[apple/icu.git] / icuSources / test / testdata / break_rules / line.txt

#
# 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.txt
#
# Reference Line Break rules for intltest rbbi/RBBIMonkeyTest.
# Rules derived from Unicode Standard Annex #14 Revision 44 for Unicode 13.0,
# with the following modification:
#
#         Boundaries between hyphens and following letters are suppressed when
#         there is a boundary preceding the hyphen. See rule 20.9
#
#         This corresponds to CSS line-break=strict (BCP47 -u-lb-strict).
#         It sets characters of class CJ to behave like NS.
#
# 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.


type = line;
locale = en;


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:];
CMS = [:LineBreak =  Combining_Mark:];
CP = [:LineBreak =  Close_Parenthesis:];
CR = [:LineBreak =  Carriage_Return:];
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:];
IN = [:LineBreak =  Inseperable:];
IS = [:LineBreak =  Infix_Numeric:];
JL = [:LineBreak =  JL:];
JV = [:LineBreak =  JV:];
JT = [:LineBreak =  JT:];
LF = [:LineBreak =  Line_Feed:];
NL = [:LineBreak =  Next_Line:];
NS = [[:LineBreak =  Nonstarter:] CJ];   # CSS Strict tailoring: CJ resolves to NS.
NU = [:LineBreak =  Numeric:];
OP = [:LineBreak =  Open_Punctuation:];
PO = [:LineBreak =  Postfix_Numeric:];
PR = [:LineBreak =  Prefix_Numeric:];
QU = [:LineBreak =  Quotation:];
RI = [:LineBreak =  Regional_Indicator:];
SA = [:LineBreak =  Complex_Context:];
SG = [:LineBreak =  Surrogate:];
SP = [:LineBreak =  Space:];
SY = [:LineBreak =  Break_Symbols:];
WJ = [:LineBreak =  Word_Joiner:];
XX = [:LineBreak =  Unknown:];
ZW = [:LineBreak =  ZWSpace:];
ZWJ = [:LineBreak =  ZWJ:];

# OP30 and CP30 are variants of OP and CP that appear in rule LB30 from UAX 14.
# Limitations of this monkey test rule parser require that these definitions be pulled out
# rather than appearing in-line in LB 30.

OP30 = [OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
CP30 = [CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];

# LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
AL = [AL AI SG XX ];
dictionary = SA;

# 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;
LB5.1:      CR ÷;
LB5.2:      LF ÷;
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 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];

# 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
# LB10: Unattached CM -> AL

#LB11:       × WJ;
#            WJ ×

LB11.1:      [^SP] CM* WJ;
LB11.2:      SP WJ;
LB11.3:      WJ CM* [^CM];

LB12:        GL CM* [^CM];

LB12a:       [^SP BA HY] CM* GL;

# 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.

LB18:        SP ÷;

LB19:        . CM* QU;
LB19.1:      QU CM* [^CM];

# LB 20   Break before and after CB.
#         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.
LB21a:       HL CM* (HY | BA) CM* [^CM CB];

LB21.1:      . CM* [BA HY NS];
LB21.2:      BB CM* [^CM CB];

LB21b:       SY CM* HL;

LB22:        . CM* IN;

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;

LB24.2:      (PR | PO) CM* (AL | HL);
LB24.3:      (AL | HL | CM) CM* (PR | PO);


LB26.1:      JL CM* (JL | JV | H2 | H3);
LB26.2:      (JV | H2) CM* (JV | JT);
LB26.3:      (JT | H3) CM* JT;

LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
LB27.3:      PR CM* (JL | JV | JT | H2 | H3);

# LB28 Do not break between Alphabetics.
#      Unattached (leading) CM treated as AL.
LB28:        (AL | HL | CM)CM* (AL | HL);

LB29:        IS CM* (AL | HL);

# LB30  is adjusted for unattached leading CM being treated as AL.
LB30.1:      (AL | CM | HL | NU) CM* OP30;
LB30.2:      CP30 CM* (AL | HL | NU);

# 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 IN 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 IN CM];
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]?;

# LB30b Do not break between Emoji Base and Emoji Modifier
LB30b:       EB CM* EM;

# LB31 Break Everywhere Else.
#      Include combining marks
LB31.1:        . CM* ZWJ [^CM];
LB31.2:        . CM* ÷;
Commit	Line	Data
2ca993e8	1	#
f3c0d7a5 A	2	# Copyright (C) 2016 and later: Unicode, Inc. and others.
f3c0d7a5 A	3	# License & terms of use: http://www.unicode.org/copyright.html
2ca993e8 A	4	# Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved.
	5
	6	# file: line.txt
	7	#
0f5d89e8	8	# Reference Line Break rules for intltest rbbi/RBBIMonkeyTest.
340931cb A	9	# Rules derived from Unicode Standard Annex #14 Revision 44 for Unicode 13.0,
	10	# with the following modification:
	11	#
	12	# Boundaries between hyphens and following letters are suppressed when
	13	# there is a boundary preceding the hyphen. See rule 20.9
	14	#
	15	# This corresponds to CSS line-break=strict (BCP47 -u-lb-strict).
	16	# It sets characters of class CJ to behave like NS.
2ca993e8 A	17	#
	18	# Note: Rule syntax and the monkey test itself are still a work in progress.
	19	# They are expected to change with review and the addition of support for rule tailoring.
	20
	21
	22	type = line;
	23	locale = en;
	24
	25
f3c0d7a5 A	26	AI = [:LineBreak = Ambiguous:];
f3c0d7a5 A	27	AL = [:LineBreak = Alphabetic:];
2ca993e8	28	BA = [:LineBreak = Break_After:];
3d1f044b	29	HH = [\u2010]; # \u2010 is HYPHEN, default line break is BA.
2ca993e8 A	30	BB = [:LineBreak = Break_Before:];
	31	BK = [:LineBreak = Mandatory_Break:];
	32	B2 = [:LineBreak = Break_Both:];
	33	CB = [:LineBreak = Contingent_Break:];
	34	CJ = [:LineBreak = Conditional_Japanese_Starter:];
	35	CL = [:LineBreak = Close_Punctuation:];
3d1f044b	36	CMS = [:LineBreak = Combining_Mark:];
2ca993e8 A	37	CP = [:LineBreak = Close_Parenthesis:];
2ca993e8 A	38	CR = [:LineBreak = Carriage_Return:];
f3c0d7a5 A	39	EB = [[:LineBreak = EB:] \U0001F46A-\U0001F46D\U0001F46F\U0001F91D\U0001F93C];
f3c0d7a5 A	40	EM = [:LineBreak = EM:];
2ca993e8 A	41	EX = [:LineBreak = Exclamation:];
	42	GL = [:LineBreak = Glue:];
	43	HL = [:LineBreak = Hebrew_Letter:];
	44	HY = [:LineBreak = Hyphen:];
	45	H2 = [:LineBreak = H2:];
	46	H3 = [:LineBreak = H3:];
f3c0d7a5	47	ID = [:LineBreak = Ideographic:];
2ca993e8 A	48	IN = [:LineBreak = Inseperable:];
	49	IS = [:LineBreak = Infix_Numeric:];
	50	JL = [:LineBreak = JL:];
	51	JV = [:LineBreak = JV:];
	52	JT = [:LineBreak = JT:];
	53	LF = [:LineBreak = Line_Feed:];
	54	NL = [:LineBreak = Next_Line:];
f3c0d7a5	55	NS = [[:LineBreak = Nonstarter:] CJ]; # CSS Strict tailoring: CJ resolves to NS.
2ca993e8 A	56	NU = [:LineBreak = Numeric:];
	57	OP = [:LineBreak = Open_Punctuation:];
	58	PO = [:LineBreak = Postfix_Numeric:];
	59	PR = [:LineBreak = Prefix_Numeric:];
	60	QU = [:LineBreak = Quotation:];
	61	RI = [:LineBreak = Regional_Indicator:];
	62	SA = [:LineBreak = Complex_Context:];
	63	SG = [:LineBreak = Surrogate:];
	64	SP = [:LineBreak = Space:];
	65	SY = [:LineBreak = Break_Symbols:];
	66	WJ = [:LineBreak = Word_Joiner:];
	67	XX = [:LineBreak = Unknown:];
	68	ZW = [:LineBreak = ZWSpace:];
f3c0d7a5	69	ZWJ = [:LineBreak = ZWJ:];
2ca993e8	70
340931cb A	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.
	74
	75	OP30 = [OP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
	76	CP30 = [CP - [\p{ea=F}\p{ea=W}\p{ea=H}]];
	77
f3c0d7a5 A	78	# LB1 - Resolve AI, CB, CJ, SA, SG, and XX into other line breaking classes
	79	AL = [AL AI SG XX ];
	80	dictionary = SA;
2ca993e8	81
f3c0d7a5 A	82	# By LB9, a ZWJ also behaves as a CM. Including it in the definition of CM avoids having to explicitly
f3c0d7a5 A	83	# list it in the numerous rules that use CM.
3d1f044b	84	CM = [CMS ZWJ];
2ca993e8 A	85
	86	LB4: BK ÷;
	87	LB5: CR LF;
	88	LB5.1: CR ÷;
	89	LB5.2: LF ÷;
	90	LB5.3: NL ÷;
	91
	92	LB6: . (BK \| CR \| LF \| NL);
	93	LB6.1: [^BK CR LF NL SP ZW] CM* (BK \| CR \| LF \| NL);
	94
3d1f044b A	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.
	98
	99	LB8: ZW SP* ÷ [^ZW SP BK CR LF NL];
	100
	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))?;
	104
2ca993e8	105	# Rules LB14 - LB17.
3d1f044b A	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
2ca993e8	109	LB14: OP CM* SP* .;
3d1f044b A	110
	111	# LB 14a Break before an IS that begins a number and follows a space.
	112	LB14a: SP ÷ IS CM* NU;
	113
	114	# LB14b × IS
	115	LB14b.1: [^SP] CM* IS;
	116	LB14b.2: SP IS;
	117
2ca993e8 A	118	LB15: QU CM* SP* OP;
	119	LB16: (CL \| CP)CM* SP* NS;
	120	LB17: B2 CM* SP* B2;
	121
3d1f044b A	122
	123	# LB7 Do not break before spaces or zero width space.
	124
2ca993e8 A	125	LB7.1: [^ZW SP] CM* [SP ZW];
	126	LB7.2: [ZW SP] [SP ZW];
	127
f3c0d7a5	128	# LB8a
0f5d89e8 A	129	# ZWJ x
	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.
	132	LB8a: ZWJ [^CM];
2ca993e8 A	133
	134
	135	# LB9: X CM -> X
	136	# LB10: Unattached CM -> AL
	137
	138	#LB11: × WJ;
	139	# WJ ×
	140
0f5d89e8	141	LB11.1: [^SP] CM* WJ;
2ca993e8 A	142	LB11.2: SP WJ;
	143	LB11.3: WJ CM* [^CM];
	144
	145	LB12: GL CM* [^CM];
	146
	147	LB12a: [^SP BA HY] CM* GL;
	148
3d1f044b A	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];
2ca993e8 A	152
	153	# LB 14-17 are moved above LB 7.
	154
	155	LB18: SP ÷;
	156
	157	LB19: . CM* QU;
	158	LB19.1: QU CM* [^CM];
	159
	160	# LB 20 Break before and after CB.
0f5d89e8	161	# Interaction with LB8a: ZWJ x . is tricky because CM includes ZWJ.
f3c0d7a5	162	# ZWJ acts like a CM to the left, combining with CB.
0f5d89e8 A	163	# ZWJ acts independently to the right, no break after by LB8a.
	164	LB20.1: . CM* ZWJ CB;
	165	LB20.2: . CM* ÷ CB;
	166
	167	LB20.3: CB CM* ZWJ [^CM];
	168	LB20.4: CB CM* ÷;
2ca993e8	169
3d1f044b A	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;
	172
2ca993e8 A	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];
	176
	177	LB21.1: . CM* [BA HY NS];
	178	LB21.2: BB CM* [^CM CB];
	179
	180	LB21b: SY CM* HL;
	181
340931cb	182	LB22: . CM* IN;
2ca993e8	183
f3c0d7a5 A	184	LB23.1: (AL \| HL \| CM) CM* NU;
	185	LB23.2: NU CM* (AL \| HL);
	186
	187	LB23a.1: PR CM* (ID \| EB \| EM);
	188	LB23a.2: (ID \| EB \| EM) CM* PO;
2ca993e8	189
f3c0d7a5 A	190	LB24.2: (PR \| PO) CM* (AL \| HL);
f3c0d7a5 A	191	LB24.3: (AL \| HL \| CM) CM* (PR \| PO);
2ca993e8	192
2ca993e8 A	193
	194	LB26.1: JL CM* (JL \| JV \| H2 \| H3);
	195	LB26.2: (JV \| H2) CM* (JV \| JT);
	196	LB26.3: (JT \| H3) CM* JT;
	197
	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);
	201
	202	# LB28 Do not break between Alphabetics.
	203	# Unattached (leading) CM treated as AL.
	204	LB28: (AL \| HL \| CM)CM* (AL \| HL);
	205
	206	LB29: IS CM* (AL \| HL);
	207
	208	# LB30 is adjusted for unattached leading CM being treated as AL.
340931cb A	209	LB30.1: (AL \| CM \| HL \| NU) CM* OP30;
340931cb A	210	LB30.2: CP30 CM* (AL \| HL \| NU);
2ca993e8	211
0f5d89e8	212	# LB30a keep pairs of RI together.
340931cb A	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]?;
2ca993e8 A	216
	217	# LB30b Do not break between Emoji Base and Emoji Modifier
	218	LB30b: EB CM* EM;
	219
	220	# LB31 Break Everywhere Else.
	221	# Include combining marks
0f5d89e8	222	LB31.1: . CM* ZWJ [^CM];
2ca993e8	223	LB31.2: . CM* ÷;