[apple/icu.git] / icuSources / data / brkitr / rules / word_fi_sv.txt

#
# Copyright (C) 2002-2016, International Business Machines Corporation
# and others. All Rights Reserved.
#
# file:  word_fi_sv.txt
#
# ICU Word Break Rules, fi/sv locales (these are actually the standard UAX #29 rules)
#      See Unicode Standard Annex #29.
#      These rules are based on UAX #29 Revision 27 for Unicode Version 8.0
#      with additions from L2/16-011R3 for Emoji sequences.
#
# Note:  Updates to word.txt will usually need to be merged into
#        word_POSIX.txt also.

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

!!chain;


#
#  Character Class Definitions.
#

# Apple $EmojiForMods becomes $E_Base here
$E_Base      = [\u261D\u26F9\u270A-\u270D\U0001F385\U0001F3C2-\U0001F3C4\U0001F3C7\U0001F3CA-\U0001F3CC\U0001F442-\U0001F443\U0001F446-\U0001F450\U0001F466-\U0001F478\U0001F47C\U0001F481-\U0001F483\U0001F485-\U0001F487\U0001F4AA\U0001F574-\U0001F575\U0001F57A\U0001F590\U0001F595-\U0001F596\U0001F645-\U0001F647\U0001F64B-\U0001F64F\U0001F6A3\U0001F6B4-\U0001F6B6\U0001F6C0\U0001F6CC\U0001F918-\U0001F91E\U0001F926\U0001F930\U0001F933-\U0001F939\U0001F93C-\U0001F93E];
# Apple $EmojiMods becomes $E_Modifier here, same chars
$E_Modifier  = [\U0001F3FB-\U0001F3FF];
$ZWJ = [\u200D];
# Apple $EmojiForSeqs becomes $GAZ here (only emoji that follow a ZWJ)
$GAZ         = [\u2640\u2642\u2764\u2695-\u2696\u2708\U0001F308\U0001F33E\U0001F373\U0001F393\U0001F3A4\U0001F3A8\U0001F3EB\U0001F3ED\U0001F466-\U0001F469\U0001F48B\U0001F4BB-\U0001F4BC\U0001F527\U0001F52C\U0001F5E8\U0001F680\U0001F692];

$CR                 = [\p{Word_Break = CR}];
$LF                 = [\p{Word_Break = LF}];
$Newline            = [\p{Word_Break = Newline} ];
$Extend             = [[\p{Word_Break = Extend}][:Block=Tags:]];
$Regional_Indicator = [\p{Word_Break = Regional_Indicator}];
$Format             = [[\p{Word_Break = Format}] - [:Block=Tags:]];
$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:];


#   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 | $ZWJ)*;
$Hebrew_LetterEx      = $Hebrew_Letter      ($Extend |  $Format | $ZWJ)*;
$ALetterEx            = $ALetterPlus        ($Extend |  $Format | $ZWJ)*;
$Single_QuoteEx       = $Single_Quote       ($Extend |  $Format | $ZWJ)*;
$Double_QuoteEx       = $Double_Quote       ($Extend |  $Format | $ZWJ)*;
$MidNumLetEx          = $MidNumLet          ($Extend |  $Format | $ZWJ)*;
$MidLetterEx          = $MidLetter          ($Extend |  $Format | $ZWJ)*;
$MidNumEx             = $MidNum             ($Extend |  $Format | $ZWJ)*;
$NumericEx            = $Numeric            ($Extend |  $Format | $ZWJ)*;
$ExtendNumLetEx       = $ExtendNumLet       ($Extend |  $Format | $ZWJ)*;
$Regional_IndicatorEx = $Regional_Indicator ($Extend |  $Format | $ZWJ)*;

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

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

!!forward;


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

# Rule 3c   ZWJ x GAZ.  Preceeds WB4, so no intervening Extend chars allowed.
#
$ZWJ $GAZ;


# 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 | $ZWJ)+;

$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};    #

$E_Base ($Extend | $Format | $ZWJ)*;
$E_Modifier ($Extend | $Format | $ZWJ)*;
$GAZ ($Extend | $Format | $ZWJ)*;

#
# 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
#    Pairs of Regional Indicators stay together.
#    With rule chaining disabled by ^, this rule will match exactly two of them.
#    No other rule begins with a Regional_Indicator, so chaining cannot extend the match.
#
^$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

# rule 13d
#    E_Base x E_Modifier
#
($E_Base | $GAZ) ($Format | $Extend | $ZWJ)* $E_Modifier;


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

!!reverse;

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

# rule 3
$LF $CR;

# Rule 3c   ZWJ x GAZ.  Preceeds WB4, so no intervening Extend chars allowed.
#
$GAZ $ZWJ;

# rule 4
($Format | $Extend | $ZWJ)*  [^$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

^$BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
        ($Format | $Extend | $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
^$BackRegional_IndicatorEx $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
        ($Format | $Extend | $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];

$GAZ $ZWJ $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
        ($Format | $Extend | $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
$GAZ $ZWJ $BackRegional_IndicatorEx $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
        ($Format | $Extend | $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];

# special handling for CJK characters: chain for later dictionary segmentation
$HangulSyllable $HangulSyllable;
$KanaKanji $KanaKanji; #different rule status if both kanji and kana found

# rule 13d

$E_Modifier ($Format | $Extend | $ZWJ)* ($E_Base | $GAZ);


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

!!safe_reverse;

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

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

# rule 7b
$Double_Quote $BackHebrew_LetterEx;


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

# rule 13c
$BackRegional_IndicatorEx*;

# For dictionary-based break
$dictionary $dictionary;

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

!!safe_forward;

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

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

# rule 7b
$Double_QuoteEx $Hebrew_LetterEx;

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

# rule 13c
$Regional_IndicatorEx*;

# For dictionary-based break
$dictionary $dictionary;
Commit	Line	Data
b75a7d8f	1	#
2ca993e8	2	# Copyright (C) 2002-2016, International Business Machines Corporation
374ca955	3	# and others. All Rights Reserved.
b75a7d8f	4	#
2ca993e8	5	# file: word_fi_sv.txt
b75a7d8f	6	#
2ca993e8	7	# ICU Word Break Rules, fi/sv locales (these are actually the standard UAX #29 rules)
b75a7d8f	8	# See Unicode Standard Annex #29.
2ca993e8 A	9	# These rules are based on UAX #29 Revision 27 for Unicode Version 8.0
2ca993e8 A	10	# with additions from L2/16-011R3 for Emoji sequences.
b75a7d8f	11	#
73c04bcf	12	# Note: Updates to word.txt will usually need to be merged into
51004dcb	13	# word_POSIX.txt also.
b75a7d8f	14
374ca955	15	##############################################################################
b75a7d8f A	16	#
	17	# Character class definitions from TR 29
	18	#
374ca955 A	19	##############################################################################
	20
	21	!!chain;
	22
b75a7d8f A	23
	24	#
	25	# Character Class Definitions.
b75a7d8f	26	#
b75a7d8f	27
2ca993e8	28	# Apple $EmojiForMods becomes $E_Base here
a62d09fc	29	$E_Base = [\u261D\u26F9\u270A-\u270D\U0001F385\U0001F3C2-\U0001F3C4\U0001F3C7\U0001F3CA-\U0001F3CC\U0001F442-\U0001F443\U0001F446-\U0001F450\U0001F466-\U0001F478\U0001F47C\U0001F481-\U0001F483\U0001F485-\U0001F487\U0001F4AA\U0001F574-\U0001F575\U0001F57A\U0001F590\U0001F595-\U0001F596\U0001F645-\U0001F647\U0001F64B-\U0001F64F\U0001F6A3\U0001F6B4-\U0001F6B6\U0001F6C0\U0001F6CC\U0001F918-\U0001F91E\U0001F926\U0001F930\U0001F933-\U0001F939\U0001F93C-\U0001F93E];
2ca993e8 A	30	# Apple $EmojiMods becomes $E_Modifier here, same chars
	31	$E_Modifier = [\U0001F3FB-\U0001F3FF];
	32	$ZWJ = [\u200D];
	33	# Apple $EmojiForSeqs becomes $GAZ here (only emoji that follow a ZWJ)
a62d09fc	34	$GAZ = [\u2640\u2642\u2764\u2695-\u2696\u2708\U0001F308\U0001F33E\U0001F373\U0001F393\U0001F3A4\U0001F3A8\U0001F3EB\U0001F3ED\U0001F466-\U0001F469\U0001F48B\U0001F4BB-\U0001F4BC\U0001F527\U0001F52C\U0001F5E8\U0001F680\U0001F692];
2ca993e8	35
57a6839d A	36	$CR = [\p{Word_Break = CR}];
57a6839d A	37	$LF = [\p{Word_Break = LF}];
2ca993e8 A	38	$Newline = [\p{Word_Break = Newline} ];
	39	$Extend = [[\p{Word_Break = Extend}][:Block=Tags:]];
	40	$Regional_Indicator = [\p{Word_Break = Regional_Indicator}];
	41	$Format = [[\p{Word_Break = Format}] - [:Block=Tags:]];
57a6839d A	42	$Katakana = [\p{Word_Break = Katakana}];
	43	$Hebrew_Letter = [\p{Word_Break = Hebrew_Letter}];
	44	$ALetter = [\p{Word_Break = ALetter}];
	45	$Single_Quote = [\p{Word_Break = Single_Quote}];
	46	$Double_Quote = [\p{Word_Break = Double_Quote}];
	47	$MidNumLet = [\p{Word_Break = MidNumLet}];
	48	$MidLetter = [\p{Word_Break = MidLetter}];
	49	$MidNum = [\p{Word_Break = MidNum}];
	50	$Numeric = [\p{Word_Break = Numeric}];
	51	$ExtendNumLet = [\p{Word_Break = ExtendNumLet}];
	52
	53	$Han = [:Han:];
	54	$Hiragana = [:Hiragana:];
73c04bcf A	55
73c04bcf A	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
2ca993e8	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	#
2ca993e8	74	# Rules 4 Ignore Format and Extend characters,
46f4442e	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
2ca993e8 A	78	$KatakanaEx = $Katakana ($Extend \| $Format \| $ZWJ)*;
	79	$Hebrew_LetterEx = $Hebrew_Letter ($Extend \| $Format \| $ZWJ)*;
	80	$ALetterEx = $ALetterPlus ($Extend \| $Format \| $ZWJ)*;
	81	$Single_QuoteEx = $Single_Quote ($Extend \| $Format \| $ZWJ)*;
	82	$Double_QuoteEx = $Double_Quote ($Extend \| $Format \| $ZWJ)*;
	83	$MidNumLetEx = $MidNumLet ($Extend \| $Format \| $ZWJ)*;
	84	$MidLetterEx = $MidLetter ($Extend \| $Format \| $ZWJ)*;
	85	$MidNumEx = $MidNum ($Extend \| $Format \| $ZWJ)*;
	86	$NumericEx = $Numeric ($Extend \| $Format \| $ZWJ)*;
	87	$ExtendNumLetEx = $ExtendNumLet ($Extend \| $Format \| $ZWJ)*;
	88	$Regional_IndicatorEx = $Regional_Indicator ($Extend \| $Format \| $ZWJ)*;
73c04bcf	89
46f4442e	90	$Ideographic = [\p{Ideographic}];
2ca993e8 A	91	$HiraganaEx = $Hiragana ($Extend \| $Format \| $ZWJ)*;
2ca993e8 A	92	$IdeographicEx = $Ideographic ($Extend \| $Format \| $ZWJ)*;
b75a7d8f	93
374ca955	94	## -------------------------------------------------
b75a7d8f	95
374ca955	96	!!forward;
b75a7d8f	97
b75a7d8f	98
73c04bcf	99	# Rule 3 - CR x LF
46f4442e A	100	#
46f4442e A	101	$CR $LF;
73c04bcf	102
2ca993e8 A	103	# Rule 3c ZWJ x GAZ. Preceeds WB4, so no intervening Extend chars allowed.
	104	#
	105	$ZWJ $GAZ;
	106
	107
73c04bcf A	108	# Rule 4 - ignore Format and Extend characters, except when they appear at the beginning
	109	# of a region of Text. The rule here comes into play when the start of text
	110	# begins with a group of Format chars, or with a "word" consisting of a single
	111	# char that is not in any of the listed word break categories followed by
51004dcb	112	# format char(s), or is not a CJK dictionary character.
2ca993e8	113	[^$CR $LF $Newline]? ($Extend \| $Format \| $ZWJ)+;
b75a7d8f	114
374ca955 A	115	$NumericEx {100};
374ca955 A	116	$ALetterEx {200};
51004dcb	117	$HangulSyllable {200};
57a6839d	118	$Hebrew_LetterEx{200};
51004dcb A	119	$KatakanaEx {400}; # note: these status values override those from rule 5
51004dcb A	120	$HiraganaEx {400}; # by virtue of being numerically larger.
46f4442e	121	$IdeographicEx {400}; #
b75a7d8f	122
2ca993e8 A	123	$E_Base ($Extend \| $Format \| $ZWJ)*;
	124	$E_Modifier ($Extend \| $Format \| $ZWJ)*;
	125	$GAZ ($Extend \| $Format \| $ZWJ)*;
	126
46f4442e	127	#
374ca955	128	# rule 5
46f4442e A	129	# Do not break between most letters.
46f4442e A	130	#
57a6839d	131	($ALetterEx \| $Hebrew_LetterEx) ($ALetterEx \| $Hebrew_LetterEx) {200};
b75a7d8f	132
374ca955	133	# rule 6 and 7
57a6839d A	134	($ALetterEx \| $Hebrew_LetterEx) ($MidLetterEx \| $MidNumLetEx \| $Single_QuoteEx) ($ALetterEx \| $Hebrew_LetterEx) {200};
	135
	136	# rule 7a
	137	$Hebrew_LetterEx $Single_QuoteEx {200};
	138
	139	# rule 7b and 7c
	140	$Hebrew_LetterEx $Double_QuoteEx $Hebrew_LetterEx {200};
b75a7d8f	141
374ca955	142	# rule 8
b75a7d8f	143
73c04bcf	144	$NumericEx $NumericEx {100};
b75a7d8f	145
374ca955	146	# rule 9
b75a7d8f	147
57a6839d	148	($ALetterEx \| $Hebrew_LetterEx) $NumericEx {200};
b75a7d8f	149
374ca955	150	# rule 10
b75a7d8f	151
57a6839d	152	$NumericEx ($ALetterEx \| $Hebrew_LetterEx) {200};
b75a7d8f	153
2ca993e8	154	# rule 11 and 12
374ca955	155
57a6839d	156	$NumericEx ($MidNumEx \| $MidNumLetEx \| $Single_QuoteEx) $NumericEx {100};
374ca955 A	157
374ca955 A	158	# rule 13
51004dcb A	159	# to be consistent with $KanaKanji $KanaKanhi, changed
	160	# from 300 to 400.
	161	# See also TestRuleStatus in intltest/rbbiapts.cpp
	162	$KatakanaEx $KatakanaEx {400};
374ca955 A	163
	164	# rule 13a/b
	165
57a6839d A	166	$ALetterEx $ExtendNumLetEx {200}; # (13a)
	167	$Hebrew_LetterEx $ExtendNumLetEx {200}; # (13a)
	168	$NumericEx $ExtendNumLetEx {100}; # (13a)
	169	$KatakanaEx $ExtendNumLetEx {400}; # (13a)
	170	$ExtendNumLetEx $ExtendNumLetEx {200}; # (13a)
374ca955	171
57a6839d A	172	$ExtendNumLetEx $ALetterEx {200}; # (13b)
	173	$ExtendNumLetEx $Hebrew_Letter {200}; # (13b)
	174	$ExtendNumLetEx $NumericEx {100}; # (13b)
	175	$ExtendNumLetEx $KatakanaEx {400}; # (13b)
51004dcb A	176
51004dcb A	177	# rule 13c
2ca993e8 A	178	# Pairs of Regional Indicators stay together.
	179	# With rule chaining disabled by ^, this rule will match exactly two of them.
	180	# No other rule begins with a Regional_Indicator, so chaining cannot extend the match.
	181	#
	182	^$Regional_IndicatorEx $Regional_IndicatorEx;
51004dcb A	183
	184	# special handling for CJK characters: chain for later dictionary segmentation
	185	$HangulSyllable $HangulSyllable {200};
2ca993e8 A	186	$KanaKanji $KanaKanji {400}; # different rule status if both kana and kanji found
	187
	188	# rule 13d
	189	# E_Base x E_Modifier
	190	#
	191	($E_Base \| $GAZ) ($Format \| $Extend \| $ZWJ)* $E_Modifier;
374ca955 A	192
	193
	194	## -------------------------------------------------
b75a7d8f	195
374ca955 A	196	!!reverse;
374ca955 A	197
2ca993e8 A	198	$BackHebrew_LetterEx = ($Format \| $Extend \| $ZWJ)* $Hebrew_Letter;
	199	$BackALetterEx = ($Format \| $Extend \| $ZWJ)* $ALetterPlus;
	200	$BackSingle_QuoteEx = ($Format \| $Extend \| $ZWJ)* $Single_Quote;
	201	$BackDouble_QuoteEx = ($Format \| $Extend \| $ZWJ)* $Double_Quote;
	202	$BackMidNumLetEx = ($Format \| $Extend \| $ZWJ)* $MidNumLet;
	203	$BackNumericEx = ($Format \| $Extend \| $ZWJ)* $Numeric;
	204	$BackMidNumEx = ($Format \| $Extend \| $ZWJ)* $MidNum;
	205	$BackMidLetterEx = ($Format \| $Extend \| $ZWJ)* $MidLetter;
	206	$BackKatakanaEx = ($Format \| $Extend \| $ZWJ)* $Katakana;
	207	$BackHiraganaEx = ($Format \| $Extend \| $ZWJ)* $Hiragana;
	208	$BackExtendNumLetEx = ($Format \| $Extend \| $ZWJ)* $ExtendNumLet;
	209	$BackRegional_IndicatorEx = ($Format \| $Extend \| $ZWJ)* $Regional_Indicator;
374ca955	210
73c04bcf	211	# rule 3
46f4442e	212	$LF $CR;
374ca955	213
2ca993e8 A	214	# Rule 3c ZWJ x GAZ. Preceeds WB4, so no intervening Extend chars allowed.
	215	#
	216	$GAZ $ZWJ;
	217
73c04bcf	218	# rule 4
2ca993e8	219	($Format \| $Extend \| $ZWJ)* [^$CR $LF $Newline]?;
374ca955 A	220
	221	# rule 5
	222
57a6839d	223	($BackALetterEx \| $BackHebrew_LetterEx) ($BackALetterEx \| $BackHebrew_LetterEx);
374ca955 A	224
	225	# rule 6 and 7
	226
57a6839d A	227	($BackALetterEx \| $BackHebrew_LetterEx) ($BackMidLetterEx \| $BackMidNumLetEx \| $BackSingle_QuoteEx) ($BackALetterEx \| $BackHebrew_LetterEx);
	228
	229	# rule 7a
	230	$BackSingle_QuoteEx $BackHebrew_LetterEx;
374ca955	231
57a6839d A	232	# Rule 7b and 7c
57a6839d A	233	$BackHebrew_LetterEx $BackDouble_QuoteEx $BackHebrew_LetterEx;
374ca955 A	234
	235	# rule 8
	236
73c04bcf	237	$BackNumericEx $BackNumericEx;
374ca955 A	238
	239	# rule 9
	240
57a6839d	241	$BackNumericEx ($BackALetterEx \| $BackHebrew_LetterEx);
374ca955 A	242
	243	# rule 10
	244
57a6839d	245	($BackALetterEx \| $BackHebrew_LetterEx) $BackNumericEx;
374ca955 A	246
	247	# rule 11 and 12
	248
57a6839d	249	$BackNumericEx ($BackMidNumEx \| $BackMidNumLetEx \| $BackSingle_QuoteEx) $BackNumericEx;
374ca955 A	250
	251	# rule 13
	252
73c04bcf	253	$BackKatakanaEx $BackKatakanaEx;
374ca955 A	254
374ca955 A	255	# rules 13 a/b
b75a7d8f	256	#
57a6839d	257	$BackExtendNumLetEx ($BackALetterEx \| $BackHebrew_LetterEx \| $BackNumericEx \| $BackKatakanaEx \| $BackExtendNumLetEx);
2ca993e8	258	($BackALetterEx \| $BackHebrew_LetterEx \| $BackNumericEx \| $BackKatakanaEx) $BackExtendNumLetEx;
374ca955	259
51004dcb A	260	# rule 13c
51004dcb A	261
2ca993e8 A	262	^$BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
	263	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
	264	^$BackRegional_IndicatorEx $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
	265	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
08b89b0a	266
2ca993e8 A	267	$GAZ $ZWJ $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
	268	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
	269	$GAZ $ZWJ $BackRegional_IndicatorEx $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
	270	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
51004dcb A	271
	272	# special handling for CJK characters: chain for later dictionary segmentation
	273	$HangulSyllable $HangulSyllable;
	274	$KanaKanji $KanaKanji; #different rule status if both kanji and kana found
	275
2ca993e8 A	276	# rule 13d
	277
	278	$E_Modifier ($Format \| $Extend \| $ZWJ)* ($E_Base \| $GAZ);
	279
	280
	281
374ca955 A	282	## -------------------------------------------------
	283
	284	!!safe_reverse;
	285
	286	# rule 3
2ca993e8	287	($Extend \| $Format \| $ZWJ)+ .?;
374ca955 A	288
374ca955 A	289	# rule 6
57a6839d A	290	($MidLetter \| $MidNumLet \| $Single_Quote) ($BackALetterEx \| $BackHebrew_LetterEx);
	291
	292	# rule 7b
	293	$Double_Quote $BackHebrew_LetterEx;
	294
374ca955 A	295
374ca955 A	296	# rule 11
57a6839d	297	($MidNum \| $MidNumLet \| $Single_Quote) $BackNumericEx;
73c04bcf	298
2ca993e8 A	299	# rule 13c
	300	$BackRegional_IndicatorEx*;
	301
73c04bcf A	302	# For dictionary-based break
73c04bcf A	303	$dictionary $dictionary;
374ca955 A	304
	305	## -------------------------------------------------
	306
	307	!!safe_forward;
	308
374ca955	309	# rule 4
2ca993e8	310	($Extend \| $Format \| $ZWJ)+ .?;
374ca955 A	311
374ca955 A	312	# rule 6
57a6839d A	313	($MidLetterEx \| $MidNumLetEx \| $Single_QuoteEx) ($ALetterEx \| $Hebrew_LetterEx);
	314
	315	# rule 7b
	316	$Double_QuoteEx $Hebrew_LetterEx;
b75a7d8f	317
374ca955	318	# rule 11
57a6839d	319	($MidNumEx \| $MidNumLetEx \| $Single_QuoteEx) $NumericEx;
73c04bcf	320
2ca993e8 A	321	# rule 13c
	322	$Regional_IndicatorEx*;
	323
73c04bcf A	324	# For dictionary-based break
73c04bcf A	325	$dictionary $dictionary;