5 perlebcdic - Considerations for running Perl on EBCDIC platforms
9 An exploration of some of the issues facing Perl programmers
10 on EBCDIC based computers. We do not cover localization,
11 internationalization, or multi-byte character set issues other
12 than some discussion of UTF-8 and UTF-EBCDIC.
14 Portions that are still incomplete are marked with XXX.
16 Perl used to work on EBCDIC machines, but there are now areas of the code where
17 it doesn't. If you want to use Perl on an EBCDIC machine, please let us know
18 by sending mail to perlbug@perl.org
20 =head1 COMMON CHARACTER CODE SETS
24 The American Standard Code for Information Interchange (ASCII or US-ASCII) is a
26 integers running from 0 to 127 (decimal) that imply character
27 interpretation by the display and other systems of computers.
28 The range 0..127 can be covered by setting the bits in a 7-bit binary
29 digit, hence the set is sometimes referred to as "7-bit ASCII".
30 ASCII was described by the American National Standards Institute
31 document ANSI X3.4-1986. It was also described by ISO 646:1991
32 (with localization for currency symbols). The full ASCII set is
33 given in the table below as the first 128 elements. Languages that
34 can be written adequately with the characters in ASCII include
35 English, Hawaiian, Indonesian, Swahili and some Native American
38 There are many character sets that extend the range of integers
39 from 0..2**7-1 up to 2**8-1, or 8 bit bytes (octets if you prefer).
40 One common one is the ISO 8859-1 character set.
44 The ISO 8859-$n are a collection of character code sets from the
45 International Organization for Standardization (ISO), each of which
46 adds characters to the ASCII set that are typically found in European
47 languages, many of which are based on the Roman, or Latin, alphabet.
49 =head2 Latin 1 (ISO 8859-1)
51 A particular 8-bit extension to ASCII that includes grave and acute
52 accented Latin characters. Languages that can employ ISO 8859-1
53 include all the languages covered by ASCII as well as Afrikaans,
54 Albanian, Basque, Catalan, Danish, Faroese, Finnish, Norwegian,
55 Portuguese, Spanish, and Swedish. Dutch is covered albeit without
56 the ij ligature. French is covered too but without the oe ligature.
57 German can use ISO 8859-1 but must do so without German-style
58 quotation marks. This set is based on Western European extensions
59 to ASCII and is commonly encountered in world wide web work.
60 In IBM character code set identification terminology ISO 8859-1 is
61 also known as CCSID 819 (or sometimes 0819 or even 00819).
65 The Extended Binary Coded Decimal Interchange Code refers to a
66 large collection of single- and multi-byte coded character sets that are
67 different from ASCII or ISO 8859-1 and are all slightly different from each
68 other; they typically run on host computers. The EBCDIC encodings derive from
69 8-bit byte extensions of Hollerith punched card encodings. The layout on the
70 cards was such that high bits were set for the upper and lower case alphabet
71 characters [a-z] and [A-Z], but there were gaps within each Latin alphabet
74 Some IBM EBCDIC character sets may be known by character code set
75 identification numbers (CCSID numbers) or code page numbers.
77 Perl can be compiled on platforms that run any of three commonly used EBCDIC
78 character sets, listed below.
80 =head2 The 13 variant characters
82 Among IBM EBCDIC character code sets there are 13 characters that
83 are often mapped to different integer values. Those characters
84 are known as the 13 "variant" characters and are:
86 \ [ ] { } ^ ~ ! # | $ @ `
88 When Perl is compiled for a platform, it looks at some of these characters to
89 guess which EBCDIC character set the platform uses, and adapts itself
90 accordingly to that platform. If the platform uses a character set that is not
91 one of the three Perl knows about, Perl will either fail to compile, or
92 mistakenly and silently choose one of the three.
97 Character code set ID 0037 is a mapping of the ASCII plus Latin-1
98 characters (i.e. ISO 8859-1) to an EBCDIC set. 0037 is used
99 in North American English locales on the OS/400 operating system
100 that runs on AS/400 computers. CCSID 0037 differs from ISO 8859-1
101 in 237 places, in other words they agree on only 19 code point values.
105 Character code set ID 1047 is also a mapping of the ASCII plus
106 Latin-1 characters (i.e. ISO 8859-1) to an EBCDIC set. 1047 is
107 used under Unix System Services for OS/390 or z/OS, and OpenEdition
108 for VM/ESA. CCSID 1047 differs from CCSID 0037 in eight places.
112 The EBCDIC code page in use on Siemens' BS2000 system is distinct from
113 1047 and 0037. It is identified below as the POSIX-BC set.
115 =head2 Unicode code points versus EBCDIC code points
117 In Unicode terminology a I<code point> is the number assigned to a
118 character: for example, in EBCDIC the character "A" is usually assigned
119 the number 193. In Unicode the character "A" is assigned the number 65.
120 This causes a problem with the semantics of the pack/unpack "U", which
121 are supposed to pack Unicode code points to characters and back to numbers.
122 The problem is: which code points to use for code points less than 256?
123 (for 256 and over there's no problem: Unicode code points are used)
124 In EBCDIC, for the low 256 the EBCDIC code points are used. This
125 means that the equivalences
127 pack("U", ord($character)) eq $character
128 unpack("U", $character) == ord $character
130 will hold. (If Unicode code points were applied consistently over
131 all the possible code points, pack("U",ord("A")) would in EBCDIC
132 equal I<A with acute> or chr(101), and unpack("U", "A") would equal
133 65, or I<non-breaking space>, not 193, or ord "A".)
135 =head2 Remaining Perl Unicode problems in EBCDIC
141 Many of the remaining problems seem to be related to case-insensitive matching
145 The extensions Unicode::Collate and Unicode::Normalized are not
146 supported under EBCDIC, likewise for the encoding pragma.
150 =head2 Unicode and UTF
152 UTF stands for C<Unicode Transformation Format>.
153 UTF-8 is an encoding of Unicode into a sequence of 8-bit byte chunks, based on
155 The length of a sequence required to represent a Unicode code point
156 depends on the ordinal number of that code point,
157 with larger numbers requiring more bytes.
158 UTF-EBCDIC is like UTF-8, but based on EBCDIC.
160 You may see the term C<invariant> character or code point.
161 This simply means that the character has the same numeric
162 value when encoded as when not.
163 (Note that this is a very different concept from L</The 13 variant characters>
165 For example, the ordinal value of 'A' is 193 in most EBCDIC code pages,
166 and also is 193 when encoded in UTF-EBCDIC.
167 All variant code points occupy at least two bytes when encoded.
168 In UTF-8, the code points corresponding to the lowest 128
169 ordinal numbers (0 - 127: the ASCII characters) are invariant.
170 In UTF-EBCDIC, there are 160 invariant characters.
171 (If you care, the EBCDIC invariants are those characters
172 which have ASCII equivalents, plus those that correspond to
173 the C1 controls (80..9f on ASCII platforms).)
175 A string encoded in UTF-EBCDIC may be longer (but never shorter) than
176 one encoded in UTF-8.
180 Starting from Perl 5.8 you can use the standard new module Encode
181 to translate from EBCDIC to Latin-1 code points.
182 Encode knows about more EBCDIC character sets than Perl can currently
183 be compiled to run on.
185 use Encode 'from_to';
187 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' );
189 # $a is in EBCDIC code points
190 from_to($a, $ebcdic{ord '^'}, 'latin1');
191 # $a is ISO 8859-1 code points
193 and from Latin-1 code points to EBCDIC code points
195 use Encode 'from_to';
197 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' );
199 # $a is ISO 8859-1 code points
200 from_to($a, 'latin1', $ebcdic{ord '^'});
201 # $a is in EBCDIC code points
203 For doing I/O it is suggested that you use the autotranslating features
204 of PerlIO, see L<perluniintro>.
206 Since version 5.8 Perl uses the new PerlIO I/O library. This enables
207 you to use different encodings per IO channel. For example you may use
210 open($f, ">:encoding(ascii)", "test.ascii");
211 print $f "Hello World!\n";
212 open($f, ">:encoding(cp37)", "test.ebcdic");
213 print $f "Hello World!\n";
214 open($f, ">:encoding(latin1)", "test.latin1");
215 print $f "Hello World!\n";
216 open($f, ">:encoding(utf8)", "test.utf8");
217 print $f "Hello World!\n";
219 to get four files containing "Hello World!\n" in ASCII, CP 0037 EBCDIC,
220 ISO 8859-1 (Latin-1) (in this example identical to ASCII since only ASCII
221 characters were printed), and
222 UTF-EBCDIC (in this example identical to normal EBCDIC since only characters
223 that don't differ between EBCDIC and UTF-EBCDIC were printed). See the
224 documentation of Encode::PerlIO for details.
226 As the PerlIO layer uses raw IO (bytes) internally, all this totally
227 ignores things like the type of your filesystem (ASCII or EBCDIC).
229 =head1 SINGLE OCTET TABLES
231 The following tables list the ASCII and Latin 1 ordered sets including
232 the subsets: C0 controls (0..31), ASCII graphics (32..7e), delete (7f),
233 C1 controls (80..9f), and Latin-1 (a.k.a. ISO 8859-1) (a0..ff). In the
234 table names of the Latin 1
235 extensions to ASCII have been labelled with character names roughly
236 corresponding to I<The Unicode Standard, Version 6.1> albeit with
237 substitutions such as s/LATIN// and s/VULGAR// in all cases, s/CAPITAL
238 LETTER// in some cases, and s/SMALL LETTER ([A-Z])/\l$1/ in some other
239 cases. Controls are listed using their Unicode 6.1 abbreviatons.
240 The differences between the 0037 and 1047 sets are
241 flagged with **. The differences between the 1047 and POSIX-BC sets
242 are flagged with ##. All ord() numbers listed are decimal. If you
243 would rather see this table listing octal values, then run the table
244 (that is, the pod source text of this document, since this recipe may not
245 work with a pod2_other_format translation) through:
253 perl -ne 'if(/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
254 -e '{printf("%s%-5.03o%-5.03o%-5.03o%.03o\n",$1,$2,$3,$4,$5)}' \
257 If you want to retain the UTF-x code points then in script form you
266 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!";
268 if (/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*)\s+(\d+)\.?(\d*)/)
270 if ($7 ne '' && $9 ne '') {
272 "%s%-5.03o%-5.03o%-5.03o%-5.03o%-3o.%-5o%-3o.%.03o\n",
273 $1,$2,$3,$4,$5,$6,$7,$8,$9);
276 printf("%s%-5.03o%-5.03o%-5.03o%-5.03o%-3o.%-5o%.03o\n",
277 $1,$2,$3,$4,$5,$6,$7,$8);
280 printf("%s%-5.03o%-5.03o%-5.03o%-5.03o%-5.03o%.03o\n",
281 $1,$2,$3,$4,$5,$6,$8);
286 If you would rather see this table listing hexadecimal values then
287 run the table through:
295 perl -ne 'if(/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
296 -e '{printf("%s%-5.02X%-5.02X%-5.02X%.02X\n",$1,$2,$3,$4,$5)}' \
299 Or, in order to retain the UTF-x code points in hexadecimal:
307 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!";
309 if (/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*)\s+(\d+)\.?(\d*)/)
311 if ($7 ne '' && $9 ne '') {
313 "%s%-5.02X%-5.02X%-5.02X%-5.02X%-2X.%-6.02X%02X.%02X\n",
314 $1,$2,$3,$4,$5,$6,$7,$8,$9);
317 printf("%s%-5.02X%-5.02X%-5.02X%-5.02X%-2X.%-6.02X%02X\n",
318 $1,$2,$3,$4,$5,$6,$7,$8);
321 printf("%s%-5.02X%-5.02X%-5.02X%-5.02X%-5.02X%02X\n",
322 $1,$2,$3,$4,$5,$6,$8);
330 CCSID CCSID CCSID IX-
331 chr 0819 0037 1047 BC UTF-8 UTF-EBCDIC
332 ---------------------------------------------------------------------
337 <EOT> 4 55 55 55 4 55
338 <ENQ> 5 45 45 45 5 45
339 <ACK> 6 46 46 46 6 46
340 <BEL> 7 47 47 47 7 47
343 <LF> 10 37 21 21 10 21 **
344 <VT> 11 11 11 11 11 11
345 <FF> 12 12 12 12 12 12
346 <CR> 13 13 13 13 13 13
347 <SO> 14 14 14 14 14 14
348 <SI> 15 15 15 15 15 15
349 <DLE> 16 16 16 16 16 16
350 <DC1> 17 17 17 17 17 17
351 <DC2> 18 18 18 18 18 18
352 <DC3> 19 19 19 19 19 19
353 <DC4> 20 60 60 60 20 60
354 <NAK> 21 61 61 61 21 61
355 <SYN> 22 50 50 50 22 50
356 <ETB> 23 38 38 38 23 38
357 <CAN> 24 24 24 24 24 24
358 <EOM> 25 25 25 25 25 25
359 <SUB> 26 63 63 63 26 63
360 <ESC> 27 39 39 39 27 39
361 <FS> 28 28 28 28 28 28
362 <GS> 29 29 29 29 29 29
363 <RS> 30 30 30 30 30 30
364 <US> 31 31 31 31 31 31
365 <SPACE> 32 64 64 64 32 64
367 " 34 127 127 127 34 127
368 # 35 123 123 123 35 123
370 % 37 108 108 108 37 108
372 ' 39 125 125 125 39 125
377 , 44 107 107 107 44 107
381 0 48 240 240 240 48 240
382 1 49 241 241 241 49 241
383 2 50 242 242 242 50 242
384 3 51 243 243 243 51 243
385 4 52 244 244 244 52 244
386 5 53 245 245 245 53 245
387 6 54 246 246 246 54 246
388 7 55 247 247 247 55 247
389 8 56 248 248 248 56 248
390 9 57 249 249 249 57 249
391 : 58 122 122 122 58 122
394 = 61 126 126 126 61 126
395 > 62 110 110 110 62 110
396 ? 63 111 111 111 63 111
397 @ 64 124 124 124 64 124
398 A 65 193 193 193 65 193
399 B 66 194 194 194 66 194
400 C 67 195 195 195 67 195
401 D 68 196 196 196 68 196
402 E 69 197 197 197 69 197
403 F 70 198 198 198 70 198
404 G 71 199 199 199 71 199
405 H 72 200 200 200 72 200
406 I 73 201 201 201 73 201
407 J 74 209 209 209 74 209
408 K 75 210 210 210 75 210
409 L 76 211 211 211 76 211
410 M 77 212 212 212 77 212
411 N 78 213 213 213 78 213
412 O 79 214 214 214 79 214
413 P 80 215 215 215 80 215
414 Q 81 216 216 216 81 216
415 R 82 217 217 217 82 217
416 S 83 226 226 226 83 226
417 T 84 227 227 227 84 227
418 U 85 228 228 228 85 228
419 V 86 229 229 229 86 229
420 W 87 230 230 230 87 230
421 X 88 231 231 231 88 231
422 Y 89 232 232 232 89 232
423 Z 90 233 233 233 90 233
424 [ 91 186 173 187 91 173 ** ##
425 \ 92 224 224 188 92 224 ##
426 ] 93 187 189 189 93 189 **
427 ^ 94 176 95 106 94 95 ** ##
428 _ 95 109 109 109 95 109
429 ` 96 121 121 74 96 121 ##
430 a 97 129 129 129 97 129
431 b 98 130 130 130 98 130
432 c 99 131 131 131 99 131
433 d 100 132 132 132 100 132
434 e 101 133 133 133 101 133
435 f 102 134 134 134 102 134
436 g 103 135 135 135 103 135
437 h 104 136 136 136 104 136
438 i 105 137 137 137 105 137
439 j 106 145 145 145 106 145
440 k 107 146 146 146 107 146
441 l 108 147 147 147 108 147
442 m 109 148 148 148 109 148
443 n 110 149 149 149 110 149
444 o 111 150 150 150 111 150
445 p 112 151 151 151 112 151
446 q 113 152 152 152 113 152
447 r 114 153 153 153 114 153
448 s 115 162 162 162 115 162
449 t 116 163 163 163 116 163
450 u 117 164 164 164 117 164
451 v 118 165 165 165 118 165
452 w 119 166 166 166 119 166
453 x 120 167 167 167 120 167
454 y 121 168 168 168 121 168
455 z 122 169 169 169 122 169
456 { 123 192 192 251 123 192 ##
457 | 124 79 79 79 124 79
458 } 125 208 208 253 125 208 ##
459 ~ 126 161 161 255 126 161 ##
460 <DEL> 127 7 7 7 127 7
461 <PAD> 128 32 32 32 194.128 32
462 <HOP> 129 33 33 33 194.129 33
463 <BPH> 130 34 34 34 194.130 34
464 <NBH> 131 35 35 35 194.131 35
465 <IND> 132 36 36 36 194.132 36
466 <NEL> 133 21 37 37 194.133 37 **
467 <SSA> 134 6 6 6 194.134 6
468 <ESA> 135 23 23 23 194.135 23
469 <HTS> 136 40 40 40 194.136 40
470 <HTJ> 137 41 41 41 194.137 41
471 <VTS> 138 42 42 42 194.138 42
472 <PLD> 139 43 43 43 194.139 43
473 <PLU> 140 44 44 44 194.140 44
474 <RI> 141 9 9 9 194.141 9
475 <SS2> 142 10 10 10 194.142 10
476 <SS3> 143 27 27 27 194.143 27
477 <DCS> 144 48 48 48 194.144 48
478 <PU1> 145 49 49 49 194.145 49
479 <PU2> 146 26 26 26 194.146 26
480 <STS> 147 51 51 51 194.147 51
481 <CCH> 148 52 52 52 194.148 52
482 <MW> 149 53 53 53 194.149 53
483 <SPA> 150 54 54 54 194.150 54
484 <EPA> 151 8 8 8 194.151 8
485 <SOS> 152 56 56 56 194.152 56
486 <SGC> 153 57 57 57 194.153 57
487 <SCI> 154 58 58 58 194.154 58
488 <CSI> 155 59 59 59 194.155 59
489 <ST> 156 4 4 4 194.156 4
490 <OSC> 157 20 20 20 194.157 20
491 <PM> 158 62 62 62 194.158 62
492 <APC> 159 255 255 95 194.159 255 ##
493 <NON-BREAKING SPACE> 160 65 65 65 194.160 128.65
494 <INVERTED "!" > 161 170 170 170 194.161 128.66
495 <CENT SIGN> 162 74 74 176 194.162 128.67 ##
496 <POUND SIGN> 163 177 177 177 194.163 128.68
497 <CURRENCY SIGN> 164 159 159 159 194.164 128.69
498 <YEN SIGN> 165 178 178 178 194.165 128.70
499 <BROKEN BAR> 166 106 106 208 194.166 128.71 ##
500 <SECTION SIGN> 167 181 181 181 194.167 128.72
501 <DIAERESIS> 168 189 187 121 194.168 128.73 ** ##
502 <COPYRIGHT SIGN> 169 180 180 180 194.169 128.74
503 <FEMININE ORDINAL> 170 154 154 154 194.170 128.81
504 <LEFT POINTING GUILLEMET> 171 138 138 138 194.171 128.82
505 <NOT SIGN> 172 95 176 186 194.172 128.83 ** ##
506 <SOFT HYPHEN> 173 202 202 202 194.173 128.84
507 <REGISTERED TRADE MARK> 174 175 175 175 194.174 128.85
508 <MACRON> 175 188 188 161 194.175 128.86 ##
509 <DEGREE SIGN> 176 144 144 144 194.176 128.87
510 <PLUS-OR-MINUS SIGN> 177 143 143 143 194.177 128.88
511 <SUPERSCRIPT TWO> 178 234 234 234 194.178 128.89
512 <SUPERSCRIPT THREE> 179 250 250 250 194.179 128.98
513 <ACUTE ACCENT> 180 190 190 190 194.180 128.99
514 <MICRO SIGN> 181 160 160 160 194.181 128.100
515 <PARAGRAPH SIGN> 182 182 182 182 194.182 128.101
516 <MIDDLE DOT> 183 179 179 179 194.183 128.102
517 <CEDILLA> 184 157 157 157 194.184 128.103
518 <SUPERSCRIPT ONE> 185 218 218 218 194.185 128.104
519 <MASC. ORDINAL INDICATOR> 186 155 155 155 194.186 128.105
520 <RIGHT POINTING GUILLEMET> 187 139 139 139 194.187 128.106
521 <FRACTION ONE QUARTER> 188 183 183 183 194.188 128.112
522 <FRACTION ONE HALF> 189 184 184 184 194.189 128.113
523 <FRACTION THREE QUARTERS> 190 185 185 185 194.190 128.114
524 <INVERTED QUESTION MARK> 191 171 171 171 194.191 128.115
525 <A WITH GRAVE> 192 100 100 100 195.128 138.65
526 <A WITH ACUTE> 193 101 101 101 195.129 138.66
527 <A WITH CIRCUMFLEX> 194 98 98 98 195.130 138.67
528 <A WITH TILDE> 195 102 102 102 195.131 138.68
529 <A WITH DIAERESIS> 196 99 99 99 195.132 138.69
530 <A WITH RING ABOVE> 197 103 103 103 195.133 138.70
531 <CAPITAL LIGATURE AE> 198 158 158 158 195.134 138.71
532 <C WITH CEDILLA> 199 104 104 104 195.135 138.72
533 <E WITH GRAVE> 200 116 116 116 195.136 138.73
534 <E WITH ACUTE> 201 113 113 113 195.137 138.74
535 <E WITH CIRCUMFLEX> 202 114 114 114 195.138 138.81
536 <E WITH DIAERESIS> 203 115 115 115 195.139 138.82
537 <I WITH GRAVE> 204 120 120 120 195.140 138.83
538 <I WITH ACUTE> 205 117 117 117 195.141 138.84
539 <I WITH CIRCUMFLEX> 206 118 118 118 195.142 138.85
540 <I WITH DIAERESIS> 207 119 119 119 195.143 138.86
541 <CAPITAL LETTER ETH> 208 172 172 172 195.144 138.87
542 <N WITH TILDE> 209 105 105 105 195.145 138.88
543 <O WITH GRAVE> 210 237 237 237 195.146 138.89
544 <O WITH ACUTE> 211 238 238 238 195.147 138.98
545 <O WITH CIRCUMFLEX> 212 235 235 235 195.148 138.99
546 <O WITH TILDE> 213 239 239 239 195.149 138.100
547 <O WITH DIAERESIS> 214 236 236 236 195.150 138.101
548 <MULTIPLICATION SIGN> 215 191 191 191 195.151 138.102
549 <O WITH STROKE> 216 128 128 128 195.152 138.103
550 <U WITH GRAVE> 217 253 253 224 195.153 138.104 ##
551 <U WITH ACUTE> 218 254 254 254 195.154 138.105
552 <U WITH CIRCUMFLEX> 219 251 251 221 195.155 138.106 ##
553 <U WITH DIAERESIS> 220 252 252 252 195.156 138.112
554 <Y WITH ACUTE> 221 173 186 173 195.157 138.113 ** ##
555 <CAPITAL LETTER THORN> 222 174 174 174 195.158 138.114
556 <SMALL LETTER SHARP S> 223 89 89 89 195.159 138.115
557 <a WITH GRAVE> 224 68 68 68 195.160 139.65
558 <a WITH ACUTE> 225 69 69 69 195.161 139.66
559 <a WITH CIRCUMFLEX> 226 66 66 66 195.162 139.67
560 <a WITH TILDE> 227 70 70 70 195.163 139.68
561 <a WITH DIAERESIS> 228 67 67 67 195.164 139.69
562 <a WITH RING ABOVE> 229 71 71 71 195.165 139.70
563 <SMALL LIGATURE ae> 230 156 156 156 195.166 139.71
564 <c WITH CEDILLA> 231 72 72 72 195.167 139.72
565 <e WITH GRAVE> 232 84 84 84 195.168 139.73
566 <e WITH ACUTE> 233 81 81 81 195.169 139.74
567 <e WITH CIRCUMFLEX> 234 82 82 82 195.170 139.81
568 <e WITH DIAERESIS> 235 83 83 83 195.171 139.82
569 <i WITH GRAVE> 236 88 88 88 195.172 139.83
570 <i WITH ACUTE> 237 85 85 85 195.173 139.84
571 <i WITH CIRCUMFLEX> 238 86 86 86 195.174 139.85
572 <i WITH DIAERESIS> 239 87 87 87 195.175 139.86
573 <SMALL LETTER eth> 240 140 140 140 195.176 139.87
574 <n WITH TILDE> 241 73 73 73 195.177 139.88
575 <o WITH GRAVE> 242 205 205 205 195.178 139.89
576 <o WITH ACUTE> 243 206 206 206 195.179 139.98
577 <o WITH CIRCUMFLEX> 244 203 203 203 195.180 139.99
578 <o WITH TILDE> 245 207 207 207 195.181 139.100
579 <o WITH DIAERESIS> 246 204 204 204 195.182 139.101
580 <DIVISION SIGN> 247 225 225 225 195.183 139.102
581 <o WITH STROKE> 248 112 112 112 195.184 139.103
582 <u WITH GRAVE> 249 221 221 192 195.185 139.104 ##
583 <u WITH ACUTE> 250 222 222 222 195.186 139.105
584 <u WITH CIRCUMFLEX> 251 219 219 219 195.187 139.106
585 <u WITH DIAERESIS> 252 220 220 220 195.188 139.112
586 <y WITH ACUTE> 253 141 141 141 195.189 139.113
587 <SMALL LETTER thorn> 254 142 142 142 195.190 139.114
588 <y WITH DIAERESIS> 255 223 223 223 195.191 139.115
590 If you would rather see the above table in CCSID 0037 order rather than
591 ASCII + Latin-1 order then run the table through:
600 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\
602 -e 'END{print map{$_->[0]}' \
603 -e ' sort{$a->[1] <=> $b->[1]}' \
604 -e ' map{[$_,substr($_,34,3)]}@l;}' perlebcdic.pod
606 If you would rather see it in CCSID 1047 order then change the number
607 34 in the last line to 39, like this:
616 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\
618 -e 'END{print map{$_->[0]}' \
619 -e ' sort{$a->[1] <=> $b->[1]}' \
620 -e ' map{[$_,substr($_,39,3)]}@l;}' perlebcdic.pod
622 If you would rather see it in POSIX-BC order then change the number
623 39 in the last line to 44, like this:
632 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\
634 -e 'END{print map{$_->[0]}' \
635 -e ' sort{$a->[1] <=> $b->[1]}' \
636 -e ' map{[$_,substr($_,44,3)]}@l;}' perlebcdic.pod
639 =head1 IDENTIFYING CHARACTER CODE SETS
641 To determine the character set you are running under from perl one
642 could use the return value of ord() or chr() to test one or more
643 character values. For example:
645 $is_ascii = "A" eq chr(65);
646 $is_ebcdic = "A" eq chr(193);
648 Also, "\t" is a C<HORIZONTAL TABULATION> character so that:
650 $is_ascii = ord("\t") == 9;
651 $is_ebcdic = ord("\t") == 5;
653 To distinguish EBCDIC code pages try looking at one or more of
654 the characters that differ between them. For example:
656 $is_ebcdic_37 = "\n" eq chr(37);
657 $is_ebcdic_1047 = "\n" eq chr(21);
659 Or better still choose a character that is uniquely encoded in any
660 of the code sets, e.g.:
662 $is_ascii = ord('[') == 91;
663 $is_ebcdic_37 = ord('[') == 186;
664 $is_ebcdic_1047 = ord('[') == 173;
665 $is_ebcdic_POSIX_BC = ord('[') == 187;
667 However, it would be unwise to write tests such as:
669 $is_ascii = "\r" ne chr(13); # WRONG
670 $is_ascii = "\n" ne chr(10); # ILL ADVISED
672 Obviously the first of these will fail to distinguish most ASCII platforms
673 from either a CCSID 0037, a 1047, or a POSIX-BC EBCDIC platform since "\r" eq
674 chr(13) under all of those coded character sets. But note too that
675 because "\n" is chr(13) and "\r" is chr(10) on the Macintosh (which is an
676 ASCII platform) the second C<$is_ascii> test will lead to trouble there.
678 To determine whether or not perl was built under an EBCDIC
679 code page you can use the Config module like so:
682 $is_ebcdic = $Config{'ebcdic'} eq 'define';
688 In order to convert a string of characters from one character set to
689 another a simple list of numbers, such as in the right columns in the
690 above table, along with perl's tr/// operator is all that is needed.
691 The data in the table are in ASCII/Latin1 order, hence the EBCDIC columns
692 provide easy-to-use ASCII/Latin1 to EBCDIC operations that are also easily
695 For example, to convert ASCII/Latin1 to code page 037 take the output of the
696 second numbers column from the output of recipe 2 (modified to add '\'
697 characters), and use it in tr/// like so:
700 '\x00\x01\x02\x03\x37\x2D\x2E\x2F\x16\x05\x25\x0B\x0C\x0D\x0E\x0F' .
701 '\x10\x11\x12\x13\x3C\x3D\x32\x26\x18\x19\x3F\x27\x1C\x1D\x1E\x1F' .
702 '\x40\x5A\x7F\x7B\x5B\x6C\x50\x7D\x4D\x5D\x5C\x4E\x6B\x60\x4B\x61' .
703 '\xF0\xF1\xF2\xF3\xF4\xF5\xF6\xF7\xF8\xF9\x7A\x5E\x4C\x7E\x6E\x6F' .
704 '\x7C\xC1\xC2\xC3\xC4\xC5\xC6\xC7\xC8\xC9\xD1\xD2\xD3\xD4\xD5\xD6' .
705 '\xD7\xD8\xD9\xE2\xE3\xE4\xE5\xE6\xE7\xE8\xE9\xBA\xE0\xBB\xB0\x6D' .
706 '\x79\x81\x82\x83\x84\x85\x86\x87\x88\x89\x91\x92\x93\x94\x95\x96' .
707 '\x97\x98\x99\xA2\xA3\xA4\xA5\xA6\xA7\xA8\xA9\xC0\x4F\xD0\xA1\x07' .
708 '\x20\x21\x22\x23\x24\x15\x06\x17\x28\x29\x2A\x2B\x2C\x09\x0A\x1B' .
709 '\x30\x31\x1A\x33\x34\x35\x36\x08\x38\x39\x3A\x3B\x04\x14\x3E\xFF' .
710 '\x41\xAA\x4A\xB1\x9F\xB2\x6A\xB5\xBD\xB4\x9A\x8A\x5F\xCA\xAF\xBC' .
711 '\x90\x8F\xEA\xFA\xBE\xA0\xB6\xB3\x9D\xDA\x9B\x8B\xB7\xB8\xB9\xAB' .
712 '\x64\x65\x62\x66\x63\x67\x9E\x68\x74\x71\x72\x73\x78\x75\x76\x77' .
713 '\xAC\x69\xED\xEE\xEB\xEF\xEC\xBF\x80\xFD\xFE\xFB\xFC\xAD\xAE\x59' .
714 '\x44\x45\x42\x46\x43\x47\x9C\x48\x54\x51\x52\x53\x58\x55\x56\x57' .
715 '\x8C\x49\xCD\xCE\xCB\xCF\xCC\xE1\x70\xDD\xDE\xDB\xDC\x8D\x8E\xDF';
717 my $ebcdic_string = $ascii_string;
718 eval '$ebcdic_string =~ tr/\000-\377/' . $cp_037 . '/';
720 To convert from EBCDIC 037 to ASCII just reverse the order of the tr///
723 my $ascii_string = $ebcdic_string;
724 eval '$ascii_string =~ tr/' . $cp_037 . '/\000-\377/';
726 Similarly one could take the output of the third numbers column from recipe 2
727 to obtain a C<$cp_1047> table. The fourth numbers column of the output from
728 recipe 2 could provide a C<$cp_posix_bc> table suitable for transcoding as
731 If you wanted to see the inverse tables, you would first have to sort on the
732 desired numbers column as in recipes 4, 5 or 6, then take the output of the
733 first numbers column.
737 XPG operability often implies the presence of an I<iconv> utility
738 available from the shell or from the C library. Consult your system's
739 documentation for information on iconv.
741 On OS/390 or z/OS see the iconv(1) manpage. One way to invoke the iconv
742 shell utility from within perl would be to:
744 # OS/390 or z/OS example
745 $ascii_data = `echo '$ebcdic_data'| iconv -f IBM-1047 -t ISO8859-1`
749 # OS/390 or z/OS example
750 $ebcdic_data = `echo '$ascii_data'| iconv -f ISO8859-1 -t IBM-1047`
752 For other perl-based conversion options see the Convert::* modules on CPAN.
756 The OS/390 and z/OS C run-time libraries provide _atoe() and _etoa() functions.
758 =head1 OPERATOR DIFFERENCES
760 The C<..> range operator treats certain character ranges with
761 care on EBCDIC platforms. For example the following array
762 will have twenty six elements on either an EBCDIC platform
763 or an ASCII platform:
765 @alphabet = ('A'..'Z'); # $#alphabet == 25
767 The bitwise operators such as & ^ | may return different results
768 when operating on string or character data in a perl program running
769 on an EBCDIC platform than when run on an ASCII platform. Here is
770 an example adapted from the one in L<perlop>:
772 # EBCDIC-based examples
773 print "j p \n" ^ " a h"; # prints "JAPH\n"
774 print "JA" | " ph\n"; # prints "japh\n"
775 print "JAPH\nJunk" & "\277\277\277\277\277"; # prints "japh\n";
776 print 'p N$' ^ " E<H\n"; # prints "Perl\n";
778 An interesting property of the 32 C0 control characters
779 in the ASCII table is that they can "literally" be constructed
780 as control characters in perl, e.g. C<(chr(0)> eq C<\c@>)>
781 C<(chr(1)> eq C<\cA>)>, and so on. Perl on EBCDIC platforms has been
782 ported to take C<\c@> to chr(0) and C<\cA> to chr(1), etc. as well, but the
783 thirty three characters that result depend on which code page you are
784 using. The table below uses the standard acronyms for the controls.
785 The POSIX-BC and 1047 sets are
786 identical throughout this range and differ from the 0037 set at only
787 one spot (21 decimal). Note that the C<LINE FEED> character
788 may be generated by C<\cJ> on ASCII platforms but by C<\cU> on 1047 or POSIX-BC
789 platforms and cannot be generated as a C<"\c.letter."> control character on
790 0037 platforms. Note also that C<\c\> cannot be the final element in a string
791 or regex, as it will absorb the terminator. But C<\c\I<X>> is a C<FILE
792 SEPARATOR> concatenated with I<X> for all I<X>.
794 chr ord 8859-1 0037 1047 && POSIX-BC
795 -----------------------------------------------------------------------
797 \c@ 0 <NUL> <NUL> <NUL>
798 \cA 1 <SOH> <SOH> <SOH>
799 \cB 2 <STX> <STX> <STX>
800 \cC 3 <ETX> <ETX> <ETX>
801 \cD 4 <EOT> <ST> <ST>
802 \cE 5 <ENQ> <HT> <HT>
803 \cF 6 <ACK> <SSA> <SSA>
804 \cG 7 <BEL> <DEL> <DEL>
805 \cH 8 <BS> <EPA> <EPA>
807 \cJ 10 <LF> <SS2> <SS2>
808 \cK 11 <VT> <VT> <VT>
809 \cL 12 <FF> <FF> <FF>
810 \cM 13 <CR> <CR> <CR>
811 \cN 14 <SO> <SO> <SO>
812 \cO 15 <SI> <SI> <SI>
813 \cP 16 <DLE> <DLE> <DLE>
814 \cQ 17 <DC1> <DC1> <DC1>
815 \cR 18 <DC2> <DC2> <DC2>
816 \cS 19 <DC3> <DC3> <DC3>
817 \cT 20 <DC4> <OSC> <OSC>
818 \cU 21 <NAK> <NEL> <LF> **
819 \cV 22 <SYN> <BS> <BS>
820 \cW 23 <ETB> <ESA> <ESA>
821 \cX 24 <CAN> <CAN> <CAN>
822 \cY 25 <EOM> <EOM> <EOM>
823 \cZ 26 <SUB> <PU2> <PU2>
824 \c[ 27 <ESC> <SS3> <SS3>
825 \c\X 28 <FS>X <FS>X <FS>X
826 \c] 29 <GS> <GS> <GS>
827 \c^ 30 <RS> <RS> <RS>
828 \c_ 31 <US> <US> <US>
830 =head1 FUNCTION DIFFERENCES
836 chr() must be given an EBCDIC code number argument to yield a desired
837 character return value on an EBCDIC platform. For example:
839 $CAPITAL_LETTER_A = chr(193);
843 ord() will return EBCDIC code number values on an EBCDIC platform.
846 $the_number_193 = ord("A");
850 The c and C templates for pack() are dependent upon character set
851 encoding. Examples of usage on EBCDIC include:
853 $foo = pack("CCCC",193,194,195,196);
855 $foo = pack("C4",193,194,195,196);
858 $foo = pack("ccxxcc",193,194,195,196);
863 One must be careful with scalars and strings that are passed to
864 print that contain ASCII encodings. One common place
865 for this to occur is in the output of the MIME type header for
866 CGI script writing. For example, many perl programming guides
867 recommend something similar to:
869 print "Content-type:\ttext/html\015\012\015\012";
870 # this may be wrong on EBCDIC
872 Under the IBM OS/390 USS Web Server or WebSphere on z/OS for example
873 you should instead write that as:
875 print "Content-type:\ttext/html\r\n\r\n"; # OK for DGW et al
877 That is because the translation from EBCDIC to ASCII is done
878 by the web server in this case (such code will not be appropriate for
879 the Macintosh however). Consult your web server's documentation for
884 The formats that can convert characters to numbers and vice versa
885 will be different from their ASCII counterparts when executed
886 on an EBCDIC platform. Examples include:
888 printf("%c%c%c",193,194,195); # prints ABC
892 EBCDIC sort results may differ from ASCII sort results especially for
893 mixed case strings. This is discussed in more detail below.
897 See the discussion of printf() above. An example of the use
900 $CAPITAL_LETTER_A = sprintf("%c",193);
904 See the discussion of pack() above.
908 =head1 REGULAR EXPRESSION DIFFERENCES
910 As of perl 5.005_03 the letter range regular expressions such as
911 [A-Z] and [a-z] have been especially coded to not pick up gap
912 characters. For example, characters such as E<ocirc> C<o WITH CIRCUMFLEX>
913 that lie between I and J would not be matched by the
914 regular expression range C</[H-K]/>. This works in
915 the other direction, too, if either of the range end points is
916 explicitly numeric: C<[\x89-\x91]> will match C<\x8e>, even
917 though C<\x89> is C<i> and C<\x91 > is C<j>, and C<\x8e>
918 is a gap character from the alphabetic viewpoint.
920 If you do want to match the alphabet gap characters in a single octet
921 regular expression try matching the hex or octal code such
922 as C</\313/> on EBCDIC or C</\364/> on ASCII platforms to
923 have your regular expression match C<o WITH CIRCUMFLEX>.
925 Another construct to be wary of is the inappropriate use of hex or
926 octal constants in regular expressions. Consider the following
930 my $char = substr(shift,0,1);
931 $char =~ /[\000-\037]/;
935 my $char = substr(shift,0,1);
936 $char =~ /[\040-\176]/;
940 my $char = substr(shift,0,1);
945 my $char = substr(shift,0,1);
946 $char =~ /[\200-\237]/;
950 my $char = substr(shift,0,1);
951 $char =~ /[\240-\377]/;
954 The above would be adequate if the concern was only with numeric code points.
955 However, the concern may be with characters rather than code points
956 and on an EBCDIC platform it may be desirable for constructs such as
957 C<if (is_print_ascii("A")) {print "A is a printable character\n";}> to print
958 out the expected message. One way to represent the above collection
959 of character classification subs that is capable of working across the
960 four coded character sets discussed in this document is as follows:
963 my $char = substr(shift,0,1);
964 if (ord('^')==94) { # ascii
965 return $char =~ /[\000-\037]/;
967 if (ord('^')==176) { # 0037
968 return $char =~ /[\000-\003\067\055-\057\026\005\045\013-\023\074\075\062\046\030\031\077\047\034-\037]/;
970 if (ord('^')==95 || ord('^')==106) { # 1047 || posix-bc
971 return $char =~ /[\000-\003\067\055-\057\026\005\025\013-\023\074\075\062\046\030\031\077\047\034-\037]/;
976 my $char = substr(shift,0,1);
977 $char =~ /[ !"\#\$%&'()*+,\-.\/0-9:;<=>?\@A-Z[\\\]^_`a-z{|}~]/;
981 my $char = substr(shift,0,1);
982 if (ord('^')==94) { # ascii
983 return $char eq "\177";
986 return $char eq "\007";
991 my $char = substr(shift,0,1);
992 if (ord('^')==94) { # ascii
993 return $char =~ /[\200-\237]/;
995 if (ord('^')==176) { # 0037
996 return $char =~ /[\040-\044\025\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\377]/;
998 if (ord('^')==95) { # 1047
999 return $char =~ /[\040-\045\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\377]/;
1001 if (ord('^')==106) { # posix-bc
1003 /[\040-\045\006\027\050-\054\011\012\033\060\061\032\063-\066\010\070-\073\040\024\076\137]/;
1008 my $char = substr(shift,0,1);
1009 if (ord('^')==94) { # ascii
1010 return $char =~ /[\240-\377]/;
1012 if (ord('^')==176) { # 0037
1014 /[\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]/;
1016 if (ord('^')==95) { # 1047
1018 /[\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]/;
1020 if (ord('^')==106) { # posix-bc
1022 /[\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]/;
1026 Note however that only the C<Is_ascii_print()> sub is really independent
1027 of coded character set. Another way to write C<Is_latin_1()> would be
1028 to use the characters in the range explicitly:
1031 my $char = substr(shift,0,1);
1032 $char =~ /[ ¡¢£¤¥¦§¨©ª«¬®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ]/;
1035 Although that form may run into trouble in network transit (due to the
1036 presence of 8 bit characters) or on non ISO-Latin character sets.
1040 Most socket programming assumes ASCII character encodings in network
1041 byte order. Exceptions can include CGI script writing under a
1042 host web server where the server may take care of translation for you.
1043 Most host web servers convert EBCDIC data to ISO-8859-1 or Unicode on
1048 One big difference between ASCII-based character sets and EBCDIC ones
1049 are the relative positions of upper and lower case letters and the
1050 letters compared to the digits. If sorted on an ASCII-based platform the
1051 two-letter abbreviation for a physician comes before the two letter
1052 abbreviation for drive; that is:
1054 @sorted = sort(qw(Dr. dr.)); # @sorted holds ('Dr.','dr.') on ASCII,
1055 # but ('dr.','Dr.') on EBCDIC
1057 The property of lowercase before uppercase letters in EBCDIC is
1058 even carried to the Latin 1 EBCDIC pages such as 0037 and 1047.
1059 An example would be that E<Euml> C<E WITH DIAERESIS> (203) comes
1060 before E<euml> C<e WITH DIAERESIS> (235) on an ASCII platform, but
1061 the latter (83) comes before the former (115) on an EBCDIC platform.
1062 (Astute readers will note that the uppercase version of E<szlig>
1063 C<SMALL LETTER SHARP S> is simply "SS" and that the upper case version of
1064 E<yuml> C<y WITH DIAERESIS> is not in the 0..255 range but it is
1065 at U+x0178 in Unicode, or C<"\x{178}"> in a Unicode enabled Perl).
1067 The sort order will cause differences between results obtained on
1068 ASCII platforms versus EBCDIC platforms. What follows are some suggestions
1069 on how to deal with these differences.
1071 =head2 Ignore ASCII vs. EBCDIC sort differences.
1073 This is the least computationally expensive strategy. It may require
1074 some user education.
1076 =head2 MONO CASE then sort data.
1078 In order to minimize the expense of mono casing mixed-case text, try to
1079 C<tr///> towards the character set case most employed within the data.
1080 If the data are primarily UPPERCASE non Latin 1 then apply tr/[a-z]/[A-Z]/
1081 then sort(). If the data are primarily lowercase non Latin 1 then
1082 apply tr/[A-Z]/[a-z]/ before sorting. If the data are primarily UPPERCASE
1083 and include Latin-1 characters then apply:
1086 tr/[àáâãäåæçèéêëìíîïðñòóôõöøùúûüýþ]/[ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞ/;
1089 then sort(). Do note however that such Latin-1 manipulation does not
1090 address the E<yuml> C<y WITH DIAERESIS> character that will remain at
1091 code point 255 on ASCII platforms, but 223 on most EBCDIC platforms
1092 where it will sort to a place less than the EBCDIC numerals. With a
1093 Unicode-enabled Perl you might try:
1097 The strategy of mono casing data before sorting does not preserve the case
1098 of the data and may not be acceptable for that reason.
1100 =head2 Convert, sort data, then re convert.
1102 This is the most expensive proposition that does not employ a network
1105 =head2 Perform sorting on one type of platform only.
1107 This strategy can employ a network connection. As such
1108 it would be computationally expensive.
1110 =head1 TRANSFORMATION FORMATS
1112 There are a variety of ways of transforming data with an intra character set
1113 mapping that serve a variety of purposes. Sorting was discussed in the
1114 previous section and a few of the other more popular mapping techniques are
1117 =head2 URL decoding and encoding
1119 Note that some URLs have hexadecimal ASCII code points in them in an
1120 attempt to overcome character or protocol limitation issues. For example
1121 the tilde character is not on every keyboard hence a URL of the form:
1123 http://www.pvhp.com/~pvhp/
1125 may also be expressed as either of:
1127 http://www.pvhp.com/%7Epvhp/
1129 http://www.pvhp.com/%7epvhp/
1131 where 7E is the hexadecimal ASCII code point for '~'. Here is an example
1132 of decoding such a URL under CCSID 1047:
1134 $url = 'http://www.pvhp.com/%7Epvhp/';
1135 # this array assumes code page 1047
1137 0, 1, 2, 3, 55, 45, 46, 47, 22, 5, 21, 11, 12, 13, 14, 15,
1138 16, 17, 18, 19, 60, 61, 50, 38, 24, 25, 63, 39, 28, 29, 30, 31,
1139 64, 90,127,123, 91,108, 80,125, 77, 93, 92, 78,107, 96, 75, 97,
1140 240,241,242,243,244,245,246,247,248,249,122, 94, 76,126,110,111,
1141 124,193,194,195,196,197,198,199,200,201,209,210,211,212,213,214,
1142 215,216,217,226,227,228,229,230,231,232,233,173,224,189, 95,109,
1143 121,129,130,131,132,133,134,135,136,137,145,146,147,148,149,150,
1144 151,152,153,162,163,164,165,166,167,168,169,192, 79,208,161, 7,
1145 32, 33, 34, 35, 36, 37, 6, 23, 40, 41, 42, 43, 44, 9, 10, 27,
1146 48, 49, 26, 51, 52, 53, 54, 8, 56, 57, 58, 59, 4, 20, 62,255,
1147 65,170, 74,177,159,178,106,181,187,180,154,138,176,202,175,188,
1148 144,143,234,250,190,160,182,179,157,218,155,139,183,184,185,171,
1149 100,101, 98,102, 99,103,158,104,116,113,114,115,120,117,118,119,
1150 172,105,237,238,235,239,236,191,128,253,254,251,252,186,174, 89,
1151 68, 69, 66, 70, 67, 71,156, 72, 84, 81, 82, 83, 88, 85, 86, 87,
1152 140, 73,205,206,203,207,204,225,112,221,222,219,220,141,142,223
1154 $url =~ s/%([0-9a-fA-F]{2})/pack("c",$a2e_1047[hex($1)])/ge;
1156 Conversely, here is a partial solution for the task of encoding such
1157 a URL under the 1047 code page:
1159 $url = 'http://www.pvhp.com/~pvhp/';
1160 # this array assumes code page 1047
1162 0, 1, 2, 3,156, 9,134,127,151,141,142, 11, 12, 13, 14, 15,
1163 16, 17, 18, 19,157, 10, 8,135, 24, 25,146,143, 28, 29, 30, 31,
1164 128,129,130,131,132,133, 23, 27,136,137,138,139,140, 5, 6, 7,
1165 144,145, 22,147,148,149,150, 4,152,153,154,155, 20, 21,158, 26,
1166 32,160,226,228,224,225,227,229,231,241,162, 46, 60, 40, 43,124,
1167 38,233,234,235,232,237,238,239,236,223, 33, 36, 42, 41, 59, 94,
1168 45, 47,194,196,192,193,195,197,199,209,166, 44, 37, 95, 62, 63,
1169 248,201,202,203,200,205,206,207,204, 96, 58, 35, 64, 39, 61, 34,
1170 216, 97, 98, 99,100,101,102,103,104,105,171,187,240,253,254,177,
1171 176,106,107,108,109,110,111,112,113,114,170,186,230,184,198,164,
1172 181,126,115,116,117,118,119,120,121,122,161,191,208, 91,222,174,
1173 172,163,165,183,169,167,182,188,189,190,221,168,175, 93,180,215,
1174 123, 65, 66, 67, 68, 69, 70, 71, 72, 73,173,244,246,242,243,245,
1175 125, 74, 75, 76, 77, 78, 79, 80, 81, 82,185,251,252,249,250,255,
1176 92,247, 83, 84, 85, 86, 87, 88, 89, 90,178,212,214,210,211,213,
1177 48, 49, 50, 51, 52, 53, 54, 55, 56, 57,179,219,220,217,218,159
1179 # The following regular expression does not address the
1180 # mappings for: ('.' => '%2E', '/' => '%2F', ':' => '%3A')
1181 $url =~ s/([\t "#%&\(\),;<=>\?\@\[\\\]^`{|}~])/sprintf("%%%02X",$e2a_1047[ord($1)])/ge;
1183 where a more complete solution would split the URL into components
1184 and apply a full s/// substitution only to the appropriate parts.
1186 In the remaining examples a @e2a or @a2e array may be employed
1187 but the assignment will not be shown explicitly. For code page 1047
1188 you could use the @a2e_1047 or @e2a_1047 arrays just shown.
1190 =head2 uu encoding and decoding
1192 The C<u> template to pack() or unpack() will render EBCDIC data in EBCDIC
1193 characters equivalent to their ASCII counterparts. For example, the
1194 following will print "Yes indeed\n" on either an ASCII or EBCDIC computer:
1196 $all_byte_chrs = '';
1197 for (0..255) { $all_byte_chrs .= chr($_); }
1198 $uuencode_byte_chrs = pack('u', $all_byte_chrs);
1199 ($uu = <<'ENDOFHEREDOC') =~ s/^\s*//gm;
1200 M``$"`P0%!@<("0H+#`T.#Q`1$A,4%187&!D:&QP='A\@(2(C)"4F)R@I*BLL
1201 M+2XO,#$R,S0U-C<X.3H[/#T^/T!!0D-$149'2$E*2TQ-3D]045)35%565UA9
1202 M6EM<75Y?8&%B8V1E9F=H:6IK;&UN;W!Q<G-T=79W>'EZ>WQ]?G^`@8*#A(6&
1203 MAXB)BHN,C8Z/D)&2DY25EI>8F9J;G)V>GZ"AHJ.DI::GJ*FJJZRMKJ^PL;*S
1204 MM+6VM[BYNKN\O;Z_P,'"P\3%QL?(R<K+S,W.S]#1TM/4U=;7V-G:V]S=WM_@
1205 ?X>+CY.7FY^CIZNOL[>[O\/'R\_3U]O?X^?K[_/W^_P``
1207 if ($uuencode_byte_chrs eq $uu) {
1210 $uudecode_byte_chrs = unpack('u', $uuencode_byte_chrs);
1211 if ($uudecode_byte_chrs eq $all_byte_chrs) {
1215 Here is a very spartan uudecoder that will work on EBCDIC provided
1216 that the @e2a array is filled in appropriately:
1218 #!/usr/local/bin/perl
1219 @e2a = ( # this must be filled in
1221 $_ = <> until ($mode,$file) = /^begin\s*(\d*)\s*(\S*)/;
1222 open(OUT, "> $file") if $file ne "";
1226 next unless int(((($e2a[ord()] - 32 ) & 077) + 2) / 3) ==
1228 print OUT unpack("u", $_);
1231 chmod oct($mode), $file;
1234 =head2 Quoted-Printable encoding and decoding
1236 On ASCII-encoded platforms it is possible to strip characters outside of
1237 the printable set using:
1239 # This QP encoder works on ASCII only
1240 $qp_string =~ s/([=\x00-\x1F\x80-\xFF])/sprintf("=%02X",ord($1))/ge;
1242 Whereas a QP encoder that works on both ASCII and EBCDIC platforms
1243 would look somewhat like the following (where the EBCDIC branch @e2a
1244 array is omitted for brevity):
1246 if (ord('A') == 65) { # ASCII
1247 $delete = "\x7F"; # ASCII
1248 @e2a = (0 .. 255) # ASCII to ASCII identity map
1251 $delete = "\x07"; # EBCDIC
1252 @e2a = # EBCDIC to ASCII map (as shown above)
1255 s/([^ !"\#\$%&'()*+,\-.\/0-9:;<>?\@A-Z[\\\]^_`a-z{|}~$delete])/sprintf("=%02X",$e2a[ord($1)])/ge;
1257 (although in production code the substitutions might be done
1258 in the EBCDIC branch with the @e2a array and separately in the
1259 ASCII branch without the expense of the identity map).
1261 Such QP strings can be decoded with:
1263 # This QP decoder is limited to ASCII only
1264 $string =~ s/=([0-9A-Fa-f][0-9A-Fa-f])/chr hex $1/ge;
1265 $string =~ s/=[\n\r]+$//;
1267 Whereas a QP decoder that works on both ASCII and EBCDIC platforms
1268 would look somewhat like the following (where the @a2e array is
1269 omitted for brevity):
1271 $string =~ s/=([0-9A-Fa-f][0-9A-Fa-f])/chr $a2e[hex $1]/ge;
1272 $string =~ s/=[\n\r]+$//;
1274 =head2 Caesarean ciphers
1276 The practice of shifting an alphabet one or more characters for encipherment
1277 dates back thousands of years and was explicitly detailed by Gaius Julius
1278 Caesar in his B<Gallic Wars> text. A single alphabet shift is sometimes
1279 referred to as a rotation and the shift amount is given as a number $n after
1280 the string 'rot' or "rot$n". Rot0 and rot26 would designate identity maps
1281 on the 26-letter English version of the Latin alphabet. Rot13 has the
1282 interesting property that alternate subsequent invocations are identity maps
1283 (thus rot13 is its own non-trivial inverse in the group of 26 alphabet
1284 rotations). Hence the following is a rot13 encoder and decoder that will
1285 work on ASCII and EBCDIC platforms:
1287 #!/usr/local/bin/perl
1290 tr/n-za-mN-ZA-M/a-zA-Z/;
1296 perl -ne 'tr/n-za-mN-ZA-M/a-zA-Z/;print'
1299 =head1 Hashing order and checksums
1301 To the extent that it is possible to write code that depends on
1302 hashing order there may be differences between hashes as stored
1303 on an ASCII-based platform and hashes stored on an EBCDIC-based platform.
1306 =head1 I18N AND L10N
1308 Internationalization (I18N) and localization (L10N) are supported at least
1309 in principle even on EBCDIC platforms. The details are system-dependent
1310 and discussed under the L<perlebcdic/OS ISSUES> section below.
1312 =head1 MULTI-OCTET CHARACTER SETS
1314 Perl may work with an internal UTF-EBCDIC encoding form for wide characters
1315 on EBCDIC platforms in a manner analogous to the way that it works with
1316 the UTF-8 internal encoding form on ASCII based platforms.
1318 Legacy multi byte EBCDIC code pages XXX.
1322 There may be a few system-dependent issues
1323 of concern to EBCDIC Perl programmers.
1331 The PASE environment is a runtime environment for OS/400 that can run
1332 executables built for PowerPC AIX in OS/400; see L<perlos400>. PASE
1333 is ASCII-based, not EBCDIC-based as the ILE.
1343 Perl runs under Unix Systems Services or USS.
1349 B<chcp> is supported as a shell utility for displaying and changing
1350 one's code page. See also L<chcp(1)>.
1352 =item dataset access
1354 For sequential data set access try:
1356 my @ds_records = `cat //DSNAME`;
1360 my @ds_records = `cat //'HLQ.DSNAME'`;
1362 See also the OS390::Stdio module on CPAN.
1364 =item OS/390, z/OS iconv
1366 B<iconv> is supported as both a shell utility and a C RTL routine.
1367 See also the iconv(1) and iconv(3) manual pages.
1371 On OS/390 or z/OS see L<locale> for information on locales. The L10N files
1372 are in F</usr/nls/locale>. $Config{d_setlocale} is 'define' on OS/390
1383 This pod document contains literal Latin 1 characters and may encounter
1384 translation difficulties. In particular one popular nroff implementation
1385 was known to strip accented characters to their unaccented counterparts
1386 while attempting to view this document through the B<pod2man> program
1387 (for example, you may see a plain C<y> rather than one with a diaeresis
1388 as in E<yuml>). Another nroff truncated the resultant manpage at
1389 the first occurrence of 8 bit characters.
1391 Not all shells will allow multiple C<-e> string arguments to perl to
1392 be concatenated together properly as recipes 0, 2, 4, 5, and 6 might
1397 L<perllocale>, L<perlfunc>, L<perlunicode>, L<utf8>.
1401 L<http://anubis.dkuug.dk/i18n/charmaps>
1403 L<http://www.unicode.org/>
1405 L<http://www.unicode.org/unicode/reports/tr16/>
1407 L<http://www.wps.com/projects/codes/>
1408 B<ASCII: American Standard Code for Information Infiltration> Tom Jennings,
1411 B<The Unicode Standard, Version 3.0> The Unicode Consortium, Lisa Moore ed.,
1412 ISBN 0-201-61633-5, Addison Wesley Developers Press, February 2000.
1414 B<CDRA: IBM - Character Data Representation Architecture -
1415 Reference and Registry>, IBM SC09-2190-00, December 1996.
1417 "Demystifying Character Sets", Andrea Vine, Multilingual Computing
1418 & Technology, B<#26 Vol. 10 Issue 4>, August/September 1999;
1419 ISSN 1523-0309; Multilingual Computing Inc. Sandpoint ID, USA.
1421 B<Codes, Ciphers, and Other Cryptic and Clandestine Communication>
1422 Fred B. Wrixon, ISBN 1-57912-040-7, Black Dog & Leventhal Publishers,
1425 L<http://www.bobbemer.com/P-BIT.HTM>
1426 B<IBM - EBCDIC and the P-bit; The biggest Computer Goof Ever> Robert Bemer.
1430 15 April 2001: added UTF-8 and UTF-EBCDIC to main table, pvhp.
1434 Peter Prymmer pvhp@best.com wrote this in 1999 and 2000
1435 with CCSID 0819 and 0037 help from Chris Leach and
1436 AndrE<eacute> Pirard A.Pirard@ulg.ac.be as well as POSIX-BC
1437 help from Thomas Dorner Thomas.Dorner@start.de.
1438 Thanks also to Vickie Cooper, Philip Newton, William Raffloer, and
1439 Joe Smith. Trademarks, registered trademarks, service marks and
1440 registered service marks used in this document are the property of
1441 their respective owners.