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

#
# Copyright (C) 2016 and later: Unicode, Inc. and others.
# License & terms of use: http://www.unicode.org/copyright.html
# 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 29 for Unicode Version 9.0
#      with additions for Emoji Sequences from https://goo.gl/cluFCn
#
# 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.
#

$CR                 = [\p{Word_Break = CR}];
$LF                 = [\p{Word_Break = LF}];
$Newline            = [\p{Word_Break = Newline} ];
$Extend             = [\p{Word_Break = Extend}];
$ZWJ                = [\p{Word_Break = ZWJ}];
$Regional_Indicator = [\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}];
$E_Base             = [[\p{Word_Break = EB}] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
$E_Modifier         = [\p{Word_Break = EM}];

# Data for Extended Pictographic scraped from CLDR common/properties/ExtendedPictographic.txt, r13267
$Extended_Pict = [\U0001F774-\U0001F77F\U00002700-\U00002701\U00002703-\U00002704\U0000270E\U00002710-\U00002711\U00002765-\U00002767\U0001F030-\U0001F093\U0001F094-\U0001F09F\U0001F10D-\U0001F10F\U0001F12F\U0001F16C-\U0001F16F\U0001F1AD-\U0001F1E5\U0001F260-\U0001F265\U0001F203-\U0001F20F\U0001F23C-\U0001F23F\U0001F249-\U0001F24F\U0001F252-\U0001F25F\U0001F266-\U0001F2FF\U0001F7D5-\U0001F7FF\U0001F000-\U0001F003\U0001F005-\U0001F02B\U0001F02C-\U0001F02F\U0001F322-\U0001F323\U0001F394-\U0001F395\U0001F398\U0001F39C-\U0001F39D\U0001F3F1-\U0001F3F2\U0001F3F6\U0001F4FE\U0001F53E-\U0001F548\U0001F54F\U0001F568-\U0001F56E\U0001F571-\U0001F572\U0001F57B-\U0001F586\U0001F588-\U0001F589\U0001F58E-\U0001F58F\U0001F591-\U0001F594\U0001F597-\U0001F5A3\U0001F5A6-\U0001F5A7\U0001F5A9-\U0001F5B0\U0001F5B3-\U0001F5BB\U0001F5BD-\U0001F5C1\U0001F5C5-\U0001F5D0\U0001F5D4-\U0001F5DB\U0001F5DF-\U0001F5E0\U0001F5E2\U0001F5E4-\U0001F5E7\U0001F5E9-\U0001F5EE\U0001F5F0-\U0001F5F2\U0001F5F4-\U0001F5F9\U00002605\U00002607-\U0000260D\U0000260F-\U00002610\U00002612\U00002616-\U00002617\U00002619-\U0000261C\U0000261E-\U0000261F\U00002621\U00002624-\U00002625\U00002627-\U00002629\U0000262B-\U0000262D\U00002630-\U00002637\U0000263B-\U00002647\U00002654-\U0000265F\U00002661-\U00002662\U00002664\U00002667\U00002669-\U0000267A\U0000267C-\U0000267E\U00002680-\U00002691\U00002695\U00002698\U0000269A\U0000269D-\U0000269F\U000026A2-\U000026A9\U000026AC-\U000026AF\U000026B2-\U000026BC\U000026BF-\U000026C3\U000026C6-\U000026C7\U000026C9-\U000026CD\U000026D0\U000026D2\U000026D5-\U000026E8\U000026EB-\U000026EF\U000026F6\U000026FB-\U000026FC\U000026FE-\U000026FF\U00002388\U0001FA00-\U0001FFFD\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F0AF-\U0001F0B0\U0001F0C0\U0001F0D0\U0001F0F6-\U0001F0FF\U0001F80C-\U0001F80F\U0001F848-\U0001F84F\U0001F85A-\U0001F85F\U0001F888-\U0001F88F\U0001F8AE-\U0001F8FF\U0001F900-\U0001F90B\U0001F91F\U0001F928-\U0001F92F\U0001F931-\U0001F932\U0001F94C\U0001F95F-\U0001F96B\U0001F992-\U0001F997\U0001F9D0-\U0001F9E6\U0001F90C-\U0001F90F\U0001F93F\U0001F94D-\U0001F94F\U0001F96C-\U0001F97F\U0001F998-\U0001F9BF\U0001F9C1-\U0001F9CF\U0001F9E7-\U0001F9FF\U0001F6C6-\U0001F6CA\U0001F6D3-\U0001F6D4\U0001F6E6-\U0001F6E8\U0001F6EA\U0001F6F1-\U0001F6F2\U0001F6F7-\U0001F6F8\U0001F6D5-\U0001F6DF\U0001F6ED-\U0001F6EF\U0001F6F9-\U0001F6FF];
$EBG                = [\p{Word_Break = EBG}];
$EmojiNRK           = [[\p{Emoji}] - [\p{Word_Break = Regional_Indicator}\u002a\u00230-9©®™〰〽]];

$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 (Extended_Pict | EmojiNRK).  Precedes WB4, so no intervening Extend chars allowed.
#
$ZWJ ($Extended_Pict | $EmojiNRK);


# 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)*;
$Extended_Pict ($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 14
#    Do not break within emoji modifier sequences

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

# rules 15 - 17
#    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 


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

!!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 (Extended_Pict | EmojiNRK).  Precedes WB4, so no intervening Extend chars allowed.
#
($Extended_Pict | $EmojiNRK) $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; 

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

# rule 14

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

# rule 15 - 17
#    Pairs of Regional Indicators stay together.

^$BackRegional_IndicatorEx / ($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}];

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


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

!!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
f3c0d7a5 A	1	#
	2	# Copyright (C) 2016 and later: Unicode, Inc. and others.
	3	# License & terms of use: http://www.unicode.org/copyright.html
	4	# Copyright (C) 2002-2016, International Business Machines Corporation
374ca955	5	# and others. All Rights Reserved.
b75a7d8f	6	#
2ca993e8	7	# file: word_fi_sv.txt
b75a7d8f	8	#
2ca993e8	9	# ICU Word Break Rules, fi/sv locales (these are actually the standard UAX #29 rules)
b75a7d8f	10	# See Unicode Standard Annex #29.
f3c0d7a5 A	11	# These rules are based on UAX #29 Revision 29 for Unicode Version 9.0
f3c0d7a5 A	12	# with additions for Emoji Sequences from https://goo.gl/cluFCn
b75a7d8f	13	#
73c04bcf	14	# Note: Updates to word.txt will usually need to be merged into
51004dcb	15	# word_POSIX.txt also.
b75a7d8f	16
374ca955	17	##############################################################################
b75a7d8f A	18	#
	19	# Character class definitions from TR 29
	20	#
374ca955 A	21	##############################################################################
	22
	23	!!chain;
	24
b75a7d8f A	25
	26	#
	27	# Character Class Definitions.
b75a7d8f	28	#
b75a7d8f	29
57a6839d A	30	$CR = [\p{Word_Break = CR}];
57a6839d A	31	$LF = [\p{Word_Break = LF}];
2ca993e8	32	$Newline = [\p{Word_Break = Newline} ];
f3c0d7a5 A	33	$Extend = [\p{Word_Break = Extend}];
f3c0d7a5 A	34	$ZWJ = [\p{Word_Break = ZWJ}];
2ca993e8	35	$Regional_Indicator = [\p{Word_Break = Regional_Indicator}];
f3c0d7a5	36	$Format = [\p{Word_Break = Format}];
57a6839d A	37	$Katakana = [\p{Word_Break = Katakana}];
	38	$Hebrew_Letter = [\p{Word_Break = Hebrew_Letter}];
	39	$ALetter = [\p{Word_Break = ALetter}];
	40	$Single_Quote = [\p{Word_Break = Single_Quote}];
	41	$Double_Quote = [\p{Word_Break = Double_Quote}];
	42	$MidNumLet = [\p{Word_Break = MidNumLet}];
	43	$MidLetter = [\p{Word_Break = MidLetter}];
	44	$MidNum = [\p{Word_Break = MidNum}];
	45	$Numeric = [\p{Word_Break = Numeric}];
	46	$ExtendNumLet = [\p{Word_Break = ExtendNumLet}];
f3c0d7a5 A	47	$E_Base = [[\p{Word_Break = EB}] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
	48	$E_Modifier = [\p{Word_Break = EM}];
	49
	50	# Data for Extended Pictographic scraped from CLDR common/properties/ExtendedPictographic.txt, r13267
	51	$Extended_Pict = [\U0001F774-\U0001F77F\U00002700-\U00002701\U00002703-\U00002704\U0000270E\U00002710-\U00002711\U00002765-\U00002767\U0001F030-\U0001F093\U0001F094-\U0001F09F\U0001F10D-\U0001F10F\U0001F12F\U0001F16C-\U0001F16F\U0001F1AD-\U0001F1E5\U0001F260-\U0001F265\U0001F203-\U0001F20F\U0001F23C-\U0001F23F\U0001F249-\U0001F24F\U0001F252-\U0001F25F\U0001F266-\U0001F2FF\U0001F7D5-\U0001F7FF\U0001F000-\U0001F003\U0001F005-\U0001F02B\U0001F02C-\U0001F02F\U0001F322-\U0001F323\U0001F394-\U0001F395\U0001F398\U0001F39C-\U0001F39D\U0001F3F1-\U0001F3F2\U0001F3F6\U0001F4FE\U0001F53E-\U0001F548\U0001F54F\U0001F568-\U0001F56E\U0001F571-\U0001F572\U0001F57B-\U0001F586\U0001F588-\U0001F589\U0001F58E-\U0001F58F\U0001F591-\U0001F594\U0001F597-\U0001F5A3\U0001F5A6-\U0001F5A7\U0001F5A9-\U0001F5B0\U0001F5B3-\U0001F5BB\U0001F5BD-\U0001F5C1\U0001F5C5-\U0001F5D0\U0001F5D4-\U0001F5DB\U0001F5DF-\U0001F5E0\U0001F5E2\U0001F5E4-\U0001F5E7\U0001F5E9-\U0001F5EE\U0001F5F0-\U0001F5F2\U0001F5F4-\U0001F5F9\U00002605\U00002607-\U0000260D\U0000260F-\U00002610\U00002612\U00002616-\U00002617\U00002619-\U0000261C\U0000261E-\U0000261F\U00002621\U00002624-\U00002625\U00002627-\U00002629\U0000262B-\U0000262D\U00002630-\U00002637\U0000263B-\U00002647\U00002654-\U0000265F\U00002661-\U00002662\U00002664\U00002667\U00002669-\U0000267A\U0000267C-\U0000267E\U00002680-\U00002691\U00002695\U00002698\U0000269A\U0000269D-\U0000269F\U000026A2-\U000026A9\U000026AC-\U000026AF\U000026B2-\U000026BC\U000026BF-\U000026C3\U000026C6-\U000026C7\U000026C9-\U000026CD\U000026D0\U000026D2\U000026D5-\U000026E8\U000026EB-\U000026EF\U000026F6\U000026FB-\U000026FC\U000026FE-\U000026FF\U00002388\U0001FA00-\U0001FFFD\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F0AF-\U0001F0B0\U0001F0C0\U0001F0D0\U0001F0F6-\U0001F0FF\U0001F80C-\U0001F80F\U0001F848-\U0001F84F\U0001F85A-\U0001F85F\U0001F888-\U0001F88F\U0001F8AE-\U0001F8FF\U0001F900-\U0001F90B\U0001F91F\U0001F928-\U0001F92F\U0001F931-\U0001F932\U0001F94C\U0001F95F-\U0001F96B\U0001F992-\U0001F997\U0001F9D0-\U0001F9E6\U0001F90C-\U0001F90F\U0001F93F\U0001F94D-\U0001F94F\U0001F96C-\U0001F97F\U0001F998-\U0001F9BF\U0001F9C1-\U0001F9CF\U0001F9E7-\U0001F9FF\U0001F6C6-\U0001F6CA\U0001F6D3-\U0001F6D4\U0001F6E6-\U0001F6E8\U0001F6EA\U0001F6F1-\U0001F6F2\U0001F6F7-\U0001F6F8\U0001F6D5-\U0001F6DF\U0001F6ED-\U0001F6EF\U0001F6F9-\U0001F6FF];
	52	$EBG = [\p{Word_Break = EBG}];
	53	$EmojiNRK = [[\p{Emoji}] - [\p{Word_Break = Regional_Indicator}\u002a\u00230-9©®™〰〽]];
57a6839d A	54
	55	$Han = [:Han:];
	56	$Hiragana = [:Hiragana:];
73c04bcf A	57
73c04bcf A	58
73c04bcf A	59	# Dictionary character set, for triggering language-based break engines. Currently
	60	# limited to LineBreak=Complex_Context. Note that this set only works in Unicode
	61	# 5.0 or later as the definition of Complex_Context was corrected to include all
	62	# characters requiring dictionary break.
	63
f3c0d7a5	64	$Control = [\p{Grapheme_Cluster_Break = Control}];
51004dcb A	65	$HangulSyllable = [\uac00-\ud7a3];
	66	$ComplexContext = [:LineBreak = Complex_Context:];
	67	$KanaKanji = [$Han $Hiragana $Katakana];
	68	$dictionaryCJK = [$KanaKanji $HangulSyllable];
	69	$dictionary = [$ComplexContext $dictionaryCJK];
	70
	71	# leave CJK scripts out of ALetterPlus
	72	$ALetterPlus = [$ALetter-$dictionaryCJK [$ComplexContext-$Extend-$Control]];
	73
73c04bcf A	74
73c04bcf A	75	#
f3c0d7a5	76	# Rules 4 Ignore Format and Extend characters,
46f4442e	77	# except when they appear at the beginning of a region of text.
73c04bcf	78	#
51004dcb	79	# TODO: check if handling of katakana in dictionary makes rules incorrect/void
2ca993e8 A	80	$KatakanaEx = $Katakana ($Extend \| $Format \| $ZWJ)*;
	81	$Hebrew_LetterEx = $Hebrew_Letter ($Extend \| $Format \| $ZWJ)*;
	82	$ALetterEx = $ALetterPlus ($Extend \| $Format \| $ZWJ)*;
	83	$Single_QuoteEx = $Single_Quote ($Extend \| $Format \| $ZWJ)*;
	84	$Double_QuoteEx = $Double_Quote ($Extend \| $Format \| $ZWJ)*;
	85	$MidNumLetEx = $MidNumLet ($Extend \| $Format \| $ZWJ)*;
	86	$MidLetterEx = $MidLetter ($Extend \| $Format \| $ZWJ)*;
	87	$MidNumEx = $MidNum ($Extend \| $Format \| $ZWJ)*;
	88	$NumericEx = $Numeric ($Extend \| $Format \| $ZWJ)*;
	89	$ExtendNumLetEx = $ExtendNumLet ($Extend \| $Format \| $ZWJ)*;
	90	$Regional_IndicatorEx = $Regional_Indicator ($Extend \| $Format \| $ZWJ)*;
73c04bcf	91
46f4442e	92	$Ideographic = [\p{Ideographic}];
2ca993e8 A	93	$HiraganaEx = $Hiragana ($Extend \| $Format \| $ZWJ)*;
2ca993e8 A	94	$IdeographicEx = $Ideographic ($Extend \| $Format \| $ZWJ)*;
b75a7d8f	95
374ca955	96	## -------------------------------------------------
b75a7d8f	97
374ca955	98	!!forward;
b75a7d8f	99
b75a7d8f	100
73c04bcf	101	# Rule 3 - CR x LF
46f4442e A	102	#
46f4442e A	103	$CR $LF;
73c04bcf	104
f3c0d7a5	105	# Rule 3c ZWJ x (Extended_Pict \| EmojiNRK). Precedes WB4, so no intervening Extend chars allowed.
2ca993e8	106	#
f3c0d7a5	107	$ZWJ ($Extended_Pict \| $EmojiNRK);
2ca993e8 A	108
2ca993e8 A	109
73c04bcf A	110	# Rule 4 - ignore Format and Extend characters, except when they appear at the beginning
	111	# of a region of Text. The rule here comes into play when the start of text
	112	# begins with a group of Format chars, or with a "word" consisting of a single
	113	# char that is not in any of the listed word break categories followed by
51004dcb	114	# format char(s), or is not a CJK dictionary character.
2ca993e8	115	[^$CR $LF $Newline]? ($Extend \| $Format \| $ZWJ)+;
b75a7d8f	116
374ca955 A	117	$NumericEx {100};
374ca955 A	118	$ALetterEx {200};
51004dcb	119	$HangulSyllable {200};
57a6839d	120	$Hebrew_LetterEx{200};
51004dcb A	121	$KatakanaEx {400}; # note: these status values override those from rule 5
51004dcb A	122	$HiraganaEx {400}; # by virtue of being numerically larger.
46f4442e	123	$IdeographicEx {400}; #
b75a7d8f	124
2ca993e8 A	125	$E_Base ($Extend \| $Format \| $ZWJ)*;
2ca993e8 A	126	$E_Modifier ($Extend \| $Format \| $ZWJ)*;
f3c0d7a5	127	$Extended_Pict ($Extend \| $Format \| $ZWJ)*;
2ca993e8	128
46f4442e	129	#
374ca955	130	# rule 5
46f4442e A	131	# Do not break between most letters.
46f4442e A	132	#
57a6839d	133	($ALetterEx \| $Hebrew_LetterEx) ($ALetterEx \| $Hebrew_LetterEx) {200};
b75a7d8f	134
374ca955	135	# rule 6 and 7
57a6839d A	136	($ALetterEx \| $Hebrew_LetterEx) ($MidLetterEx \| $MidNumLetEx \| $Single_QuoteEx) ($ALetterEx \| $Hebrew_LetterEx) {200};
	137
	138	# rule 7a
	139	$Hebrew_LetterEx $Single_QuoteEx {200};
	140
	141	# rule 7b and 7c
	142	$Hebrew_LetterEx $Double_QuoteEx $Hebrew_LetterEx {200};
b75a7d8f	143
374ca955	144	# rule 8
b75a7d8f	145
73c04bcf	146	$NumericEx $NumericEx {100};
b75a7d8f	147
374ca955	148	# rule 9
b75a7d8f	149
57a6839d	150	($ALetterEx \| $Hebrew_LetterEx) $NumericEx {200};
b75a7d8f	151
374ca955	152	# rule 10
b75a7d8f	153
57a6839d	154	$NumericEx ($ALetterEx \| $Hebrew_LetterEx) {200};
b75a7d8f	155
f3c0d7a5	156	# rule 11 and 12
374ca955	157
57a6839d	158	$NumericEx ($MidNumEx \| $MidNumLetEx \| $Single_QuoteEx) $NumericEx {100};
374ca955 A	159
374ca955 A	160	# rule 13
51004dcb A	161	# to be consistent with $KanaKanji $KanaKanhi, changed
	162	# from 300 to 400.
	163	# See also TestRuleStatus in intltest/rbbiapts.cpp
	164	$KatakanaEx $KatakanaEx {400};
374ca955 A	165
	166	# rule 13a/b
	167
57a6839d A	168	$ALetterEx $ExtendNumLetEx {200}; # (13a)
	169	$Hebrew_LetterEx $ExtendNumLetEx {200}; # (13a)
	170	$NumericEx $ExtendNumLetEx {100}; # (13a)
	171	$KatakanaEx $ExtendNumLetEx {400}; # (13a)
	172	$ExtendNumLetEx $ExtendNumLetEx {200}; # (13a)
374ca955	173
57a6839d A	174	$ExtendNumLetEx $ALetterEx {200}; # (13b)
	175	$ExtendNumLetEx $Hebrew_Letter {200}; # (13b)
	176	$ExtendNumLetEx $NumericEx {100}; # (13b)
	177	$ExtendNumLetEx $KatakanaEx {400}; # (13b)
51004dcb	178
f3c0d7a5 A	179	# rule 14
	180	# Do not break within emoji modifier sequences
	181
	182	($E_Base \| $EBG) ($Format \| $Extend \| $ZWJ)* $E_Modifier;
	183
	184	# rules 15 - 17
2ca993e8 A	185	# Pairs of Regional Indicators stay together.
	186	# With rule chaining disabled by ^, this rule will match exactly two of them.
	187	# No other rule begins with a Regional_Indicator, so chaining cannot extend the match.
	188	#
	189	^$Regional_IndicatorEx $Regional_IndicatorEx;
51004dcb A	190
	191	# special handling for CJK characters: chain for later dictionary segmentation
	192	$HangulSyllable $HangulSyllable {200};
f3c0d7a5	193	$KanaKanji $KanaKanji {400}; # different rule status if both kana and kanji found
374ca955 A	194
	195
	196	## -------------------------------------------------
b75a7d8f	197
374ca955 A	198	!!reverse;
374ca955 A	199
2ca993e8 A	200	$BackHebrew_LetterEx = ($Format \| $Extend \| $ZWJ)* $Hebrew_Letter;
	201	$BackALetterEx = ($Format \| $Extend \| $ZWJ)* $ALetterPlus;
	202	$BackSingle_QuoteEx = ($Format \| $Extend \| $ZWJ)* $Single_Quote;
	203	$BackDouble_QuoteEx = ($Format \| $Extend \| $ZWJ)* $Double_Quote;
	204	$BackMidNumLetEx = ($Format \| $Extend \| $ZWJ)* $MidNumLet;
	205	$BackNumericEx = ($Format \| $Extend \| $ZWJ)* $Numeric;
	206	$BackMidNumEx = ($Format \| $Extend \| $ZWJ)* $MidNum;
	207	$BackMidLetterEx = ($Format \| $Extend \| $ZWJ)* $MidLetter;
	208	$BackKatakanaEx = ($Format \| $Extend \| $ZWJ)* $Katakana;
	209	$BackHiraganaEx = ($Format \| $Extend \| $ZWJ)* $Hiragana;
	210	$BackExtendNumLetEx = ($Format \| $Extend \| $ZWJ)* $ExtendNumLet;
	211	$BackRegional_IndicatorEx = ($Format \| $Extend \| $ZWJ)* $Regional_Indicator;
374ca955	212
73c04bcf	213	# rule 3
46f4442e	214	$LF $CR;
374ca955	215
f3c0d7a5	216	# Rule 3c ZWJ x (Extended_Pict \| EmojiNRK). Precedes WB4, so no intervening Extend chars allowed.
2ca993e8	217	#
f3c0d7a5	218	($Extended_Pict \| $EmojiNRK) $ZWJ;
2ca993e8	219
73c04bcf	220	# rule 4
2ca993e8	221	($Format \| $Extend \| $ZWJ)* [^$CR $LF $Newline]?;
374ca955 A	222
	223	# rule 5
	224
57a6839d	225	($BackALetterEx \| $BackHebrew_LetterEx) ($BackALetterEx \| $BackHebrew_LetterEx);
374ca955 A	226
	227	# rule 6 and 7
	228
57a6839d A	229	($BackALetterEx \| $BackHebrew_LetterEx) ($BackMidLetterEx \| $BackMidNumLetEx \| $BackSingle_QuoteEx) ($BackALetterEx \| $BackHebrew_LetterEx);
	230
	231	# rule 7a
	232	$BackSingle_QuoteEx $BackHebrew_LetterEx;
374ca955	233
57a6839d A	234	# Rule 7b and 7c
57a6839d A	235	$BackHebrew_LetterEx $BackDouble_QuoteEx $BackHebrew_LetterEx;
374ca955 A	236
	237	# rule 8
	238
73c04bcf	239	$BackNumericEx $BackNumericEx;
374ca955 A	240
	241	# rule 9
	242
57a6839d	243	$BackNumericEx ($BackALetterEx \| $BackHebrew_LetterEx);
374ca955 A	244
	245	# rule 10
	246
57a6839d	247	($BackALetterEx \| $BackHebrew_LetterEx) $BackNumericEx;
374ca955 A	248
	249	# rule 11 and 12
	250
57a6839d	251	$BackNumericEx ($BackMidNumEx \| $BackMidNumLetEx \| $BackSingle_QuoteEx) $BackNumericEx;
374ca955 A	252
	253	# rule 13
	254
73c04bcf	255	$BackKatakanaEx $BackKatakanaEx;
374ca955 A	256
374ca955 A	257	# rules 13 a/b
b75a7d8f	258	#
57a6839d	259	$BackExtendNumLetEx ($BackALetterEx \| $BackHebrew_LetterEx \| $BackNumericEx \| $BackKatakanaEx \| $BackExtendNumLetEx);
f3c0d7a5	260	($BackALetterEx \| $BackHebrew_LetterEx \| $BackNumericEx \| $BackKatakanaEx) $BackExtendNumLetEx;
374ca955	261
f3c0d7a5 A	262	# special handling for CJK characters: chain for later dictionary segmentation
	263	$HangulSyllable $HangulSyllable;
	264	$KanaKanji $KanaKanji; #different rule status if both kanji and kana found
	265
	266	# rule 14
	267
	268	$E_Modifier ($Format \| $Extend \| $ZWJ)* ($E_Base \| $EBG);
51004dcb	269
f3c0d7a5 A	270	# rule 15 - 17
	271	# Pairs of Regional Indicators stay together.
	272
	273	^$BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
2ca993e8	274	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
f3c0d7a5	275	^$BackRegional_IndicatorEx $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
2ca993e8	276	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
08b89b0a	277
f3c0d7a5	278	($Extended_Pict \| $EmojiNRK) $ZWJ $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
2ca993e8	279	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
f3c0d7a5	280	($Extended_Pict \| $EmojiNRK) $ZWJ $BackRegional_IndicatorEx $BackRegional_IndicatorEx / ($BackRegional_IndicatorEx $BackRegional_IndicatorEx)*
2ca993e8	281	($Format \| $Extend \| $ZWJ)* [[^$Regional_Indicator $Format $Extend $ZWJ] {eof}];
51004dcb	282
2ca993e8 A	283
2ca993e8 A	284
374ca955 A	285	## -------------------------------------------------
	286
	287	!!safe_reverse;
	288
	289	# rule 3
2ca993e8	290	($Extend \| $Format \| $ZWJ)+ .?;
374ca955 A	291
374ca955 A	292	# rule 6
57a6839d A	293	($MidLetter \| $MidNumLet \| $Single_Quote) ($BackALetterEx \| $BackHebrew_LetterEx);
	294
	295	# rule 7b
	296	$Double_Quote $BackHebrew_LetterEx;
	297
374ca955 A	298
374ca955 A	299	# rule 11
57a6839d	300	($MidNum \| $MidNumLet \| $Single_Quote) $BackNumericEx;
73c04bcf	301
2ca993e8 A	302	# rule 13c
	303	$BackRegional_IndicatorEx*;
	304
73c04bcf A	305	# For dictionary-based break
73c04bcf A	306	$dictionary $dictionary;
374ca955 A	307
	308	## -------------------------------------------------
	309
	310	!!safe_forward;
	311
374ca955	312	# rule 4
2ca993e8	313	($Extend \| $Format \| $ZWJ)+ .?;
374ca955 A	314
374ca955 A	315	# rule 6
57a6839d A	316	($MidLetterEx \| $MidNumLetEx \| $Single_QuoteEx) ($ALetterEx \| $Hebrew_LetterEx);
	317
	318	# rule 7b
	319	$Double_QuoteEx $Hebrew_LetterEx;
b75a7d8f	320
374ca955	321	# rule 11
57a6839d	322	($MidNumEx \| $MidNumLetEx \| $Single_QuoteEx) $NumericEx;
73c04bcf	323
2ca993e8 A	324	# rule 13c
	325	$Regional_IndicatorEx*;
	326
73c04bcf A	327	# For dictionary-based break
73c04bcf A	328	$dictionary $dictionary;