2 # Copyright (C) 2016 and later: Unicode, Inc. and others.
3 # License & terms of use: http://www.unicode.org/copyright.html
4 # Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved.
8 # Reference Line Break rules for intltest rbbi/RBBIMonkeyTest.
9 # Rules derived from Unicode Standard Annex #14 Revision 44 for Unicode 13.0
11 # Note: Rule syntax and the monkey test itself are still a work in progress.
12 # They are expected to change with review and the addition of support for rule tailoring.
19 AI = [:LineBreak = Ambiguous:];
20 AL = [:LineBreak = Alphabetic:];
21 BA = [:LineBreak = Break_After:];
22 HH = [\u2010]; # \u2010 is HYPHEN, default line break is BA.
23 BB = [:LineBreak = Break_Before:];
24 BK = [:LineBreak = Mandatory_Break:];
25 B2 = [:LineBreak = Break_Both:];
26 CB = [:LineBreak = Contingent_Break:];
27 CJ = [:LineBreak = Conditional_Japanese_Starter:];
28 CL = [[:LineBreak = Close_Punctuation:] \u201d];
29 CMS = [:LineBreak = Combining_Mark:];
30 CP = [:LineBreak = Close_Parenthesis:];
31 CR = [:LineBreak = Carriage_Return:];
32 EB = [:LineBreak = EB:];
33 EM = [:LineBreak = EM:];
34 EX = [:LineBreak = Exclamation:];
35 GL = [:LineBreak = Glue:];
36 HL = [:LineBreak = Hebrew_Letter:];
37 HY = [:LineBreak = Hyphen:];
38 H2 = [:LineBreak = H2:];
39 H3 = [:LineBreak = H3:];
40 ID = [:LineBreak = Ideographic:];
41 IN = [:LineBreak = Inseperable:];
42 IS = [:LineBreak = Infix_Numeric:];
43 JL = [:LineBreak = JL:];
44 JV = [:LineBreak = JV:];
45 JT = [:LineBreak = JT:];
46 LF = [:LineBreak = Line_Feed:];
47 NL = [:LineBreak = Next_Line:];
48 NS = [[:LineBreak = Nonstarter:] CJ]; # CSS Strict tailoring: CJ resolves to NS.
49 NU = [:LineBreak = Numeric:];
50 OP = [[:LineBreak = Open_Punctuation:] \u201c];
51 PO = [:LineBreak = Postfix_Numeric:];
52 PR = [:LineBreak = Prefix_Numeric:];
53 QU = [[:LineBreak = Quotation:] - [\u201c\u201d]];
54 RI = [:LineBreak = Regional_Indicator:];
55 SA = [:LineBreak = Complex_Context:];
56 SG = [:LineBreak = Surrogate:];
57 SP = [:LineBreak = Space:];
58 SY = [:LineBreak = Break_Symbols:];
59 WJ = [:LineBreak = Word_Joiner:];
60 XX = [:LineBreak = Unknown:];
61 ZW = [:LineBreak = ZWSpace:];
62 ZWJ = [:LineBreak = ZWJ:];
64 # OP30 and CP30 are variants of OP and CP that appear in rule LB30 from UAX 14.
65 # Limitations of this monkey test rule parser require that these definitions be pulled out
66 # rather than appearing in-line in LB 30.
68 OP30 = [OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
69 CP30 = [CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
71 # OP30 and CP30 are variants of OP and CP that appear in rule LB30 from UAX 14.
72 # Limitations of this monkey test rule parser require that these definitions be pulled out
73 # rather than appearing in-line in LB 30.
75 OP30 = [OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
76 CP30 = [CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
78 # LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
82 # By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
83 # list it in the numerous rules that use CM.
92 LB6: . (BK | CR | LF | NL);
93 LB6.1: [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
95 # LB8 break after ZW SP*.
96 # Precedes LB7 because both rules will match the sequences like ZW SP,
97 # and LB8 must take precedence.
99 LB8: ZW SP* ÷ [^ZW SP BK CR LF NL];
101 # Numbers. Equivalent to Tailoring example 8 from UAX 14.
102 # Moved up, before LB14, because it can match longer sequences which must take precedence.
103 LB25: ((PR | PO)CM*)? ((OP | HY)CM*)? (IS CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?;
106 # Moved up, before LB7, because they can match a longer sequence that would also match LB7.
107 # For example, the sequence "OP CM SP AL" matches LB14
108 # while the prefix of it, "OP CM SP" matches LB7.1
111 # LB 14a Break before an IS that begins a number and follows a space.
112 LB14a: SP ÷ IS CM* NU;
115 LB14b.1: [^SP] CM* IS;
119 LB16: (CL | CP)CM* SP* NS;
123 # LB7 Do not break before spaces or zero width space.
125 LB7.1: [^ZW SP] CM* [SP ZW];
126 LB7.2: [ZW SP] [SP ZW];
130 # Don't match a CM on the right - let other rules pick up CM sequences, where
131 # the ZWJ behaves as just another generic CM.
136 # LB10: Unattached CM -> AL
141 LB11.1: [^SP] CM* WJ;
143 LB11.3: WJ CM* [^CM];
147 LB12a: [^SP BA HY] CM* GL;
149 # LB 13 Do not break before ‘]’ or ‘!’ or ‘/’, even after spaces.
150 LB13.1: [^SP] CM* [CL CP EX SY];
151 LB13.2: SP [CL CP EX SY];
153 # LB 14-17 are moved above LB 7.
158 LB19.1: QU CM* [^CM];
160 # LB 20 Break before and after CB.
161 # Interaction with LB8a: ZWJ x . is tricky because CM includes ZWJ.
162 # ZWJ acts like a CM to the left, combining with CB.
163 # ZWJ acts independently to the right, no break after by LB8a.
164 LB20.1: . CM* ZWJ CB;
167 LB20.3: CB CM* ZWJ [^CM];
170 # LB 20.09 Don't break between Hyphens and Letters when there is a break preceding the hyphen.
171 LB20.09: ^(HY | HH) CM* AL;
173 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
174 # not picking up the continuing match after the BA from 21a.
175 LB21a: HL CM* (HY | BA) CM* [^CM CB];
177 LB21.1: . CM* [BA HY NS];
178 LB21.2: BB CM* [^CM CB];
184 LB23.1: (AL | HL | CM) CM* NU;
185 LB23.2: NU CM* (AL | HL);
187 LB23a.1: PR CM* (ID | EB | EM);
188 LB23a.2: (ID | EB | EM) CM* PO;
190 LB24.2: (PR | PO) CM* (AL | HL);
191 LB24.3: (AL | HL | CM) CM* (PR | PO);
194 LB26.1: JL CM* (JL | JV | H2 | H3);
195 LB26.2: (JV | H2) CM* (JV | JT);
196 LB26.3: (JT | H3) CM* JT;
198 LB27.1: (JL | JV | JT | H2 | H3) CM* IN;
199 LB27.2: (JL | JV | JT | H2 | H3) CM* PO;
200 LB27.3: PR CM* (JL | JV | JT | H2 | H3);
202 # LB28 Do not break between Alphabetics.
203 # Unattached (leading) CM treated as AL.
204 LB28: (AL | HL | CM)CM* (AL | HL);
206 LB29: IS CM* (AL | HL);
208 # LB30 is adjusted for unattached leading CM being treated as AL.
209 LB30.1: (AL | CM | HL | NU) CM* OP30;
210 LB30.2: CP30 CM* (AL | HL | NU);
212 # LB30a keep pairs of RI together.
213 LB30a.1: RI CM* RI ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS IN CM];
214 LB30a.2: RI CM* RI CM* CMS ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS IN CM];
215 LB30a.3: RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS IN ZWJ]?;
217 # LB30b Do not break between Emoji Base and Emoji Modifier
220 # LB31 Break Everywhere Else.
221 # Include combining marks
222 LB31.1: . CM* ZWJ [^CM];