X-Git-Url: https://git.saurik.com/apple/icu.git/blobdiff_plain/a62d09fcbc8ca9da27887e04112ec143e19b1caf..3d1f044b704633e2e541231cd17ae9ecf9ad5c7a:/icuSources/test/testdata/break_rules/line_normal.txt diff --git a/icuSources/test/testdata/break_rules/line_normal.txt b/icuSources/test/testdata/break_rules/line_normal.txt index fc978fa2..86d35826 100644 --- a/icuSources/test/testdata/break_rules/line_normal.txt +++ b/icuSources/test/testdata/break_rules/line_normal.txt @@ -1,25 +1,28 @@ # +# Copyright (C) 2016 and later: Unicode, Inc. and others. +# License & terms of use: http://www.unicode.org/copyright.html # Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved. # # file: line_normal.txt # -# Reference Line Break rules for intltest rbbi/RBBIMonkeyTest +# Reference Line Break rules for intltest rbbi/RBBIMonkeyTest. +# Rules derived from Unicode Standard Annex #14 Revision 40 for Unicode 11.0 # # Note: Rule syntax and the monkey test itself are still a work in progress. # They are expected to change with review and the addition of support for rule tailoring. # # Line Breaking Rules -# Implement default line breaking as defined by +# Implement default line breaking as defined by # Unicode Standard Annex #14 Revision 34 for Unicode 8.0 # http://www.unicode.org/reports/tr14/ -# tailored as noted in 2nd paragraph below.. +# tailored as noted in 2nd paragraph below. # # 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 tailors the line break behavior to correspond to CSS -# line-break=normal (BCP47 -u-lb-normal) as defined for languages other than +# line-break=normal (BCP47 -u-lb-normal) as defined for languages other than # Chinese & Japanese. # It sets characters of class CJ to behave like ID. @@ -27,29 +30,28 @@ type = line; locale = en@lb=normal; -AI = [[:LineBreak = Ambiguous:] - [\u2640\u2642]]; -AL = [[:LineBreak = Alphabetic:]-[\u2695\u2696\u2764]]; +AI = [:LineBreak = Ambiguous:]; +AL = [:LineBreak = Alphabetic:]; BA = [:LineBreak = Break_After:]; +HH = [\u2010]; # \u2010 is HYPHEN, default line break is BA. 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:]; -CM = [[:LineBreak = Combining_Mark:] \u200D]; +CMS = [:LineBreak = Combining_Mark:]; CP = [:LineBreak = Close_Parenthesis:]; CR = [:LineBreak = Carriage_Return:]; - -EB = [\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]; -EM = [\U0001F3FB-\U0001F3FF]; - +EB = [[:LineBreak = EB:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C]; +EM = [:LineBreak = EM:]; EX = [:LineBreak = Exclamation:]; GL = [:LineBreak = Glue:]; HL = [:LineBreak = Hebrew_Letter:]; HY = [:LineBreak = Hyphen:]; H2 = [:LineBreak = H2:]; H3 = [:LineBreak = H3:]; -ID = [[:LineBreak = Ideographic:] CJ [\u2640\u2642\u2695\u2696\u2764]]; +ID = [[:LineBreak = Ideographic:] CJ]; # CSS Normal tailoring: CJ resolves to ID IN = [:LineBreak = Inseperable:]; IS = [:LineBreak = Infix_Numeric:]; JL = [:LineBreak = JL:]; @@ -71,16 +73,15 @@ SY = [:LineBreak = Break_Symbols:]; WJ = [:LineBreak = Word_Joiner:]; XX = [:LineBreak = Unknown:]; ZW = [:LineBreak = ZWSpace:]; -ZJ = [\u200D]; - -# TODO: adjustment to sets needed only until Unicode properties are updated for Emoji. -ID = [ID - EB]; -AL = [AL - EM]; +ZWJ = [:LineBreak = ZWJ:]; -dictionary = [:LineBreak = Complex_Context:]; +# LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes +AL = [AL AI SG XX ]; +dictionary = SA; -# Redfine AL. LB1. TODO: refine according to latest UAX. -AL = [ AL AI SA SG XX ]; +# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly +# list it in the numerous rules that use CM. +CM = [CMS ZWJ]; LB4: BK ÷; LB5: CR LF; @@ -91,26 +92,44 @@ LB5.3: NL ÷; LB6: . (BK | CR | LF | NL); LB6.1: [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL); +# LB8 break after ZW SP*. +# Precedes LB7 because both rules will match the sequences like ZW SP, +# and LB8 must take precedence. + +LB8: ZW SP* ÷ [^ZW SP BK CR LF NL]; + +# Numbers. Equivalent to Tailoring example 8 from UAX 14. +# Moved up, before LB14, because it can match longer sequences which must take precedence. +LB25: ((PR | PO)CM*)? ((OP | HY)CM*)? (IS CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?; + # Rules LB14 - LB17. -# Moved before LB7, because they can match a longer sequence that would also match LB7, -# for example, the sequence "OP CM SP AL" matches LB14 while the prefix of it, -# "while only the prefix "OP CM SP" matches LB7.1 +# Moved up, before LB7, because they can match a longer sequence that would also match LB7. +# For example, the sequence "OP CM SP AL" matches LB14 +# while the prefix of it, "OP CM SP" matches LB7.1 LB14: OP CM* SP* .; + +# LB 14a Break before an IS that begins a number and follows a space. +LB14a: SP ÷ IS CM* NU; + +# LB14b × IS +LB14b.1: [^SP] CM* IS; +LB14b.2: SP IS; + LB15: QU CM* SP* OP; LB16: (CL | CP)CM* SP* NS; LB17: B2 CM* SP* B2; + +# LB7 Do not break before spaces or zero width space. + LB7.1: [^ZW SP] CM* [SP ZW]; LB7.2: [ZW SP] [SP ZW]; -# LB8, ICU differs from UAX-14, -# ICU: ZW ÷; -# UAX 14: ZW SP* ÷; -LB8: ZW ÷; - -# LB8a, from Emoji proposal L2/16-011R3 -# ZWJ x ID -LB8a: ZJ (ID | EB | EM); +# LB8a +# ZWJ x +# Don't match a CM on the right - let other rules pick up CM sequences, where +# the ZWJ behaves as just another generic CM. +LB8a: ZWJ [^CM]; # LB9: X CM -> X @@ -119,7 +138,7 @@ LB8a: ZJ (ID | EB | EM); #LB11: × WJ; # WJ × -LB11.1: [^BK CR LF NL SP ZW] CM* WJ; +LB11.1: [^SP] CM* WJ; LB11.2: SP WJ; LB11.3: WJ CM* [^CM]; @@ -127,15 +146,9 @@ LB12: GL CM* [^CM]; LB12a: [^SP BA HY] CM* GL; -# LB 13 ICU Tailoring, matches tailoring exmaple 8 from UAX 14. -# -# LB13.1 [^SP] CM* [CL CP EX IS SY] # original UAX 14 rule. -# LB13.2 SP CM* [CL CP EX IS SY] - -LB13.1: [^NU SP] CM* [CL CP IS SY]; -LB13.2: [^SP] CM* EX; -LB13.2: SP [CL CP EX IS SY]; - +# LB 13 Do not break before ‘]’ or ‘!’ or ‘/’, even after spaces. +LB13.1: [^SP] CM* [CL CP EX SY]; +LB13.2: SP [CL CP EX SY]; # LB 14-17 are moved above LB 7. @@ -145,12 +158,17 @@ LB19: . CM* QU; LB19.1: QU CM* [^CM]; # LB 20 Break before and after CB. -# Interaction with LB8a: ZJ x ID is tricky because CM includes ZJ. -# ZJ acts like a CM to the left, combining with CB. -# ZJ acts independently to the right, no break from ID by LB8a. -LB20: . CM* ÷ CB; -LB20.1a: CB CM* ZJ (ID | EB | EM); -LB20.1b: CB CM* ÷; +# Interaction with LB8a: ZWJ x . is tricky because CM includes ZWJ. +# ZWJ acts like a CM to the left, combining with CB. +# ZWJ acts independently to the right, no break after by LB8a. +LB20.1: . CM* ZWJ CB; +LB20.2: . CM* ÷ CB; + +LB20.3: CB CM* ZWJ [^CM]; +LB20.4: CB CM* ÷; + +# LB 20.09 Don't break between Hyphens and Letters when there is a break preceding the hyphen. +LB20.09: ^(HY | HH) CM* AL; # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then # not picking up the continuing match after the BA from 21a. @@ -167,17 +185,15 @@ LB22.3: (ID | EB | EM) CM* IN; LB22.4: IN CM* IN; LB22.5: NU CM* IN; -LB23.1: (ID | EB | EM) CM* PO; -LB23.2: (AL | HL | CM) CM* NU; -LB23.3: NU CM* (AL | HL); +LB23.1: (AL | HL | CM) CM* NU; +LB23.2: NU CM* (AL | HL); + +LB23a.1: PR CM* (ID | EB | EM); +LB23a.2: (ID | EB | EM) CM* PO; -LB24.1: PR CM* (ID | EB | EM); -LB24.2: PR CM* (AL | HL); -LB24.3: PO CM* (AL | HL); -LB24.4: (AL | HL | CM) CM* (PR | PO); +LB24.2: (PR | PO) CM* (AL | HL); +LB24.3: (AL | HL | CM) CM* (PR | PO); -# Numbers. Equivalent to Tailoring example 8 from UAx 14. -LB25: ((PR | PO)CM*)? ((OP | HY)CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?; LB26.1: JL CM* (JL | JV | H2 | H3); LB26.2: (JV | H2) CM* (JV | JT); @@ -197,15 +213,15 @@ LB29: IS CM* (AL | HL); LB30.1: (AL | CM | HL | NU) CM* OP; LB30.2: CP CM* (AL | HL | NU); -# LB31 keep pairs of RI together. -LB30a.1: RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS]; -LB30a.2: RI CM* RI CM* ZJ (ID | EB | EM); -LB30a.3: RI CM* RI CM* ÷; +# LB30a keep pairs of RI together. +LB30a.1: RI CM* RI ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS CM]; +LB30a.2: RI CM* RI CM* CMS ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS CM]; +LB30a.3: RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS ZWJ]?; # LB30b Do not break between Emoji Base and Emoji Modifier LB30b: EB CM* EM; # LB31 Break Everywhere Else. # Include combining marks -LB31.1: . CM* ZJ (ID | EB | EM); +LB31.1: . CM* ZWJ [^CM]; LB31.2: . CM* ÷;