This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
OS/400 PASE port
[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,
395f5a0c
PK
9internationalization, or multi byte character set issues other
10than some discussion of UTF-8 and UTF-EBCDIC.
d396a558
JH
11
12Portions that are still incomplete are marked with XXX.
13
14=head1 COMMON CHARACTER CODE SETS
15
16=head2 ASCII
17
18The American Standard Code for Information Interchange is a set of
19integers running from 0 to 127 (decimal) that imply character
20interpretation by the display and other system(s) of computers.
51b5cecb 21The range 0..127 can be covered by setting the bits in a 7-bit binary
d396a558 22digit, hence the set is sometimes referred to as a "7-bit ASCII".
51b5cecb 23ASCII was described by the American National Standards Institute
d396a558
JH
24document ANSI X3.4-1986. It was also described by ISO 646:1991
25(with localization for currency symbols). The full ASCII set is
26given in the table below as the first 128 elements. Languages that
27can be written adequately with the characters in ASCII include
28English, Hawaiian, Indonesian, Swahili and some Native American
29languages.
30
51b5cecb
PP
31There are many character sets that extend the range of integers
32from 0..2**7-1 up to 2**8-1, or 8 bit bytes (octets if you prefer).
33One common one is the ISO 8859-1 character set.
34
d396a558
JH
35=head2 ISO 8859
36
37The ISO 8859-$n are a collection of character code sets from the
38International Organization for Standardization (ISO) each of which
39adds characters to the ASCII set that are typically found in European
40languages many of which are based on the Roman, or Latin, alphabet.
41
42=head2 Latin 1 (ISO 8859-1)
43
44A particular 8-bit extension to ASCII that includes grave and acute
45accented Latin characters. Languages that can employ ISO 8859-1
46include all the languages covered by ASCII as well as Afrikaans,
47Albanian, Basque, Catalan, Danish, Faroese, Finnish, Norwegian,
3958b146 48Portuguese, Spanish, and Swedish. Dutch is covered albeit without
d396a558
JH
49the ij ligature. French is covered too but without the oe ligature.
50German can use ISO 8859-1 but must do so without German-style
51quotation marks. This set is based on Western European extensions
52to ASCII and is commonly encountered in world wide web work.
53In IBM character code set identification terminology ISO 8859-1 is
51b5cecb 54also known as CCSID 819 (or sometimes 0819 or even 00819).
d396a558
JH
55
56=head2 EBCDIC
57
395f5a0c 58The Extended Binary Coded Decimal Interchange Code refers to a
51b5cecb
PP
59large collection of slightly different single and multi byte
60coded character sets that are different from ASCII or ISO 8859-1
61and typically run on host computers. The EBCDIC encodings derive
62from 8 bit byte extensions of Hollerith punched card encodings.
d396a558
JH
63The layout on the cards was such that high bits were set for the
64upper and lower case alphabet characters [a-z] and [A-Z], but there
65were gaps within each latin alphabet range.
66
51b5cecb
PP
67Some IBM EBCDIC character sets may be known by character code set
68identification numbers (CCSID numbers) or code page numbers. Leading
69zero digits in CCSID numbers within this document are insignificant.
70E.g. CCSID 0037 may be referred to as 37 in places.
71
1e054b24
PP
72=head2 13 variant characters
73
51b5cecb
PP
74Among IBM EBCDIC character code sets there are 13 characters that
75are often mapped to different integer values. Those characters
76are known as the 13 "variant" characters and are:
d396a558 77
51b5cecb 78 \ [ ] { } ^ ~ ! # | $ @ `
d396a558
JH
79
80=head2 0037
81
82Character code set ID 0037 is a mapping of the ASCII plus Latin-1
83characters (i.e. ISO 8859-1) to an EBCDIC set. 0037 is used
51b5cecb
PP
84in North American English locales on the OS/400 operating system
85that runs on AS/400 computers. CCSID 37 differs from ISO 8859-1
86in 237 places, in other words they agree on only 19 code point values.
d396a558
JH
87
88=head2 1047
89
90Character code set ID 1047 is also a mapping of the ASCII plus
91Latin-1 characters (i.e. ISO 8859-1) to an EBCDIC set. 1047 is
395f5a0c
PK
92used under Unix System Services for OS/390 or z/OS, and OpenEdition
93for VM/ESA. CCSID 1047 differs from CCSID 0037 in eight places.
d396a558
JH
94
95=head2 POSIX-BC
96
97The EBCDIC code page in use on Siemens' BS2000 system is distinct from
981047 and 0037. It is identified below as the POSIX-BC set.
99
64c66fb6
JH
100=head2 Unicode code points versus EBCDIC code points
101
102In Unicode terminology a I<code point> is the number assigned to a
103character: for example, in EBCDIC the character "A" is usually assigned
104the number 193. In Unicode the character "A" is assigned the number 65.
105This causes a problem with the semantics of the pack/unpack "U", which
106are supposed to pack Unicode code points to characters and back to numbers.
107The problem is: which code points to use for code points less than 256?
108(for 256 and over there's no problem: Unicode code points are used)
109In EBCDIC, for the low 256 the EBCDIC code points are used. This
110means that the equivalences
111
112 pack("U", ord($character)) eq $character
113 unpack("U", $character) == ord $character
114
115will hold. (If Unicode code points were applied consistently over
116all the possible code points, pack("U",ord("A")) would in EBCDIC
117equal I<A with acute> or chr(101), and unpack("U", "A") would equal
11865, or I<non-breaking space>, not 193, or ord "A".)
119
dc4af4bb
JH
120=head2 Remaining Perl Unicode problems in EBCDIC
121
122=over 4
123
124=item *
125
126Many of the remaining seem to be related to case-insensitive matching:
127for example, C<< /[\x{131}]/ >> (LATIN SMALL LETTER DOTLESS I) does
128not match "I" case-insensitively, as it should under Unicode.
129(The match succeeds in ASCII-derived platforms.)
130
131=item *
132
133The extensions Unicode::Collate and Unicode::Normalized are not
134supported under EBCDIC, likewise for the encoding pragma.
135
136=back
137
395f5a0c
PK
138=head2 Unicode and UTF
139
140UTF is a Unicode Transformation Format. UTF-8 is a Unicode conforming
141representation of the Unicode standard that looks very much like ASCII.
142UTF-EBCDIC is an attempt to represent Unicode characters in an EBCDIC
143transparent manner.
144
8704cfd1 145=head2 Using Encode
8f94de01
JH
146
147Starting from Perl 5.8 you can use the standard new module Encode
148to translate from EBCDIC to Latin-1 code points
149
150 use Encode 'from_to';
151
152 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' );
153
154 # $a is in EBCDIC code points
155 from_to($a, $ebcdic{ord '^'}, 'latin1');
156 # $a is ISO 8859-1 code points
157
158and from Latin-1 code points to EBCDIC code points
159
160 use Encode 'from_to';
161
162 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' );
163
164 # $a is ISO 8859-1 code points
165 from_to($a, 'latin1', $ebcdic{ord '^'});
166 # $a is in EBCDIC code points
167
168For doing I/O it is suggested that you use the autotranslating features
169of PerlIO, see L<perluniintro>.
170
d396a558
JH
171=head1 SINGLE OCTET TABLES
172
173The following tables list the ASCII and Latin 1 ordered sets including
174the subsets: C0 controls (0..31), ASCII graphics (32..7e), delete (7f),
175C1 controls (80..9f), and Latin-1 (a.k.a. ISO 8859-1) (a0..ff). In the
176table non-printing control character names as well as the Latin 1
177extensions to ASCII have been labelled with character names roughly
395f5a0c 178corresponding to I<The Unicode Standard, Version 3.0> albeit with
d396a558
JH
179substitutions such as s/LATIN// and s/VULGAR// in all cases,
180s/CAPITAL LETTER// in some cases, and s/SMALL LETTER ([A-Z])/\l$1/
1e054b24
PP
181in some other cases (the C<charnames> pragma names unfortunately do
182not list explicit names for the C0 or C1 control characters). The
183"names" of the C1 control set (128..159 in ISO 8859-1) listed here are
184somewhat arbitrary. The differences between the 0037 and 1047 sets are
185flagged with ***. The differences between the 1047 and POSIX-BC sets
186are flagged with ###. All ord() numbers listed are decimal. If you
187would rather see this table listing octal values then run the table
188(that is, the pod version of this document since this recipe may not
189work with a pod2_other_format translation) through:
d396a558
JH
190
191=over 4
192
193=item recipe 0
194
195=back
196
84f709e7
JH
197 perl -ne 'if(/(.{33})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
198 -e '{printf("%s%-9o%-9o%-9o%o\n",$1,$2,$3,$4,$5)}' perlebcdic.pod
395f5a0c
PK
199
200If you want to retain the UTF-x code points then in script form you
201might want to write:
202
203=over 4
204
205=item recipe 1
206
207=back
208
84f709e7
JH
209 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!";
210 while (<FH>) {
211 if (/(.{33})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*)\s+(\d+)\.?(\d*)/) {
212 if ($7 ne '' && $9 ne '') {
213 printf("%s%-9o%-9o%-9o%-9o%-3o.%-5o%-3o.%o\n",$1,$2,$3,$4,$5,$6,$7,$8,$9);
214 }
215 elsif ($7 ne '') {
216 printf("%s%-9o%-9o%-9o%-9o%-3o.%-5o%o\n",$1,$2,$3,$4,$5,$6,$7,$8);
217 }
218 else {
219 printf("%s%-9o%-9o%-9o%-9o%-9o%o\n",$1,$2,$3,$4,$5,$6,$8);
395f5a0c
PK
220 }
221 }
222 }
d396a558
JH
223
224If you would rather see this table listing hexadecimal values then
225run the table through:
226
227=over 4
228
395f5a0c 229=item recipe 2
d396a558
JH
230
231=back
232
84f709e7
JH
233 perl -ne 'if(/(.{33})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
234 -e '{printf("%s%-9X%-9X%-9X%X\n",$1,$2,$3,$4,$5)}' perlebcdic.pod
395f5a0c
PK
235
236Or, in order to retain the UTF-x code points in hexadecimal:
237
238=over 4
239
240=item recipe 3
241
242=back
243
84f709e7 244 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!";
395f5a0c 245 while (<FH>) {
84f709e7
JH
246 if (/(.{33})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*)\s+(\d+)\.?(\d*)/) {
247 if ($7 ne '' && $9 ne '') {
248 printf("%s%-9X%-9X%-9X%-9X%-2X.%-6X%-2X.%X\n",$1,$2,$3,$4,$5,$6,$7,$8,$9);
395f5a0c 249 }
84f709e7
JH
250 elsif ($7 ne '') {
251 printf("%s%-9X%-9X%-9X%-9X%-2X.%-6X%X\n",$1,$2,$3,$4,$5,$6,$7,$8);
395f5a0c
PK
252 }
253 else {
84f709e7 254 printf("%s%-9X%-9X%-9X%-9X%-9X%X\n",$1,$2,$3,$4,$5,$6,$8);
395f5a0c
PK
255 }
256 }
257 }
258
259
260 incomp- incomp-
261 8859-1 lete lete
262 chr 0819 0037 1047 POSIX-BC UTF-8 UTF-EBCDIC
263 ------------------------------------------------------------------------------------
264 <NULL> 0 0 0 0 0 0
265 <START OF HEADING> 1 1 1 1 1 1
266 <START OF TEXT> 2 2 2 2 2 2
267 <END OF TEXT> 3 3 3 3 3 3
268 <END OF TRANSMISSION> 4 55 55 55 4 55
269 <ENQUIRY> 5 45 45 45 5 45
270 <ACKNOWLEDGE> 6 46 46 46 6 46
271 <BELL> 7 47 47 47 7 47
272 <BACKSPACE> 8 22 22 22 8 22
273 <HORIZONTAL TABULATION> 9 5 5 5 9 5
274 <LINE FEED> 10 37 21 21 10 21 ***
275 <VERTICAL TABULATION> 11 11 11 11 11 11
276 <FORM FEED> 12 12 12 12 12 12
277 <CARRIAGE RETURN> 13 13 13 13 13 13
278 <SHIFT OUT> 14 14 14 14 14 14
279 <SHIFT IN> 15 15 15 15 15 15
280 <DATA LINK ESCAPE> 16 16 16 16 16 16
281 <DEVICE CONTROL ONE> 17 17 17 17 17 17
282 <DEVICE CONTROL TWO> 18 18 18 18 18 18
283 <DEVICE CONTROL THREE> 19 19 19 19 19 19
284 <DEVICE CONTROL FOUR> 20 60 60 60 20 60
285 <NEGATIVE ACKNOWLEDGE> 21 61 61 61 21 61
286 <SYNCHRONOUS IDLE> 22 50 50 50 22 50
287 <END OF TRANSMISSION BLOCK> 23 38 38 38 23 38
288 <CANCEL> 24 24 24 24 24 24
289 <END OF MEDIUM> 25 25 25 25 25 25
290 <SUBSTITUTE> 26 63 63 63 26 63
291 <ESCAPE> 27 39 39 39 27 39
292 <FILE SEPARATOR> 28 28 28 28 28 28
293 <GROUP SEPARATOR> 29 29 29 29 29 29
294 <RECORD SEPARATOR> 30 30 30 30 30 30
295 <UNIT SEPARATOR> 31 31 31 31 31 31
296 <SPACE> 32 64 64 64 32 64
297 ! 33 90 90 90 33 90
298 " 34 127 127 127 34 127
299 # 35 123 123 123 35 123
300 $ 36 91 91 91 36 91
301 % 37 108 108 108 37 108
302 & 38 80 80 80 38 80
303 ' 39 125 125 125 39 125
304 ( 40 77 77 77 40 77
305 ) 41 93 93 93 41 93
306 * 42 92 92 92 42 92
307 + 43 78 78 78 43 78
308 , 44 107 107 107 44 107
309 - 45 96 96 96 45 96
310 . 46 75 75 75 46 75
311 / 47 97 97 97 47 97
312 0 48 240 240 240 48 240
313 1 49 241 241 241 49 241
314 2 50 242 242 242 50 242
315 3 51 243 243 243 51 243
316 4 52 244 244 244 52 244
317 5 53 245 245 245 53 245
318 6 54 246 246 246 54 246
319 7 55 247 247 247 55 247
320 8 56 248 248 248 56 248
321 9 57 249 249 249 57 249
322 : 58 122 122 122 58 122
323 ; 59 94 94 94 59 94
324 < 60 76 76 76 60 76
325 = 61 126 126 126 61 126
326 > 62 110 110 110 62 110
327 ? 63 111 111 111 63 111
328 @ 64 124 124 124 64 124
329 A 65 193 193 193 65 193
330 B 66 194 194 194 66 194
331 C 67 195 195 195 67 195
332 D 68 196 196 196 68 196
333 E 69 197 197 197 69 197
334 F 70 198 198 198 70 198
335 G 71 199 199 199 71 199
336 H 72 200 200 200 72 200
337 I 73 201 201 201 73 201
338 J 74 209 209 209 74 209
339 K 75 210 210 210 75 210
340 L 76 211 211 211 76 211
341 M 77 212 212 212 77 212
342 N 78 213 213 213 78 213
343 O 79 214 214 214 79 214
344 P 80 215 215 215 80 215
345 Q 81 216 216 216 81 216
346 R 82 217 217 217 82 217
347 S 83 226 226 226 83 226
348 T 84 227 227 227 84 227
349 U 85 228 228 228 85 228
350 V 86 229 229 229 86 229
351 W 87 230 230 230 87 230
352 X 88 231 231 231 88 231
353 Y 89 232 232 232 89 232
354 Z 90 233 233 233 90 233
355 [ 91 186 173 187 91 173 *** ###
356 \ 92 224 224 188 92 224 ###
357 ] 93 187 189 189 93 189 ***
358 ^ 94 176 95 106 94 95 *** ###
359 _ 95 109 109 109 95 109
360 ` 96 121 121 74 96 121 ###
361 a 97 129 129 129 97 129
362 b 98 130 130 130 98 130
363 c 99 131 131 131 99 131
364 d 100 132 132 132 100 132
365 e 101 133 133 133 101 133
366 f 102 134 134 134 102 134
367 g 103 135 135 135 103 135
368 h 104 136 136 136 104 136
369 i 105 137 137 137 105 137
370 j 106 145 145 145 106 145
371 k 107 146 146 146 107 146
372 l 108 147 147 147 108 147
373 m 109 148 148 148 109 148
374 n 110 149 149 149 110 149
375 o 111 150 150 150 111 150
376 p 112 151 151 151 112 151
377 q 113 152 152 152 113 152
378 r 114 153 153 153 114 153
379 s 115 162 162 162 115 162
380 t 116 163 163 163 116 163
381 u 117 164 164 164 117 164
382 v 118 165 165 165 118 165
383 w 119 166 166 166 119 166
384 x 120 167 167 167 120 167
385 y 121 168 168 168 121 168
386 z 122 169 169 169 122 169
387 { 123 192 192 251 123 192 ###
388 | 124 79 79 79 124 79
389 } 125 208 208 253 125 208 ###
390 ~ 126 161 161 255 126 161 ###
391 <DELETE> 127 7 7 7 127 7
392 <C1 0> 128 32 32 32 194.128 32
393 <C1 1> 129 33 33 33 194.129 33
394 <C1 2> 130 34 34 34 194.130 34
395 <C1 3> 131 35 35 35 194.131 35
396 <C1 4> 132 36 36 36 194.132 36
397 <C1 5> 133 21 37 37 194.133 37 ***
398 <C1 6> 134 6 6 6 194.134 6
399 <C1 7> 135 23 23 23 194.135 23
400 <C1 8> 136 40 40 40 194.136 40
401 <C1 9> 137 41 41 41 194.137 41
402 <C1 10> 138 42 42 42 194.138 42
403 <C1 11> 139 43 43 43 194.139 43
404 <C1 12> 140 44 44 44 194.140 44
405 <C1 13> 141 9 9 9 194.141 9
406 <C1 14> 142 10 10 10 194.142 10
407 <C1 15> 143 27 27 27 194.143 27
408 <C1 16> 144 48 48 48 194.144 48
409 <C1 17> 145 49 49 49 194.145 49
410 <C1 18> 146 26 26 26 194.146 26
411 <C1 19> 147 51 51 51 194.147 51
412 <C1 20> 148 52 52 52 194.148 52
413 <C1 21> 149 53 53 53 194.149 53
414 <C1 22> 150 54 54 54 194.150 54
415 <C1 23> 151 8 8 8 194.151 8
416 <C1 24> 152 56 56 56 194.152 56
417 <C1 25> 153 57 57 57 194.153 57
418 <C1 26> 154 58 58 58 194.154 58
419 <C1 27> 155 59 59 59 194.155 59
420 <C1 28> 156 4 4 4 194.156 4
421 <C1 29> 157 20 20 20 194.157 20
422 <C1 30> 158 62 62 62 194.158 62
423 <C1 31> 159 255 255 95 194.159 255 ###
424 <NON-BREAKING SPACE> 160 65 65 65 194.160 128.65
425 <INVERTED EXCLAMATION MARK> 161 170 170 170 194.161 128.66
426 <CENT SIGN> 162 74 74 176 194.162 128.67 ###
427 <POUND SIGN> 163 177 177 177 194.163 128.68
428 <CURRENCY SIGN> 164 159 159 159 194.164 128.69
429 <YEN SIGN> 165 178 178 178 194.165 128.70
430 <BROKEN BAR> 166 106 106 208 194.166 128.71 ###
431 <SECTION SIGN> 167 181 181 181 194.167 128.72
432 <DIAERESIS> 168 189 187 121 194.168 128.73 *** ###
433 <COPYRIGHT SIGN> 169 180 180 180 194.169 128.74
434 <FEMININE ORDINAL INDICATOR> 170 154 154 154 194.170 128.81
435 <LEFT POINTING GUILLEMET> 171 138 138 138 194.171 128.82
436 <NOT SIGN> 172 95 176 186 194.172 128.83 *** ###
437 <SOFT HYPHEN> 173 202 202 202 194.173 128.84
438 <REGISTERED TRADE MARK SIGN> 174 175 175 175 194.174 128.85
439 <MACRON> 175 188 188 161 194.175 128.86 ###
440 <DEGREE SIGN> 176 144 144 144 194.176 128.87
441 <PLUS-OR-MINUS SIGN> 177 143 143 143 194.177 128.88
442 <SUPERSCRIPT TWO> 178 234 234 234 194.178 128.89
443 <SUPERSCRIPT THREE> 179 250 250 250 194.179 128.98
444 <ACUTE ACCENT> 180 190 190 190 194.180 128.99
445 <MICRO SIGN> 181 160 160 160 194.181 128.100
446 <PARAGRAPH SIGN> 182 182 182 182 194.182 128.101
447 <MIDDLE DOT> 183 179 179 179 194.183 128.102
448 <CEDILLA> 184 157 157 157 194.184 128.103
449 <SUPERSCRIPT ONE> 185 218 218 218 194.185 128.104
450 <MASC. ORDINAL INDICATOR> 186 155 155 155 194.186 128.105
451 <RIGHT POINTING GUILLEMET> 187 139 139 139 194.187 128.106
452 <FRACTION ONE QUARTER> 188 183 183 183 194.188 128.112
453 <FRACTION ONE HALF> 189 184 184 184 194.189 128.113
454 <FRACTION THREE QUARTERS> 190 185 185 185 194.190 128.114
455 <INVERTED QUESTION MARK> 191 171 171 171 194.191 128.115
456 <A WITH GRAVE> 192 100 100 100 195.128 138.65
457 <A WITH ACUTE> 193 101 101 101 195.129 138.66
458 <A WITH CIRCUMFLEX> 194 98 98 98 195.130 138.67
459 <A WITH TILDE> 195 102 102 102 195.131 138.68
460 <A WITH DIAERESIS> 196 99 99 99 195.132 138.69
461 <A WITH RING ABOVE> 197 103 103 103 195.133 138.70
462 <CAPITAL LIGATURE AE> 198 158 158 158 195.134 138.71
463 <C WITH CEDILLA> 199 104 104 104 195.135 138.72
464 <E WITH GRAVE> 200 116 116 116 195.136 138.73
465 <E WITH ACUTE> 201 113 113 113 195.137 138.74
466 <E WITH CIRCUMFLEX> 202 114 114 114 195.138 138.81
467 <E WITH DIAERESIS> 203 115 115 115 195.139 138.82
468 <I WITH GRAVE> 204 120 120 120 195.140 138.83
469 <I WITH ACUTE> 205 117 117 117 195.141 138.84
470 <I WITH CIRCUMFLEX> 206 118 118 118 195.142 138.85
471 <I WITH DIAERESIS> 207 119 119 119 195.143 138.86
472 <CAPITAL LETTER ETH> 208 172 172 172 195.144 138.87
473 <N WITH TILDE> 209 105 105 105 195.145 138.88
474 <O WITH GRAVE> 210 237 237 237 195.146 138.89
475 <O WITH ACUTE> 211 238 238 238 195.147 138.98
476 <O WITH CIRCUMFLEX> 212 235 235 235 195.148 138.99
477 <O WITH TILDE> 213 239 239 239 195.149 138.100
478 <O WITH DIAERESIS> 214 236 236 236 195.150 138.101
479 <MULTIPLICATION SIGN> 215 191 191 191 195.151 138.102
480 <O WITH STROKE> 216 128 128 128 195.152 138.103
481 <U WITH GRAVE> 217 253 253 224 195.153 138.104 ###
482 <U WITH ACUTE> 218 254 254 254 195.154 138.105
483 <U WITH CIRCUMFLEX> 219 251 251 221 195.155 138.106 ###
484 <U WITH DIAERESIS> 220 252 252 252 195.156 138.112
485 <Y WITH ACUTE> 221 173 186 173 195.157 138.113 *** ###
486 <CAPITAL LETTER THORN> 222 174 174 174 195.158 138.114
487 <SMALL LETTER SHARP S> 223 89 89 89 195.159 138.115
488 <a WITH GRAVE> 224 68 68 68 195.160 139.65
489 <a WITH ACUTE> 225 69 69 69 195.161 139.66
490 <a WITH CIRCUMFLEX> 226 66 66 66 195.162 139.67
491 <a WITH TILDE> 227 70 70 70 195.163 139.68
492 <a WITH DIAERESIS> 228 67 67 67 195.164 139.69
493 <a WITH RING ABOVE> 229 71 71 71 195.165 139.70
494 <SMALL LIGATURE ae> 230 156 156 156 195.166 139.71
495 <c WITH CEDILLA> 231 72 72 72 195.167 139.72
496 <e WITH GRAVE> 232 84 84 84 195.168 139.73
497 <e WITH ACUTE> 233 81 81 81 195.169 139.74
498 <e WITH CIRCUMFLEX> 234 82 82 82 195.170 139.81
499 <e WITH DIAERESIS> 235 83 83 83 195.171 139.82
500 <i WITH GRAVE> 236 88 88 88 195.172 139.83
501 <i WITH ACUTE> 237 85 85 85 195.173 139.84
502 <i WITH CIRCUMFLEX> 238 86 86 86 195.174 139.85
503 <i WITH DIAERESIS> 239 87 87 87 195.175 139.86
504 <SMALL LETTER eth> 240 140 140 140 195.176 139.87
505 <n WITH TILDE> 241 73 73 73 195.177 139.88
506 <o WITH GRAVE> 242 205 205 205 195.178 139.89
507 <o WITH ACUTE> 243 206 206 206 195.179 139.98
508 <o WITH CIRCUMFLEX> 244 203 203 203 195.180 139.99
509 <o WITH TILDE> 245 207 207 207 195.181 139.100
510 <o WITH DIAERESIS> 246 204 204 204 195.182 139.101
511 <DIVISION SIGN> 247 225 225 225 195.183 139.102
512 <o WITH STROKE> 248 112 112 112 195.184 139.103
513 <u WITH GRAVE> 249 221 221 192 195.185 139.104 ###
514 <u WITH ACUTE> 250 222 222 222 195.186 139.105
515 <u WITH CIRCUMFLEX> 251 219 219 219 195.187 139.106
516 <u WITH DIAERESIS> 252 220 220 220 195.188 139.112
517 <y WITH ACUTE> 253 141 141 141 195.189 139.113
518 <SMALL LETTER thorn> 254 142 142 142 195.190 139.114
519 <y WITH DIAERESIS> 255 223 223 223 195.191 139.115
d396a558
JH
520
521If you would rather see the above table in CCSID 0037 order rather than
522ASCII + Latin-1 order then run the table through:
523
524=over 4
525
395f5a0c 526=item recipe 4
d396a558
JH
527
528=back
529
84f709e7
JH
530 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}/)'\
531 -e '{push(@l,$_)}' \
532 -e 'END{print map{$_->[0]}' \
533 -e ' sort{$a->[1] <=> $b->[1]}' \
534 -e ' map{[$_,substr($_,42,3)]}@l;}' perlebcdic.pod
d396a558
JH
535
536If you would rather see it in CCSID 1047 order then change the digit
53742 in the last line to 51, like this:
538
539=over 4
540
395f5a0c 541=item recipe 5
d396a558
JH
542
543=back
544
84f709e7
JH
545 perl -ne 'if(/.{33}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}/)'\
546 -e '{push(@l,$_)}' \
547 -e 'END{print map{$_->[0]}' \
548 -e ' sort{$a->[1] <=> $b->[1]}' \
549 -e ' map{[$_,substr($_,51,3)]}@l;}' perlebcdic.pod
d396a558
JH
550
551If you would rather see it in POSIX-BC order then change the digit
55251 in the last line to 60, like this:
553
554=over 4
555
395f5a0c 556=item recipe 6
d396a558
JH
557
558=back
559
84f709e7
JH
560 perl -ne 'if(/.{33}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}\s{6,8}\d{1,3}/)'\
561 -e '{push(@l,$_)}' \
562 -e 'END{print map{$_->[0]}' \
563 -e ' sort{$a->[1] <=> $b->[1]}' \
564 -e ' map{[$_,substr($_,60,3)]}@l;}' perlebcdic.pod
d396a558
JH
565
566
567=head1 IDENTIFYING CHARACTER CODE SETS
568
569To determine the character set you are running under from perl one
570could use the return value of ord() or chr() to test one or more
571character values. For example:
572
84f709e7
JH
573 $is_ascii = "A" eq chr(65);
574 $is_ebcdic = "A" eq chr(193);
d396a558 575
51b5cecb 576Also, "\t" is a C<HORIZONTAL TABULATION> character so that:
d396a558 577
84f709e7
JH
578 $is_ascii = ord("\t") == 9;
579 $is_ebcdic = ord("\t") == 5;
d396a558
JH
580
581To distinguish EBCDIC code pages try looking at one or more of
582the characters that differ between them. For example:
583
84f709e7
JH
584 $is_ebcdic_37 = "\n" eq chr(37);
585 $is_ebcdic_1047 = "\n" eq chr(21);
d396a558
JH
586
587Or better still choose a character that is uniquely encoded in any
588of the code sets, e.g.:
589
84f709e7
JH
590 $is_ascii = ord('[') == 91;
591 $is_ebcdic_37 = ord('[') == 186;
592 $is_ebcdic_1047 = ord('[') == 173;
593 $is_ebcdic_POSIX_BC = ord('[') == 187;
d396a558
JH
594
595However, it would be unwise to write tests such as:
596
84f709e7
JH
597 $is_ascii = "\r" ne chr(13); # WRONG
598 $is_ascii = "\n" ne chr(10); # ILL ADVISED
d396a558
JH
599
600Obviously the first of these will fail to distinguish most ASCII machines
84f709e7
JH
601from either a CCSID 0037, a 1047, or a POSIX-BC EBCDIC machine since "\r" eq
602chr(13) under all of those coded character sets. But note too that
603because "\n" is chr(13) and "\r" is chr(10) on the MacIntosh (which is an
d396a558
JH
604ASCII machine) the second C<$is_ascii> test will lead to trouble there.
605
84f709e7 606To determine whether or not perl was built under an EBCDIC
d396a558
JH
607code page you can use the Config module like so:
608
609 use Config;
84f709e7 610 $is_ebcdic = $Config{'ebcdic'} eq 'define';
d396a558
JH
611
612=head1 CONVERSIONS
613
1e054b24
PP
614=head2 tr///
615
d396a558
JH
616In order to convert a string of characters from one character set to
617another a simple list of numbers, such as in the right columns in the
618above table, along with perl's tr/// operator is all that is needed.
619The data in the table are in ASCII order hence the EBCDIC columns
620provide easy to use ASCII to EBCDIC operations that are also easily
621reversed.
622
84f709e7
JH
623For example, to convert ASCII to code page 037 take the output of the second
624column from the output of recipe 0 (modified to add \\ characters) and use
d5d9880c 625it in tr/// like so:
d396a558 626
84f709e7
JH
627 $cp_037 =
628 '\000\001\002\003\234\011\206\177\227\215\216\013\014\015\016\017' .
629 '\020\021\022\023\235\205\010\207\030\031\222\217\034\035\036\037' .
630 '\200\201\202\203\204\012\027\033\210\211\212\213\214\005\006\007' .
631 '\220\221\026\223\224\225\226\004\230\231\232\233\024\025\236\032' .
632 '\040\240\342\344\340\341\343\345\347\361\242\056\074\050\053\174' .
633 '\046\351\352\353\350\355\356\357\354\337\041\044\052\051\073\254' .
634 '\055\057\302\304\300\301\303\305\307\321\246\054\045\137\076\077' .
635 '\370\311\312\313\310\315\316\317\314\140\072\043\100\047\075\042' .
636 '\330\141\142\143\144\145\146\147\150\151\253\273\360\375\376\261' .
637 '\260\152\153\154\155\156\157\160\161\162\252\272\346\270\306\244' .
638 '\265\176\163\164\165\166\167\170\171\172\241\277\320\335\336\256' .
639 '\136\243\245\267\251\247\266\274\275\276\133\135\257\250\264\327' .
640 '\173\101\102\103\104\105\106\107\110\111\255\364\366\362\363\365' .
641 '\175\112\113\114\115\116\117\120\121\122\271\373\374\371\372\377' .
642 '\134\367\123\124\125\126\127\130\131\132\262\324\326\322\323\325' .
643 '\060\061\062\063\064\065\066\067\070\071\263\333\334\331\332\237' ;
d396a558
JH
644
645 my $ebcdic_string = $ascii_string;
1e054b24 646 eval '$ebcdic_string =~ tr/\000-\377/' . $cp_037 . '/';
d396a558 647
d5d9880c 648To convert from EBCDIC 037 to ASCII just reverse the order of the tr///
d396a558
JH
649arguments like so:
650
651 my $ascii_string = $ebcdic_string;
d5d9880c
PK
652 eval '$ascii_string = tr/' . $cp_037 . '/\000-\377/';
653
654Similarly one could take the output of the third column from recipe 0 to
655obtain a C<$cp_1047> table. The fourth column of the output from recipe
6560 could provide a C<$cp_posix_bc> table suitable for transcoding as well.
1e054b24
PP
657
658=head2 iconv
d396a558 659
d5d9880c 660XPG operability often implies the presence of an I<iconv> utility
d396a558
JH
661available from the shell or from the C library. Consult your system's
662documentation for information on iconv.
663
3958b146 664On OS/390 or z/OS see the iconv(1) manpage. One way to invoke the iconv
d396a558
JH
665shell utility from within perl would be to:
666
395f5a0c 667 # OS/390 or z/OS example
84f709e7 668 $ascii_data = `echo '$ebcdic_data'| iconv -f IBM-1047 -t ISO8859-1`
d396a558
JH
669
670or the inverse map:
671
395f5a0c 672 # OS/390 or z/OS example
84f709e7 673 $ebcdic_data = `echo '$ascii_data'| iconv -f ISO8859-1 -t IBM-1047`
d396a558 674
d396a558
JH
675For other perl based conversion options see the Convert::* modules on CPAN.
676
1e054b24
PP
677=head2 C RTL
678
395f5a0c 679The OS/390 and z/OS C run time libraries provide _atoe() and _etoa() functions.
1e054b24 680
d396a558
JH
681=head1 OPERATOR DIFFERENCES
682
683The C<..> range operator treats certain character ranges with
684care on EBCDIC machines. For example the following array
685will have twenty six elements on either an EBCDIC machine
686or an ASCII machine:
687
84f709e7 688 @alphabet = ('A'..'Z'); # $#alphabet == 25
d396a558
JH
689
690The bitwise operators such as & ^ | may return different results
691when operating on string or character data in a perl program running
692on an EBCDIC machine than when run on an ASCII machine. Here is
693an example adapted from the one in L<perlop>:
694
695 # EBCDIC-based examples
84f709e7
JH
696 print "j p \n" ^ " a h"; # prints "JAPH\n"
697 print "JA" | " ph\n"; # prints "japh\n"
698 print "JAPH\nJunk" & "\277\277\277\277\277"; # prints "japh\n";
699 print 'p N$' ^ " E<H\n"; # prints "Perl\n";
d396a558
JH
700
701An interesting property of the 32 C0 control characters
702in the ASCII table is that they can "literally" be constructed
51b5cecb
PP
703as control characters in perl, e.g. C<(chr(0) eq "\c@")>
704C<(chr(1) eq "\cA")>, and so on. Perl on EBCDIC machines has been
705ported to take "\c@" to chr(0) and "\cA" to chr(1) as well, but the
d396a558
JH
706thirty three characters that result depend on which code page you are
707using. The table below uses the character names from the previous table
51b5cecb 708but with substitutions such as s/START OF/S.O./; s/END OF /E.O./;
d396a558
JH
709s/TRANSMISSION/TRANS./; s/TABULATION/TAB./; s/VERTICAL/VERT./;
710s/HORIZONTAL/HORIZ./; s/DEVICE CONTROL/D.C./; s/SEPARATOR/SEP./;
711s/NEGATIVE ACKNOWLEDGE/NEG. ACK./;. The POSIX-BC and 1047 sets are
712identical throughout this range and differ from the 0037 set at only
51b5cecb
PP
713one spot (21 decimal). Note that the C<LINE FEED> character
714may be generated by "\cJ" on ASCII machines but by "\cU" on 1047 or POSIX-BC
715machines and cannot be generated as a C<"\c.letter."> control character on
7160037 machines. Note also that "\c\\" maps to two characters
d396a558
JH
717not one.
718
719 chr ord 8859-1 0037 1047 && POSIX-BC
720 ------------------------------------------------------------------------
721 "\c?" 127 <DELETE> " " ***><
722 "\c@" 0 <NULL> <NULL> <NULL> ***><
723 "\cA" 1 <S.O. HEADING> <S.O. HEADING> <S.O. HEADING>
724 "\cB" 2 <S.O. TEXT> <S.O. TEXT> <S.O. TEXT>
725 "\cC" 3 <E.O. TEXT> <E.O. TEXT> <E.O. TEXT>
726 "\cD" 4 <E.O. TRANS.> <C1 28> <C1 28>
727 "\cE" 5 <ENQUIRY> <HORIZ. TAB.> <HORIZ. TAB.>
728 "\cF" 6 <ACKNOWLEDGE> <C1 6> <C1 6>
729 "\cG" 7 <BELL> <DELETE> <DELETE>
730 "\cH" 8 <BACKSPACE> <C1 23> <C1 23>
731 "\cI" 9 <HORIZ. TAB.> <C1 13> <C1 13>
732 "\cJ" 10 <LINE FEED> <C1 14> <C1 14>
733 "\cK" 11 <VERT. TAB.> <VERT. TAB.> <VERT. TAB.>
734 "\cL" 12 <FORM FEED> <FORM FEED> <FORM FEED>
735 "\cM" 13 <CARRIAGE RETURN> <CARRIAGE RETURN> <CARRIAGE RETURN>
736 "\cN" 14 <SHIFT OUT> <SHIFT OUT> <SHIFT OUT>
737 "\cO" 15 <SHIFT IN> <SHIFT IN> <SHIFT IN>
738 "\cP" 16 <DATA LINK ESCAPE> <DATA LINK ESCAPE> <DATA LINK ESCAPE>
739 "\cQ" 17 <D.C. ONE> <D.C. ONE> <D.C. ONE>
740 "\cR" 18 <D.C. TWO> <D.C. TWO> <D.C. TWO>
741 "\cS" 19 <D.C. THREE> <D.C. THREE> <D.C. THREE>
742 "\cT" 20 <D.C. FOUR> <C1 29> <C1 29>
743 "\cU" 21 <NEG. ACK.> <C1 5> <LINE FEED> ***
744 "\cV" 22 <SYNCHRONOUS IDLE> <BACKSPACE> <BACKSPACE>
745 "\cW" 23 <E.O. TRANS. BLOCK> <C1 7> <C1 7>
746 "\cX" 24 <CANCEL> <CANCEL> <CANCEL>
747 "\cY" 25 <E.O. MEDIUM> <E.O. MEDIUM> <E.O. MEDIUM>
748 "\cZ" 26 <SUBSTITUTE> <C1 18> <C1 18>
749 "\c[" 27 <ESCAPE> <C1 15> <C1 15>
750 "\c\\" 28 <FILE SEP.>\ <FILE SEP.>\ <FILE SEP.>\
751 "\c]" 29 <GROUP SEP.> <GROUP SEP.> <GROUP SEP.>
752 "\c^" 30 <RECORD SEP.> <RECORD SEP.> <RECORD SEP.> ***><
753 "\c_" 31 <UNIT SEP.> <UNIT SEP.> <UNIT SEP.> ***><
754
755
756=head1 FUNCTION DIFFERENCES
757
758=over 8
759
760=item chr()
761
762chr() must be given an EBCDIC code number argument to yield a desired
763character return value on an EBCDIC machine. For example:
764
84f709e7 765 $CAPITAL_LETTER_A = chr(193);
d396a558
JH
766
767=item ord()
768
769ord() will return EBCDIC code number values on an EBCDIC machine.
770For example:
771
84f709e7 772 $the_number_193 = ord("A");
d396a558
JH
773
774=item pack()
775
776The c and C templates for pack() are dependent upon character set
777encoding. Examples of usage on EBCDIC include:
778
779 $foo = pack("CCCC",193,194,195,196);
780 # $foo eq "ABCD"
84f709e7 781 $foo = pack("C4",193,194,195,196);
d396a558
JH
782 # same thing
783
784 $foo = pack("ccxxcc",193,194,195,196);
785 # $foo eq "AB\0\0CD"
786
787=item print()
788
789One must be careful with scalars and strings that are passed to
790print that contain ASCII encodings. One common place
791for this to occur is in the output of the MIME type header for
792CGI script writing. For example, many perl programming guides
793recommend something similar to:
794
795 print "Content-type:\ttext/html\015\012\015\012";
796 # this may be wrong on EBCDIC
797
395f5a0c
PK
798Under the IBM OS/390 USS Web Server or WebSphere on z/OS for example
799you should instead write that as:
d396a558
JH
800
801 print "Content-type:\ttext/html\r\n\r\n"; # OK for DGW et alia
802
803That is because the translation from EBCDIC to ASCII is done
804by the web server in this case (such code will not be appropriate for
805the Macintosh however). Consult your web server's documentation for
806further details.
807
808=item printf()
809
810The formats that can convert characters to numbers and vice versa
811will be different from their ASCII counterparts when executed
812on an EBCDIC machine. Examples include:
813
814 printf("%c%c%c",193,194,195); # prints ABC
815
816=item sort()
817
818EBCDIC sort results may differ from ASCII sort results especially for
819mixed case strings. This is discussed in more detail below.
820
821=item sprintf()
822
823See the discussion of printf() above. An example of the use
824of sprintf would be:
825
84f709e7 826 $CAPITAL_LETTER_A = sprintf("%c",193);
d396a558
JH
827
828=item unpack()
829
830See the discussion of pack() above.
831
832=back
833
834=head1 REGULAR EXPRESSION DIFFERENCES
835
836As of perl 5.005_03 the letter range regular expression such as
837[A-Z] and [a-z] have been especially coded to not pick up gap
b3b6085d
PP
838characters. For example, characters such as E<ocirc> C<o WITH CIRCUMFLEX>
839that lie between I and J would not be matched by the
1b2d223b
JH
840regular expression range C</[H-K]/>. This works in
841the other direction, too, if either of the range end points is
842explicitly numeric: C<[\x89-\x91]> will match C<\x8e>, even
843though C<\x89> is C<i> and C<\x91 > is C<j>, and C<\x8e>
844is a gap character from the alphabetic viewpoint.
51b5cecb
PP
845
846If you do want to match the alphabet gap characters in a single octet
d396a558
JH
847regular expression try matching the hex or octal code such
848as C</\313/> on EBCDIC or C</\364/> on ASCII machines to
51b5cecb 849have your regular expression match C<o WITH CIRCUMFLEX>.
d396a558 850
51b5cecb 851Another construct to be wary of is the inappropriate use of hex or
d396a558
JH
852octal constants in regular expressions. Consider the following
853set of subs:
854
855 sub is_c0 {
856 my $char = substr(shift,0,1);
857 $char =~ /[\000-\037]/;
858 }
859
860 sub is_print_ascii {
861 my $char = substr(shift,0,1);
862 $char =~ /[\040-\176]/;
863 }
864
865 sub is_delete {
866 my $char = substr(shift,0,1);
867 $char eq "\177";
868 }
869
870 sub is_c1 {
871 my $char = substr(shift,0,1);
872 $char =~ /[\200-\237]/;
873 }
874
875 sub is_latin_1 {
876 my $char = substr(shift,0,1);
877 $char =~ /[\240-\377]/;
878 }
879
51b5cecb
PP
880The above would be adequate if the concern was only with numeric code points.
881However, the concern may be with characters rather than code points
882and on an EBCDIC machine it may be desirable for constructs such as
d396a558
JH
883C<if (is_print_ascii("A")) {print "A is a printable character\n";}> to print
884out the expected message. One way to represent the above collection
885of character classification subs that is capable of working across the
886four coded character sets discussed in this document is as follows:
887
888 sub Is_c0 {
889 my $char = substr(shift,0,1);
84f709e7 890 if (ord('^')==94) { # ascii
d396a558 891 return $char =~ /[\000-\037]/;
84f709e7
JH
892 }
893 if (ord('^')==176) { # 37
d396a558
JH
894 return $char =~ /[\000-\003\067\055-\057\026\005\045\013-\023\074\075\062\046\030\031\077\047\034-\037]/;
895 }
84f709e7 896 if (ord('^')==95 || ord('^')==106) { # 1047 || posix-bc
d396a558
JH
897 return $char =~ /[\000-\003\067\055-\057\026\005\025\013-\023\074\075\062\046\030\031\077\047\034-\037]/;
898 }
899 }
900
901 sub Is_print_ascii {
902 my $char = substr(shift,0,1);
903 $char =~ /[ !"\#\$%&'()*+,\-.\/0-9:;<=>?\@A-Z[\\\]^_`a-z{|}~]/;
904 }
905
906 sub Is_delete {
907 my $char = substr(shift,0,1);
84f709e7 908 if (ord('^')==94) { # ascii
d396a558 909 return $char eq "\177";
84f709e7
JH
910 }
911 else { # ebcdic
d396a558
JH
912 return $char eq "\007";
913 }
914 }
915
916 sub Is_c1 {
917 my $char = substr(shift,0,1);
84f709e7 918 if (ord('^')==94) { # ascii
d396a558
JH
919 return $char =~ /[\200-\237]/;
920 }
84f709e7 921 if (ord('^')==176) { # 37
d396a558
JH
922 return $char =~ /[\040-\044\025\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\377]/;
923 }
84f709e7 924 if (ord('^')==95) { # 1047
d396a558
JH
925 return $char =~ /[\040-\045\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\377]/;
926 }
84f709e7
JH
927 if (ord('^')==106) { # posix-bc
928 return $char =~
d396a558
JH
929 /[\040-\045\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\137]/;
930 }
931 }
932
933 sub Is_latin_1 {
934 my $char = substr(shift,0,1);
84f709e7 935 if (ord('^')==94) { # ascii
d396a558
JH
936 return $char =~ /[\240-\377]/;
937 }
84f709e7
JH
938 if (ord('^')==176) { # 37
939 return $char =~
d396a558
JH
940 /[\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]/;
941 }
84f709e7 942 if (ord('^')==95) { # 1047
d396a558
JH
943 return $char =~
944 /[\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]/;
945 }
84f709e7
JH
946 if (ord('^')==106) { # posix-bc
947 return $char =~
d396a558
JH
948 /[\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]/;
949 }
950 }
951
952Note however that only the C<Is_ascii_print()> sub is really independent
953of coded character set. Another way to write C<Is_latin_1()> would be
954to use the characters in the range explicitly:
955
956 sub Is_latin_1 {
957 my $char = substr(shift,0,1);
958 $char =~ /[ ¡¢£¤¥¦§¨©ª«¬­®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ]/;
959 }
960
961Although that form may run into trouble in network transit (due to the
962presence of 8 bit characters) or on non ISO-Latin character sets.
d396a558
JH
963
964=head1 SOCKETS
965
966Most socket programming assumes ASCII character encodings in network
967byte order. Exceptions can include CGI script writing under a
968host web server where the server may take care of translation for you.
969Most host web servers convert EBCDIC data to ISO-8859-1 or Unicode on
970output.
971
972=head1 SORTING
973
974One big difference between ASCII based character sets and EBCDIC ones
975are the relative positions of upper and lower case letters and the
976letters compared to the digits. If sorted on an ASCII based machine the
977two letter abbreviation for a physician comes before the two letter
978for drive, that is:
979
84f709e7
JH
980 @sorted = sort(qw(Dr. dr.)); # @sorted holds ('Dr.','dr.') on ASCII,
981 # but ('dr.','Dr.') on EBCDIC
d396a558
JH
982
983The property of lower case before uppercase letters in EBCDIC is
984even carried to the Latin 1 EBCDIC pages such as 0037 and 1047.
b3b6085d
PP
985An example would be that E<Euml> C<E WITH DIAERESIS> (203) comes
986before E<euml> C<e WITH DIAERESIS> (235) on an ASCII machine, but
51b5cecb 987the latter (83) comes before the former (115) on an EBCDIC machine.
b3b6085d 988(Astute readers will note that the upper case version of E<szlig>
51b5cecb 989C<SMALL LETTER SHARP S> is simply "SS" and that the upper case version of
b3b6085d 990E<yuml> C<y WITH DIAERESIS> is not in the 0..255 range but it is
51b5cecb 991at U+x0178 in Unicode, or C<"\x{178}"> in a Unicode enabled Perl).
d396a558
JH
992
993The sort order will cause differences between results obtained on
994ASCII machines versus EBCDIC machines. What follows are some suggestions
995on how to deal with these differences.
996
51b5cecb 997=head2 Ignore ASCII vs. EBCDIC sort differences.
d396a558
JH
998
999This is the least computationally expensive strategy. It may require
1000some user education.
1001
51b5cecb 1002=head2 MONO CASE then sort data.
d396a558 1003
51b5cecb 1004In order to minimize the expense of mono casing mixed test try to
d396a558
JH
1005C<tr///> towards the character set case most employed within the data.
1006If the data are primarily UPPERCASE non Latin 1 then apply tr/[a-z]/[A-Z]/
1007then sort(). If the data are primarily lowercase non Latin 1 then
1008apply tr/[A-Z]/[a-z]/ before sorting. If the data are primarily UPPERCASE
51b5cecb
PP
1009and include Latin-1 characters then apply:
1010
84f709e7 1011 tr/[a-z]/[A-Z]/;
51b5cecb 1012 tr/[àáâãäåæçèéêëìíîïðñòóôõöøùúûüýþ]/[ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞ]/;
84f709e7 1013 s/ß/SS/g;
d396a558 1014
51b5cecb 1015then sort(). Do note however that such Latin-1 manipulation does not
b3b6085d
PP
1016address the E<yuml> C<y WITH DIAERESIS> character that will remain at
1017code point 255 on ASCII machines, but 223 on most EBCDIC machines
51b5cecb
PP
1018where it will sort to a place less than the EBCDIC numerals. With a
1019Unicode enabled Perl you might try:
d396a558 1020
51b5cecb
PP
1021 tr/^?/\x{178}/;
1022
1023The strategy of mono casing data before sorting does not preserve the case
1024of the data and may not be acceptable for that reason.
1025
1026=head2 Convert, sort data, then re convert.
d396a558
JH
1027
1028This is the most expensive proposition that does not employ a network
1029connection.
1030
1031=head2 Perform sorting on one type of machine only.
1032
1033This strategy can employ a network connection. As such
1034it would be computationally expensive.
1035
395f5a0c 1036=head1 TRANSFORMATION FORMATS
1e054b24
PP
1037
1038There are a variety of ways of transforming data with an intra character set
1039mapping that serve a variety of purposes. Sorting was discussed in the
1040previous section and a few of the other more popular mapping techniques are
1041discussed next.
1042
1043=head2 URL decoding and encoding
d396a558 1044
51b5cecb 1045Note that some URLs have hexadecimal ASCII code points in them in an
1e054b24
PP
1046attempt to overcome character or protocol limitation issues. For example
1047the tilde character is not on every keyboard hence a URL of the form:
d396a558
JH
1048
1049 http://www.pvhp.com/~pvhp/
1050
1051may also be expressed as either of:
1052
1053 http://www.pvhp.com/%7Epvhp/
1054
1055 http://www.pvhp.com/%7epvhp/
1056
51b5cecb 1057where 7E is the hexadecimal ASCII code point for '~'. Here is an example
d396a558
JH
1058of decoding such a URL under CCSID 1047:
1059
84f709e7 1060 $url = 'http://www.pvhp.com/%7Epvhp/';
d396a558
JH
1061 # this array assumes code page 1047
1062 my @a2e_1047 = (
1063 0, 1, 2, 3, 55, 45, 46, 47, 22, 5, 21, 11, 12, 13, 14, 15,
1064 16, 17, 18, 19, 60, 61, 50, 38, 24, 25, 63, 39, 28, 29, 30, 31,
1065 64, 90,127,123, 91,108, 80,125, 77, 93, 92, 78,107, 96, 75, 97,
1066 240,241,242,243,244,245,246,247,248,249,122, 94, 76,126,110,111,
1067 124,193,194,195,196,197,198,199,200,201,209,210,211,212,213,214,
1068 215,216,217,226,227,228,229,230,231,232,233,173,224,189, 95,109,
1069 121,129,130,131,132,133,134,135,136,137,145,146,147,148,149,150,
1070 151,152,153,162,163,164,165,166,167,168,169,192, 79,208,161, 7,
1071 32, 33, 34, 35, 36, 37, 6, 23, 40, 41, 42, 43, 44, 9, 10, 27,
1072 48, 49, 26, 51, 52, 53, 54, 8, 56, 57, 58, 59, 4, 20, 62,255,
1073 65,170, 74,177,159,178,106,181,187,180,154,138,176,202,175,188,
1074 144,143,234,250,190,160,182,179,157,218,155,139,183,184,185,171,
1075 100,101, 98,102, 99,103,158,104,116,113,114,115,120,117,118,119,
1076 172,105,237,238,235,239,236,191,128,253,254,251,252,186,174, 89,
1077 68, 69, 66, 70, 67, 71,156, 72, 84, 81, 82, 83, 88, 85, 86, 87,
1078 140, 73,205,206,203,207,204,225,112,221,222,219,220,141,142,223
1079 );
1080 $url =~ s/%([0-9a-fA-F]{2})/pack("c",$a2e_1047[hex($1)])/ge;
1081
1e054b24
PP
1082Conversely, here is a partial solution for the task of encoding such
1083a URL under the 1047 code page:
1084
84f709e7 1085 $url = 'http://www.pvhp.com/~pvhp/';
1e054b24
PP
1086 # this array assumes code page 1047
1087 my @e2a_1047 = (
1088 0, 1, 2, 3,156, 9,134,127,151,141,142, 11, 12, 13, 14, 15,
1089 16, 17, 18, 19,157, 10, 8,135, 24, 25,146,143, 28, 29, 30, 31,
1090 128,129,130,131,132,133, 23, 27,136,137,138,139,140, 5, 6, 7,
1091 144,145, 22,147,148,149,150, 4,152,153,154,155, 20, 21,158, 26,
1092 32,160,226,228,224,225,227,229,231,241,162, 46, 60, 40, 43,124,
1093 38,233,234,235,232,237,238,239,236,223, 33, 36, 42, 41, 59, 94,
1094 45, 47,194,196,192,193,195,197,199,209,166, 44, 37, 95, 62, 63,
1095 248,201,202,203,200,205,206,207,204, 96, 58, 35, 64, 39, 61, 34,
1096 216, 97, 98, 99,100,101,102,103,104,105,171,187,240,253,254,177,
1097 176,106,107,108,109,110,111,112,113,114,170,186,230,184,198,164,
1098 181,126,115,116,117,118,119,120,121,122,161,191,208, 91,222,174,
1099 172,163,165,183,169,167,182,188,189,190,221,168,175, 93,180,215,
1100 123, 65, 66, 67, 68, 69, 70, 71, 72, 73,173,244,246,242,243,245,
1101 125, 74, 75, 76, 77, 78, 79, 80, 81, 82,185,251,252,249,250,255,
1102 92,247, 83, 84, 85, 86, 87, 88, 89, 90,178,212,214,210,211,213,
1103 48, 49, 50, 51, 52, 53, 54, 55, 56, 57,179,219,220,217,218,159
1104 );
84f709e7 1105 # The following regular expression does not address the
1e054b24
PP
1106 # mappings for: ('.' => '%2E', '/' => '%2F', ':' => '%3A')
1107 $url =~ s/([\t "#%&\(\),;<=>\?\@\[\\\]^`{|}~])/sprintf("%%%02X",$e2a_1047[ord($1)])/ge;
1108
1109where a more complete solution would split the URL into components
1110and apply a full s/// substitution only to the appropriate parts.
1111
1112In the remaining examples a @e2a or @a2e array may be employed
1113but the assignment will not be shown explicitly. For code page 1047
1114you could use the @a2e_1047 or @e2a_1047 arrays just shown.
1115
1116=head2 uu encoding and decoding
1117
1118The C<u> template to pack() or unpack() will render EBCDIC data in EBCDIC
1119characters equivalent to their ASCII counterparts. For example, the
1120following will print "Yes indeed\n" on either an ASCII or EBCDIC computer:
1121
84f709e7
JH
1122 $all_byte_chrs = '';
1123 for (0..255) { $all_byte_chrs .= chr($_); }
1124 $uuencode_byte_chrs = pack('u', $all_byte_chrs);
210b36aa 1125 ($uu = <<'ENDOFHEREDOC') =~ s/^\s*//gm;
1e054b24
PP
1126 M``$"`P0%!@<("0H+#`T.#Q`1$A,4%187&!D:&QP='A\@(2(C)"4F)R@I*BLL
1127 M+2XO,#$R,S0U-C<X.3H[/#T^/T!!0D-$149'2$E*2TQ-3D]045)35%565UA9
1128 M6EM<75Y?8&%B8V1E9F=H:6IK;&UN;W!Q<G-T=79W>'EZ>WQ]?G^`@8*#A(6&
1129 MAXB)BHN,C8Z/D)&2DY25EI>8F9J;G)V>GZ"AHJ.DI::GJ*FJJZRMKJ^PL;*S
1130 MM+6VM[BYNKN\O;Z_P,'"P\3%QL?(R<K+S,W.S]#1TM/4U=;7V-G:V]S=WM_@
1131 ?X>+CY.7FY^CIZNOL[>[O\/'R\_3U]O?X^?K[_/W^_P``
1132 ENDOFHEREDOC
84f709e7 1133 if ($uuencode_byte_chrs eq $uu) {
1e054b24
PP
1134 print "Yes ";
1135 }
1136 $uudecode_byte_chrs = unpack('u', $uuencode_byte_chrs);
84f709e7 1137 if ($uudecode_byte_chrs eq $all_byte_chrs) {
1e054b24
PP
1138 print "indeed\n";
1139 }
1140
1141Here is a very spartan uudecoder that will work on EBCDIC provided
1142that the @e2a array is filled in appropriately:
1143
84f709e7
JH
1144 #!/usr/local/bin/perl
1145 @e2a = ( # this must be filled in
1146 );
1147 $_ = <> until ($mode,$file) = /^begin\s*(\d*)\s*(\S*)/;
1e054b24
PP
1148 open(OUT, "> $file") if $file ne "";
1149 while(<>) {
1150 last if /^end/;
1151 next if /[a-z]/;
1152 next unless int(((($e2a[ord()] - 32 ) & 077) + 2) / 3) ==
1153 int(length() / 4);
1154 print OUT unpack("u", $_);
1155 }
1156 close(OUT);
1157 chmod oct($mode), $file;
1158
1159
1160=head2 Quoted-Printable encoding and decoding
1161
1162On ASCII encoded machines it is possible to strip characters outside of
1163the printable set using:
1164
1165 # This QP encoder works on ASCII only
84f709e7 1166 $qp_string =~ s/([=\x00-\x1F\x80-\xFF])/sprintf("=%02X",ord($1))/ge;
1e054b24
PP
1167
1168Whereas a QP encoder that works on both ASCII and EBCDIC machines
1169would look somewhat like the following (where the EBCDIC branch @e2a
1170array is omitted for brevity):
1171
1172 if (ord('A') == 65) { # ASCII
1173 $delete = "\x7F"; # ASCII
1174 @e2a = (0 .. 255) # ASCII to ASCII identity map
84f709e7
JH
1175 }
1176 else { # EBCDIC
1e054b24 1177 $delete = "\x07"; # EBCDIC
84f709e7 1178 @e2a = # EBCDIC to ASCII map (as shown above)
1e054b24 1179 }
84f709e7 1180 $qp_string =~
1e054b24
PP
1181 s/([^ !"\#\$%&'()*+,\-.\/0-9:;<>?\@A-Z[\\\]^_`a-z{|}~$delete])/sprintf("=%02X",$e2a[ord($1)])/ge;
1182
1183(although in production code the substitutions might be done
1184in the EBCDIC branch with the @e2a array and separately in the
1185ASCII branch without the expense of the identity map).
1186
1187Such QP strings can be decoded with:
1188
1189 # This QP decoder is limited to ASCII only
1190 $string =~ s/=([0-9A-Fa-f][0-9A-Fa-f])/chr hex $1/ge;
1191 $string =~ s/=[\n\r]+$//;
1192
1193Whereas a QP decoder that works on both ASCII and EBCDIC machines
1194would look somewhat like the following (where the @a2e array is
1195omitted for brevity):
1196
1197 $string =~ s/=([0-9A-Fa-f][0-9A-Fa-f])/chr $a2e[hex $1]/ge;
1198 $string =~ s/=[\n\r]+$//;
1199
395f5a0c 1200=head2 Caesarian ciphers
1e054b24
PP
1201
1202The practice of shifting an alphabet one or more characters for encipherment
1203dates back thousands of years and was explicitly detailed by Gaius Julius
1204Caesar in his B<Gallic Wars> text. A single alphabet shift is sometimes
1205referred to as a rotation and the shift amount is given as a number $n after
1206the string 'rot' or "rot$n". Rot0 and rot26 would designate identity maps
1207on the 26 letter English version of the Latin alphabet. Rot13 has the
1208interesting property that alternate subsequent invocations are identity maps
1209(thus rot13 is its own non-trivial inverse in the group of 26 alphabet
1210rotations). Hence the following is a rot13 encoder and decoder that will
1211work on ASCII and EBCDIC machines:
1212
1213 #!/usr/local/bin/perl
1214
84f709e7 1215 while(<>){
1e054b24
PP
1216 tr/n-za-mN-ZA-M/a-zA-Z/;
1217 print;
1218 }
1219
1220In one-liner form:
1221
84f709e7 1222 perl -ne 'tr/n-za-mN-ZA-M/a-zA-Z/;print'
1e054b24
PP
1223
1224
1225=head1 Hashing order and checksums
1226
395f5a0c
PK
1227To the extent that it is possible to write code that depends on
1228hashing order there may be differences between hashes as stored
1229on an ASCII based machine and hashes stored on an EBCDIC based machine.
1e054b24
PP
1230XXX
1231
d396a558
JH
1232=head1 I18N AND L10N
1233
1234Internationalization(I18N) and localization(L10N) are supported at least
1235in principle even on EBCDIC machines. The details are system dependent
1236and discussed under the L<perlebcdic/OS ISSUES> section below.
1237
1238=head1 MULTI OCTET CHARACTER SETS
1239
395f5a0c
PK
1240Perl may work with an internal UTF-EBCDIC encoding form for wide characters
1241on EBCDIC platforms in a manner analogous to the way that it works with
1242the UTF-8 internal encoding form on ASCII based platforms.
1243
1244Legacy multi byte EBCDIC code pages XXX.
d396a558
JH
1245
1246=head1 OS ISSUES
1247
1248There may be a few system dependent issues
1249of concern to EBCDIC Perl programmers.
1250
522b859a 1251=head2 OS/400
51b5cecb 1252
d396a558
JH
1253=over 8
1254
522b859a
JH
1255=item PASE
1256
1257The PASE environment is runtime environment for OS/400 that can run
1258executables built for PowerPC AIX in OS/400, see L<perlos400>. PASE
1259is ASCII-based, not EBCDIC-based as the ILE.
1260
d396a558
JH
1261=item IFS access
1262
1263XXX.
1264
1265=back
1266
395f5a0c 1267=head2 OS/390, z/OS
d396a558 1268
51b5cecb
PP
1269Perl runs under Unix Systems Services or USS.
1270
d396a558
JH
1271=over 8
1272
51b5cecb
PP
1273=item chcp
1274
1e054b24
PP
1275B<chcp> is supported as a shell utility for displaying and changing
1276one's code page. See also L<chcp>.
51b5cecb 1277
d396a558
JH
1278=item dataset access
1279
1280For sequential data set access try:
1281
1282 my @ds_records = `cat //DSNAME`;
1283
1284or:
1285
1286 my @ds_records = `cat //'HLQ.DSNAME'`;
1287
1288See also the OS390::Stdio module on CPAN.
1289
395f5a0c 1290=item OS/390, z/OS iconv
51b5cecb 1291
1e054b24
PP
1292B<iconv> is supported as both a shell utility and a C RTL routine.
1293See also the iconv(1) and iconv(3) manual pages.
51b5cecb 1294
d396a558
JH
1295=item locales
1296
395f5a0c
PK
1297On OS/390 or z/OS see L<locale> for information on locales. The L10N files
1298are in F</usr/nls/locale>. $Config{d_setlocale} is 'define' on OS/390
1299or z/OS.
d396a558
JH
1300
1301=back
1302
1303=head2 VM/ESA?
1304
1305XXX.
1306
1307=head2 POSIX-BC?
1308
1309XXX.
1310
51b5cecb
PP
1311=head1 BUGS
1312
1313This pod document contains literal Latin 1 characters and may encounter
b1866b2d 1314translation difficulties. In particular one popular nroff implementation
51b5cecb
PP
1315was known to strip accented characters to their unaccented counterparts
1316while attempting to view this document through the B<pod2man> program
1317(for example, you may see a plain C<y> rather than one with a diaeresis
3958b146 1318as in E<yuml>). Another nroff truncated the resultant manpage at
395f5a0c 1319the first occurrence of 8 bit characters.
51b5cecb
PP
1320
1321Not all shells will allow multiple C<-e> string arguments to perl to
395f5a0c
PK
1322be concatenated together properly as recipes 0, 2, 4, 5, and 6 might
1323seem to imply.
51b5cecb 1324
b3b6085d
PP
1325=head1 SEE ALSO
1326
395f5a0c 1327L<perllocale>, L<perlfunc>, L<perlunicode>, L<utf8>.
b3b6085d 1328
d396a558
JH
1329=head1 REFERENCES
1330
1331http://anubis.dkuug.dk/i18n/charmaps
1332
d396a558
JH
1333http://www.unicode.org/
1334
1335http://www.unicode.org/unicode/reports/tr16/
1336
51b5cecb
PP
1337http://www.wps.com/texts/codes/
1338B<ASCII: American Standard Code for Information Infiltration> Tom Jennings,
1339September 1999.
1340
395f5a0c 1341B<The Unicode Standard, Version 3.0> The Unicode Consortium, Lisa Moore ed.,
51b5cecb
PP
1342ISBN 0-201-61633-5, Addison Wesley Developers Press, February 2000.
1343
d396a558
JH
1344B<CDRA: IBM - Character Data Representation Architecture -
1345Reference and Registry>, IBM SC09-2190-00, December 1996.
1346
1347"Demystifying Character Sets", Andrea Vine, Multilingual Computing
1348& Technology, B<#26 Vol. 10 Issue 4>, August/September 1999;
1349ISSN 1523-0309; Multilingual Computing Inc. Sandpoint ID, USA.
1350
1e054b24
PP
1351B<Codes, Ciphers, and Other Cryptic and Clandestine Communication>
1352Fred B. Wrixon, ISBN 1-57912-040-7, Black Dog & Leventhal Publishers,
13531998.
1354
395f5a0c
PK
1355http://www.bobbemer.com/P-BIT.HTM
1356B<IBM - EBCDIC and the P-bit; The biggest Computer Goof Ever> Robert Bemer.
1357
1358=head1 HISTORY
1359
136015 April 2001: added UTF-8 and UTF-EBCDIC to main table, pvhp.
1361
d396a558
JH
1362=head1 AUTHOR
1363
b3b6085d 1364Peter Prymmer pvhp@best.com wrote this in 1999 and 2000
d396a558 1365with CCSID 0819 and 0037 help from Chris Leach and
b3b6085d
PP
1366AndrE<eacute> Pirard A.Pirard@ulg.ac.be as well as POSIX-BC
1367help from Thomas Dorner Thomas.Dorner@start.de.
1e054b24
PP
1368Thanks also to Vickie Cooper, Philip Newton, William Raffloer, and
1369Joe Smith. Trademarks, registered trademarks, service marks and
1370registered service marks used in this document are the property of
1371their respective owners.
84f709e7
JH
1372
1373