrect/void
$KatakanaEx           = $Katakana           ($Extend |  $Format)*;
$Hebrew_LetterEx      = $Hebrew_Letter      ($Extend |  $Format)*;
$ALetterEx            = $ALetterPlus        ($Extend |  $Format)*;
$Single_QuoteEx       = $Single_Quote       ($Extend |  $Format)*;
$Double_QuoteEx       = $Double_Quote       ($Extend |  $Format)*;
$MidNumLetEx          = $MidNumLet          ($Extend |  $Format)*;
$MidLetterEx          = $MidLetter          ($Extend |  $Format)*;
$MidNumEx             = $MidNum             ($Extend |  $Format)*;
$NumericEx            = $Numeric            ($Extend |  $Format)*;
$ExtendNumLetEx       = $ExtendNumLet       ($Extend |  $Format)*;

$Ideographic    = [\p{Ideographic}];
$HiraganaEx     = $Hiragana     ($Extend |  $Format)*;
$IdeographicEx  = $Ideographic  ($Extend |  $Format)*;

## -------------------------------------------------

!!forward;


# Rule 3 - CR x LF
#
$CR $LF;

# Rule 4 - ignore Format and Extend characters, except when they appear at the beginning
#          of a region of Text.   The rule here comes into play when the start of text
#          begins with a group of Format chars, or with a "word" consisting of a single
#          char that is not in any of the listed word break categories followed by
#          format char(s), or is not a CJK dictionary character.
[^$CR $LF $Newline]? ($Extend |  $Format)+;

$NumericEx {100};
$ALetterEx {200};
$HangulSyllable {200};
$Hebrew_LetterEx{200};
$KatakanaEx {400};       # note:  these status values override those from rule 5
$HiraganaEx {400};       #        by virtue of being numerically larger.
$IdeographicEx {400};    #

#
# rule 5
#    Do not break between most letters.
#
($ALetterEx | $Hebrew_LetterEx)  ($ALetterEx | $Hebrew_LetterEx) {200};

# rule 6 and 7
($ALetterEx | $Hebrew_LetterEx) ($MidLetterEx | $MidNumLetEx | $Single_QuoteEx) ($ALetterEx | $Hebrew_LetterEx) {200};

# rule 7a
$Hebrew_LetterEx $Single_QuoteEx {200};

# rule 7b and 7c
$Hebrew_LetterEx $Double_QuoteEx $Hebrew_LetterEx {200};

# rule 8

$NumericEx $NumericEx {100};

# rule 9

($ALetterEx | $Hebrew_LetterEx) $NumericEx {200};

# rule 10

$NumericEx ($ALetterEx | $Hebrew_LetterEx) {200};

# rule 11 and 12 

$NumericEx ($MidNumEx | $MidNumLetEx | $Single_QuoteEx) $NumericEx {100};

# rule 13
# to be consistent with $KanaKanji $KanaKanhi, changed
# from 300 to 400.
# See also TestRuleStatus in intltest/rbbiapts.cpp
$KatakanaEx  $KatakanaEx {400};

# rule 13a/b

$ALetterEx       $ExtendNumLetEx {200};    #  (13a)
$Hebrew_LetterEx $ExtendNumLetEx {200};    #  (13a)
$NumericEx       $ExtendNumLetEx {100};    #  (13a)
$KatakanaEx      $ExtendNumLetEx {400};    #  (13a)
$ExtendNumLetEx  $ExtendNumLetEx {200};    #  (13a)

$ExtendNumLetEx  $ALetterEx      {200};    #  (13b)
$ExtendNumLetEx  $Hebrew_Letter  {200};    #  (13b)
$ExtendNumLetEx  $NumericEx      {100};    #  (13b)
$ExtendNumLetEx  $KatakanaEx     {400};    #  (13b)

# rule 13c

$RI $RI $Extend* / $RI;
$RI $RI $Extend*;

# Special forward rules for people & body part emoji:
# don't break $ZWJ from subsequent $EmojiForSeqs; don't break between relevant emoji and $EmojiMods
$ZWJ $EmojiForSeqs;
$EmojiForMods $EmojiVar? $EmojiMods;

# 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;

$BackHebrew_LetterEx      = ($Format | $Extend)* $Hebrew_Letter;
$BackALetterEx            = ($Format | $Extend)* $ALetterPlus;
$BackSingle_QuoteEx       = ($Format | $Extend)* $Single_Quote;
$BackDouble_QuoteEx       = ($Format | $Extend)* $Double_Quote;
$BackMidNumLetEx          = ($Format | $Extend)* $MidNumLet;
$BackNumericEx            = ($Format | $Extend)* $Numeric;
$BackMidNumEx             = ($Format | $Extend)* $MidNum;
$BackMidLetterEx          = ($Format | $Extend)* $MidLetter;
$BackKatakanaEx           = ($Format | $Extend)* $Katakana;
$BackHiraganaEx           = ($Format | $Extend)* $Hiragana;
$BackExtendNumLetEx       = ($Format | $Extend)* $ExtendNumLet;

# rule 3
$LF $CR;

# rule 4
($Format | $Extend)*  [^$CR $LF $Newline]?;

# rule 5

($BackALetterEx | $BackHebrew_LetterEx) ($BackALetterEx | $BackHebrew_LetterEx);

# rule 6 and 7

($BackALetterEx | $BackHebrew_LetterEx) ($BackMidLetterEx | $BackMidNumLetEx | $BackSingle_QuoteEx) ($BackALetterEx | $BackHebrew_LetterEx);

# rule 7a
$BackSingle_QuoteEx $BackHebrew_LetterEx;

# Rule 7b and 7c
$BackHebrew_LetterEx $BackDouble_QuoteEx $BackHebrew_LetterEx;

# rule 8

$BackNumericEx $BackNumericEx;

# rule 9

$BackNumericEx ($BackALetterEx | $BackHebrew_LetterEx);

# rule 10

($BackALetterEx | $BackHebrew_LetterEx) $BackNumericEx;

# rule 11 and 12

$BackNumericEx ($BackMidNumEx | $BackMidNumLetEx | $BackSingle_QuoteEx) $BackNumericEx;

# rule 13

$BackKatakanaEx $BackKatakanaEx;

# rules 13 a/b
#
$BackExtendNumLetEx ($BackALetterEx | $BackHebrew_LetterEx | $BackNumericEx | $BackKatakanaEx | $BackExtendNumLetEx);
($BackALetterEx | $BackHebrew_LetterEx | $BackNumericEx | $BackKatakanaEx) $BackExtendNumLetEx; 

# rule 13c

$Extend* $RI $RI / $Extend* $RI $RI;
$Extend* $RI $RI;

# Special reverse rules for people & body part emoji:
# don't break $ZWJ from subsequent $EmojiForSeqs; don't break between relevant emoji and $EmojiMods
$EmojiForSeqs $ZWJ;
$EmojiMods $EmojiVar? $EmojiForMods;

# 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;

# rule 3
($Extend | $Format)+ .?;

# rule 6
($MidLetter | $MidNumLet | $Single_Quote) ($BackALetterEx | $BackHebrew_LetterEx);

# rule 7b
$Double_Quote $BackHebrew_LetterEx;


# rule 11
($MidNum | $MidNumLet | $Single_Quote) $BackNumericEx;

$RI $RI+;
# For dictionary-based break
$dictionary $dictionary;

## -------------------------------------------------

!!safe_forward;

# rule 4
($Extend | $Format)+ .?;

# rule 6
($MidLetterEx | $MidNumLetEx | $Single_QuoteEx) ($ALetterEx | $Hebrew_LetterEx);

# rule 7b
$Double_QuoteEx $Hebrew_LetterEx;

# rule 11
($MidNumEx | $MidNumLetEx | $Single_QuoteEx) $NumericEx;

$RI $RI+;
# For dictionary-based break
$dictionary $dictionary;
Commit	Line	Data
b75a7d8f	1	#
08b89b0a	2	# Copyright (C) 2002-2015, International Business Machines Corporation
374ca955	3	# and others. All Rights Reserved.
b75a7d8f	4	#
374ca955	5	# file: word.txt
b75a7d8f	6	#
374ca955	7	# ICU Word Break Rules
b75a7d8f	8	# See Unicode Standard Annex #29.
57a6839d	9	# These rules are based on UAX #29 Revision 22 for Unicode Version 6.3
b75a7d8f	10	#
73c04bcf	11	# Note: Updates to word.txt will usually need to be merged into
51004dcb	12	# word_POSIX.txt also.
b75a7d8f	13
374ca955	14	##############################################################################
b75a7d8f A	15	#
	16	# Character class definitions from TR 29
	17	#
374ca955 A	18	##############################################################################
	19
	20	!!chain;
b331163b	21	!!RINoChain;
374ca955	22
b75a7d8f A	23
	24	#
	25	# Character Class Definitions.
b75a7d8f	26	#
b75a7d8f	27
57a6839d A	28	$CR = [\p{Word_Break = CR}];
	29	$LF = [\p{Word_Break = LF}];
	30	$Newline = [\p{Word_Break = Newline}];
	31	$Extend = [\p{Word_Break = Extend}];
b331163b	32	$RI = [\p{Word_Break = Regional_Indicator}];
57a6839d A	33	$Format = [\p{Word_Break = Format}];
	34	$Katakana = [\p{Word_Break = Katakana}];
	35	$Hebrew_Letter = [\p{Word_Break = Hebrew_Letter}];
	36	$ALetter = [\p{Word_Break = ALetter}];
	37	$Single_Quote = [\p{Word_Break = Single_Quote}];
	38	$Double_Quote = [\p{Word_Break = Double_Quote}];
	39	$MidNumLet = [\p{Word_Break = MidNumLet}];
	40	$MidLetter = [\p{Word_Break = MidLetter}];
	41	$MidNum = [\p{Word_Break = MidNum}];
	42	$Numeric = [\p{Word_Break = Numeric}];
	43	$ExtendNumLet = [\p{Word_Break = ExtendNumLet}];
	44
	45	$Han = [:Han:];
	46	$Hiragana = [:Hiragana:];
73c04bcf	47
08b89b0a A	48	# Special character classes for people & body part emoji:
	49	# Subsets of $Extend:
	50	$ZWJ = \u200D;
	51	$EmojiVar = \uFE0F;
	52	# The following are subsets of \p{Word_Break = Other}
b801cf36 A	53	$EmojiForSeqs = [\u2764 \U0001F441 \U0001F466-\U0001F469 \U0001F48B \U0001F5E8];
b801cf36 A	54	$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];
08b89b0a	55	$EmojiMods = [\U0001F3FB-\U0001F3FF];
73c04bcf	56
73c04bcf A	57	# Dictionary character set, for triggering language-based break engines. Currently
	58	# limited to LineBreak=Complex_Context. Note that this set only works in Unicode
	59	# 5.0 or later as the definition of Complex_Context was corrected to include all
	60	# characters requiring dictionary break.
	61
46f4442e	62	$Control = [\p{Grapheme_Cluster_Break = Control}];
51004dcb A	63	$HangulSyllable = [\uac00-\ud7a3];
	64	$ComplexContext = [:LineBreak = Complex_Context:];
	65	$KanaKanji = [$Han $Hiragana $Katakana];
	66	$dictionaryCJK = [$KanaKanji $HangulSyllable];
	67	$dictionary = [$ComplexContext $dictionaryCJK];
	68
	69	# leave CJK scripts out of ALetterPlus
	70	$ALetterPlus = [$ALetter-$dictionaryCJK [$ComplexContext-$Extend-$Control]];
	71
73c04bcf A	72
73c04bcf A	73	#
46f4442e A	74	# Rules 4 Ignore Format and Extend characters,
46f4442e A	75	# except when they appear at the beginning of a region of text.
73c04bcf	76	#
51004dcb	77	# TODO: check if handling of katakana in dictionary makes rules incorrect/void
57a6839d A	78	$KatakanaEx = $Katakana ($Extend \| $Format)*;
	79	$Hebrew_LetterEx = $Hebrew_Letter ($Extend \| $Format)*;
	80	$ALetterEx = $ALetterPlus ($Extend \| $Format)*;
	81	$Single_QuoteEx = $Single_Quote ($Extend \| $Format)*;
	82	$Double_QuoteEx = $Double_Quote ($Extend \| $Format)*;
	83	$MidNumLetEx = $MidNumLet ($Extend \| $Format)*;
	84	$MidLetterEx = $MidLetter ($Extend \| $Format)*;
	85	$MidNumEx = $MidNum ($Extend \| $Format)*;
	86	$NumericEx = $Numeric ($Extend \| $Format)*;
	87	$ExtendNumLetEx = $ExtendNumLet ($Extend \| $Format)*;
73c04bcf	88
46f4442e	89	$Ideographic = [\p{Ideographic}];
73c04bcf A	90	$HiraganaEx = $Hiragana ($Extend \| $Format)*;
73c04bcf A	91	$IdeographicEx = $Ideographic ($Extend \| $Format)*;
b75a7d8f	92
374ca955	93	## -------------------------------------------------
b75a7d8f	94
374ca955	95	!!forward;
b75a7d8f	96
b75a7d8f	97
73c04bcf	98	# Rule 3 - CR x LF
46f4442e A	99	#
46f4442e A	100	$CR $LF;
73c04bcf A	101
	102	# Rule 4 - ignore Format and Extend characters, except when they appear at the beginning
	103	# of a region of Text. The rule here comes into play when the start of text
	104	# begins with a group of Format chars, or with a "word" consisting of a single
	105	# char that is not in any of the listed word break categories followed by
51004dcb	106	# format char(s), or is not a CJK dictionary character.
46f4442e	107	[^$CR $LF $Newline]? ($Extend \| $Format)+;
b75a7d8f	108
374ca955 A	109	$NumericEx {100};
374ca955 A	110	$ALetterEx {200};
51004dcb	111	$HangulSyllable {200};
57a6839d	112	$Hebrew_LetterEx{200};
51004dcb A	113	$KatakanaEx {400}; # note: these status values override those from rule 5
51004dcb A	114	$HiraganaEx {400}; # by virtue of being numerically larger.
46f4442e	115	$IdeographicEx {400}; #
b75a7d8f	116
46f4442e	117	#
374ca955	118	# rule 5
46f4442e A	119	# Do not break between most letters.
46f4442e A	120	#
57a6839d	121	($ALetterEx \| $Hebrew_LetterEx) ($ALetterEx \| $Hebrew_LetterEx) {200};
b75a7d8f	122
374ca955	123	# rule 6 and 7
57a6839d A	124	($ALetterEx \| $Hebrew_LetterEx) ($MidLetterEx \| $MidNumLetEx \| $Single_QuoteEx) ($ALetterEx \| $Hebrew_LetterEx) {200};
	125
	126	# rule 7a
	127	$Hebrew_LetterEx $Single_QuoteEx {200};
	128
	129	# rule 7b and 7c
	130	$Hebrew_LetterEx $Double_QuoteEx $Hebrew_LetterEx {200};
b75a7d8f	131
374ca955	132	# rule 8
b75a7d8f	133
73c04bcf	134	$NumericEx $NumericEx {100};
b75a7d8f	135
374ca955	136	# rule 9
b75a7d8f	137
57a6839d	138	($ALetterEx \| $Hebrew_LetterEx) $NumericEx {200};
b75a7d8f	139
374ca955	140	# rule 10
b75a7d8f	141
57a6839d	142	$NumericEx ($ALetterEx \| $Hebrew_LetterEx) {200};
b75a7d8f	143
374ca955 A	144	# rule 11 and 12
374ca955 A	145
57a6839d	146	$NumericEx ($MidNumEx \| $MidNumLetEx \| $Single_QuoteEx) $NumericEx {100};
374ca955 A	147
374ca955 A	148	# rule 13
51004dcb A	149	# to be consistent with $KanaKanji $KanaKanhi, changed
	150	# from 300 to 400.
	151	# See also TestRuleStatus in intltest/rbbiapts.cpp
	152	$KatakanaEx $KatakanaEx {400};
374ca955 A	153
	154	# rule 13a/b
	155
57a6839d A	156	$ALetterEx $ExtendNumLetEx {200}; # (13a)
	157	$Hebrew_LetterEx $ExtendNumLetEx {200}; # (13a)
	158	$NumericEx $ExtendNumLetEx {100}; # (13a)
	159	$KatakanaEx $ExtendNumLetEx {400}; # (13a)
	160	$ExtendNumLetEx $ExtendNumLetEx {200}; # (13a)
374ca955	161
57a6839d A	162	$ExtendNumLetEx $ALetterEx {200}; # (13b)
	163	$ExtendNumLetEx $Hebrew_Letter {200}; # (13b)
	164	$ExtendNumLetEx $NumericEx {100}; # (13b)
	165	$ExtendNumLetEx $KatakanaEx {400}; # (13b)
51004dcb A	166
	167	# rule 13c
	168
b331163b A	169	$RI $RI $Extend* / $RI;
b331163b A	170	$RI $RI $Extend*;
08b89b0a A	171
	172	# Special forward rules for people & body part emoji:
	173	# don't break $ZWJ from subsequent $EmojiForSeqs; don't break between relevant emoji and $EmojiMods
	174	$ZWJ $EmojiForSeqs;
	175	$EmojiForMods $EmojiVar? $EmojiMods;
51004dcb A	176
	177	# special handling for CJK characters: chain for later dictionary segmentation
	178	$HangulSyllable $HangulSyllable {200};
	179	$KanaKanji $KanaKanji {400}; # different rule status if both kana and kanji found
374ca955 A	180
	181
	182	## -------------------------------------------------
b75a7d8f	183
374ca955 A	184	!!reverse;
374ca955 A	185
57a6839d	186	$BackHebrew_LetterEx = ($Format \| $Extend)* $Hebrew_Letter;
51004dcb	187	$BackALetterEx = ($Format \| $Extend)* $ALetterPlus;
57a6839d A	188	$BackSingle_QuoteEx = ($Format \| $Extend)* $Single_Quote;
57a6839d A	189	$BackDouble_QuoteEx = ($Format \| $Extend)* $Double_Quote;
51004dcb A	190	$BackMidNumLetEx = ($Format \| $Extend)* $MidNumLet;
	191	$BackNumericEx = ($Format \| $Extend)* $Numeric;
	192	$BackMidNumEx = ($Format \| $Extend)* $MidNum;
	193	$BackMidLetterEx = ($Format \| $Extend)* $MidLetter;
	194	$BackKatakanaEx = ($Format \| $Extend)* $Katakana;
	195	$BackHiraganaEx = ($Format \| $Extend)* $Hiragana;
	196	$BackExtendNumLetEx = ($Format \| $Extend)* $ExtendNumLet;
374ca955	197
73c04bcf	198	# rule 3
46f4442e	199	$LF $CR;
374ca955	200
73c04bcf	201	# rule 4
46f4442e	202	($Format \| $Extend)* [^$CR $LF $Newline]?;
374ca955 A	203
	204	# rule 5
	205
57a6839d	206	($BackALetterEx \| $BackHebrew_LetterEx) ($BackALetterEx \| $BackHebrew_LetterEx);
374ca955 A	207
	208	# rule 6 and 7
	209
57a6839d A	210	($BackALetterEx \| $BackHebrew_LetterEx) ($BackMidLetterEx \| $BackMidNumLetEx \| $BackSingle_QuoteEx) ($BackALetterEx \| $BackHebrew_LetterEx);
	211
	212	# rule 7a
	213	$BackSingle_QuoteEx $BackHebrew_LetterEx;
374ca955	214
57a6839d A	215	# Rule 7b and 7c
57a6839d A	216	$BackHebrew_LetterEx $BackDouble_QuoteEx $BackHebrew_LetterEx;
374ca955 A	217
	218	# rule 8
	219
73c04bcf	220	$BackNumericEx $BackNumericEx;
374ca955 A	221
	222	# rule 9
	223
57a6839d	224	$BackNumericEx ($BackALetterEx \| $BackHebrew_LetterEx);
374ca955 A	225
	226	# rule 10
	227
57a6839d	228	($BackALetterEx \| $BackHebrew_LetterEx) $BackNumericEx;
374ca955 A	229
	230	# rule 11 and 12
	231
57a6839d	232	$BackNumericEx ($BackMidNumEx \| $BackMidNumLetEx \| $BackSingle_QuoteEx) $BackNumericEx;
374ca955 A	233
	234	# rule 13
	235
73c04bcf	236	$BackKatakanaEx $BackKatakanaEx;
374ca955 A	237
374ca955 A	238	# rules 13 a/b
b75a7d8f	239	#
57a6839d A	240	$BackExtendNumLetEx ($BackALetterEx \| $BackHebrew_LetterEx \| $BackNumericEx \| $BackKatakanaEx \| $BackExtendNumLetEx);
57a6839d A	241	($BackALetterEx \| $BackHebrew_LetterEx \| $BackNumericEx \| $BackKatakanaEx) $BackExtendNumLetEx;
374ca955	242
51004dcb A	243	# rule 13c
51004dcb A	244
b331163b A	245	$Extend* $RI $RI / $Extend* $RI $RI;
b331163b A	246	$Extend* $RI $RI;
08b89b0a A	247
	248	# Special reverse rules for people & body part emoji:
	249	# don't break $ZWJ from subsequent $EmojiForSeqs; don't break between relevant emoji and $EmojiMods
	250	$EmojiForSeqs $ZWJ;
	251	$EmojiMods $EmojiVar? $EmojiForMods;
51004dcb A	252
	253	# special handling for CJK characters: chain for later dictionary segmentation
	254	$HangulSyllable $HangulSyllable;
	255	$KanaKanji $KanaKanji; #different rule status if both kanji and kana found
	256
374ca955 A	257	## -------------------------------------------------
	258
	259	!!safe_reverse;
	260
	261	# rule 3
73c04bcf	262	($Extend \| $Format)+ .?;
374ca955 A	263
374ca955 A	264	# rule 6
57a6839d A	265	($MidLetter \| $MidNumLet \| $Single_Quote) ($BackALetterEx \| $BackHebrew_LetterEx);
	266
	267	# rule 7b
	268	$Double_Quote $BackHebrew_LetterEx;
	269
374ca955 A	270
374ca955 A	271	# rule 11
57a6839d	272	($MidNum \| $MidNumLet \| $Single_Quote) $BackNumericEx;
73c04bcf	273
b331163b	274	$RI $RI+;
73c04bcf A	275	# For dictionary-based break
73c04bcf A	276	$dictionary $dictionary;
374ca955 A	277
	278	## -------------------------------------------------
	279
	280	!!safe_forward;
	281
374ca955	282	# rule 4
73c04bcf	283	($Extend \| $Format)+ .?;
374ca955 A	284
374ca955 A	285	# rule 6
57a6839d A	286	($MidLetterEx \| $MidNumLetEx \| $Single_QuoteEx) ($ALetterEx \| $Hebrew_LetterEx);
	287
	288	# rule 7b
	289	$Double_QuoteEx $Hebrew_LetterEx;
b75a7d8f	290
374ca955	291	# rule 11
57a6839d	292	($MidNumEx \| $MidNumLetEx \| $Single_QuoteEx) $NumericEx;
73c04bcf	293
b331163b	294	$RI $RI+;
73c04bcf A	295	# For dictionary-based break
73c04bcf A	296	$dictionary $dictionary;