This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
charnames.pm: return ord not chr
[perl5.git] / lib / charnames.pm
CommitLineData
423cee85 1package charnames;
b177ca84
JF
2use strict;
3use warnings;
51cf30b6 4use File::Spec;
63098191 5our $VERSION = '1.11';
b75c8c73 6
d5448623 7use bytes (); # for $bytes::hint_bits
423cee85 8
232cbbee 9my %system_aliases = (
16036bcd
KW
10 # Icky 3.2 names with parentheses.
11 'LINE FEED' => 0x0A, # LINE FEED (LF)
12 'FORM FEED' => 0x0C, # FORM FEED (FF)
13 'CARRIAGE RETURN' => 0x0D, # CARRIAGE RETURN (CR)
14 'NEXT LINE' => 0x85, # NEXT LINE (NEL)
15
16 # Some variant names from Wikipedia
17 'SINGLE-SHIFT 2' => 0x8E,
18 'SINGLE-SHIFT 3' => 0x8F,
19 'PRIVATE USE 1' => 0x91,
20 'PRIVATE USE 2' => 0x92,
21 'START OF PROTECTED AREA' => 0x96,
22 'END OF PROTECTED AREA' => 0x97,
23
24 # Convenience. Standard abbreviations for the controls
25 'NUL' => 0x00, # NULL
26 'SOH' => 0x01, # START OF HEADING
27 'STX' => 0x02, # START OF TEXT
28 'ETX' => 0x03, # END OF TEXT
29 'EOT' => 0x04, # END OF TRANSMISSION
30 'ENQ' => 0x05, # ENQUIRY
31 'ACK' => 0x06, # ACKNOWLEDGE
32 'BEL' => 0x07, # BELL
33 'BS' => 0x08, # BACKSPACE
34 'HT' => 0x09, # HORIZONTAL TABULATION
35 'LF' => 0x0A, # LINE FEED (LF)
36 'VT' => 0x0B, # VERTICAL TABULATION
37 'FF' => 0x0C, # FORM FEED (FF)
38 'CR' => 0x0D, # CARRIAGE RETURN (CR)
39 'SO' => 0x0E, # SHIFT OUT
40 'SI' => 0x0F, # SHIFT IN
41 'DLE' => 0x10, # DATA LINK ESCAPE
42 'DC1' => 0x11, # DEVICE CONTROL ONE
43 'DC2' => 0x12, # DEVICE CONTROL TWO
44 'DC3' => 0x13, # DEVICE CONTROL THREE
45 'DC4' => 0x14, # DEVICE CONTROL FOUR
46 'NAK' => 0x15, # NEGATIVE ACKNOWLEDGE
47 'SYN' => 0x16, # SYNCHRONOUS IDLE
48 'ETB' => 0x17, # END OF TRANSMISSION BLOCK
49 'CAN' => 0x18, # CANCEL
50 'EOM' => 0x19, # END OF MEDIUM
51 'SUB' => 0x1A, # SUBSTITUTE
52 'ESC' => 0x1B, # ESCAPE
53 'FS' => 0x1C, # FILE SEPARATOR
54 'GS' => 0x1D, # GROUP SEPARATOR
55 'RS' => 0x1E, # RECORD SEPARATOR
56 'US' => 0x1F, # UNIT SEPARATOR
57 'DEL' => 0x7F, # DELETE
58 'BPH' => 0x82, # BREAK PERMITTED HERE
59 'NBH' => 0x83, # NO BREAK HERE
60 'NEL' => 0x85, # NEXT LINE (NEL)
61 'SSA' => 0x86, # START OF SELECTED AREA
62 'ESA' => 0x87, # END OF SELECTED AREA
63 'HTS' => 0x88, # CHARACTER TABULATION SET
64 'HTJ' => 0x89, # CHARACTER TABULATION WITH JUSTIFICATION
65 'VTS' => 0x8A, # LINE TABULATION SET
66 'PLD' => 0x8B, # PARTIAL LINE FORWARD
67 'PLU' => 0x8C, # PARTIAL LINE BACKWARD
68 'RI ' => 0x8D, # REVERSE LINE FEED
69 'SS2' => 0x8E, # SINGLE SHIFT TWO
70 'SS3' => 0x8F, # SINGLE SHIFT THREE
71 'DCS' => 0x90, # DEVICE CONTROL STRING
72 'PU1' => 0x91, # PRIVATE USE ONE
73 'PU2' => 0x92, # PRIVATE USE TWO
74 'STS' => 0x93, # SET TRANSMIT STATE
75 'CCH' => 0x94, # CANCEL CHARACTER
76 'MW ' => 0x95, # MESSAGE WAITING
77 'SPA' => 0x96, # START OF GUARDED AREA
78 'EPA' => 0x97, # END OF GUARDED AREA
79 'SOS' => 0x98, # START OF STRING
80 'SCI' => 0x9A, # SINGLE CHARACTER INTRODUCER
81 'CSI' => 0x9B, # CONTROL SEQUENCE INTRODUCER
82 'ST ' => 0x9C, # STRING TERMINATOR
83 'OSC' => 0x9D, # OPERATING SYSTEM COMMAND
84 'PM ' => 0x9E, # PRIVACY MESSAGE
85 'APC' => 0x9F, # APPLICATION PROGRAM COMMAND
86
87 # There are no names for these in the Unicode standard;
88 # perhaps should be deprecated, but then again there are
89 # no alternative names, so am not deprecating. And if
90 # did, the code would have to change to not recommend an
91 # alternative for these.
92 'PADDING CHARACTER' => 0x80,
93 'PAD' => 0x80,
94 'HIGH OCTET PRESET' => 0x81,
95 'HOP' => 0x81,
96 'INDEX' => 0x84,
97 'IND' => 0x84,
98 'SINGLE GRAPHIC CHARACTER INTRODUCER' => 0x99,
99 'SGC' => 0x99,
100
101 # More convenience. For further convenience,
102 # it is suggested some way of using the NamesList
103 # aliases be implemented, but there are ambiguities in
232cbbee 104 # NamesList.txt
16036bcd
KW
105 'BOM' => 0xFEFF, # BYTE ORDER MARK
106 'BYTE ORDER MARK'=> 0xFEFF,
107 'CGJ' => 0x034F, # COMBINING GRAPHEME JOINER
108 'FVS1' => 0x180B, # MONGOLIAN FREE VARIATION SELECTOR ONE
109 'FVS2' => 0x180C, # MONGOLIAN FREE VARIATION SELECTOR TWO
110 'FVS3' => 0x180D, # MONGOLIAN FREE VARIATION SELECTOR THREE
111 'LRE' => 0x202A, # LEFT-TO-RIGHT EMBEDDING
112 'LRM' => 0x200E, # LEFT-TO-RIGHT MARK
113 'LRO' => 0x202D, # LEFT-TO-RIGHT OVERRIDE
114 'MMSP' => 0x205F, # MEDIUM MATHEMATICAL SPACE
115 'MVS' => 0x180E, # MONGOLIAN VOWEL SEPARATOR
116 'NBSP' => 0x00A0, # NO-BREAK SPACE
117 'NNBSP' => 0x202F, # NARROW NO-BREAK SPACE
118 'PDF' => 0x202C, # POP DIRECTIONAL FORMATTING
119 'RLE' => 0x202B, # RIGHT-TO-LEFT EMBEDDING
120 'RLM' => 0x200F, # RIGHT-TO-LEFT MARK
121 'RLO' => 0x202E, # RIGHT-TO-LEFT OVERRIDE
122 'SHY' => 0x00AD, # SOFT HYPHEN
123 'VS1' => 0xFE00, # VARIATION SELECTOR-1
124 'VS2' => 0xFE01, # VARIATION SELECTOR-2
125 'VS3' => 0xFE02, # VARIATION SELECTOR-3
126 'VS4' => 0xFE03, # VARIATION SELECTOR-4
127 'VS5' => 0xFE04, # VARIATION SELECTOR-5
128 'VS6' => 0xFE05, # VARIATION SELECTOR-6
129 'VS7' => 0xFE06, # VARIATION SELECTOR-7
130 'VS8' => 0xFE07, # VARIATION SELECTOR-8
131 'VS9' => 0xFE08, # VARIATION SELECTOR-9
132 'VS10' => 0xFE09, # VARIATION SELECTOR-10
133 'VS11' => 0xFE0A, # VARIATION SELECTOR-11
134 'VS12' => 0xFE0B, # VARIATION SELECTOR-12
135 'VS13' => 0xFE0C, # VARIATION SELECTOR-13
136 'VS14' => 0xFE0D, # VARIATION SELECTOR-14
137 'VS15' => 0xFE0E, # VARIATION SELECTOR-15
138 'VS16' => 0xFE0F, # VARIATION SELECTOR-16
139 'VS17' => 0xE0100, # VARIATION SELECTOR-17
140 'VS18' => 0xE0101, # VARIATION SELECTOR-18
141 'VS19' => 0xE0102, # VARIATION SELECTOR-19
142 'VS20' => 0xE0103, # VARIATION SELECTOR-20
143 'VS21' => 0xE0104, # VARIATION SELECTOR-21
144 'VS22' => 0xE0105, # VARIATION SELECTOR-22
145 'VS23' => 0xE0106, # VARIATION SELECTOR-23
146 'VS24' => 0xE0107, # VARIATION SELECTOR-24
147 'VS25' => 0xE0108, # VARIATION SELECTOR-25
148 'VS26' => 0xE0109, # VARIATION SELECTOR-26
149 'VS27' => 0xE010A, # VARIATION SELECTOR-27
150 'VS28' => 0xE010B, # VARIATION SELECTOR-28
151 'VS29' => 0xE010C, # VARIATION SELECTOR-29
152 'VS30' => 0xE010D, # VARIATION SELECTOR-30
153 'VS31' => 0xE010E, # VARIATION SELECTOR-31
154 'VS32' => 0xE010F, # VARIATION SELECTOR-32
155 'VS33' => 0xE0110, # VARIATION SELECTOR-33
156 'VS34' => 0xE0111, # VARIATION SELECTOR-34
157 'VS35' => 0xE0112, # VARIATION SELECTOR-35
158 'VS36' => 0xE0113, # VARIATION SELECTOR-36
159 'VS37' => 0xE0114, # VARIATION SELECTOR-37
160 'VS38' => 0xE0115, # VARIATION SELECTOR-38
161 'VS39' => 0xE0116, # VARIATION SELECTOR-39
162 'VS40' => 0xE0117, # VARIATION SELECTOR-40
163 'VS41' => 0xE0118, # VARIATION SELECTOR-41
164 'VS42' => 0xE0119, # VARIATION SELECTOR-42
165 'VS43' => 0xE011A, # VARIATION SELECTOR-43
166 'VS44' => 0xE011B, # VARIATION SELECTOR-44
167 'VS45' => 0xE011C, # VARIATION SELECTOR-45
168 'VS46' => 0xE011D, # VARIATION SELECTOR-46
169 'VS47' => 0xE011E, # VARIATION SELECTOR-47
170 'VS48' => 0xE011F, # VARIATION SELECTOR-48
171 'VS49' => 0xE0120, # VARIATION SELECTOR-49
172 'VS50' => 0xE0121, # VARIATION SELECTOR-50
173 'VS51' => 0xE0122, # VARIATION SELECTOR-51
174 'VS52' => 0xE0123, # VARIATION SELECTOR-52
175 'VS53' => 0xE0124, # VARIATION SELECTOR-53
176 'VS54' => 0xE0125, # VARIATION SELECTOR-54
177 'VS55' => 0xE0126, # VARIATION SELECTOR-55
178 'VS56' => 0xE0127, # VARIATION SELECTOR-56
179 'VS57' => 0xE0128, # VARIATION SELECTOR-57
180 'VS58' => 0xE0129, # VARIATION SELECTOR-58
181 'VS59' => 0xE012A, # VARIATION SELECTOR-59
182 'VS60' => 0xE012B, # VARIATION SELECTOR-60
183 'VS61' => 0xE012C, # VARIATION SELECTOR-61
184 'VS62' => 0xE012D, # VARIATION SELECTOR-62
185 'VS63' => 0xE012E, # VARIATION SELECTOR-63
186 'VS64' => 0xE012F, # VARIATION SELECTOR-64
187 'VS65' => 0xE0130, # VARIATION SELECTOR-65
188 'VS66' => 0xE0131, # VARIATION SELECTOR-66
189 'VS67' => 0xE0132, # VARIATION SELECTOR-67
190 'VS68' => 0xE0133, # VARIATION SELECTOR-68
191 'VS69' => 0xE0134, # VARIATION SELECTOR-69
192 'VS70' => 0xE0135, # VARIATION SELECTOR-70
193 'VS71' => 0xE0136, # VARIATION SELECTOR-71
194 'VS72' => 0xE0137, # VARIATION SELECTOR-72
195 'VS73' => 0xE0138, # VARIATION SELECTOR-73
196 'VS74' => 0xE0139, # VARIATION SELECTOR-74
197 'VS75' => 0xE013A, # VARIATION SELECTOR-75
198 'VS76' => 0xE013B, # VARIATION SELECTOR-76
199 'VS77' => 0xE013C, # VARIATION SELECTOR-77
200 'VS78' => 0xE013D, # VARIATION SELECTOR-78
201 'VS79' => 0xE013E, # VARIATION SELECTOR-79
202 'VS80' => 0xE013F, # VARIATION SELECTOR-80
203 'VS81' => 0xE0140, # VARIATION SELECTOR-81
204 'VS82' => 0xE0141, # VARIATION SELECTOR-82
205 'VS83' => 0xE0142, # VARIATION SELECTOR-83
206 'VS84' => 0xE0143, # VARIATION SELECTOR-84
207 'VS85' => 0xE0144, # VARIATION SELECTOR-85
208 'VS86' => 0xE0145, # VARIATION SELECTOR-86
209 'VS87' => 0xE0146, # VARIATION SELECTOR-87
210 'VS88' => 0xE0147, # VARIATION SELECTOR-88
211 'VS89' => 0xE0148, # VARIATION SELECTOR-89
212 'VS90' => 0xE0149, # VARIATION SELECTOR-90
213 'VS91' => 0xE014A, # VARIATION SELECTOR-91
214 'VS92' => 0xE014B, # VARIATION SELECTOR-92
215 'VS93' => 0xE014C, # VARIATION SELECTOR-93
216 'VS94' => 0xE014D, # VARIATION SELECTOR-94
217 'VS95' => 0xE014E, # VARIATION SELECTOR-95
218 'VS96' => 0xE014F, # VARIATION SELECTOR-96
219 'VS97' => 0xE0150, # VARIATION SELECTOR-97
220 'VS98' => 0xE0151, # VARIATION SELECTOR-98
221 'VS99' => 0xE0152, # VARIATION SELECTOR-99
222 'VS100' => 0xE0153, # VARIATION SELECTOR-100
223 'VS101' => 0xE0154, # VARIATION SELECTOR-101
224 'VS102' => 0xE0155, # VARIATION SELECTOR-102
225 'VS103' => 0xE0156, # VARIATION SELECTOR-103
226 'VS104' => 0xE0157, # VARIATION SELECTOR-104
227 'VS105' => 0xE0158, # VARIATION SELECTOR-105
228 'VS106' => 0xE0159, # VARIATION SELECTOR-106
229 'VS107' => 0xE015A, # VARIATION SELECTOR-107
230 'VS108' => 0xE015B, # VARIATION SELECTOR-108
231 'VS109' => 0xE015C, # VARIATION SELECTOR-109
232 'VS110' => 0xE015D, # VARIATION SELECTOR-110
233 'VS111' => 0xE015E, # VARIATION SELECTOR-111
234 'VS112' => 0xE015F, # VARIATION SELECTOR-112
235 'VS113' => 0xE0160, # VARIATION SELECTOR-113
236 'VS114' => 0xE0161, # VARIATION SELECTOR-114
237 'VS115' => 0xE0162, # VARIATION SELECTOR-115
238 'VS116' => 0xE0163, # VARIATION SELECTOR-116
239 'VS117' => 0xE0164, # VARIATION SELECTOR-117
240 'VS118' => 0xE0165, # VARIATION SELECTOR-118
241 'VS119' => 0xE0166, # VARIATION SELECTOR-119
242 'VS120' => 0xE0167, # VARIATION SELECTOR-120
243 'VS121' => 0xE0168, # VARIATION SELECTOR-121
244 'VS122' => 0xE0169, # VARIATION SELECTOR-122
245 'VS123' => 0xE016A, # VARIATION SELECTOR-123
246 'VS124' => 0xE016B, # VARIATION SELECTOR-124
247 'VS125' => 0xE016C, # VARIATION SELECTOR-125
248 'VS126' => 0xE016D, # VARIATION SELECTOR-126
249 'VS127' => 0xE016E, # VARIATION SELECTOR-127
250 'VS128' => 0xE016F, # VARIATION SELECTOR-128
251 'VS129' => 0xE0170, # VARIATION SELECTOR-129
252 'VS130' => 0xE0171, # VARIATION SELECTOR-130
253 'VS131' => 0xE0172, # VARIATION SELECTOR-131
254 'VS132' => 0xE0173, # VARIATION SELECTOR-132
255 'VS133' => 0xE0174, # VARIATION SELECTOR-133
256 'VS134' => 0xE0175, # VARIATION SELECTOR-134
257 'VS135' => 0xE0176, # VARIATION SELECTOR-135
258 'VS136' => 0xE0177, # VARIATION SELECTOR-136
259 'VS137' => 0xE0178, # VARIATION SELECTOR-137
260 'VS138' => 0xE0179, # VARIATION SELECTOR-138
261 'VS139' => 0xE017A, # VARIATION SELECTOR-139
262 'VS140' => 0xE017B, # VARIATION SELECTOR-140
263 'VS141' => 0xE017C, # VARIATION SELECTOR-141
264 'VS142' => 0xE017D, # VARIATION SELECTOR-142
265 'VS143' => 0xE017E, # VARIATION SELECTOR-143
266 'VS144' => 0xE017F, # VARIATION SELECTOR-144
267 'VS145' => 0xE0180, # VARIATION SELECTOR-145
268 'VS146' => 0xE0181, # VARIATION SELECTOR-146
269 'VS147' => 0xE0182, # VARIATION SELECTOR-147
270 'VS148' => 0xE0183, # VARIATION SELECTOR-148
271 'VS149' => 0xE0184, # VARIATION SELECTOR-149
272 'VS150' => 0xE0185, # VARIATION SELECTOR-150
273 'VS151' => 0xE0186, # VARIATION SELECTOR-151
274 'VS152' => 0xE0187, # VARIATION SELECTOR-152
275 'VS153' => 0xE0188, # VARIATION SELECTOR-153
276 'VS154' => 0xE0189, # VARIATION SELECTOR-154
277 'VS155' => 0xE018A, # VARIATION SELECTOR-155
278 'VS156' => 0xE018B, # VARIATION SELECTOR-156
279 'VS157' => 0xE018C, # VARIATION SELECTOR-157
280 'VS158' => 0xE018D, # VARIATION SELECTOR-158
281 'VS159' => 0xE018E, # VARIATION SELECTOR-159
282 'VS160' => 0xE018F, # VARIATION SELECTOR-160
283 'VS161' => 0xE0190, # VARIATION SELECTOR-161
284 'VS162' => 0xE0191, # VARIATION SELECTOR-162
285 'VS163' => 0xE0192, # VARIATION SELECTOR-163
286 'VS164' => 0xE0193, # VARIATION SELECTOR-164
287 'VS165' => 0xE0194, # VARIATION SELECTOR-165
288 'VS166' => 0xE0195, # VARIATION SELECTOR-166
289 'VS167' => 0xE0196, # VARIATION SELECTOR-167
290 'VS168' => 0xE0197, # VARIATION SELECTOR-168
291 'VS169' => 0xE0198, # VARIATION SELECTOR-169
292 'VS170' => 0xE0199, # VARIATION SELECTOR-170
293 'VS171' => 0xE019A, # VARIATION SELECTOR-171
294 'VS172' => 0xE019B, # VARIATION SELECTOR-172
295 'VS173' => 0xE019C, # VARIATION SELECTOR-173
296 'VS174' => 0xE019D, # VARIATION SELECTOR-174
297 'VS175' => 0xE019E, # VARIATION SELECTOR-175
298 'VS176' => 0xE019F, # VARIATION SELECTOR-176
299 'VS177' => 0xE01A0, # VARIATION SELECTOR-177
300 'VS178' => 0xE01A1, # VARIATION SELECTOR-178
301 'VS179' => 0xE01A2, # VARIATION SELECTOR-179
302 'VS180' => 0xE01A3, # VARIATION SELECTOR-180
303 'VS181' => 0xE01A4, # VARIATION SELECTOR-181
304 'VS182' => 0xE01A5, # VARIATION SELECTOR-182
305 'VS183' => 0xE01A6, # VARIATION SELECTOR-183
306 'VS184' => 0xE01A7, # VARIATION SELECTOR-184
307 'VS185' => 0xE01A8, # VARIATION SELECTOR-185
308 'VS186' => 0xE01A9, # VARIATION SELECTOR-186
309 'VS187' => 0xE01AA, # VARIATION SELECTOR-187
310 'VS188' => 0xE01AB, # VARIATION SELECTOR-188
311 'VS189' => 0xE01AC, # VARIATION SELECTOR-189
312 'VS190' => 0xE01AD, # VARIATION SELECTOR-190
313 'VS191' => 0xE01AE, # VARIATION SELECTOR-191
314 'VS192' => 0xE01AF, # VARIATION SELECTOR-192
315 'VS193' => 0xE01B0, # VARIATION SELECTOR-193
316 'VS194' => 0xE01B1, # VARIATION SELECTOR-194
317 'VS195' => 0xE01B2, # VARIATION SELECTOR-195
318 'VS196' => 0xE01B3, # VARIATION SELECTOR-196
319 'VS197' => 0xE01B4, # VARIATION SELECTOR-197
320 'VS198' => 0xE01B5, # VARIATION SELECTOR-198
321 'VS199' => 0xE01B6, # VARIATION SELECTOR-199
322 'VS200' => 0xE01B7, # VARIATION SELECTOR-200
323 'VS201' => 0xE01B8, # VARIATION SELECTOR-201
324 'VS202' => 0xE01B9, # VARIATION SELECTOR-202
325 'VS203' => 0xE01BA, # VARIATION SELECTOR-203
326 'VS204' => 0xE01BB, # VARIATION SELECTOR-204
327 'VS205' => 0xE01BC, # VARIATION SELECTOR-205
328 'VS206' => 0xE01BD, # VARIATION SELECTOR-206
329 'VS207' => 0xE01BE, # VARIATION SELECTOR-207
330 'VS208' => 0xE01BF, # VARIATION SELECTOR-208
331 'VS209' => 0xE01C0, # VARIATION SELECTOR-209
332 'VS210' => 0xE01C1, # VARIATION SELECTOR-210
333 'VS211' => 0xE01C2, # VARIATION SELECTOR-211
334 'VS212' => 0xE01C3, # VARIATION SELECTOR-212
335 'VS213' => 0xE01C4, # VARIATION SELECTOR-213
336 'VS214' => 0xE01C5, # VARIATION SELECTOR-214
337 'VS215' => 0xE01C6, # VARIATION SELECTOR-215
338 'VS216' => 0xE01C7, # VARIATION SELECTOR-216
339 'VS217' => 0xE01C8, # VARIATION SELECTOR-217
340 'VS218' => 0xE01C9, # VARIATION SELECTOR-218
341 'VS219' => 0xE01CA, # VARIATION SELECTOR-219
342 'VS220' => 0xE01CB, # VARIATION SELECTOR-220
343 'VS221' => 0xE01CC, # VARIATION SELECTOR-221
344 'VS222' => 0xE01CD, # VARIATION SELECTOR-222
345 'VS223' => 0xE01CE, # VARIATION SELECTOR-223
346 'VS224' => 0xE01CF, # VARIATION SELECTOR-224
347 'VS225' => 0xE01D0, # VARIATION SELECTOR-225
348 'VS226' => 0xE01D1, # VARIATION SELECTOR-226
349 'VS227' => 0xE01D2, # VARIATION SELECTOR-227
350 'VS228' => 0xE01D3, # VARIATION SELECTOR-228
351 'VS229' => 0xE01D4, # VARIATION SELECTOR-229
352 'VS230' => 0xE01D5, # VARIATION SELECTOR-230
353 'VS231' => 0xE01D6, # VARIATION SELECTOR-231
354 'VS232' => 0xE01D7, # VARIATION SELECTOR-232
355 'VS233' => 0xE01D8, # VARIATION SELECTOR-233
356 'VS234' => 0xE01D9, # VARIATION SELECTOR-234
357 'VS235' => 0xE01DA, # VARIATION SELECTOR-235
358 'VS236' => 0xE01DB, # VARIATION SELECTOR-236
359 'VS237' => 0xE01DC, # VARIATION SELECTOR-237
360 'VS238' => 0xE01DD, # VARIATION SELECTOR-238
361 'VS239' => 0xE01DE, # VARIATION SELECTOR-239
362 'VS240' => 0xE01DF, # VARIATION SELECTOR-240
363 'VS241' => 0xE01E0, # VARIATION SELECTOR-241
364 'VS242' => 0xE01E1, # VARIATION SELECTOR-242
365 'VS243' => 0xE01E2, # VARIATION SELECTOR-243
366 'VS244' => 0xE01E3, # VARIATION SELECTOR-244
367 'VS245' => 0xE01E4, # VARIATION SELECTOR-245
368 'VS246' => 0xE01E5, # VARIATION SELECTOR-246
369 'VS247' => 0xE01E6, # VARIATION SELECTOR-247
370 'VS248' => 0xE01E7, # VARIATION SELECTOR-248
371 'VS249' => 0xE01E8, # VARIATION SELECTOR-249
372 'VS250' => 0xE01E9, # VARIATION SELECTOR-250
373 'VS251' => 0xE01EA, # VARIATION SELECTOR-251
374 'VS252' => 0xE01EB, # VARIATION SELECTOR-252
375 'VS253' => 0xE01EC, # VARIATION SELECTOR-253
376 'VS254' => 0xE01ED, # VARIATION SELECTOR-254
377 'VS255' => 0xE01EE, # VARIATION SELECTOR-255
378 'VS256' => 0xE01EF, # VARIATION SELECTOR-256
379 'WJ' => 0x2060, # WORD JOINER
380 'ZWJ' => 0x200D, # ZERO WIDTH JOINER
381 'ZWNJ' => 0x200C, # ZERO WIDTH NON-JOINER
382 'ZWSP' => 0x200B, # ZERO WIDTH SPACE
383 );
52ea3e69 384
232cbbee 385my %deprecated_aliases = (
16036bcd
KW
386 # Pre-3.2 compatibility (only for the first 256 characters).
387 # Use of these gives deprecated message.
388 'HORIZONTAL TABULATION' => 0x09, # CHARACTER TABULATION
389 'VERTICAL TABULATION' => 0x0B, # LINE TABULATION
390 'FILE SEPARATOR' => 0x1C, # INFORMATION SEPARATOR FOUR
391 'GROUP SEPARATOR' => 0x1D, # INFORMATION SEPARATOR THREE
392 'RECORD SEPARATOR' => 0x1E, # INFORMATION SEPARATOR TWO
393 'UNIT SEPARATOR' => 0x1F, # INFORMATION SEPARATOR ONE
394 'HORIZONTAL TABULATION SET' => 0x88, # CHARACTER TABULATION SET
395 'HORIZONTAL TABULATION WITH JUSTIFICATION' => 0x89, # CHARACTER TABULATION WITH JUSTIFICATION
396 'PARTIAL LINE DOWN' => 0x8B, # PARTIAL LINE FORWARD
397 'PARTIAL LINE UP' => 0x8C, # PARTIAL LINE BACKWARD
398 'VERTICAL TABULATION SET' => 0x8A, # LINE TABULATION SET
399 'REVERSE INDEX' => 0x8D, # REVERSE LINE FEED
400 );
52ea3e69 401
423cee85 402my $txt;
281aa49e
KW
403
404# Designed so that test decimal first, and then hex. Leading zeros
405# imply non-decimal, as do non-[0-9]
232cbbee
KW
406my $decimal_qr = qr/^[1-9]\d*$/;
407
408# Returns the hex number in $1.
409my $hex_qr = qr/^(?:[Uu]\+|0[xX])?([[:xdigit:]]+)$/;
423cee85 410
8878f897
T
411sub croak
412{
413 require Carp; goto &Carp::croak;
414} # croak
415
416sub carp
417{
418 require Carp; goto &Carp::carp;
419} # carp
420
35c0985d
MB
421sub alias (@)
422{
35c0985d 423 my $alias = ref $_[0] ? $_[0] : { @_ };
232cbbee
KW
424 foreach my $name (keys %$alias) {
425 my $value = $alias->{$name};
426 if ($value =~ $decimal_qr) {
427 $user_numeric_aliases{$name} = $value;
428
429 # Use a canonical form.
430 $inverse_user_aliases{sprintf("%04X", $value)} = $name;
431 }
432 elsif ($value =~ $hex_qr) {
b342e77e 433 my $decimal = CORE::hex $1;
232cbbee
KW
434 $user_numeric_aliases{$name} = $decimal;
435
436 # Must convert to decimal and back to guarantee canonical form
437 $inverse_user_aliases{sprintf("%04X", $decimal)} = $name;
438 }
439 else {
440 $user_name_aliases{$name} = $value;
441 }
442 }
35c0985d
MB
443} # alias
444
5a7fb30a
KW
445sub not_legal_use_bytes_msg {
446 my ($name, $ord) = @_;
447 return sprintf("Character 0x%04x with name '$name' is above 0xFF with 'use bytes' in effect", $ord);
448}
449
281aa49e 450sub alias_file ($) # Reads a file containing alias definitions
35c0985d 451{
51cf30b6
MB
452 my ($arg, $file) = @_;
453 if (-f $arg && File::Spec->file_name_is_absolute ($arg)) {
454 $file = $arg;
455 }
456 elsif ($arg =~ m/^\w+$/) {
457 $file = "unicore/${arg}_alias.pl";
458 }
459 else {
460 croak "Charnames alias files can only have identifier characters";
461 }
35c0985d 462 if (my @alias = do $file) {
51cf30b6
MB
463 @alias == 1 && !defined $alias[0] and
464 croak "$file cannot be used as alias file for charnames";
465 @alias % 2 and
466 croak "$file did not return a (valid) list of alias pairs";
35c0985d
MB
467 alias (@alias);
468 return (1);
469 }
470 0;
471} # alias_file
472
63098191
KW
473
474sub lookup_name {
b177ca84 475 my $name = shift;
63098191
KW
476 my $runtime = shift; # compile vs run time
477
478 # Finds the ordinal of a character name, first in the aliases, then in
479 # the large table. If not found, returns undef if runtime; complains
480 # and returns the Unicode replacement if compile.
481 # This is not optimized in any way yet
482
16036bcd 483 my $ord;
b177ca84 484
232cbbee
KW
485 # User alias should be checked first or else can't override ours, and if we
486 # add any, could conflict with theirs.
487 if (exists $user_numeric_aliases{$name}) {
488 $ord = $user_numeric_aliases{$name};
16036bcd 489 }
232cbbee
KW
490 elsif (exists $user_name_aliases{$name}) {
491 $name = $user_name_aliases{$name};
492 }
493 elsif (exists $system_aliases{$name}) {
494 $ord = $system_aliases{$name};
52ea3e69 495 }
232cbbee 496 elsif (exists $deprecated_aliases{$name}) {
35c0985d 497 require warnings;
232cbbee
KW
498 warnings::warnif('deprecated', "Unicode character name \"$name\" is deprecated, use \"" . viacode($deprecated_aliases{$name}) . "\" instead");
499 $ord = $deprecated_aliases{$name};
52ea3e69 500 }
b177ca84 501
423cee85 502 my @off;
52ea3e69 503
16036bcd 504 if (! defined $ord) {
35c0985d
MB
505 ## Suck in the code/name list as a big string.
506 ## Lines look like:
507 ## "0052\t\tLATIN CAPITAL LETTER R\n"
508 $txt = do "unicore/Name.pl" unless $txt;
509
510 ## @off will hold the index into the code/name string of the start and
511 ## end of the name as we find it.
512
63098191
KW
513 ## If :full, look for the name exactly; runtime implies full
514 if (($runtime || $^H{charnames_full}) && $txt =~ /\t\t\Q$name\E$/m) {
515 @off = ($-[0] + 2, $+[0]); # The 2 is for the 2 tabs
35c0985d
MB
516 }
517
518 ## If we didn't get above, and :short allowed, look for the short name.
519 ## The short name is like "greek:Sigma"
520 unless (@off) {
63098191
KW
521 if (($runtime || $^H{charnames_short}) && $name =~ /^(.+?):(.+)/s) {
522 my ($script, $cname) = ($1, $2);
523 my $case = $cname =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL";
524 if ($txt =~ m/\t\t\U$script\E (?:$case )?LETTER \U\Q$cname\E$/m) {
525 @off = ($-[0] + 2, $+[0]);
526 }
423cee85 527 }
35c0985d 528 }
b177ca84 529
35c0985d
MB
530 ## If we still don't have it, check for the name among the loaded
531 ## scripts.
63098191 532 if (! $runtime && not @off) {
35c0985d
MB
533 my $case = $name =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL";
534 for my $script (@{$^H{charnames_scripts}}) {
63098191
KW
535 if ($txt =~ m/\t\t$script (?:$case )?LETTER \U\Q$name\E$/m) {
536 @off = ($-[0] + 2, $+[0]);
537 last;
538 }
52ea3e69 539 }
35c0985d
MB
540 }
541
542 ## If we don't have it by now, give up.
543 unless (@off) {
63098191 544 return if $runtime;
35c0985d 545 carp "Unknown charname '$name'";
29eb6cbd 546 return 0xFFFD;
35c0985d
MB
547 }
548
549 ##
550 ## Now know where in the string the name starts.
551 ## The code, in hex, is before that.
552 ##
553 ## The code can be 4-6 characters long, so we've got to sort of
554 ## go look for it, just after the newline that comes before $off[0].
555 ##
556 ## This would be much easier if unicore/Name.pl had info in
557 ## a name/code order, instead of code/name order.
558 ##
559 ## The +1 after the rindex() is to skip past the newline we're finding,
560 ## or, if the rindex() fails, to put us to an offset of zero.
561 ##
562 my $hexstart = rindex($txt, "\n", $off[0]) + 1;
563
564 ## we know where it starts, so turn into number -
565 ## the ordinal for the char.
63098191 566 $ord = CORE::hex substr($txt, $hexstart, $off[0] - 2 - $hexstart);
423cee85 567 }
b177ca84 568
63098191
KW
569 return $ord if $runtime || $ord <= 255 || ! ($^H & $bytes::hint_bits);
570
571 # Here is compile time, "use bytes" is in effect, and the character
572 # won't fit in a byte
281aa49e 573 # Get the official name if have one
5a7fb30a 574 $name = substr($txt, $off[0], $off[1] - $off[0]) if @off;
5a7fb30a 575 croak not_legal_use_bytes_msg($name, $ord);
63098191
KW
576} # lookup_name
577
578sub charnames {
579 my $name = shift;
580
581 # For \N{...}. Looks up the character name and returns its ordinal if
582 # found, undef otherwise. If not in 'use bytes', forces into utf8
583
584 my $ord = lookup_name($name, 0); # 0 means compile-time
585 return unless defined $ord;
586 return chr $ord if $^H & $bytes::hint_bits;
f0175764 587
52ea3e69 588 no warnings 'utf8'; # allow even illegal characters
bfa383d6 589 return pack "U", $ord;
63098191 590}
423cee85 591
b177ca84
JF
592sub import
593{
594 shift; ## ignore class name
595
35c0985d
MB
596 if (not @_) {
597 carp("`use charnames' needs explicit imports list");
b177ca84 598 }
423cee85 599 $^H{charnames} = \&charnames ;
b177ca84
JF
600
601 ##
602 ## fill %h keys with our @_ args.
603 ##
35c0985d 604 my ($promote, %h, @args) = (0);
e5c3f898
MG
605 while (my $arg = shift) {
606 if ($arg eq ":alias") {
51cf30b6
MB
607 @_ or
608 croak ":alias needs an argument in charnames";
35c0985d
MB
609 my $alias = shift;
610 if (ref $alias) {
611 ref $alias eq "HASH" or
51cf30b6 612 croak "Only HASH reference supported as argument to :alias";
35c0985d
MB
613 alias ($alias);
614 next;
615 }
51cf30b6
MB
616 if ($alias =~ m{:(\w+)$}) {
617 $1 eq "full" || $1 eq "short" and
618 croak ":alias cannot use existing pragma :$1 (reversed order?)";
619 alias_file ($1) and $promote = 1;
620 next;
35c0985d 621 }
51cf30b6
MB
622 alias_file ($alias);
623 next;
624 }
e5c3f898
MG
625 if (substr($arg, 0, 1) eq ':' and ! ($arg eq ":full" || $arg eq ":short")) {
626 warn "unsupported special '$arg' in charnames";
51cf30b6 627 next;
35c0985d 628 }
e5c3f898 629 push @args, $arg;
35c0985d
MB
630 }
631 @args == 0 && $promote and @args = (":full");
632 @h{@args} = (1) x @args;
b177ca84 633
423cee85
JH
634 $^H{charnames_full} = delete $h{':full'};
635 $^H{charnames_short} = delete $h{':short'};
636 $^H{charnames_scripts} = [map uc, keys %h];
b177ca84
JF
637
638 ##
639 ## If utf8? warnings are enabled, and some scripts were given,
281aa49e 640 ## see if at least we can find one letter from each script.
b177ca84 641 ##
35c0985d
MB
642 if (warnings::enabled('utf8') && @{$^H{charnames_scripts}}) {
643 $txt = do "unicore/Name.pl" unless $txt;
644
645 for my $script (@{$^H{charnames_scripts}}) {
646 if (not $txt =~ m/\t\t$script (?:CAPITAL |SMALL )?LETTER /) {
647 warnings::warn('utf8', "No such script: '$script'");
b177ca84 648 }
35c0985d 649 }
bd62941a 650 }
35c0985d 651} # import
423cee85 652
63098191
KW
653my %viacode; # Cache of already-found codes
654
655sub viacode {
656
657 # Returns the name of the code point argument
4e2cda5d 658
35c0985d
MB
659 if (@_ != 1) {
660 carp "charnames::viacode() expects one argument";
bd5c3bd9 661 return;
35c0985d 662 }
f0175764 663
35c0985d 664 my $arg = shift;
b177ca84 665
e5432b89
KW
666 # This is derived from Unicode::UCD, where it is nearly the same as the
667 # function _getcode(), but here it makes sure that even a hex argument
668 # has the proper number of leading zeros, which is critical in
669 # matching against $txt below
281aa49e 670 # Must check if decimal first; see comments at that definition
35c0985d 671 my $hex;
232cbbee 672 if ($arg =~ $decimal_qr) {
35c0985d 673 $hex = sprintf "%04X", $arg;
232cbbee 674 } elsif ($arg =~ $hex_qr) {
e10d7780 675 # Below is the line that differs from the _getcode() source
c8002005 676 $hex = sprintf "%04X", hex $1;
35c0985d
MB
677 } else {
678 carp("unexpected arg \"$arg\" to charnames::viacode()");
679 return;
680 }
b177ca84 681
35c0985d 682 return $viacode{$hex} if exists $viacode{$hex};
4e2cda5d 683
ac046fe1
KW
684 # If the code point is above the max in the table, there's no point
685 # looking through it. Checking the length first is slightly faster
686 if (length($hex) <= 5 || CORE::hex($hex) <= 0x10FFFF) {
687 $txt = do "unicore/Name.pl" unless $txt;
b177ca84 688
ac046fe1
KW
689 # Return the official name, if exists. It's unclear to me (khw) at
690 # this juncture if it is better to return a user-defined override, so
691 # leaving it as is for now.
f3227b74
KW
692 if ($txt =~ m/^$hex\t\t/m) {
693
694 # The name starts with the next character and goes up to the
695 # next new-line. Using capturing parentheses above instead of
696 # @$+ more than doubles the execution time in Perl 5.13
697 $viacode{$hex} = substr($txt, $+[0], index($txt, "\n", $+[0]) - $+[0]);
698 return $viacode{$hex};
ac046fe1 699 }
232cbbee
KW
700 }
701
702 # See if there is a user name for it, before giving up completely.
ac046fe1
KW
703 if (! exists $inverse_user_aliases{$hex}) {
704 if (CORE::hex($hex) > 0x10FFFF) {
705 carp "Unicode characters only allocated up to U+10FFFF (you asked for U+$hex)";
706 }
707 return;
708 }
bd5c3bd9 709
232cbbee
KW
710 $viacode{$hex} = $inverse_user_aliases{$hex};
711 return $inverse_user_aliases{$hex};
35c0985d 712} # viacode
daf0d493 713
63098191 714my %vianame; # Cache of already-found names
4e2cda5d 715
daf0d493
JH
716sub vianame
717{
35c0985d
MB
718 if (@_ != 1) {
719 carp "charnames::vianame() expects one name argument";
720 return ()
721 }
daf0d493 722
63098191
KW
723 # Looks up the character name and returns its ordinal if
724 # found, undef otherwise.
daf0d493 725
63098191 726 my $arg = shift;
dbc0d4f2 727
63098191 728 if ($arg =~ /^U\+([0-9a-fA-F]+)$/) {
4e2cda5d 729
63098191
KW
730 # khw claims that this is bad. The function should return either a
731 # an ord or a chr for all inputs; not be bipolar. Also, under 'use
732 # bytes', can create a chr above 255.
5a7fb30a
KW
733 my $ord = CORE::hex $1;
734 return chr $ord if $ord <= 255 || ! ((caller 0)[8] & $bytes::hint_bits);
735 carp not_legal_use_bytes_msg($arg, $ord);
736 return;
63098191 737 }
daf0d493 738
63098191
KW
739 if (! exists $vianame{$arg}) {
740 $vianame{$arg} = lookup_name($arg, 1); # 1 means run-time
35c0985d 741 }
63098191
KW
742
743 return $vianame{$arg};
35c0985d 744} # vianame
b177ca84 745
423cee85
JH
746
7471;
748__END__
749
750=head1 NAME
751
da9dec57 752charnames - access to Unicode character names and define character names for C<\N{named}> string literal escapes
423cee85
JH
753
754=head1 SYNOPSIS
755
756 use charnames ':full';
4a2d328f 757 print "\N{GREEK SMALL LETTER SIGMA} is called sigma.\n";
423cee85
JH
758
759 use charnames ':short';
4a2d328f 760 print "\N{greek:Sigma} is an upper-case sigma.\n";
423cee85
JH
761
762 use charnames qw(cyrillic greek);
4a2d328f 763 print "\N{sigma} is Greek sigma, and \N{be} is Cyrillic b.\n";
423cee85 764
35c0985d
MB
765 use charnames ":full", ":alias" => {
766 e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
e5432b89 767 mychar => 0xE8000, # Private use area
76ae0c45 768 };
35c0985d 769 print "\N{e_ACUTE} is a small letter e with an acute.\n";
da9dec57 770 print "\\N{mychar} allows me to name private use characters.\n";
35c0985d 771
76ae0c45 772 use charnames ();
a23c04e4 773 print charnames::viacode(0x1234); # prints "ETHIOPIC SYLLABLE SEE"
16036bcd
KW
774 printf "%04X", charnames::vianame("GOTHIC LETTER AHSA"); # prints
775 # "10330"
b177ca84 776
423cee85
JH
777=head1 DESCRIPTION
778
da9dec57
KW
779Pragma C<use charnames> is used to gain access to the names of the
780Unicode characters, and to allow you to define your own character names.
781
782All forms of the pragma enable use of the
783L</charnames::vianame(I<name>)> function for run-time lookup of a
784character name to get its ordinal (code point), and the inverse
785function, L</charnames::viacode(I<code>)>.
786
787Forms other than C<S<"use charnames ();">> enable the use of of
788C<\N{I<CHARNAME>}> sequences to compile a Unicode character into a
789string based on its name.
790
791Note that C<\N{U+I<...>}>, where the I<...> is a hexadecimal number,
792also inserts a character into a string, but doesn't require the use of
793this pragma. The character it inserts is the one whose code point
794(ordinal value) is equal to the number. For example, C<"\N{U+263a}"> is
795the Unicode (white background, black foreground) smiley face; it doesn't
796require this pragma, whereas the equivalent, C<"\N{WHITE SMILING FACE}">
797does.
798Also, C<\N{I<...>}> can mean a regex quantifier instead of a character
799name, when the I<...> is a number (or comma separated pair of numbers;
800see L<perlreref/QUANTIFIERS>), and is not related to this pragma.
801
802The C<charnames> pragma supports arguments C<:full>, C<:short>, script
803names and customized aliases. If C<:full> is present, for expansion of
804C<\N{I<CHARNAME>}>, the string I<CHARNAME> is first looked up in the list of
76ae0c45 805standard Unicode character names. If C<:short> is present, and
da9dec57
KW
806I<CHARNAME> has the form C<I<SCRIPT>:I<CNAME>>, then I<CNAME> is looked up
807as a letter in script I<SCRIPT>. If C<use charnames> is used
808with script name arguments, then for C<\N{I<CHARNAME>}> the name
809I<CHARNAME> is looked up as a letter in the given scripts (in the
16036bcd
KW
810specified order). Customized aliases can override these, and are explained in
811L</CUSTOM ALIASES>.
423cee85 812
da9dec57 813For lookup of I<CHARNAME> inside a given script I<SCRIPTNAME>
d5448623 814this pragma looks for the names
423cee85
JH
815
816 SCRIPTNAME CAPITAL LETTER CHARNAME
817 SCRIPTNAME SMALL LETTER CHARNAME
818 SCRIPTNAME LETTER CHARNAME
819
da9dec57 820in the table of standard Unicode names. If I<CHARNAME> is lowercase,
daf0d493
JH
821then the C<CAPITAL> variant is ignored, otherwise the C<SMALL> variant
822is ignored.
823
da9dec57
KW
824Note that C<\N{...}> is compile-time; it's a special form of string
825constant used inside double-quotish strings; this means that you cannot
4e2cda5d 826use variables inside the C<\N{...}>. If you want similar run-time
da9dec57 827functionality, use L<charnames::vianame()|/charnames::vianame(I<name>)>.
423cee85 828
301a3cda 829For the C0 and C1 control characters (U+0000..U+001F, U+0080..U+009F)
da9dec57
KW
830there are no official Unicode names but you can use instead the ISO 6429
831names (LINE FEED, ESCAPE, and so forth, and their abbreviations, LF,
1f31fcd4
KW
832ESC, ...). In Unicode 3.2 (as of Perl 5.8) some naming changes took
833place, and ISO 6429 was updated, see L</ALIASES>.
301a3cda 834
e5432b89
KW
835If the input name is unknown, C<\N{NAME}> raises a warning and
836substitutes the Unicode REPLACEMENT CHARACTER (U+FFFD).
837
838It is a fatal error if C<use bytes> is in effect and the input name is
839that of a character that won't fit into a byte (i.e., whose ordinal is
840above 255).
841
da9dec57
KW
842Otherwise, any string that includes a C<\N{I<charname>}> or
843C<S<\N{U+I<code point>}>> will automatically have Unicode semantics (see
844L<perlunicode/Byte and Character Semantics>).
845
5ffe0e96 846=head1 ALIASES
423cee85 847
5ffe0e96
MB
848A few aliases have been defined for convenience: instead of having
849to use the official names
423cee85 850
5ffe0e96
MB
851 LINE FEED (LF)
852 FORM FEED (FF)
853 CARRIAGE RETURN (CR)
854 NEXT LINE (NEL)
423cee85 855
e5432b89 856(yes, with parentheses), one can use
d5448623 857
5ffe0e96
MB
858 LINE FEED
859 FORM FEED
860 CARRIAGE RETURN
861 NEXT LINE
862 LF
863 FF
864 CR
865 NEL
866
16036bcd
KW
867All the other standard abbreviations for the controls, such as C<ACK> for
868C<ACKNOWLEDGE> also can be used.
869
5ffe0e96
MB
870One can also use
871
872 BYTE ORDER MARK
873 BOM
874
16036bcd
KW
875and these abbreviations
876
877 Abbreviation Full Name
878
879 CGJ COMBINING GRAPHEME JOINER
880 FVS1 MONGOLIAN FREE VARIATION SELECTOR ONE
881 FVS2 MONGOLIAN FREE VARIATION SELECTOR TWO
882 FVS3 MONGOLIAN FREE VARIATION SELECTOR THREE
883 LRE LEFT-TO-RIGHT EMBEDDING
884 LRM LEFT-TO-RIGHT MARK
885 LRO LEFT-TO-RIGHT OVERRIDE
886 MMSP MEDIUM MATHEMATICAL SPACE
887 MVS MONGOLIAN VOWEL SEPARATOR
888 NBSP NO-BREAK SPACE
889 NNBSP NARROW NO-BREAK SPACE
890 PDF POP DIRECTIONAL FORMATTING
891 RLE RIGHT-TO-LEFT EMBEDDING
892 RLM RIGHT-TO-LEFT MARK
893 RLO RIGHT-TO-LEFT OVERRIDE
894 SHY SOFT HYPHEN
895 VS1 VARIATION SELECTOR-1
896 .
897 .
898 .
899 VS256 VARIATION SELECTOR-256
900 WJ WORD JOINER
901 ZWJ ZERO WIDTH JOINER
902 ZWNJ ZERO WIDTH NON-JOINER
903 ZWSP ZERO WIDTH SPACE
5ffe0e96
MB
904
905For backward compatibility one can use the old names for
906certain C0 and C1 controls
907
908 old new
909
5ffe0e96
MB
910 FILE SEPARATOR INFORMATION SEPARATOR FOUR
911 GROUP SEPARATOR INFORMATION SEPARATOR THREE
16036bcd
KW
912 HORIZONTAL TABULATION CHARACTER TABULATION
913 HORIZONTAL TABULATION SET CHARACTER TABULATION SET
914 HORIZONTAL TABULATION WITH JUSTIFICATION CHARACTER TABULATION
915 WITH JUSTIFICATION
5ffe0e96
MB
916 PARTIAL LINE DOWN PARTIAL LINE FORWARD
917 PARTIAL LINE UP PARTIAL LINE BACKWARD
16036bcd
KW
918 RECORD SEPARATOR INFORMATION SEPARATOR TWO
919 REVERSE INDEX REVERSE LINE FEED
920 UNIT SEPARATOR INFORMATION SEPARATOR ONE
921 VERTICAL TABULATION LINE TABULATION
922 VERTICAL TABULATION SET LINE TABULATION SET
5ffe0e96
MB
923
924but the old names in addition to giving the character
925will also give a warning about being deprecated.
423cee85 926
16036bcd
KW
927And finally, certain published variants are usable, including some for
928controls that have no Unicode names:
929
1f31fcd4
KW
930 name character
931
932 END OF PROTECTED AREA END OF GUARDED AREA, U+0097
933 HIGH OCTET PRESET U+0081
934 HOP U+0081
935 IND U+0084
936 INDEX U+0084
937 PAD U+0080
938 PADDING CHARACTER U+0080
939 PRIVATE USE 1 PRIVATE USE ONE, U+0091
940 PRIVATE USE 2 PRIVATE USE TWO, U+0092
941 SGC U+0099
942 SINGLE GRAPHIC CHARACTER INTRODUCER U+0099
943 SINGLE-SHIFT 2 SINGLE SHIFT TWO, U+008E
944 SINGLE-SHIFT 3 SINGLE SHIFT THREE, U+008F
945 START OF PROTECTED AREA START OF GUARDED AREA, U+0096
16036bcd 946
35c0985d
MB
947=head1 CUSTOM ALIASES
948
1f31fcd4
KW
949You can add customized aliases to standard (C<:full>) Unicode naming
950conventions. The aliases override any standard definitions, so, if
da9dec57
KW
951you're twisted enough, you can change C<"\N{LATIN CAPITAL LETTER A}"> to
952mean C<"B">, etc.
55bc7d3c
KW
953
954Note that an alias should not be something that is a legal curly
955brace-enclosed quantifier (see L<perlreref/QUANTIFIERS>). For example
e5432b89
KW
956C<\N{123}> means to match 123 non-newline characters, and is not treated as a
957charnames alias. Aliases are discouraged from beginning with anything
958other than an alphabetic character and from containing anything other
959than alphanumerics, spaces, dashes, parentheses, and underscores.
960Currently they must be ASCII.
961
962An alias can map to either an official Unicode character name or to a
963numeric code point (ordinal). The latter is useful for assigning names
964to code points in Unicode private use areas such as U+E800 through
965U+F8FF. The number must look like an unsigned decimal integer, or a
966hexadecimal constant beginning with C<0x>, or C<U+>.
232cbbee 967
da9dec57 968Aliases are added either by the use of anonymous hashes:
35c0985d 969
da9dec57 970 use charnames ":alias" => {
35c0985d 971 e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
232cbbee 972 mychar1 => 0xE8000,
35c0985d
MB
973 };
974 my $str = "\N{e_ACUTE}";
975
da9dec57 976or by using a file containing aliases:
35c0985d 977
da9dec57 978 use charnames ":alias" => "pro";
35c0985d 979
da9dec57
KW
980will try to read C<"unicore/pro_alias.pl"> from the C<@INC> path. This
981file should return a list in plain perl:
35c0985d
MB
982
983 (
984 A_GRAVE => "LATIN CAPITAL LETTER A WITH GRAVE",
985 A_CIRCUM => "LATIN CAPITAL LETTER A WITH CIRCUMFLEX",
986 A_DIAERES => "LATIN CAPITAL LETTER A WITH DIAERESIS",
987 A_TILDE => "LATIN CAPITAL LETTER A WITH TILDE",
988 A_BREVE => "LATIN CAPITAL LETTER A WITH BREVE",
989 A_RING => "LATIN CAPITAL LETTER A WITH RING ABOVE",
990 A_MACRON => "LATIN CAPITAL LETTER A WITH MACRON",
232cbbee 991 mychar2 => U+E8001,
35c0985d
MB
992 );
993
da9dec57
KW
994Both these methods insert C<":full"> automatically as the first argument (if no
995other argument is given), and you can give the C<":full"> explicitly as
996well, like
35c0985d 997
da9dec57 998 use charnames ":full", ":alias" => "pro";
35c0985d 999
da9dec57 1000=head1 charnames::viacode(I<code>)
b177ca84
JF
1001
1002Returns the full name of the character indicated by the numeric code.
da9dec57 1003For example,
b177ca84
JF
1004
1005 print charnames::viacode(0x2722);
1006
1007prints "FOUR TEARDROP-SPOKED ASTERISK".
1008
232cbbee
KW
1009The name returned is the official name for the code point, if
1010available, otherwise your custom alias for it. This means that your
1011alias will only be returned for code points that don't have an official
1012Unicode name (nor Unicode version 1 name), such as private use code
1013points, and the 4 control characters U+0080, U+0081, U+0084, and U+0099.
da9dec57
KW
1014If you define more than one name for the code point, it is indeterminate
1015which one will be returned.
1016
1017The function returns C<undef> if no name is known for the code point.
1018In Unicode the proper name of these is the empty string, which
1019C<undef> stringifies to. (If you ask for a code point past the legal
1020Unicode maximum of U+10FFFF that you haven't assigned an alias to, you
1021get C<undef> and a warning.)
daf0d493 1022
274085e3
PN
1023Notice that the name returned for of U+FEFF is "ZERO WIDTH NO-BREAK
1024SPACE", not "BYTE ORDER MARK".
1025
da9dec57 1026=head1 charnames::vianame(I<name>)
daf0d493
JH
1027
1028Returns the code point indicated by the name.
1f31fcd4 1029For example,
daf0d493
JH
1030
1031 printf "%04X", charnames::vianame("FOUR TEARDROP-SPOKED ASTERISK");
1032
1033prints "2722".
1034
da9dec57
KW
1035C<vianame> takes the identical inputs that C<\N{...}> does under the
1036L<C<:full> and C<:short>|/DESCRIPTION> options to the C<charnames>
1037pragma, including any L<custom aliases|/CUSTOM ALIASES> you may have
1038defined.
b177ca84 1039
1f31fcd4
KW
1040There are just a few differences. The main one is that under
1041most circumstances, (see L</BUGS> for the other ones), vianame returns
1042an ord, whereas C<\\N{...}> is seamlessly placed as a chr into the
1043string in which it appears. This leads to a second difference.
1044Since an ord is returned, it can be that of any character, even one
1045that isn't legal under the C<S<use bytes>> pragma. It is up to the
1046caller to validate the return under C<S<use bytes>> before converting it
1047to chr.
1048
1049The final difference is that if the input name is unknown C<vianame>
1050returns C<undef> instead of the REPLACEMENT CHARACTER, and it does not
1051raise a warning message.
b177ca84 1052
5ffe0e96 1053=head1 CUSTOM TRANSLATORS
52ea3e69 1054
5ffe0e96
MB
1055The mechanism of translation of C<\N{...}> escapes is general and not
1056hardwired into F<charnames.pm>. A module can install custom
1057translations (inside the scope which C<use>s the module) with the
1058following magic incantation:
52ea3e69 1059
5ffe0e96
MB
1060 sub import {
1061 shift;
1062 $^H{charnames} = \&translator;
1063 }
52ea3e69 1064
da9dec57 1065Here translator() is a subroutine which takes I<CHARNAME> as an
5ffe0e96 1066argument, and returns text to insert into the string instead of the
da9dec57 1067C<\N{I<CHARNAME>}> escape. Since the text to insert should be different
5ffe0e96
MB
1068in C<bytes> mode and out of it, the function should check the current
1069state of C<bytes>-flag as in:
52ea3e69 1070
5ffe0e96
MB
1071 use bytes (); # for $bytes::hint_bits
1072 sub translator {
1073 if ($^H & $bytes::hint_bits) {
1074 return bytes_translator(@_);
1075 }
1076 else {
1077 return utf8_translator(@_);
1078 }
1079 }
52ea3e69 1080
da9dec57 1081See L</CUSTOM ALIASES> above for restrictions on I<CHARNAME>.
f0175764 1082
1f31fcd4
KW
1083Of course, C<vianame> and C<viacode> would need to be overridden as
1084well.
1085
423cee85
JH
1086=head1 BUGS
1087
55bc7d3c 1088vianame returns a chr if the input name is of the form C<U+...>, and an ord
a0a3bc7f 1089otherwise. It is proposed to change this to always return an ord. Send email
1f31fcd4
KW
1090to C<perl5-porters@perl.org> to comment on this proposal. If S<C<use
1091bytes>> is in effect when a chr is returned, and if that chr won't fit
1092into a byte, C<undef> is returned instead.
55bc7d3c 1093
da9dec57
KW
1094All the Hangul syllable characters are treated as having no names, as
1095are almost all the CJK Unicode characters that have their code points as
1096part of their names.
55bc7d3c 1097
16036bcd
KW
1098Names must be ASCII characters only, which means that you are out of luck if
1099you want to create aliases in a language where some or all the characters of
1100the desired aliases are non-ASCII.
bee80e93 1101
fe749c9a
KW
1102Unicode standard named sequences are not recognized, such as
1103C<LATIN CAPITAL LETTER A WITH MACRON AND GRAVE>
1104(which should mean C<LATIN CAPITAL LETTER A WITH MACRON> with an additional
1105C<COMBINING GRAVE ACCENT>).
1106
55bc7d3c 1107Since evaluation of the translation function happens in the middle of
423cee85 1108compilation (of a string literal), the translation function should not
da9dec57
KW
1109do any C<eval>s or C<require>s. This restriction should be lifted (but
1110is low priority) in a future version of Perl.
423cee85
JH
1111
1112=cut
0eacc33e
KW
1113
1114# ex: set ts=8 sts=2 sw=2 noet: