Commit | Line | Data |
---|---|---|
423cee85 | 1 | package charnames; |
b177ca84 JF |
2 | use strict; |
3 | use warnings; | |
51cf30b6 | 4 | use File::Spec; |
63098191 | 5 | our $VERSION = '1.11'; |
b75c8c73 | 6 | |
d5448623 | 7 | use bytes (); # for $bytes::hint_bits |
423cee85 | 8 | |
232cbbee | 9 | my %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 | 385 | my %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 | |
232cbbee | 402 | my %user_name_aliases = ( |
16036bcd | 403 | # User defined aliases. Even more convenient :) |
232cbbee KW |
404 | # These are the ones that resolved to names |
405 | ); | |
406 | ||
407 | my %user_numeric_aliases = ( | |
408 | # And these resolve directly to code points. | |
409 | ); | |
410 | my %inverse_user_aliases = ( | |
411 | # Map from code point to name | |
16036bcd | 412 | ); |
423cee85 | 413 | my $txt; |
232cbbee KW |
414 | my $decimal_qr = qr/^[1-9]\d*$/; |
415 | ||
416 | # Returns the hex number in $1. | |
417 | my $hex_qr = qr/^(?:[Uu]\+|0[xX])?([[:xdigit:]]+)$/; | |
423cee85 | 418 | |
8878f897 T |
419 | sub croak |
420 | { | |
421 | require Carp; goto &Carp::croak; | |
422 | } # croak | |
423 | ||
424 | sub carp | |
425 | { | |
426 | require Carp; goto &Carp::carp; | |
427 | } # carp | |
428 | ||
35c0985d MB |
429 | sub alias (@) |
430 | { | |
35c0985d | 431 | my $alias = ref $_[0] ? $_[0] : { @_ }; |
232cbbee KW |
432 | foreach my $name (keys %$alias) { |
433 | my $value = $alias->{$name}; | |
434 | if ($value =~ $decimal_qr) { | |
435 | $user_numeric_aliases{$name} = $value; | |
436 | ||
437 | # Use a canonical form. | |
438 | $inverse_user_aliases{sprintf("%04X", $value)} = $name; | |
439 | } | |
440 | elsif ($value =~ $hex_qr) { | |
b342e77e | 441 | my $decimal = CORE::hex $1; |
232cbbee KW |
442 | $user_numeric_aliases{$name} = $decimal; |
443 | ||
444 | # Must convert to decimal and back to guarantee canonical form | |
445 | $inverse_user_aliases{sprintf("%04X", $decimal)} = $name; | |
446 | } | |
447 | else { | |
448 | $user_name_aliases{$name} = $value; | |
449 | } | |
450 | } | |
35c0985d MB |
451 | } # alias |
452 | ||
5a7fb30a KW |
453 | sub not_legal_use_bytes_msg { |
454 | my ($name, $ord) = @_; | |
455 | return sprintf("Character 0x%04x with name '$name' is above 0xFF with 'use bytes' in effect", $ord); | |
456 | } | |
457 | ||
35c0985d MB |
458 | sub alias_file ($) |
459 | { | |
51cf30b6 MB |
460 | my ($arg, $file) = @_; |
461 | if (-f $arg && File::Spec->file_name_is_absolute ($arg)) { | |
462 | $file = $arg; | |
463 | } | |
464 | elsif ($arg =~ m/^\w+$/) { | |
465 | $file = "unicore/${arg}_alias.pl"; | |
466 | } | |
467 | else { | |
468 | croak "Charnames alias files can only have identifier characters"; | |
469 | } | |
35c0985d | 470 | if (my @alias = do $file) { |
51cf30b6 MB |
471 | @alias == 1 && !defined $alias[0] and |
472 | croak "$file cannot be used as alias file for charnames"; | |
473 | @alias % 2 and | |
474 | croak "$file did not return a (valid) list of alias pairs"; | |
35c0985d MB |
475 | alias (@alias); |
476 | return (1); | |
477 | } | |
478 | 0; | |
479 | } # alias_file | |
480 | ||
63098191 KW |
481 | |
482 | sub lookup_name { | |
b177ca84 | 483 | my $name = shift; |
63098191 KW |
484 | my $runtime = shift; # compile vs run time |
485 | ||
486 | # Finds the ordinal of a character name, first in the aliases, then in | |
487 | # the large table. If not found, returns undef if runtime; complains | |
488 | # and returns the Unicode replacement if compile. | |
489 | # This is not optimized in any way yet | |
490 | ||
16036bcd | 491 | my $ord; |
b177ca84 | 492 | |
232cbbee KW |
493 | # User alias should be checked first or else can't override ours, and if we |
494 | # add any, could conflict with theirs. | |
495 | if (exists $user_numeric_aliases{$name}) { | |
496 | $ord = $user_numeric_aliases{$name}; | |
16036bcd | 497 | } |
232cbbee KW |
498 | elsif (exists $user_name_aliases{$name}) { |
499 | $name = $user_name_aliases{$name}; | |
500 | } | |
501 | elsif (exists $system_aliases{$name}) { | |
502 | $ord = $system_aliases{$name}; | |
52ea3e69 | 503 | } |
232cbbee | 504 | elsif (exists $deprecated_aliases{$name}) { |
35c0985d | 505 | require warnings; |
232cbbee KW |
506 | warnings::warnif('deprecated', "Unicode character name \"$name\" is deprecated, use \"" . viacode($deprecated_aliases{$name}) . "\" instead"); |
507 | $ord = $deprecated_aliases{$name}; | |
52ea3e69 | 508 | } |
b177ca84 | 509 | |
423cee85 | 510 | my @off; |
52ea3e69 | 511 | |
16036bcd | 512 | if (! defined $ord) { |
35c0985d MB |
513 | ## Suck in the code/name list as a big string. |
514 | ## Lines look like: | |
515 | ## "0052\t\tLATIN CAPITAL LETTER R\n" | |
516 | $txt = do "unicore/Name.pl" unless $txt; | |
517 | ||
518 | ## @off will hold the index into the code/name string of the start and | |
519 | ## end of the name as we find it. | |
520 | ||
63098191 KW |
521 | ## If :full, look for the name exactly; runtime implies full |
522 | if (($runtime || $^H{charnames_full}) && $txt =~ /\t\t\Q$name\E$/m) { | |
523 | @off = ($-[0] + 2, $+[0]); # The 2 is for the 2 tabs | |
35c0985d MB |
524 | } |
525 | ||
526 | ## If we didn't get above, and :short allowed, look for the short name. | |
527 | ## The short name is like "greek:Sigma" | |
528 | unless (@off) { | |
63098191 KW |
529 | if (($runtime || $^H{charnames_short}) && $name =~ /^(.+?):(.+)/s) { |
530 | my ($script, $cname) = ($1, $2); | |
531 | my $case = $cname =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL"; | |
532 | if ($txt =~ m/\t\t\U$script\E (?:$case )?LETTER \U\Q$cname\E$/m) { | |
533 | @off = ($-[0] + 2, $+[0]); | |
534 | } | |
423cee85 | 535 | } |
35c0985d | 536 | } |
b177ca84 | 537 | |
35c0985d MB |
538 | ## If we still don't have it, check for the name among the loaded |
539 | ## scripts. | |
63098191 | 540 | if (! $runtime && not @off) { |
35c0985d MB |
541 | my $case = $name =~ /[[:upper:]]/ ? "CAPITAL" : "SMALL"; |
542 | for my $script (@{$^H{charnames_scripts}}) { | |
63098191 KW |
543 | if ($txt =~ m/\t\t$script (?:$case )?LETTER \U\Q$name\E$/m) { |
544 | @off = ($-[0] + 2, $+[0]); | |
545 | last; | |
546 | } | |
52ea3e69 | 547 | } |
35c0985d MB |
548 | } |
549 | ||
550 | ## If we don't have it by now, give up. | |
551 | unless (@off) { | |
63098191 | 552 | return if $runtime; |
35c0985d MB |
553 | carp "Unknown charname '$name'"; |
554 | return "\x{FFFD}"; | |
555 | } | |
556 | ||
557 | ## | |
558 | ## Now know where in the string the name starts. | |
559 | ## The code, in hex, is before that. | |
560 | ## | |
561 | ## The code can be 4-6 characters long, so we've got to sort of | |
562 | ## go look for it, just after the newline that comes before $off[0]. | |
563 | ## | |
564 | ## This would be much easier if unicore/Name.pl had info in | |
565 | ## a name/code order, instead of code/name order. | |
566 | ## | |
567 | ## The +1 after the rindex() is to skip past the newline we're finding, | |
568 | ## or, if the rindex() fails, to put us to an offset of zero. | |
569 | ## | |
570 | my $hexstart = rindex($txt, "\n", $off[0]) + 1; | |
571 | ||
572 | ## we know where it starts, so turn into number - | |
573 | ## the ordinal for the char. | |
63098191 | 574 | $ord = CORE::hex substr($txt, $hexstart, $off[0] - 2 - $hexstart); |
423cee85 | 575 | } |
b177ca84 | 576 | |
63098191 KW |
577 | return $ord if $runtime || $ord <= 255 || ! ($^H & $bytes::hint_bits); |
578 | ||
579 | # Here is compile time, "use bytes" is in effect, and the character | |
580 | # won't fit in a byte | |
581 | ||
5a7fb30a KW |
582 | |
583 | # Get the official name if have one for the message | |
584 | $name = substr($txt, $off[0], $off[1] - $off[0]) if @off; | |
585 | ||
586 | croak not_legal_use_bytes_msg($name, $ord); | |
63098191 KW |
587 | } # lookup_name |
588 | ||
589 | sub charnames { | |
590 | my $name = shift; | |
591 | ||
592 | # For \N{...}. Looks up the character name and returns its ordinal if | |
593 | # found, undef otherwise. If not in 'use bytes', forces into utf8 | |
594 | ||
595 | my $ord = lookup_name($name, 0); # 0 means compile-time | |
596 | return unless defined $ord; | |
597 | return chr $ord if $^H & $bytes::hint_bits; | |
f0175764 | 598 | |
52ea3e69 | 599 | no warnings 'utf8'; # allow even illegal characters |
bfa383d6 | 600 | return pack "U", $ord; |
63098191 | 601 | } |
423cee85 | 602 | |
b177ca84 JF |
603 | sub import |
604 | { | |
605 | shift; ## ignore class name | |
606 | ||
35c0985d MB |
607 | if (not @_) { |
608 | carp("`use charnames' needs explicit imports list"); | |
b177ca84 | 609 | } |
423cee85 | 610 | $^H{charnames} = \&charnames ; |
b177ca84 JF |
611 | |
612 | ## | |
613 | ## fill %h keys with our @_ args. | |
614 | ## | |
35c0985d | 615 | my ($promote, %h, @args) = (0); |
e5c3f898 MG |
616 | while (my $arg = shift) { |
617 | if ($arg eq ":alias") { | |
51cf30b6 MB |
618 | @_ or |
619 | croak ":alias needs an argument in charnames"; | |
35c0985d MB |
620 | my $alias = shift; |
621 | if (ref $alias) { | |
622 | ref $alias eq "HASH" or | |
51cf30b6 | 623 | croak "Only HASH reference supported as argument to :alias"; |
35c0985d MB |
624 | alias ($alias); |
625 | next; | |
626 | } | |
51cf30b6 MB |
627 | if ($alias =~ m{:(\w+)$}) { |
628 | $1 eq "full" || $1 eq "short" and | |
629 | croak ":alias cannot use existing pragma :$1 (reversed order?)"; | |
630 | alias_file ($1) and $promote = 1; | |
631 | next; | |
35c0985d | 632 | } |
51cf30b6 MB |
633 | alias_file ($alias); |
634 | next; | |
635 | } | |
e5c3f898 MG |
636 | if (substr($arg, 0, 1) eq ':' and ! ($arg eq ":full" || $arg eq ":short")) { |
637 | warn "unsupported special '$arg' in charnames"; | |
51cf30b6 | 638 | next; |
35c0985d | 639 | } |
e5c3f898 | 640 | push @args, $arg; |
35c0985d MB |
641 | } |
642 | @args == 0 && $promote and @args = (":full"); | |
643 | @h{@args} = (1) x @args; | |
b177ca84 | 644 | |
423cee85 JH |
645 | $^H{charnames_full} = delete $h{':full'}; |
646 | $^H{charnames_short} = delete $h{':short'}; | |
647 | $^H{charnames_scripts} = [map uc, keys %h]; | |
b177ca84 JF |
648 | |
649 | ## | |
650 | ## If utf8? warnings are enabled, and some scripts were given, | |
651 | ## see if at least we can find one letter of each script. | |
652 | ## | |
35c0985d MB |
653 | if (warnings::enabled('utf8') && @{$^H{charnames_scripts}}) { |
654 | $txt = do "unicore/Name.pl" unless $txt; | |
655 | ||
656 | for my $script (@{$^H{charnames_scripts}}) { | |
657 | if (not $txt =~ m/\t\t$script (?:CAPITAL |SMALL )?LETTER /) { | |
658 | warnings::warn('utf8', "No such script: '$script'"); | |
b177ca84 | 659 | } |
35c0985d | 660 | } |
bd62941a | 661 | } |
35c0985d | 662 | } # import |
423cee85 | 663 | |
63098191 KW |
664 | my %viacode; # Cache of already-found codes |
665 | ||
666 | sub viacode { | |
667 | ||
668 | # Returns the name of the code point argument | |
4e2cda5d | 669 | |
35c0985d MB |
670 | if (@_ != 1) { |
671 | carp "charnames::viacode() expects one argument"; | |
bd5c3bd9 | 672 | return; |
35c0985d | 673 | } |
f0175764 | 674 | |
35c0985d | 675 | my $arg = shift; |
b177ca84 | 676 | |
e5432b89 KW |
677 | # This is derived from Unicode::UCD, where it is nearly the same as the |
678 | # function _getcode(), but here it makes sure that even a hex argument | |
679 | # has the proper number of leading zeros, which is critical in | |
680 | # matching against $txt below | |
35c0985d | 681 | my $hex; |
232cbbee | 682 | if ($arg =~ $decimal_qr) { |
35c0985d | 683 | $hex = sprintf "%04X", $arg; |
232cbbee | 684 | } elsif ($arg =~ $hex_qr) { |
e10d7780 | 685 | # Below is the line that differs from the _getcode() source |
c8002005 | 686 | $hex = sprintf "%04X", hex $1; |
35c0985d MB |
687 | } else { |
688 | carp("unexpected arg \"$arg\" to charnames::viacode()"); | |
689 | return; | |
690 | } | |
b177ca84 | 691 | |
35c0985d | 692 | return $viacode{$hex} if exists $viacode{$hex}; |
4e2cda5d | 693 | |
ac046fe1 KW |
694 | # If the code point is above the max in the table, there's no point |
695 | # looking through it. Checking the length first is slightly faster | |
696 | if (length($hex) <= 5 || CORE::hex($hex) <= 0x10FFFF) { | |
697 | $txt = do "unicore/Name.pl" unless $txt; | |
b177ca84 | 698 | |
ac046fe1 KW |
699 | # Return the official name, if exists. It's unclear to me (khw) at |
700 | # this juncture if it is better to return a user-defined override, so | |
701 | # leaving it as is for now. | |
702 | if ($txt =~ m/^$hex\t\t(.+)/m) { | |
703 | $viacode{$hex} = $1; | |
704 | return $1; | |
705 | } | |
232cbbee KW |
706 | } |
707 | ||
708 | # See if there is a user name for it, before giving up completely. | |
ac046fe1 KW |
709 | if (! exists $inverse_user_aliases{$hex}) { |
710 | if (CORE::hex($hex) > 0x10FFFF) { | |
711 | carp "Unicode characters only allocated up to U+10FFFF (you asked for U+$hex)"; | |
712 | } | |
713 | return; | |
714 | } | |
bd5c3bd9 | 715 | |
232cbbee KW |
716 | $viacode{$hex} = $inverse_user_aliases{$hex}; |
717 | return $inverse_user_aliases{$hex}; | |
35c0985d | 718 | } # viacode |
daf0d493 | 719 | |
63098191 | 720 | my %vianame; # Cache of already-found names |
4e2cda5d | 721 | |
daf0d493 JH |
722 | sub vianame |
723 | { | |
35c0985d MB |
724 | if (@_ != 1) { |
725 | carp "charnames::vianame() expects one name argument"; | |
726 | return () | |
727 | } | |
daf0d493 | 728 | |
63098191 KW |
729 | # Looks up the character name and returns its ordinal if |
730 | # found, undef otherwise. | |
daf0d493 | 731 | |
63098191 | 732 | my $arg = shift; |
dbc0d4f2 | 733 | |
63098191 | 734 | if ($arg =~ /^U\+([0-9a-fA-F]+)$/) { |
4e2cda5d | 735 | |
63098191 KW |
736 | # khw claims that this is bad. The function should return either a |
737 | # an ord or a chr for all inputs; not be bipolar. Also, under 'use | |
738 | # bytes', can create a chr above 255. | |
5a7fb30a KW |
739 | my $ord = CORE::hex $1; |
740 | return chr $ord if $ord <= 255 || ! ((caller 0)[8] & $bytes::hint_bits); | |
741 | carp not_legal_use_bytes_msg($arg, $ord); | |
742 | return; | |
63098191 | 743 | } |
daf0d493 | 744 | |
63098191 KW |
745 | if (! exists $vianame{$arg}) { |
746 | $vianame{$arg} = lookup_name($arg, 1); # 1 means run-time | |
35c0985d | 747 | } |
63098191 KW |
748 | |
749 | return $vianame{$arg}; | |
35c0985d | 750 | } # vianame |
b177ca84 | 751 | |
423cee85 JH |
752 | |
753 | 1; | |
754 | __END__ | |
755 | ||
756 | =head1 NAME | |
757 | ||
da9dec57 | 758 | charnames - access to Unicode character names and define character names for C<\N{named}> string literal escapes |
423cee85 JH |
759 | |
760 | =head1 SYNOPSIS | |
761 | ||
762 | use charnames ':full'; | |
4a2d328f | 763 | print "\N{GREEK SMALL LETTER SIGMA} is called sigma.\n"; |
423cee85 JH |
764 | |
765 | use charnames ':short'; | |
4a2d328f | 766 | print "\N{greek:Sigma} is an upper-case sigma.\n"; |
423cee85 JH |
767 | |
768 | use charnames qw(cyrillic greek); | |
4a2d328f | 769 | print "\N{sigma} is Greek sigma, and \N{be} is Cyrillic b.\n"; |
423cee85 | 770 | |
35c0985d MB |
771 | use charnames ":full", ":alias" => { |
772 | e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE", | |
e5432b89 | 773 | mychar => 0xE8000, # Private use area |
76ae0c45 | 774 | }; |
35c0985d | 775 | print "\N{e_ACUTE} is a small letter e with an acute.\n"; |
da9dec57 | 776 | print "\\N{mychar} allows me to name private use characters.\n"; |
35c0985d | 777 | |
76ae0c45 | 778 | use charnames (); |
a23c04e4 | 779 | print charnames::viacode(0x1234); # prints "ETHIOPIC SYLLABLE SEE" |
16036bcd KW |
780 | printf "%04X", charnames::vianame("GOTHIC LETTER AHSA"); # prints |
781 | # "10330" | |
b177ca84 | 782 | |
423cee85 JH |
783 | =head1 DESCRIPTION |
784 | ||
da9dec57 KW |
785 | Pragma C<use charnames> is used to gain access to the names of the |
786 | Unicode characters, and to allow you to define your own character names. | |
787 | ||
788 | All forms of the pragma enable use of the | |
789 | L</charnames::vianame(I<name>)> function for run-time lookup of a | |
790 | character name to get its ordinal (code point), and the inverse | |
791 | function, L</charnames::viacode(I<code>)>. | |
792 | ||
793 | Forms other than C<S<"use charnames ();">> enable the use of of | |
794 | C<\N{I<CHARNAME>}> sequences to compile a Unicode character into a | |
795 | string based on its name. | |
796 | ||
797 | Note that C<\N{U+I<...>}>, where the I<...> is a hexadecimal number, | |
798 | also inserts a character into a string, but doesn't require the use of | |
799 | this pragma. The character it inserts is the one whose code point | |
800 | (ordinal value) is equal to the number. For example, C<"\N{U+263a}"> is | |
801 | the Unicode (white background, black foreground) smiley face; it doesn't | |
802 | require this pragma, whereas the equivalent, C<"\N{WHITE SMILING FACE}"> | |
803 | does. | |
804 | Also, C<\N{I<...>}> can mean a regex quantifier instead of a character | |
805 | name, when the I<...> is a number (or comma separated pair of numbers; | |
806 | see L<perlreref/QUANTIFIERS>), and is not related to this pragma. | |
807 | ||
808 | The C<charnames> pragma supports arguments C<:full>, C<:short>, script | |
809 | names and customized aliases. If C<:full> is present, for expansion of | |
810 | C<\N{I<CHARNAME>}>, the string I<CHARNAME> is first looked up in the list of | |
76ae0c45 | 811 | standard Unicode character names. If C<:short> is present, and |
da9dec57 KW |
812 | I<CHARNAME> has the form C<I<SCRIPT>:I<CNAME>>, then I<CNAME> is looked up |
813 | as a letter in script I<SCRIPT>. If C<use charnames> is used | |
814 | with script name arguments, then for C<\N{I<CHARNAME>}> the name | |
815 | I<CHARNAME> is looked up as a letter in the given scripts (in the | |
16036bcd KW |
816 | specified order). Customized aliases can override these, and are explained in |
817 | L</CUSTOM ALIASES>. | |
423cee85 | 818 | |
da9dec57 | 819 | For lookup of I<CHARNAME> inside a given script I<SCRIPTNAME> |
d5448623 | 820 | this pragma looks for the names |
423cee85 JH |
821 | |
822 | SCRIPTNAME CAPITAL LETTER CHARNAME | |
823 | SCRIPTNAME SMALL LETTER CHARNAME | |
824 | SCRIPTNAME LETTER CHARNAME | |
825 | ||
da9dec57 | 826 | in the table of standard Unicode names. If I<CHARNAME> is lowercase, |
daf0d493 JH |
827 | then the C<CAPITAL> variant is ignored, otherwise the C<SMALL> variant |
828 | is ignored. | |
829 | ||
da9dec57 KW |
830 | Note that C<\N{...}> is compile-time; it's a special form of string |
831 | constant used inside double-quotish strings; this means that you cannot | |
4e2cda5d | 832 | use variables inside the C<\N{...}>. If you want similar run-time |
da9dec57 | 833 | functionality, use L<charnames::vianame()|/charnames::vianame(I<name>)>. |
423cee85 | 834 | |
301a3cda | 835 | For the C0 and C1 control characters (U+0000..U+001F, U+0080..U+009F) |
da9dec57 KW |
836 | there are no official Unicode names but you can use instead the ISO 6429 |
837 | names (LINE FEED, ESCAPE, and so forth, and their abbreviations, LF, | |
1f31fcd4 KW |
838 | ESC, ...). In Unicode 3.2 (as of Perl 5.8) some naming changes took |
839 | place, and ISO 6429 was updated, see L</ALIASES>. | |
301a3cda | 840 | |
e5432b89 KW |
841 | If the input name is unknown, C<\N{NAME}> raises a warning and |
842 | substitutes the Unicode REPLACEMENT CHARACTER (U+FFFD). | |
843 | ||
844 | It is a fatal error if C<use bytes> is in effect and the input name is | |
845 | that of a character that won't fit into a byte (i.e., whose ordinal is | |
846 | above 255). | |
847 | ||
da9dec57 KW |
848 | Otherwise, any string that includes a C<\N{I<charname>}> or |
849 | C<S<\N{U+I<code point>}>> will automatically have Unicode semantics (see | |
850 | L<perlunicode/Byte and Character Semantics>). | |
851 | ||
5ffe0e96 | 852 | =head1 ALIASES |
423cee85 | 853 | |
5ffe0e96 MB |
854 | A few aliases have been defined for convenience: instead of having |
855 | to use the official names | |
423cee85 | 856 | |
5ffe0e96 MB |
857 | LINE FEED (LF) |
858 | FORM FEED (FF) | |
859 | CARRIAGE RETURN (CR) | |
860 | NEXT LINE (NEL) | |
423cee85 | 861 | |
e5432b89 | 862 | (yes, with parentheses), one can use |
d5448623 | 863 | |
5ffe0e96 MB |
864 | LINE FEED |
865 | FORM FEED | |
866 | CARRIAGE RETURN | |
867 | NEXT LINE | |
868 | LF | |
869 | FF | |
870 | CR | |
871 | NEL | |
872 | ||
16036bcd KW |
873 | All the other standard abbreviations for the controls, such as C<ACK> for |
874 | C<ACKNOWLEDGE> also can be used. | |
875 | ||
5ffe0e96 MB |
876 | One can also use |
877 | ||
878 | BYTE ORDER MARK | |
879 | BOM | |
880 | ||
16036bcd KW |
881 | and these abbreviations |
882 | ||
883 | Abbreviation Full Name | |
884 | ||
885 | CGJ COMBINING GRAPHEME JOINER | |
886 | FVS1 MONGOLIAN FREE VARIATION SELECTOR ONE | |
887 | FVS2 MONGOLIAN FREE VARIATION SELECTOR TWO | |
888 | FVS3 MONGOLIAN FREE VARIATION SELECTOR THREE | |
889 | LRE LEFT-TO-RIGHT EMBEDDING | |
890 | LRM LEFT-TO-RIGHT MARK | |
891 | LRO LEFT-TO-RIGHT OVERRIDE | |
892 | MMSP MEDIUM MATHEMATICAL SPACE | |
893 | MVS MONGOLIAN VOWEL SEPARATOR | |
894 | NBSP NO-BREAK SPACE | |
895 | NNBSP NARROW NO-BREAK SPACE | |
896 | PDF POP DIRECTIONAL FORMATTING | |
897 | RLE RIGHT-TO-LEFT EMBEDDING | |
898 | RLM RIGHT-TO-LEFT MARK | |
899 | RLO RIGHT-TO-LEFT OVERRIDE | |
900 | SHY SOFT HYPHEN | |
901 | VS1 VARIATION SELECTOR-1 | |
902 | . | |
903 | . | |
904 | . | |
905 | VS256 VARIATION SELECTOR-256 | |
906 | WJ WORD JOINER | |
907 | ZWJ ZERO WIDTH JOINER | |
908 | ZWNJ ZERO WIDTH NON-JOINER | |
909 | ZWSP ZERO WIDTH SPACE | |
5ffe0e96 MB |
910 | |
911 | For backward compatibility one can use the old names for | |
912 | certain C0 and C1 controls | |
913 | ||
914 | old new | |
915 | ||
5ffe0e96 MB |
916 | FILE SEPARATOR INFORMATION SEPARATOR FOUR |
917 | GROUP SEPARATOR INFORMATION SEPARATOR THREE | |
16036bcd KW |
918 | HORIZONTAL TABULATION CHARACTER TABULATION |
919 | HORIZONTAL TABULATION SET CHARACTER TABULATION SET | |
920 | HORIZONTAL TABULATION WITH JUSTIFICATION CHARACTER TABULATION | |
921 | WITH JUSTIFICATION | |
5ffe0e96 MB |
922 | PARTIAL LINE DOWN PARTIAL LINE FORWARD |
923 | PARTIAL LINE UP PARTIAL LINE BACKWARD | |
16036bcd KW |
924 | RECORD SEPARATOR INFORMATION SEPARATOR TWO |
925 | REVERSE INDEX REVERSE LINE FEED | |
926 | UNIT SEPARATOR INFORMATION SEPARATOR ONE | |
927 | VERTICAL TABULATION LINE TABULATION | |
928 | VERTICAL TABULATION SET LINE TABULATION SET | |
5ffe0e96 MB |
929 | |
930 | but the old names in addition to giving the character | |
931 | will also give a warning about being deprecated. | |
423cee85 | 932 | |
16036bcd KW |
933 | And finally, certain published variants are usable, including some for |
934 | controls that have no Unicode names: | |
935 | ||
1f31fcd4 KW |
936 | name character |
937 | ||
938 | END OF PROTECTED AREA END OF GUARDED AREA, U+0097 | |
939 | HIGH OCTET PRESET U+0081 | |
940 | HOP U+0081 | |
941 | IND U+0084 | |
942 | INDEX U+0084 | |
943 | PAD U+0080 | |
944 | PADDING CHARACTER U+0080 | |
945 | PRIVATE USE 1 PRIVATE USE ONE, U+0091 | |
946 | PRIVATE USE 2 PRIVATE USE TWO, U+0092 | |
947 | SGC U+0099 | |
948 | SINGLE GRAPHIC CHARACTER INTRODUCER U+0099 | |
949 | SINGLE-SHIFT 2 SINGLE SHIFT TWO, U+008E | |
950 | SINGLE-SHIFT 3 SINGLE SHIFT THREE, U+008F | |
951 | START OF PROTECTED AREA START OF GUARDED AREA, U+0096 | |
16036bcd | 952 | |
35c0985d MB |
953 | =head1 CUSTOM ALIASES |
954 | ||
1f31fcd4 KW |
955 | You can add customized aliases to standard (C<:full>) Unicode naming |
956 | conventions. The aliases override any standard definitions, so, if | |
da9dec57 KW |
957 | you're twisted enough, you can change C<"\N{LATIN CAPITAL LETTER A}"> to |
958 | mean C<"B">, etc. | |
55bc7d3c KW |
959 | |
960 | Note that an alias should not be something that is a legal curly | |
961 | brace-enclosed quantifier (see L<perlreref/QUANTIFIERS>). For example | |
e5432b89 KW |
962 | C<\N{123}> means to match 123 non-newline characters, and is not treated as a |
963 | charnames alias. Aliases are discouraged from beginning with anything | |
964 | other than an alphabetic character and from containing anything other | |
965 | than alphanumerics, spaces, dashes, parentheses, and underscores. | |
966 | Currently they must be ASCII. | |
967 | ||
968 | An alias can map to either an official Unicode character name or to a | |
969 | numeric code point (ordinal). The latter is useful for assigning names | |
970 | to code points in Unicode private use areas such as U+E800 through | |
971 | U+F8FF. The number must look like an unsigned decimal integer, or a | |
972 | hexadecimal constant beginning with C<0x>, or C<U+>. | |
232cbbee | 973 | |
da9dec57 | 974 | Aliases are added either by the use of anonymous hashes: |
35c0985d | 975 | |
da9dec57 | 976 | use charnames ":alias" => { |
35c0985d | 977 | e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE", |
232cbbee | 978 | mychar1 => 0xE8000, |
35c0985d MB |
979 | }; |
980 | my $str = "\N{e_ACUTE}"; | |
981 | ||
da9dec57 | 982 | or by using a file containing aliases: |
35c0985d | 983 | |
da9dec57 | 984 | use charnames ":alias" => "pro"; |
35c0985d | 985 | |
da9dec57 KW |
986 | will try to read C<"unicore/pro_alias.pl"> from the C<@INC> path. This |
987 | file should return a list in plain perl: | |
35c0985d MB |
988 | |
989 | ( | |
990 | A_GRAVE => "LATIN CAPITAL LETTER A WITH GRAVE", | |
991 | A_CIRCUM => "LATIN CAPITAL LETTER A WITH CIRCUMFLEX", | |
992 | A_DIAERES => "LATIN CAPITAL LETTER A WITH DIAERESIS", | |
993 | A_TILDE => "LATIN CAPITAL LETTER A WITH TILDE", | |
994 | A_BREVE => "LATIN CAPITAL LETTER A WITH BREVE", | |
995 | A_RING => "LATIN CAPITAL LETTER A WITH RING ABOVE", | |
996 | A_MACRON => "LATIN CAPITAL LETTER A WITH MACRON", | |
232cbbee | 997 | mychar2 => U+E8001, |
35c0985d MB |
998 | ); |
999 | ||
da9dec57 KW |
1000 | Both these methods insert C<":full"> automatically as the first argument (if no |
1001 | other argument is given), and you can give the C<":full"> explicitly as | |
1002 | well, like | |
35c0985d | 1003 | |
da9dec57 | 1004 | use charnames ":full", ":alias" => "pro"; |
35c0985d | 1005 | |
da9dec57 | 1006 | =head1 charnames::viacode(I<code>) |
b177ca84 JF |
1007 | |
1008 | Returns the full name of the character indicated by the numeric code. | |
da9dec57 | 1009 | For example, |
b177ca84 JF |
1010 | |
1011 | print charnames::viacode(0x2722); | |
1012 | ||
1013 | prints "FOUR TEARDROP-SPOKED ASTERISK". | |
1014 | ||
232cbbee KW |
1015 | The name returned is the official name for the code point, if |
1016 | available, otherwise your custom alias for it. This means that your | |
1017 | alias will only be returned for code points that don't have an official | |
1018 | Unicode name (nor Unicode version 1 name), such as private use code | |
1019 | points, and the 4 control characters U+0080, U+0081, U+0084, and U+0099. | |
da9dec57 KW |
1020 | If you define more than one name for the code point, it is indeterminate |
1021 | which one will be returned. | |
1022 | ||
1023 | The function returns C<undef> if no name is known for the code point. | |
1024 | In Unicode the proper name of these is the empty string, which | |
1025 | C<undef> stringifies to. (If you ask for a code point past the legal | |
1026 | Unicode maximum of U+10FFFF that you haven't assigned an alias to, you | |
1027 | get C<undef> and a warning.) | |
daf0d493 | 1028 | |
274085e3 PN |
1029 | Notice that the name returned for of U+FEFF is "ZERO WIDTH NO-BREAK |
1030 | SPACE", not "BYTE ORDER MARK". | |
1031 | ||
da9dec57 | 1032 | =head1 charnames::vianame(I<name>) |
daf0d493 JH |
1033 | |
1034 | Returns the code point indicated by the name. | |
1f31fcd4 | 1035 | For example, |
daf0d493 JH |
1036 | |
1037 | printf "%04X", charnames::vianame("FOUR TEARDROP-SPOKED ASTERISK"); | |
1038 | ||
1039 | prints "2722". | |
1040 | ||
da9dec57 KW |
1041 | C<vianame> takes the identical inputs that C<\N{...}> does under the |
1042 | L<C<:full> and C<:short>|/DESCRIPTION> options to the C<charnames> | |
1043 | pragma, including any L<custom aliases|/CUSTOM ALIASES> you may have | |
1044 | defined. | |
b177ca84 | 1045 | |
1f31fcd4 KW |
1046 | There are just a few differences. The main one is that under |
1047 | most circumstances, (see L</BUGS> for the other ones), vianame returns | |
1048 | an ord, whereas C<\\N{...}> is seamlessly placed as a chr into the | |
1049 | string in which it appears. This leads to a second difference. | |
1050 | Since an ord is returned, it can be that of any character, even one | |
1051 | that isn't legal under the C<S<use bytes>> pragma. It is up to the | |
1052 | caller to validate the return under C<S<use bytes>> before converting it | |
1053 | to chr. | |
1054 | ||
1055 | The final difference is that if the input name is unknown C<vianame> | |
1056 | returns C<undef> instead of the REPLACEMENT CHARACTER, and it does not | |
1057 | raise a warning message. | |
b177ca84 | 1058 | |
5ffe0e96 | 1059 | =head1 CUSTOM TRANSLATORS |
52ea3e69 | 1060 | |
5ffe0e96 MB |
1061 | The mechanism of translation of C<\N{...}> escapes is general and not |
1062 | hardwired into F<charnames.pm>. A module can install custom | |
1063 | translations (inside the scope which C<use>s the module) with the | |
1064 | following magic incantation: | |
52ea3e69 | 1065 | |
5ffe0e96 MB |
1066 | sub import { |
1067 | shift; | |
1068 | $^H{charnames} = \&translator; | |
1069 | } | |
52ea3e69 | 1070 | |
da9dec57 | 1071 | Here translator() is a subroutine which takes I<CHARNAME> as an |
5ffe0e96 | 1072 | argument, and returns text to insert into the string instead of the |
da9dec57 | 1073 | C<\N{I<CHARNAME>}> escape. Since the text to insert should be different |
5ffe0e96 MB |
1074 | in C<bytes> mode and out of it, the function should check the current |
1075 | state of C<bytes>-flag as in: | |
52ea3e69 | 1076 | |
5ffe0e96 MB |
1077 | use bytes (); # for $bytes::hint_bits |
1078 | sub translator { | |
1079 | if ($^H & $bytes::hint_bits) { | |
1080 | return bytes_translator(@_); | |
1081 | } | |
1082 | else { | |
1083 | return utf8_translator(@_); | |
1084 | } | |
1085 | } | |
52ea3e69 | 1086 | |
da9dec57 | 1087 | See L</CUSTOM ALIASES> above for restrictions on I<CHARNAME>. |
f0175764 | 1088 | |
1f31fcd4 KW |
1089 | Of course, C<vianame> and C<viacode> would need to be overridden as |
1090 | well. | |
1091 | ||
423cee85 JH |
1092 | =head1 BUGS |
1093 | ||
55bc7d3c | 1094 | vianame returns a chr if the input name is of the form C<U+...>, and an ord |
a0a3bc7f | 1095 | otherwise. It is proposed to change this to always return an ord. Send email |
1f31fcd4 KW |
1096 | to C<perl5-porters@perl.org> to comment on this proposal. If S<C<use |
1097 | bytes>> is in effect when a chr is returned, and if that chr won't fit | |
1098 | into a byte, C<undef> is returned instead. | |
55bc7d3c | 1099 | |
da9dec57 KW |
1100 | All the Hangul syllable characters are treated as having no names, as |
1101 | are almost all the CJK Unicode characters that have their code points as | |
1102 | part of their names. | |
55bc7d3c | 1103 | |
16036bcd KW |
1104 | Names must be ASCII characters only, which means that you are out of luck if |
1105 | you want to create aliases in a language where some or all the characters of | |
1106 | the desired aliases are non-ASCII. | |
bee80e93 | 1107 | |
fe749c9a KW |
1108 | Unicode standard named sequences are not recognized, such as |
1109 | C<LATIN CAPITAL LETTER A WITH MACRON AND GRAVE> | |
1110 | (which should mean C<LATIN CAPITAL LETTER A WITH MACRON> with an additional | |
1111 | C<COMBINING GRAVE ACCENT>). | |
1112 | ||
55bc7d3c | 1113 | Since evaluation of the translation function happens in the middle of |
423cee85 | 1114 | compilation (of a string literal), the translation function should not |
da9dec57 KW |
1115 | do any C<eval>s or C<require>s. This restriction should be lifted (but |
1116 | is low priority) in a future version of Perl. | |
423cee85 JH |
1117 | |
1118 | =cut | |
0eacc33e KW |
1119 | |
1120 | # ex: set ts=8 sts=2 sw=2 noet: |