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) 2016, International Business Machines Corporation and others. All Rights Reserved.
5#
6# file: line_normal.txt
7#
8# Reference Line Break rules for intltest rbbi/RBBIMonkeyTest.
9# Rules derived from Unicode Standard Annex #14 Revision 40 for Unicode 11.0
10#
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.
13#
14#         Line Breaking Rules
15#         Implement default line breaking as defined by
16#         Unicode Standard Annex #14 Revision 34 for Unicode 8.0
17#         http://www.unicode.org/reports/tr14/
18#         tailored as noted in 2nd paragraph below.
19#
20#         TODO:  Rule LB 8 remains as it was in Unicode 5.2
21#         This is only because of a limitation of ICU break engine implementation,
22#         not because the older behavior is desirable.
23#
24#         This tailors the line break behavior to correspond to CSS
25#         line-break=normal (BCP47 -u-lb-normal) as defined for languages other than
26#         Chinese & Japanese.
27#         It sets characters of class CJ to behave like ID.
28
29
30type = line;
31locale = en@lb=normal;
32
33AI = [:LineBreak =  Ambiguous:];
34AL = [:LineBreak =  Alphabetic:];
35BA = [:LineBreak =  Break_After:];
36HH = [\u2010];      # \u2010 is HYPHEN, default line break is BA.
37BB = [:LineBreak =  Break_Before:];
38BK = [:LineBreak =  Mandatory_Break:];
39B2 = [:LineBreak =  Break_Both:];
40CB = [:LineBreak =  Contingent_Break:];
41CJ = [:LineBreak =  Conditional_Japanese_Starter:];
42CL = [:LineBreak =  Close_Punctuation:];
43CMS = [:LineBreak =  Combining_Mark:];
44CP = [:LineBreak =  Close_Parenthesis:];
45CR = [:LineBreak =  Carriage_Return:];
46EB = [:LineBreak =  EB:];
47EM = [:LineBreak =  EM:];
48EX = [:LineBreak =  Exclamation:];
49GL = [:LineBreak =  Glue:];
50HL = [:LineBreak =  Hebrew_Letter:];
51HY = [:LineBreak =  Hyphen:];
52H2 = [:LineBreak =  H2:];
53H3 = [:LineBreak =  H3:];
54ID = [[:LineBreak =  Ideographic:] CJ];  # CSS Normal tailoring: CJ resolves to ID
55IN = [:LineBreak =  Inseperable:];
56IS = [:LineBreak =  Infix_Numeric:];
57JL = [:LineBreak =  JL:];
58JV = [:LineBreak =  JV:];
59JT = [:LineBreak =  JT:];
60LF = [:LineBreak =  Line_Feed:];
61NL = [:LineBreak =  Next_Line:];
62NS = [:LineBreak =  Nonstarter:];
63NU = [:LineBreak =  Numeric:];
64OP = [:LineBreak =  Open_Punctuation:];
65PO = [:LineBreak =  Postfix_Numeric:];
66PR = [:LineBreak =  Prefix_Numeric:];
67QU = [:LineBreak =  Quotation:];
68RI = [:LineBreak =  Regional_Indicator:];
69SA = [:LineBreak =  Complex_Context:];
70SG = [:LineBreak =  Surrogate:];
71SP = [:LineBreak =  Space:];
72SY = [:LineBreak =  Break_Symbols:];
73WJ = [:LineBreak =  Word_Joiner:];
74XX = [:LineBreak =  Unknown:];
75ZW = [:LineBreak =  ZWSpace:];
76ZWJ = [:LineBreak =  ZWJ:];
77
78# LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
79AL = [AL AI SG XX ];
80dictionary = SA;
81
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.
84CM = [CMS ZWJ];
85
86LB4:        BK ÷;
87LB5:        CR LF;
88LB5.1:      CR ÷;
89LB5.2:      LF ÷;
90LB5.3:      NL ÷;
91
92LB6:        . (BK | CR | LF | NL);
93LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
94
95# Rules LB14 - LB17.
96# Moved before LB7, because they can match a longer sequence that would also match LB7,
97# for example, the sequence "OP CM SP AL" matches LB14 while the prefix of it,
98#                           "while only the prefix "OP CM SP" matches LB7.1
99LB14:        OP CM* SP* .;
100LB15:        QU CM* SP* OP;
101LB16:        (CL | CP)CM* SP* NS;
102LB17:        B2 CM* SP* B2;
103
104# LB8, break after ZW SP*, precedes LB7 because they will both match the sequences like ZW SP,
105# and LB8 should take precedence.
106
107LB8:        ZW SP* ÷ [^ZW SP BK CR LF NL];
108
109# LB7 Do not break before spaces or zero width space.
110
111LB7.1:      [^ZW SP] CM* [SP ZW];
112LB7.2:      [ZW SP] [SP ZW];
113
114# LB8a
115#      ZWJ x
116#      Don't match a CM on the right - let other rules pick up CM sequences, where
117#      the ZWJ behaves as just another generic CM.
118LB8a:       ZWJ [^CM];
119
120
121# LB9:  X CM -> X
122# LB10: Unattached CM -> AL
123
124#LB11:       × WJ;
125#            WJ ×
126
127LB11.1:      [^SP] CM* WJ;
128LB11.2:      SP WJ;
129LB11.3:      WJ CM* [^CM];
130
131LB12:        GL CM* [^CM];
132
133LB12a:       [^SP BA HY] CM* GL;
134
135# LB 13 ICU Tailoring, matches tailoring exmaple 8 from UAX 14.
136#
137#   LB13.1   [^SP] CM* [CL CP EX IS SY]    # original UAX 14 rule.
138#   LB13.2   SP    CM* [CL CP EX IS SY]
139
140LB13.1: [^NU SP] CM* [CL CP IS SY];
141LB13.2: [^SP] CM* EX;
142LB13.2: SP [CL CP EX IS SY];
143
144
145# LB 14-17 are moved above LB 7.
146
147LB18:        SP ÷;
148
149LB19:        . CM* QU;
150LB19.1:      QU CM* [^CM];
151
152# LB 20   Break before and after CB.
153#         Interaction with LB8a:  ZWJ x . is tricky because CM includes ZWJ.
154#                                 ZWJ acts like a CM to the left, combining with CB.
155#                                 ZWJ acts independently to the right, no break after by LB8a.
156LB20.1:      . CM* ZWJ CB;
157LB20.2:      . CM* ÷ CB;
158
159LB20.3:      CB CM* ZWJ [^CM];
160LB20.4:      CB CM* ÷;
161
162# LB 20.09    Don't break between Hyphens and Letters when there is a break preceding the hyphen.
163LB20.09:     ^(HY | HH) CM* AL;
164
165# Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
166#       not picking up the continuing match after the BA from 21a.
167LB21a:       HL CM* (HY | BA) CM* [^CM CB];
168
169LB21.1:      . CM* [BA HY NS];
170LB21.2:      BB CM* [^CM CB];
171
172LB21b:       SY CM* HL;
173
174LB22.1:        (AL | HL | CM) CM* IN;   # The CM is from LB10, treat an unattached CM as AL.
175LB22.2:       EX CM* IN;
176LB22.3:       (ID | EB | EM) CM* IN;
177LB22.4:       IN CM* IN;
178LB22.5:       NU CM* IN;
179
180LB23.1:      (AL | HL | CM) CM* NU;
181LB23.2:      NU CM* (AL | HL);
182
183LB23a.1:     PR CM* (ID | EB | EM);
184LB23a.2:     (ID | EB | EM) CM* PO;
185
186LB24.2:      (PR | PO) CM* (AL | HL);
187LB24.3:      (AL | HL | CM) CM* (PR | PO);
188
189# Numbers. Equivalent to Tailoring example 8 from UAx 14.
190LB25:        ((PR | PO)CM*)? ((OP | HY)CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?;
191
192LB26.1:      JL CM* (JL | JV | H2 | H3);
193LB26.2:      (JV | H2) CM* (JV | JT);
194LB26.3:      (JT | H3) CM* JT;
195
196LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
197LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
198LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
199
200# LB28 Do not break between Alphabetics.
201#      Unattached (leading) CM treated as AL.
202LB28:        (AL | HL | CM)CM* (AL | HL);
203
204LB29:        IS CM* (AL | HL);
205
206# LB30  is adjusted for unattached leading CM being treated as AL.
207LB30.1:      (AL | CM | HL | NU) CM* OP;
208LB30.2:      CP CM* (AL | HL | NU);
209
210# LB30a  keep pairs of RI together.
211LB30a.1:     RI CM* RI         ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS CM];
212LB30a.2:     RI CM* RI CM* CMS ÷ [^BK CR LF NL SP ZW WJ CL CP EX IS SY GL QU BA HY NS CM];
213LB30a.3:     RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS ZWJ]?;
214
215# LB30b Do not break between Emoji Base and Emoji Modifier
216LB30b:       EB CM* EM;
217
218# LB31 Break Everywhere Else.
219#      Include combining marks
220LB31.1:        . CM* ZWJ [^CM];
221LB31.2:        . CM* ÷;
222