This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Re: utf8 concat, mg_get
[perl5.git] / pod / perlebcdic.pod
CommitLineData
d396a558
JH
1=head1 NAME
2
3perlebcdic - Considerations for running Perl on EBCDIC platforms
4
5=head1 DESCRIPTION
6
7An exploration of some of the issues facing Perl programmers
8on EBCDIC based computers. We do not cover localization,
9internationalization, or multi byte character set issues (yet).
10
11Portions that are still incomplete are marked with XXX.
12
13=head1 COMMON CHARACTER CODE SETS
14
15=head2 ASCII
16
17The American Standard Code for Information Interchange is a set of
18integers running from 0 to 127 (decimal) that imply character
19interpretation by the display and other system(s) of computers.
51b5cecb 20The range 0..127 can be covered by setting the bits in a 7-bit binary
d396a558 21digit, hence the set is sometimes referred to as a "7-bit ASCII".
51b5cecb 22ASCII was described by the American National Standards Institute
d396a558
JH
23document ANSI X3.4-1986. It was also described by ISO 646:1991
24(with localization for currency symbols). The full ASCII set is
25given in the table below as the first 128 elements. Languages that
26can be written adequately with the characters in ASCII include
27English, Hawaiian, Indonesian, Swahili and some Native American
28languages.
29
51b5cecb
PP
30There are many character sets that extend the range of integers
31from 0..2**7-1 up to 2**8-1, or 8 bit bytes (octets if you prefer).
32One common one is the ISO 8859-1 character set.
33
d396a558
JH
34=head2 ISO 8859
35
36The ISO 8859-$n are a collection of character code sets from the
37International Organization for Standardization (ISO) each of which
38adds characters to the ASCII set that are typically found in European
39languages many of which are based on the Roman, or Latin, alphabet.
40
41=head2 Latin 1 (ISO 8859-1)
42
43A particular 8-bit extension to ASCII that includes grave and acute
44accented Latin characters. Languages that can employ ISO 8859-1
45include all the languages covered by ASCII as well as Afrikaans,
46Albanian, Basque, Catalan, Danish, Faroese, Finnish, Norwegian,
47Portugese, Spanish, and Swedish. Dutch is covered albeit without
48the ij ligature. French is covered too but without the oe ligature.
49German can use ISO 8859-1 but must do so without German-style
50quotation marks. This set is based on Western European extensions
51to ASCII and is commonly encountered in world wide web work.
52In IBM character code set identification terminology ISO 8859-1 is
51b5cecb 53also known as CCSID 819 (or sometimes 0819 or even 00819).
d396a558
JH
54
55=head2 EBCDIC
56
51b5cecb
PP
57The Extended Binary Coded Decimal Interchange Code refers to a
58large collection of slightly different single and multi byte
59coded character sets that are different from ASCII or ISO 8859-1
60and typically run on host computers. The EBCDIC encodings derive
61from 8 bit byte extensions of Hollerith punched card encodings.
d396a558
JH
62The layout on the cards was such that high bits were set for the
63upper and lower case alphabet characters [a-z] and [A-Z], but there
64were gaps within each latin alphabet range.
65
66=head2 13 variant characters
67
51b5cecb
PP
68Some IBM EBCDIC character sets may be known by character code set
69identification numbers (CCSID numbers) or code page numbers. Leading
70zero digits in CCSID numbers within this document are insignificant.
71E.g. CCSID 0037 may be referred to as 37 in places.
72
73Among IBM EBCDIC character code sets there are 13 characters that
74are often mapped to different integer values. Those characters
75are known as the 13 "variant" characters and are:
d396a558 76
51b5cecb 77 \ [ ] { } ^ ~ ! # | $ @ `
d396a558
JH
78
79=head2 0037
80
81Character code set ID 0037 is a mapping of the ASCII plus Latin-1
82characters (i.e. ISO 8859-1) to an EBCDIC set. 0037 is used
51b5cecb
PP
83in North American English locales on the OS/400 operating system
84that runs on AS/400 computers. CCSID 37 differs from ISO 8859-1
85in 237 places, in other words they agree on only 19 code point values.
d396a558
JH
86
87=head2 1047
88
89Character code set ID 1047 is also a mapping of the ASCII plus
90Latin-1 characters (i.e. ISO 8859-1) to an EBCDIC set. 1047 is
91used under Unix System Services for OS/390, and OpenEdition for VM/ESA.
92CCSID 1047 differs from CCSID 0037 in eight places.
93
94=head2 POSIX-BC
95
96The EBCDIC code page in use on Siemens' BS2000 system is distinct from
971047 and 0037. It is identified below as the POSIX-BC set.
98
99=head1 SINGLE OCTET TABLES
100
101The following tables list the ASCII and Latin 1 ordered sets including
102the subsets: C0 controls (0..31), ASCII graphics (32..7e), delete (7f),
103C1 controls (80..9f), and Latin-1 (a.k.a. ISO 8859-1) (a0..ff). In the
104table non-printing control character names as well as the Latin 1
105extensions to ASCII have been labelled with character names roughly
106corresponding to I<The Unicode Standard, Version 2.0> albeit with
107substitutions such as s/LATIN// and s/VULGAR// in all cases,
108s/CAPITAL LETTER// in some cases, and s/SMALL LETTER ([A-Z])/\l$1/
109in some other cases. The "names" of the C1 control set
110(128..159 in ISO 8859-1) are somewhat arbitrary. The differences
111between the 0037 and 1047 sets are flagged with ***. The differences
112between the 1047 and POSIX-BC sets are flagged with ###.
113All ord() numbers listed are decimal. If you would rather see this
114table listing octal values then run the table (that is, the pod
115version of this document since this recipe may not work with
51b5cecb 116a pod2_other_format translation) through:
d396a558
JH
117
118=over 4
119
120=item recipe 0
121
122=back
123
124 perl -ne 'if(/(.{33})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
125 -e '{printf("%s%-9o%-9o%-9o%-9o\n",$1,$2,$3,$4,$5)}' perlebcdic.pod
126
127If you would rather see this table listing hexadecimal values then
128run the table through:
129
130=over 4
131
132=item recipe 1
133
134=back
135
136 perl -ne 'if(/(.{33})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
137 -e '{printf("%s%-9X%-9X%-9X%-9X\n",$1,$2,$3,$4,$5)}' perlebcdic.pod
138
139
140 8859-1
141 chr 0819 0037 1047 POSIX-BC
142 ----------------------------------------------------------------
143 <NULL> 0 0 0 0
144 <START OF HEADING> 1 1 1 1
145 <START OF TEXT> 2 2 2 2
146 <END OF TEXT> 3 3 3 3
147 <END OF TRANSMISSION> 4 55 55 55
148 <ENQUIRY> 5 45 45 45
149 <ACKNOWLEDGE> 6 46 46 46
150 <BELL> 7 47 47 47
151 <BACKSPACE> 8 22 22 22
152 <HORIZONTAL TABULATION> 9 5 5 5
153 <LINE FEED> 10 37 21 21 ***
154 <VERTICAL TABULATION> 11 11 11 11
155 <FORM FEED> 12 12 12 12
156 <CARRIAGE RETURN> 13 13 13 13
157 <SHIFT OUT> 14 14 14 14
158 <SHIFT IN> 15 15 15 15
159 <DATA LINK ESCAPE> 16 16 16 16
160 <DEVICE CONTROL ONE> 17 17 17 17
161 <DEVICE CONTROL TWO> 18 18 18 18
162 <DEVICE CONTROL THREE> 19 19 19 19
163 <DEVICE CONTROL FOUR> 20 60 60 60
164 <NEGATIVE ACKNOWLEDGE> 21 61 61 61
165 <SYNCHRONOUS IDLE> 22 50 50 50
166 <END OF TRANSMISSION BLOCK> 23 38 38 38
167 <CANCEL> 24 24 24 24
168 <END OF MEDIUM> 25 25 25 25
169 <SUBSTITUTE> 26 63 63 63
170 <ESCAPE> 27 39 39 39
171 <FILE SEPARATOR> 28 28 28 28
172 <GROUP SEPARATOR> 29 29 29 29
173 <RECORD SEPARATOR> 30 30 30 30
174 <UNIT SEPARATOR> 31 31 31 31
175 <SPACE> 32 64 64 64
176 ! 33 90 90 90
177 " 34 127 127 127
178 # 35 123 123 123
179 $ 36 91 91 91
180 % 37 108 108 108
181 & 38 80 80 80
182 ' 39 125 125 125
183 ( 40 77 77 77
184 ) 41 93 93 93
185 * 42 92 92 92
186 + 43 78 78 78
187 , 44 107 107 107
188 - 45 96 96 96
189 . 46 75 75 75
190 / 47 97 97 97
191 0 48 240 240 240
192 1 49 241 241 241
193 2 50 242 242 242
194 3 51 243 243 243
195 4 52 244 244 244
196 5 53 245 245 245
197 6 54 246 246 246
198 7 55 247 247 247
199 8 56 248 248 248
200 9 57 249 249 249
201 : 58 122 122 122
202 ; 59 94 94 94
203 < 60 76 76 76
204 = 61 126 126 126
205 > 62 110 110 110
206 ? 63 111 111 111
207 @ 64 124 124 124
208 A 65 193 193 193
209 B 66 194 194 194
210 C 67 195 195 195
211 D 68 196 196 196
212 E 69 197 197 197
213 F 70 198 198 198
214 G 71 199 199 199
215 H 72 200 200 200
216 I 73 201 201 201
217 J 74 209 209 209
218 K 75 210 210 210
219 L 76 211 211 211
220 M 77 212 212 212
221 N 78 213 213 213
222 O 79 214 214 214
223 P 80 215 215 215
224 Q 81 216 216 216
225 R 82 217 217 217
226 S 83 226 226 226
227 T 84 227 227 227
228 U 85 228 228 228
229 V 86 229 229 229
230 W 87 230 230 230
231 X 88 231 231 231
232 Y 89 232 232 232
233 Z 90 233 233 233
234 [ 91 186 173 187 *** ###
235 \ 92 224 224 188 ###
236 ] 93 187 189 189 ***
237 ^ 94 176 95 106 *** ###
238 _ 95 109 109 109
239 ` 96 121 121 74 ###
240 a 97 129 129 129
241 b 98 130 130 130
242 c 99 131 131 131
243 d 100 132 132 132
244 e 101 133 133 133
245 f 102 134 134 134
246 g 103 135 135 135
247 h 104 136 136 136
248 i 105 137 137 137
249 j 106 145 145 145
250 k 107 146 146 146
251 l 108 147 147 147
252 m 109 148 148 148
253 n 110 149 149 149
254 o 111 150 150 150
255 p 112 151 151 151
256 q 113 152 152 152
257 r 114 153 153 153
258 s 115 162 162 162
259 t 116 163 163 163
260 u 117 164 164 164
261 v 118 165 165 165
262 w 119 166 166 166
263 x 120 167 167 167
264 y 121 168 168 168
265 z 122 169 169 169
266 { 123 192 192 251 ###
267 | 124 79 79 79
268 } 125 208 208 253 ###
269 ~ 126 161 161 255 ###
270 <DELETE> 127 7 7 7
271 <C1 0> 128 32 32 32
272 <C1 1> 129 33 33 33
273 <C1 2> 130 34 34 34
274 <C1 3> 131 35 35 35
275 <C1 4> 132 36 36 36
276 <C1 5> 133 21 37 37 ***
277 <C1 6> 134 6 6 6
278 <C1 7> 135 23 23 23
279 <C1 8> 136 40 40 40
280 <C1 9> 137 41 41 41
281 <C1 10> 138 42 42 42
282 <C1 11> 139 43 43 43
283 <C1 12> 140 44 44 44
284 <C1 13> 141 9 9 9
285 <C1 14> 142 10 10 10
286 <C1 15> 143 27 27 27
287 <C1 16> 144 48 48 48
288 <C1 17> 145 49 49 49
289 <C1 18> 146 26 26 26
290 <C1 19> 147 51 51 51
291 <C1 20> 148 52 52 52
292 <C1 21> 149 53 53 53
293 <C1 22> 150 54 54 54
294 <C1 23> 151 8 8 8
295 <C1 24> 152 56 56 56
296 <C1 25> 153 57 57 57
297 <C1 26> 154 58 58 58
298 <C1 27> 155 59 59 59
299 <C1 28> 156 4 4 4
300 <C1 29> 157 20 20 20
301 <C1 30> 158 62 62 62
302 <C1 31> 159 255 255 95 ###
303 <NON-BREAKING SPACE> 160 65 65 65
304 <INVERTED EXCLAMATION MARK> 161 170 170 170
305 <CENT SIGN> 162 74 74 176 ###
306 <POUND SIGN> 163 177 177 177
307 <CURRENCY SIGN> 164 159 159 159
308 <YEN SIGN> 165 178 178 178
309 <BROKEN BAR> 166 106 106 208 ###
310 <SECTION SIGN> 167 181 181 181
311 <DIAERESIS> 168 189 187 121 *** ###
312 <COPYRIGHT SIGN> 169 180 180 180
313 <FEMININE ORDINAL INDICATOR> 170 154 154 154
314 <LEFT POINTING GUILLEMET> 171 138 138 138
315 <NOT SIGN> 172 95 176 186 *** ###
316 <SOFT HYPHEN> 173 202 202 202
317 <REGISTERED TRADE MARK SIGN> 174 175 175 175
318 <MACRON> 175 188 188 161 ###
319 <DEGREE SIGN> 176 144 144 144
320 <PLUS-OR-MINUS SIGN> 177 143 143 143
321 <SUPERSCRIPT TWO> 178 234 234 234
322 <SUPERSCRIPT THREE> 179 250 250 250
323 <ACUTE ACCENT> 180 190 190 190
324 <MICRO SIGN> 181 160 160 160
325 <PARAGRAPH SIGN> 182 182 182 182
326 <MIDDLE DOT> 183 179 179 179
327 <CEDILLA> 184 157 157 157
328 <SUPERSCRIPT ONE> 185 218 218 218
329 <MASC. ORDINAL INDICATOR> 186 155 155 155
330 <RIGHT POINTING GUILLEMET> 187 139 139 139
331 <FRACTION ONE QUARTER> 188 183 183 183
332 <FRACTION ONE HALF> 189 184 184 184
333 <FRACTION THREE QUARTERS> 190 185 185 185
334 <INVERTED QUESTION MARK> 191 171 171 171
335 <A WITH GRAVE> 192 100 100 100
336 <A WITH ACUTE> 193 101 101 101
337 <A WITH CIRCUMFLEX> 194 98 98 98
338 <A WITH TILDE> 195 102 102 102
339 <A WITH DIAERESIS> 196 99 99 99
340 <A WITH RING ABOVE> 197 103 103 103
341 <CAPITAL LIGATURE AE> 198 158 158 158
342 <C WITH CEDILLA> 199 104 104 104
343 <E WITH GRAVE> 200 116 116 116
344 <E WITH ACUTE> 201 113 113 113
345 <E WITH CIRCUMFLEX> 202 114 114 114
346 <E WITH DIAERESIS> 203 115 115 115
347 <I WITH GRAVE> 204 120 120 120
348 <I WITH ACUTE> 205 117 117 117
349 <I WITH CIRCUMFLEX> 206 118 118 118
350 <I WITH DIAERESIS> 207 119 119 119
351 <CAPITAL LETTER ETH> 208 172 172 172
352 <N WITH TILDE> 209 105 105 105
353 <O WITH GRAVE> 210 237 237 237
354 <O WITH ACUTE> 211 238 238 238
355 <O WITH CIRCUMFLEX> 212 235 235 235
356 <O WITH TILDE> 213 239 239 239
357 <O WITH DIAERESIS> 214 236 236 236
358 <MULTIPLICATION SIGN> 215 191 191 191
359 <O WITH STROKE> 216 128 128 128
360 <U WITH GRAVE> 217 253 253 224 ###
361 <U WITH ACUTE> 218 254 254 254
362 <U WITH CIRCUMFLEX> 219 251 251 221 ###
363 <U WITH DIAERESIS> 220 252 252 252
364 <Y WITH ACUTE> 221 173 186 173 *** ###
365 <CAPITAL LETTER THORN> 222 174 174 174
366 <SMALL LETTER SHARP S> 223 89 89 89
367 <a WITH GRAVE> 224 68 68 68
368 <a WITH ACUTE> 225 69 69 69
369 <a WITH CIRCUMFLEX> 226 66 66 66
370 <a WITH TILDE> 227 70 70 70
371 <a WITH DIAERESIS> 228 67 67 67
372 <a WITH RING ABOVE> 229 71 71 71
373 <SMALL LIGATURE ae> 230 156 156 156
374 <c WITH CEDILLA> 231 72 72 72
375 <e WITH GRAVE> 232 84 84 84
376 <e WITH ACUTE> 233 81 81 81
377 <e WITH CIRCUMFLEX> 234 82 82 82
378 <e WITH DIAERESIS> 235 83 83 83
379 <i WITH GRAVE> 236 88 88 88
380 <i WITH ACUTE> 237 85 85 85
381 <i WITH CIRCUMFLEX> 238 86 86 86
382 <i WITH DIAERESIS> 239 87 87 87
383 <SMALL LETTER eth> 240 140 140 140
384 <n WITH TILDE> 241 73 73 73
385 <o WITH GRAVE> 242 205 205 205
386 <o WITH ACUTE> 243 206 206 206
387 <o WITH CIRCUMFLEX> 244 203 203 203
388 <o WITH TILDE> 245 207 207 207
389 <o WITH DIAERESIS> 246 204 204 204
390 <DIVISION SIGN> 247 225 225 225
391 <o WITH STROKE> 248 112 112 112
392 <u WITH GRAVE> 249 221 221 192 ###
393 <u WITH ACUTE> 250 222 222 222
394 <u WITH CIRCUMFLEX> 251 219 219 219
395 <u WITH DIAERESIS> 252 220 220 220
396 <y WITH ACUTE> 253 141 141 141
397 <SMALL LETTER thorn> 254 142 142 142
398 <y WITH DIAERESIS> 255 223 223 223
399
400If you would rather see the above table in CCSID 0037 order rather than
401ASCII + Latin-1 order then run the table through:
402
403=over 4
404
405=item recipe 2
406
407=back
408
51b5cecb 409 perl -ne 'if(/.{33}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}/)'\
d396a558
JH
410 -e '{push(@l,$_)}' \
411 -e 'END{print map{$_->[0]}' \
412 -e ' sort{$a->[1] <=> $b->[1]}' \
413 -e ' map{[$_,substr($_,42,3)]}@l;}' perlebcdic.pod
414
415If you would rather see it in CCSID 1047 order then change the digit
41642 in the last line to 51, like this:
417
418=over 4
419
420=item recipe 3
421
422=back
423
51b5cecb 424 perl -ne 'if(/.{33}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}/)'\
d396a558
JH
425 -e '{push(@l,$_)}' \
426 -e 'END{print map{$_->[0]}' \
427 -e ' sort{$a->[1] <=> $b->[1]}' \
428 -e ' map{[$_,substr($_,51,3)]}@l;}' perlebcdic.pod
429
430If you would rather see it in POSIX-BC order then change the digit
43151 in the last line to 60, like this:
432
433=over 4
434
435=item recipe 4
436
437=back
438
51b5cecb 439 perl -ne 'if(/.{33}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}/)'\
d396a558
JH
440 -e '{push(@l,$_)}' \
441 -e 'END{print map{$_->[0]}' \
442 -e ' sort{$a->[1] <=> $b->[1]}' \
443 -e ' map{[$_,substr($_,60,3)]}@l;}' perlebcdic.pod
444
445
446=head1 IDENTIFYING CHARACTER CODE SETS
447
448To determine the character set you are running under from perl one
449could use the return value of ord() or chr() to test one or more
450character values. For example:
451
452 $is_ascii = "A" eq chr(65);
453 $is_ebcdic = "A" eq chr(193);
454
51b5cecb 455Also, "\t" is a C<HORIZONTAL TABULATION> character so that:
d396a558
JH
456
457 $is_ascii = ord("\t") == 9;
458 $is_ebcdic = ord("\t") == 5;
459
460To distinguish EBCDIC code pages try looking at one or more of
461the characters that differ between them. For example:
462
463 $is_ebcdic_37 = "\n" eq chr(37);
464 $is_ebcdic_1047 = "\n" eq chr(21);
465
466Or better still choose a character that is uniquely encoded in any
467of the code sets, e.g.:
468
469 $is_ascii = ord('[') == 91;
470 $is_ebcdic_37 = ord('[') == 186;
471 $is_ebcdic_1047 = ord('[') == 173;
472 $is_ebcdic_POSIX_BC = ord('[') == 187;
473
474However, it would be unwise to write tests such as:
475
476 $is_ascii = "\r" ne chr(13); # WRONG
477 $is_ascii = "\n" ne chr(10); # ILL ADVISED
478
479Obviously the first of these will fail to distinguish most ASCII machines
480from either a CCSID 0037, a 1047, or a POSIX-BC EBCDIC machine since "\r" eq
481chr(13) under all of those coded character sets. But note too that
482because "\n" is chr(13) and "\r" is chr(10) on the MacIntosh (which is an
483ASCII machine) the second C<$is_ascii> test will lead to trouble there.
484
485To determine whether or not perl was built under an EBCDIC
486code page you can use the Config module like so:
487
488 use Config;
51b5cecb 489 $is_ebcdic = $Config{'ebcdic'} eq 'define';
d396a558
JH
490
491=head1 CONVERSIONS
492
493In order to convert a string of characters from one character set to
494another a simple list of numbers, such as in the right columns in the
495above table, along with perl's tr/// operator is all that is needed.
496The data in the table are in ASCII order hence the EBCDIC columns
497provide easy to use ASCII to EBCDIC operations that are also easily
498reversed.
499
500For example, to convert ASCII to code page 037 take the output of the second
501column from the output of recipe 0 and use it in tr/// like so:
502
503 $cp_037 =
504 '\000\001\002\003\234\011\206\177\227\215\216\013\014\015\016\017' .
505 '\020\021\022\023\235\205\010\207\030\031\222\217\034\035\036\037' .
506 '\200\201\202\203\204\012\027\033\210\211\212\213\214\005\006\007' .
507 '\220\221\026\223\224\225\226\004\230\231\232\233\024\025\236\032' .
508 '\040\240\342\344\340\341\343\345\347\361\242\056\074\050\053\174' .
509 '\046\351\352\353\350\355\356\357\354\337\041\044\052\051\073\254' .
510 '\055\057\302\304\300\301\303\305\307\321\246\054\045\137\076\077' .
511 '\370\311\312\313\310\315\316\317\314\140\072\043\100\047\075\042' .
512 '\330\141\142\143\144\145\146\147\150\151\253\273\360\375\376\261' .
513 '\260\152\153\154\155\156\157\160\161\162\252\272\346\270\306\244' .
514 '\265\176\163\164\165\166\167\170\171\172\241\277\320\335\336\256' .
515 '\136\243\245\267\251\247\266\274\275\276\133\135\257\250\264\327' .
516 '\173\101\102\103\104\105\106\107\110\111\255\364\366\362\363\365' .
517 '\175\112\113\114\115\116\117\120\121\122\271\373\374\371\372\377' .
518 '\134\367\123\124\125\126\127\130\131\132\262\324\326\322\323\325' .
519 '\060\061\062\063\064\065\066\067\070\071\263\333\334\331\332\237' ;
520
521 my $ebcdic_string = $ascii_string;
522 $ebcdic_string = tr/\000-\377/$cp_037/;
523
524To convert from EBCDIC to ASCII just reverse the order of the tr///
525arguments like so:
526
527 my $ascii_string = $ebcdic_string;
528 $ascii_string = tr/$code_page_chrs/\000-\037/;
529
51b5cecb 530XPG4 operability often implies the presence of an I<iconv> utility
d396a558
JH
531available from the shell or from the C library. Consult your system's
532documentation for information on iconv.
533
534On OS/390 see the iconv(1) man page. One way to invoke the iconv
535shell utility from within perl would be to:
536
51b5cecb 537 # OS/390 example
d396a558
JH
538 $ascii_data = `echo '$ebcdic_data'| iconv -f IBM-1047 -t ISO8859-1`
539
540or the inverse map:
541
51b5cecb 542 # OS/390 example
d396a558
JH
543 $ebcdic_data = `echo '$ascii_data'| iconv -f ISO8859-1 -t IBM-1047`
544
d396a558
JH
545For other perl based conversion options see the Convert::* modules on CPAN.
546
547=head1 OPERATOR DIFFERENCES
548
549The C<..> range operator treats certain character ranges with
550care on EBCDIC machines. For example the following array
551will have twenty six elements on either an EBCDIC machine
552or an ASCII machine:
553
554 @alphabet = ('A'..'Z'); # $#alphabet == 25
555
556The bitwise operators such as & ^ | may return different results
557when operating on string or character data in a perl program running
558on an EBCDIC machine than when run on an ASCII machine. Here is
559an example adapted from the one in L<perlop>:
560
561 # EBCDIC-based examples
562 print "j p \n" ^ " a h"; # prints "JAPH\n"
563 print "JA" | " ph\n"; # prints "japh\n"
564 print "JAPH\nJunk" & "\277\277\277\277\277"; # prints "japh\n";
565 print 'p N$' ^ " E<H\n"; # prints "Perl\n";
566
567An interesting property of the 32 C0 control characters
568in the ASCII table is that they can "literally" be constructed
51b5cecb
PP
569as control characters in perl, e.g. C<(chr(0) eq "\c@")>
570C<(chr(1) eq "\cA")>, and so on. Perl on EBCDIC machines has been
571ported to take "\c@" to chr(0) and "\cA" to chr(1) as well, but the
d396a558
JH
572thirty three characters that result depend on which code page you are
573using. The table below uses the character names from the previous table
51b5cecb 574but with substitutions such as s/START OF/S.O./; s/END OF /E.O./;
d396a558
JH
575s/TRANSMISSION/TRANS./; s/TABULATION/TAB./; s/VERTICAL/VERT./;
576s/HORIZONTAL/HORIZ./; s/DEVICE CONTROL/D.C./; s/SEPARATOR/SEP./;
577s/NEGATIVE ACKNOWLEDGE/NEG. ACK./;. The POSIX-BC and 1047 sets are
578identical throughout this range and differ from the 0037 set at only
51b5cecb
PP
579one spot (21 decimal). Note that the C<LINE FEED> character
580may be generated by "\cJ" on ASCII machines but by "\cU" on 1047 or POSIX-BC
581machines and cannot be generated as a C<"\c.letter."> control character on
5820037 machines. Note also that "\c\\" maps to two characters
d396a558
JH
583not one.
584
585 chr ord 8859-1 0037 1047 && POSIX-BC
586 ------------------------------------------------------------------------
587 "\c?" 127 <DELETE> " " ***><
588 "\c@" 0 <NULL> <NULL> <NULL> ***><
589 "\cA" 1 <S.O. HEADING> <S.O. HEADING> <S.O. HEADING>
590 "\cB" 2 <S.O. TEXT> <S.O. TEXT> <S.O. TEXT>
591 "\cC" 3 <E.O. TEXT> <E.O. TEXT> <E.O. TEXT>
592 "\cD" 4 <E.O. TRANS.> <C1 28> <C1 28>
593 "\cE" 5 <ENQUIRY> <HORIZ. TAB.> <HORIZ. TAB.>
594 "\cF" 6 <ACKNOWLEDGE> <C1 6> <C1 6>
595 "\cG" 7 <BELL> <DELETE> <DELETE>
596 "\cH" 8 <BACKSPACE> <C1 23> <C1 23>
597 "\cI" 9 <HORIZ. TAB.> <C1 13> <C1 13>
598 "\cJ" 10 <LINE FEED> <C1 14> <C1 14>
599 "\cK" 11 <VERT. TAB.> <VERT. TAB.> <VERT. TAB.>
600 "\cL" 12 <FORM FEED> <FORM FEED> <FORM FEED>
601 "\cM" 13 <CARRIAGE RETURN> <CARRIAGE RETURN> <CARRIAGE RETURN>
602 "\cN" 14 <SHIFT OUT> <SHIFT OUT> <SHIFT OUT>
603 "\cO" 15 <SHIFT IN> <SHIFT IN> <SHIFT IN>
604 "\cP" 16 <DATA LINK ESCAPE> <DATA LINK ESCAPE> <DATA LINK ESCAPE>
605 "\cQ" 17 <D.C. ONE> <D.C. ONE> <D.C. ONE>
606 "\cR" 18 <D.C. TWO> <D.C. TWO> <D.C. TWO>
607 "\cS" 19 <D.C. THREE> <D.C. THREE> <D.C. THREE>
608 "\cT" 20 <D.C. FOUR> <C1 29> <C1 29>
609 "\cU" 21 <NEG. ACK.> <C1 5> <LINE FEED> ***
610 "\cV" 22 <SYNCHRONOUS IDLE> <BACKSPACE> <BACKSPACE>
611 "\cW" 23 <E.O. TRANS. BLOCK> <C1 7> <C1 7>
612 "\cX" 24 <CANCEL> <CANCEL> <CANCEL>
613 "\cY" 25 <E.O. MEDIUM> <E.O. MEDIUM> <E.O. MEDIUM>
614 "\cZ" 26 <SUBSTITUTE> <C1 18> <C1 18>
615 "\c[" 27 <ESCAPE> <C1 15> <C1 15>
616 "\c\\" 28 <FILE SEP.>\ <FILE SEP.>\ <FILE SEP.>\
617 "\c]" 29 <GROUP SEP.> <GROUP SEP.> <GROUP SEP.>
618 "\c^" 30 <RECORD SEP.> <RECORD SEP.> <RECORD SEP.> ***><
619 "\c_" 31 <UNIT SEP.> <UNIT SEP.> <UNIT SEP.> ***><
620
621
622=head1 FUNCTION DIFFERENCES
623
624=over 8
625
626=item chr()
627
628chr() must be given an EBCDIC code number argument to yield a desired
629character return value on an EBCDIC machine. For example:
630
631 $CAPITAL_LETTER_A = chr(193);
632
633=item ord()
634
635ord() will return EBCDIC code number values on an EBCDIC machine.
636For example:
637
638 $the_number_193 = ord("A");
639
640=item pack()
641
642The c and C templates for pack() are dependent upon character set
643encoding. Examples of usage on EBCDIC include:
644
645 $foo = pack("CCCC",193,194,195,196);
646 # $foo eq "ABCD"
647 $foo = pack("C4",193,194,195,196);
648 # same thing
649
650 $foo = pack("ccxxcc",193,194,195,196);
651 # $foo eq "AB\0\0CD"
652
653=item print()
654
655One must be careful with scalars and strings that are passed to
656print that contain ASCII encodings. One common place
657for this to occur is in the output of the MIME type header for
658CGI script writing. For example, many perl programming guides
659recommend something similar to:
660
661 print "Content-type:\ttext/html\015\012\015\012";
662 # this may be wrong on EBCDIC
663
664Under the IBM OS/390 USS Web Server for example you should instead
665write that as:
666
667 print "Content-type:\ttext/html\r\n\r\n"; # OK for DGW et alia
668
669That is because the translation from EBCDIC to ASCII is done
670by the web server in this case (such code will not be appropriate for
671the Macintosh however). Consult your web server's documentation for
672further details.
673
674=item printf()
675
676The formats that can convert characters to numbers and vice versa
677will be different from their ASCII counterparts when executed
678on an EBCDIC machine. Examples include:
679
680 printf("%c%c%c",193,194,195); # prints ABC
681
682=item sort()
683
684EBCDIC sort results may differ from ASCII sort results especially for
685mixed case strings. This is discussed in more detail below.
686
687=item sprintf()
688
689See the discussion of printf() above. An example of the use
690of sprintf would be:
691
692 $CAPITAL_LETTER_A = sprintf("%c",193);
693
694=item unpack()
695
696See the discussion of pack() above.
697
698=back
699
700=head1 REGULAR EXPRESSION DIFFERENCES
701
702As of perl 5.005_03 the letter range regular expression such as
703[A-Z] and [a-z] have been especially coded to not pick up gap
b3b6085d
PP
704characters. For example, characters such as E<ocirc> C<o WITH CIRCUMFLEX>
705that lie between I and J would not be matched by the
51b5cecb
PP
706regular expression range C</[H-K]/>.
707
708If you do want to match the alphabet gap characters in a single octet
d396a558
JH
709regular expression try matching the hex or octal code such
710as C</\313/> on EBCDIC or C</\364/> on ASCII machines to
51b5cecb 711have your regular expression match C<o WITH CIRCUMFLEX>.
d396a558 712
51b5cecb 713Another construct to be wary of is the inappropriate use of hex or
d396a558
JH
714octal constants in regular expressions. Consider the following
715set of subs:
716
717 sub is_c0 {
718 my $char = substr(shift,0,1);
719 $char =~ /[\000-\037]/;
720 }
721
722 sub is_print_ascii {
723 my $char = substr(shift,0,1);
724 $char =~ /[\040-\176]/;
725 }
726
727 sub is_delete {
728 my $char = substr(shift,0,1);
729 $char eq "\177";
730 }
731
732 sub is_c1 {
733 my $char = substr(shift,0,1);
734 $char =~ /[\200-\237]/;
735 }
736
737 sub is_latin_1 {
738 my $char = substr(shift,0,1);
739 $char =~ /[\240-\377]/;
740 }
741
51b5cecb
PP
742The above would be adequate if the concern was only with numeric code points.
743However, the concern may be with characters rather than code points
744and on an EBCDIC machine it may be desirable for constructs such as
d396a558
JH
745C<if (is_print_ascii("A")) {print "A is a printable character\n";}> to print
746out the expected message. One way to represent the above collection
747of character classification subs that is capable of working across the
748four coded character sets discussed in this document is as follows:
749
750 sub Is_c0 {
751 my $char = substr(shift,0,1);
752 if (ord('^')==94) { # ascii
753 return $char =~ /[\000-\037]/;
754 }
755 if (ord('^')==176) { # 37
756 return $char =~ /[\000-\003\067\055-\057\026\005\045\013-\023\074\075\062\046\030\031\077\047\034-\037]/;
757 }
758 if (ord('^')==95 || ord('^')==106) { # 1047 || posix-bc
759 return $char =~ /[\000-\003\067\055-\057\026\005\025\013-\023\074\075\062\046\030\031\077\047\034-\037]/;
760 }
761 }
762
763 sub Is_print_ascii {
764 my $char = substr(shift,0,1);
765 $char =~ /[ !"\#\$%&'()*+,\-.\/0-9:;<=>?\@A-Z[\\\]^_`a-z{|}~]/;
766 }
767
768 sub Is_delete {
769 my $char = substr(shift,0,1);
770 if (ord('^')==94) { # ascii
771 return $char eq "\177";
772 }
773 else { # ebcdic
774 return $char eq "\007";
775 }
776 }
777
778 sub Is_c1 {
779 my $char = substr(shift,0,1);
780 if (ord('^')==94) { # ascii
781 return $char =~ /[\200-\237]/;
782 }
783 if (ord('^')==176) { # 37
784 return $char =~ /[\040-\044\025\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\377]/;
785 }
786 if (ord('^')==95) { # 1047
787 return $char =~ /[\040-\045\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\377]/;
788 }
789 if (ord('^')==106) { # posix-bc
790 return $char =~
791 /[\040-\045\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\137]/;
792 }
793 }
794
795 sub Is_latin_1 {
796 my $char = substr(shift,0,1);
797 if (ord('^')==94) { # ascii
798 return $char =~ /[\240-\377]/;
799 }
800 if (ord('^')==176) { # 37
801 return $char =~
802 /[\101\252\112\261\237\262\152\265\275\264\232\212\137\312\257\274\220\217\352\372\276\240\266\263\235\332\233\213\267\270\271\253\144\145\142\146\143\147\236\150\164\161-\163\170\165-\167\254\151\355\356\353\357\354\277\200\375\376\373\374\255\256\131\104\105\102\106\103\107\234\110\124\121-\123\130\125-\127\214\111\315\316\313\317\314\341\160\335\336\333\334\215\216\337]/;
803 }
804 if (ord('^')==95) { # 1047
805 return $char =~
806 /[\101\252\112\261\237\262\152\265\273\264\232\212\260\312\257\274\220\217\352\372\276\240\266\263\235\332\233\213\267\270\271\253\144\145\142\146\143\147\236\150\164\161-\163\170\165-\167\254\151\355\356\353\357\354\277\200\375\376\373\374\272\256\131\104\105\102\106\103\107\234\110\124\121-\123\130\125-\127\214\111\315\316\313\317\314\341\160\335\336\333\334\215\216\337]/;
807 }
808 if (ord('^')==106) { # posix-bc
809 return $char =~
810 /[\101\252\260\261\237\262\320\265\171\264\232\212\272\312\257\241\220\217\352\372\276\240\266\263\235\332\233\213\267\270\271\253\144\145\142\146\143\147\236\150\164\161-\163\170\165-\167\254\151\355\356\353\357\354\277\200\340\376\335\374\255\256\131\104\105\102\106\103\107\234\110\124\121-\123\130\125-\127\214\111\315\316\313\317\314\341\160\300\336\333\334\215\216\337]/;
811 }
812 }
813
814Note however that only the C<Is_ascii_print()> sub is really independent
815of coded character set. Another way to write C<Is_latin_1()> would be
816to use the characters in the range explicitly:
817
818 sub Is_latin_1 {
819 my $char = substr(shift,0,1);
820 $char =~ /[ ¡¢£¤¥¦§¨©ª«¬­®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ]/;
821 }
822
823Although that form may run into trouble in network transit (due to the
824presence of 8 bit characters) or on non ISO-Latin character sets.
d396a558
JH
825
826=head1 SOCKETS
827
828Most socket programming assumes ASCII character encodings in network
829byte order. Exceptions can include CGI script writing under a
830host web server where the server may take care of translation for you.
831Most host web servers convert EBCDIC data to ISO-8859-1 or Unicode on
832output.
833
834=head1 SORTING
835
836One big difference between ASCII based character sets and EBCDIC ones
837are the relative positions of upper and lower case letters and the
838letters compared to the digits. If sorted on an ASCII based machine the
839two letter abbreviation for a physician comes before the two letter
840for drive, that is:
841
51b5cecb
PP
842 @sorted = sort(qw(Dr. dr.)); # @sorted holds ('Dr.','dr.') on ASCII,
843 # but ('dr.','Dr.') on EBCDIC
d396a558
JH
844
845The property of lower case before uppercase letters in EBCDIC is
846even carried to the Latin 1 EBCDIC pages such as 0037 and 1047.
b3b6085d
PP
847An example would be that E<Euml> C<E WITH DIAERESIS> (203) comes
848before E<euml> C<e WITH DIAERESIS> (235) on an ASCII machine, but
51b5cecb 849the latter (83) comes before the former (115) on an EBCDIC machine.
b3b6085d 850(Astute readers will note that the upper case version of E<szlig>
51b5cecb 851C<SMALL LETTER SHARP S> is simply "SS" and that the upper case version of
b3b6085d 852E<yuml> C<y WITH DIAERESIS> is not in the 0..255 range but it is
51b5cecb 853at U+x0178 in Unicode, or C<"\x{178}"> in a Unicode enabled Perl).
d396a558
JH
854
855The sort order will cause differences between results obtained on
856ASCII machines versus EBCDIC machines. What follows are some suggestions
857on how to deal with these differences.
858
51b5cecb 859=head2 Ignore ASCII vs. EBCDIC sort differences.
d396a558
JH
860
861This is the least computationally expensive strategy. It may require
862some user education.
863
51b5cecb 864=head2 MONO CASE then sort data.
d396a558 865
51b5cecb 866In order to minimize the expense of mono casing mixed test try to
d396a558
JH
867C<tr///> towards the character set case most employed within the data.
868If the data are primarily UPPERCASE non Latin 1 then apply tr/[a-z]/[A-Z]/
869then sort(). If the data are primarily lowercase non Latin 1 then
870apply tr/[A-Z]/[a-z]/ before sorting. If the data are primarily UPPERCASE
51b5cecb
PP
871and include Latin-1 characters then apply:
872
873 tr/[a-z]/[A-Z]/;
874 tr/[àáâãäåæçèéêëìíîïðñòóôõöøùúûüýþ]/[ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞ]/;
875 s/ß/SS/g;
d396a558 876
51b5cecb 877then sort(). Do note however that such Latin-1 manipulation does not
b3b6085d
PP
878address the E<yuml> C<y WITH DIAERESIS> character that will remain at
879code point 255 on ASCII machines, but 223 on most EBCDIC machines
51b5cecb
PP
880where it will sort to a place less than the EBCDIC numerals. With a
881Unicode enabled Perl you might try:
d396a558 882
51b5cecb
PP
883 tr/^?/\x{178}/;
884
885The strategy of mono casing data before sorting does not preserve the case
886of the data and may not be acceptable for that reason.
887
888=head2 Convert, sort data, then re convert.
d396a558
JH
889
890This is the most expensive proposition that does not employ a network
891connection.
892
893=head2 Perform sorting on one type of machine only.
894
895This strategy can employ a network connection. As such
896it would be computationally expensive.
897
898=head1 URL ENCODING and DECODING
899
51b5cecb 900Note that some URLs have hexadecimal ASCII code points in them in an
d396a558
JH
901attempt to overcome character limitation issues. For example the
902tilde character is not on every keyboard hence a URL of the form:
903
904 http://www.pvhp.com/~pvhp/
905
906may also be expressed as either of:
907
908 http://www.pvhp.com/%7Epvhp/
909
910 http://www.pvhp.com/%7epvhp/
911
51b5cecb 912where 7E is the hexadecimal ASCII code point for '~'. Here is an example
d396a558
JH
913of decoding such a URL under CCSID 1047:
914
915 $url = 'http://www.pvhp.com/%7Epvhp/';
916 # this array assumes code page 1047
917 my @a2e_1047 = (
918 0, 1, 2, 3, 55, 45, 46, 47, 22, 5, 21, 11, 12, 13, 14, 15,
919 16, 17, 18, 19, 60, 61, 50, 38, 24, 25, 63, 39, 28, 29, 30, 31,
920 64, 90,127,123, 91,108, 80,125, 77, 93, 92, 78,107, 96, 75, 97,
921 240,241,242,243,244,245,246,247,248,249,122, 94, 76,126,110,111,
922 124,193,194,195,196,197,198,199,200,201,209,210,211,212,213,214,
923 215,216,217,226,227,228,229,230,231,232,233,173,224,189, 95,109,
924 121,129,130,131,132,133,134,135,136,137,145,146,147,148,149,150,
925 151,152,153,162,163,164,165,166,167,168,169,192, 79,208,161, 7,
926 32, 33, 34, 35, 36, 37, 6, 23, 40, 41, 42, 43, 44, 9, 10, 27,
927 48, 49, 26, 51, 52, 53, 54, 8, 56, 57, 58, 59, 4, 20, 62,255,
928 65,170, 74,177,159,178,106,181,187,180,154,138,176,202,175,188,
929 144,143,234,250,190,160,182,179,157,218,155,139,183,184,185,171,
930 100,101, 98,102, 99,103,158,104,116,113,114,115,120,117,118,119,
931 172,105,237,238,235,239,236,191,128,253,254,251,252,186,174, 89,
932 68, 69, 66, 70, 67, 71,156, 72, 84, 81, 82, 83, 88, 85, 86, 87,
933 140, 73,205,206,203,207,204,225,112,221,222,219,220,141,142,223
934 );
935 $url =~ s/%([0-9a-fA-F]{2})/pack("c",$a2e_1047[hex($1)])/ge;
936
937=head1 I18N AND L10N
938
939Internationalization(I18N) and localization(L10N) are supported at least
940in principle even on EBCDIC machines. The details are system dependent
941and discussed under the L<perlebcdic/OS ISSUES> section below.
942
943=head1 MULTI OCTET CHARACTER SETS
944
51b5cecb 945Multi byte EBCDIC code pages; Unicode, UTF-8, UTF-EBCDIC, XXX.
d396a558
JH
946
947=head1 OS ISSUES
948
949There may be a few system dependent issues
950of concern to EBCDIC Perl programmers.
951
952=head2 OS/400
953
51b5cecb
PP
954The PASE environment.
955
d396a558
JH
956=over 8
957
958=item IFS access
959
960XXX.
961
962=back
963
964=head2 OS/390
965
51b5cecb
PP
966Perl runs under Unix Systems Services or USS.
967
d396a558
JH
968=over 8
969
51b5cecb
PP
970=item chcp
971
972L<chcp> is supported as a shell utility for displaying and changing
973one's code page.
974
d396a558
JH
975=item dataset access
976
977For sequential data set access try:
978
979 my @ds_records = `cat //DSNAME`;
980
981or:
982
983 my @ds_records = `cat //'HLQ.DSNAME'`;
984
985See also the OS390::Stdio module on CPAN.
986
51b5cecb
PP
987=item iconv
988
989L<iconv> is supported as both a shell utility and a C RTL routine.
990
d396a558
JH
991=item locales
992
993On OS/390 see L<locale> for information on locales. The L10N files
994are in F</usr/nls/locale>. $Config{d_setlocale} is 'define' on OS/390.
995
996=back
997
998=head2 VM/ESA?
999
1000XXX.
1001
1002=head2 POSIX-BC?
1003
1004XXX.
1005
51b5cecb
PP
1006=head1 BUGS
1007
1008This pod document contains literal Latin 1 characters and may encounter
b1866b2d 1009translation difficulties. In particular one popular nroff implementation
51b5cecb
PP
1010was known to strip accented characters to their unaccented counterparts
1011while attempting to view this document through the B<pod2man> program
1012(for example, you may see a plain C<y> rather than one with a diaeresis
b3b6085d
PP
1013as in E<yuml>). Another nroff truncated the resultant man page at
1014the first occurence of 8 bit characters.
51b5cecb
PP
1015
1016Not all shells will allow multiple C<-e> string arguments to perl to
1017be concatenated together properly as recipes 2, 3, and 4 might seem
1018to imply.
1019
1020Perl does not yet work with any Unicode features on EBCDIC platforms.
1021
b3b6085d
PP
1022=head1 SEE ALSO
1023
1024L<perllocale>, L<perlfunc>.
1025
d396a558
JH
1026=head1 REFERENCES
1027
1028http://anubis.dkuug.dk/i18n/charmaps
1029
d396a558
JH
1030http://www.unicode.org/
1031
1032http://www.unicode.org/unicode/reports/tr16/
1033
51b5cecb
PP
1034http://www.wps.com/texts/codes/
1035B<ASCII: American Standard Code for Information Infiltration> Tom Jennings,
1036September 1999.
1037
d396a558
JH
1038B<The Unicode Standard Version 2.0> The Unicode Consortium,
1039ISBN 0-201-48345-9, Addison Wesley Developers Press, July 1996.
1040
51b5cecb
PP
1041B<The Unicode Standard Version 3.0> The Unicode Consortium, Lisa Moore ed.,
1042ISBN 0-201-61633-5, Addison Wesley Developers Press, February 2000.
1043
d396a558
JH
1044B<CDRA: IBM - Character Data Representation Architecture -
1045Reference and Registry>, IBM SC09-2190-00, December 1996.
1046
1047"Demystifying Character Sets", Andrea Vine, Multilingual Computing
1048& Technology, B<#26 Vol. 10 Issue 4>, August/September 1999;
1049ISSN 1523-0309; Multilingual Computing Inc. Sandpoint ID, USA.
1050
1051=head1 AUTHOR
1052
b3b6085d 1053Peter Prymmer pvhp@best.com wrote this in 1999 and 2000
d396a558 1054with CCSID 0819 and 0037 help from Chris Leach and
b3b6085d
PP
1055AndrE<eacute> Pirard A.Pirard@ulg.ac.be as well as POSIX-BC
1056help from Thomas Dorner Thomas.Dorner@start.de.
d396a558
JH
1057Thanks also to Philip Newton and Vickie Cooper. Trademarks, registered
1058trademarks, service marks and registered service marks used in this
1059document are the property of their respective owners.
1060
1061