1=encoding utf8 2 3=head1 NAME 4 5perlebcdic - Considerations for running Perl on EBCDIC platforms 6 7=head1 DESCRIPTION 8 9An exploration of some of the issues facing Perl programmers 10on EBCDIC based computers. 11 12Portions of this document that are still incomplete are marked with XXX. 13 14Early Perl versions worked on some EBCDIC machines, but the last known 15version that ran on EBCDIC was v5.8.7, until v5.22, when the Perl core 16again works on z/OS. Theoretically, it could work on OS/400 or Siemens' 17BS2000 (or their successors), but this is untested. In v5.22 and 5.24, 18not all 19the modules found on CPAN but shipped with core Perl work on z/OS. 20 21If you want to use Perl on a non-z/OS EBCDIC machine, please let us know 22at L<https://github.com/Perl/perl5/issues>. 23 24Writing Perl on an EBCDIC platform is really no different than writing 25on an L</ASCII> one, but with different underlying numbers, as we'll see 26shortly. You'll have to know something about those L</ASCII> platforms 27because the documentation is biased and will frequently use example 28numbers that don't apply to EBCDIC. There are also very few CPAN 29modules that are written for EBCDIC and which don't work on ASCII; 30instead the vast majority of CPAN modules are written for ASCII, and 31some may happen to work on EBCDIC, while a few have been designed to 32portably work on both. 33 34If your code just uses the 52 letters A-Z and a-z, plus SPACE, the 35digits 0-9, and the punctuation characters that Perl uses, plus a few 36controls that are denoted by escape sequences like C<\n> and C<\t>, then 37there's nothing special about using Perl, and your code may very well 38work on an ASCII machine without change. 39 40But if you write code that uses C<\005> to mean a TAB or C<\xC1> to mean 41an "A", or C<\xDF> to mean a "E<yuml>" (small C<"y"> with a diaeresis), 42then your code may well work on your EBCDIC platform, but not on an 43ASCII one. That's fine to do if no one will ever want to run your code 44on an ASCII platform; but the bias in this document will be towards writing 45code portable between EBCDIC and ASCII systems. Again, if every 46character you care about is easily enterable from your keyboard, you 47don't have to know anything about ASCII, but many keyboards don't easily 48allow you to directly enter, say, the character C<\xDF>, so you have to 49specify it indirectly, such as by using the C<"\xDF"> escape sequence. 50In those cases it's easiest to know something about the ASCII/Unicode 51character sets. If you know that the small "E<yuml>" is C<U+00FF>, then 52you can instead specify it as C<"\N{U+FF}">, and have the computer 53automatically translate it to C<\xDF> on your platform, and leave it as 54C<\xFF> on ASCII ones. Or you could specify it by name, C<\N{LATIN 55SMALL LETTER Y WITH DIAERESIS> and not have to know the numbers. 56Either way works, but both require familiarity with Unicode. 57 58=head1 COMMON CHARACTER CODE SETS 59 60=head2 ASCII 61 62The American Standard Code for Information Interchange (ASCII or 63US-ASCII) is a set of 64integers running from 0 to 127 (decimal) that have standardized 65interpretations by the computers which use ASCII. For example, 65 means 66the letter "A". 67The range 0..127 can be covered by setting various bits in a 7-bit binary 68digit, hence the set is sometimes referred to as "7-bit ASCII". 69ASCII was described by the American National Standards Institute 70document ANSI X3.4-1986. It was also described by ISO 646:1991 71(with localization for currency symbols). The full ASCII set is 72given in the table L<below|/recipe 3> as the first 128 elements. 73Languages that 74can be written adequately with the characters in ASCII include 75English, Hawaiian, Indonesian, Swahili and some Native American 76languages. 77 78Most non-EBCDIC character sets are supersets of ASCII. That is the 79integers 0-127 mean what ASCII says they mean. But integers 128 and 80above are specific to the character set. 81 82Many of these fit entirely into 8 bits, using ASCII as 0-127, while 83specifying what 128-255 mean, and not using anything above 255. 84Thus, these are single-byte (or octet if you prefer) character sets. 85One important one (since Unicode is a superset of it) is the ISO 8859-1 86character set. 87 88=head2 ISO 8859 89 90The ISO 8859-I<B<$n>> are a collection of character code sets from the 91International Organization for Standardization (ISO), each of which adds 92characters to the ASCII set that are typically found in various 93languages, many of which are based on the Roman, or Latin, alphabet. 94Most are for European languages, but there are also ones for Arabic, 95Greek, Hebrew, and Thai. There are good references on the web about 96all these. 97 98=head2 Latin 1 (ISO 8859-1) 99 100A particular 8-bit extension to ASCII that includes grave and acute 101accented Latin characters. Languages that can employ ISO 8859-1 102include all the languages covered by ASCII as well as Afrikaans, 103Albanian, Basque, Catalan, Danish, Faroese, Finnish, Norwegian, 104Portuguese, Spanish, and Swedish. Dutch is covered albeit without 105the ij ligature. French is covered too but without the oe ligature. 106German can use ISO 8859-1 but must do so without German-style 107quotation marks. This set is based on Western European extensions 108to ASCII and is commonly encountered in world wide web work. 109In IBM character code set identification terminology, ISO 8859-1 is 110also known as CCSID 819 (or sometimes 0819 or even 00819). 111 112=head2 EBCDIC 113 114The Extended Binary Coded Decimal Interchange Code refers to a 115large collection of single- and multi-byte coded character sets that are 116quite different from ASCII and ISO 8859-1, and are all slightly 117different from each other; they typically run on host computers. The 118EBCDIC encodings derive from 8-bit byte extensions of Hollerith punched 119card encodings, which long predate ASCII. The layout on the 120cards was such that high bits were set for the upper and lower case 121alphabetic 122characters C<[a-z]> and C<[A-Z]>, but there were gaps within each Latin 123alphabet range, visible in the table L<below|/recipe 3>. These gaps can 124cause complications. 125 126Some IBM EBCDIC character sets may be known by character code set 127identification numbers (CCSID numbers) or code page numbers. 128 129Perl can be compiled on platforms that run any of three commonly used EBCDIC 130character sets, listed below. 131 132=head3 The 13 variant characters 133 134Among IBM EBCDIC character code sets there are 13 characters that 135are often mapped to different integer values. Those characters 136are known as the 13 "variant" characters and are: 137 138 \ [ ] { } ^ ~ ! # | $ @ ` 139 140When Perl is compiled for a platform, it looks at all of these characters to 141guess which EBCDIC character set the platform uses, and adapts itself 142accordingly to that platform. If the platform uses a character set that is not 143one of the three Perl knows about, Perl will either fail to compile, or 144mistakenly and silently choose one of the three. 145 146The Line Feed (LF) character is actually a 14th variant character, and 147Perl checks for that as well. 148 149=head3 EBCDIC code sets recognized by Perl 150 151=over 152 153=item B<0037> 154 155Character code set ID 0037 is a mapping of the ASCII plus Latin-1 156characters (i.e. ISO 8859-1) to an EBCDIC set. 0037 is used 157in North American English locales on the OS/400 operating system 158that runs on AS/400 computers. CCSID 0037 differs from ISO 8859-1 159in 236 places; in other words they agree on only 20 code point values. 160 161=item B<1047> 162 163Character code set ID 1047 is also a mapping of the ASCII plus 164Latin-1 characters (i.e. ISO 8859-1) to an EBCDIC set. 1047 is 165used under Unix System Services for OS/390 or z/OS, and OpenEdition 166for VM/ESA. CCSID 1047 differs from CCSID 0037 in eight places, 167and from ISO 8859-1 in 236. 168 169=item B<POSIX-BC> 170 171The EBCDIC code page in use on Siemens' BS2000 system is distinct from 1721047 and 0037. It is identified below as the POSIX-BC set. 173Like 0037 and 1047, it is the same as ISO 8859-1 in 20 code point 174values. 175 176=back 177 178=head2 Unicode code points versus EBCDIC code points 179 180In Unicode terminology a I<code point> is the number assigned to a 181character: for example, in EBCDIC the character "A" is usually assigned 182the number 193. In Unicode, the character "A" is assigned the number 65. 183All the code points in ASCII and Latin-1 (ISO 8859-1) have the same 184meaning in Unicode. All three of the recognized EBCDIC code sets have 185256 code points, and in each code set, all 256 code points are mapped to 186equivalent Latin1 code points. Obviously, "A" will map to "A", "B" => 187"B", "%" => "%", etc., for all printable characters in Latin1 and these 188code pages. 189 190It also turns out that EBCDIC has nearly precise equivalents for the 191ASCII/Latin1 C0 controls and the DELETE control. (The C0 controls are 192those whose ASCII code points are 0..0x1F; things like TAB, ACK, BEL, 193etc.) A mapping is set up between these ASCII/EBCDIC controls. There 194isn't such a precise mapping between the C1 controls on ASCII platforms 195and the remaining EBCDIC controls. What has been done is to map these 196controls, mostly arbitrarily, to some otherwise unmatched character in 197the other character set. Most of these are very very rarely used 198nowadays in EBCDIC anyway, and their names have been dropped, without 199much complaint. For example the EO (Eight Ones) EBCDIC control 200(consisting of eight one bits = 0xFF) is mapped to the C1 APC control 201(0x9F), and you can't use the name "EO". 202 203The EBCDIC controls provide three possible line terminator characters, 204CR (0x0D), LF (0x25), and NL (0x15). On ASCII platforms, the symbols 205"NL" and "LF" refer to the same character, but in strict EBCDIC 206terminology they are different ones. The EBCDIC NL is mapped to the C1 207control called "NEL" ("Next Line"; here's a case where the mapping makes 208quite a bit of sense, and hence isn't just arbitrary). On some EBCDIC 209platforms, this NL or NEL is the typical line terminator. This is true 210of z/OS and BS2000. In these platforms, the C compilers will swap the 211LF and NEL code points, so that C<"\n"> is 0x15, and refers to NL. Perl 212does that too; you can see it in the code chart L<below|/recipe 3>. 213This makes things generally "just work" without you even having to be 214aware that there is a swap. 215 216=head2 Unicode and UTF 217 218UTF stands for "Unicode Transformation Format". 219UTF-8 is an encoding of Unicode into a sequence of 8-bit byte chunks, based on 220ASCII and Latin-1. 221The length of a sequence required to represent a Unicode code point 222depends on the ordinal number of that code point, 223with larger numbers requiring more bytes. 224UTF-EBCDIC is like UTF-8, but based on EBCDIC. 225They are enough alike that often, casual usage will conflate the two 226terms, and use "UTF-8" to mean both the UTF-8 found on ASCII platforms, 227and the UTF-EBCDIC found on EBCDIC ones. 228 229You may see the term "invariant" character or code point. 230This simply means that the character has the same numeric 231value and representation when encoded in UTF-8 (or UTF-EBCDIC) as when 232not. (Note that this is a very different concept from L</The 13 variant 233characters> mentioned above. Careful prose will use the term "UTF-8 234invariant" instead of just "invariant", but most often you'll see just 235"invariant".) For example, the ordinal value of "A" is 193 in most 236EBCDIC code pages, and also is 193 when encoded in UTF-EBCDIC. All 237UTF-8 (or UTF-EBCDIC) variant code points occupy at least two bytes when 238encoded in UTF-8 (or UTF-EBCDIC); by definition, the UTF-8 (or 239UTF-EBCDIC) invariant code points are exactly one byte whether encoded 240in UTF-8 (or UTF-EBCDIC), or not. (By now you see why people typically 241just say "UTF-8" when they also mean "UTF-EBCDIC". For the rest of this 242document, we'll mostly be casual about it too.) 243In ASCII UTF-8, the code points corresponding to the lowest 128 244ordinal numbers (0 - 127: the ASCII characters) are invariant. 245In UTF-EBCDIC, there are 160 invariant characters. 246(If you care, the EBCDIC invariants are those characters 247which have ASCII equivalents, plus those that correspond to 248the C1 controls (128 - 159 on ASCII platforms).) 249 250A string encoded in UTF-EBCDIC may be longer (very rarely shorter) than 251one encoded in UTF-8. Perl extends both UTF-8 and UTF-EBCDIC so that 252they can encode code points above the Unicode maximum of U+10FFFF. Both 253extensions are constructed to allow encoding of any code point that fits 254in a 64-bit word. 255 256UTF-EBCDIC is defined by 257L<Unicode Technical Report #16|https://www.unicode.org/reports/tr16> 258(often referred to as just TR16). 259It is defined based on CCSID 1047, not allowing for the differences for 260other code pages. This allows for easy interchange of text between 261computers running different code pages, but makes it unusable, without 262adaptation, for Perl on those other code pages. 263 264The reason for this unusability is that a fundamental assumption of Perl 265is that the characters it cares about for parsing and lexical analysis 266are the same whether or not the text is in UTF-8. For example, Perl 267expects the character C<"["> to have the same representation, no matter 268if the string containing it (or program text) is UTF-8 encoded or not. 269To ensure this, Perl adapts UTF-EBCDIC to the particular code page so 270that all characters it expects to be UTF-8 invariant are in fact UTF-8 271invariant. This means that text generated on a computer running one 272version of Perl's UTF-EBCDIC has to be translated to be intelligible to 273a computer running another. 274 275TR16 implies a method to extend UTF-EBCDIC to encode points up through 276S<C<2 ** 31 - 1>>. Perl uses this method for code points up through 277S<C<2 ** 30 - 1>>, but uses an incompatible method for larger ones, to 278enable it to handle much larger code points than otherwise. 279 280=head2 Using Encode 281 282Starting from Perl 5.8 you can use the standard module Encode 283to translate from EBCDIC to Latin-1 code points. 284Encode knows about more EBCDIC character sets than Perl can currently 285be compiled to run on. 286 287 use Encode 'from_to'; 288 289 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' ); 290 291 # $a is in EBCDIC code points 292 from_to($a, $ebcdic{ord '^'}, 'latin1'); 293 # $a is ISO 8859-1 code points 294 295and from Latin-1 code points to EBCDIC code points 296 297 use Encode 'from_to'; 298 299 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' ); 300 301 # $a is ISO 8859-1 code points 302 from_to($a, 'latin1', $ebcdic{ord '^'}); 303 # $a is in EBCDIC code points 304 305For doing I/O it is suggested that you use the autotranslating features 306of PerlIO, see L<perluniintro>. 307 308Since version 5.8 Perl uses the PerlIO I/O library. This enables 309you to use different encodings per IO channel. For example you may use 310 311 use Encode; 312 open($f, ">:encoding(ascii)", "test.ascii"); 313 print $f "Hello World!\n"; 314 open($f, ">:encoding(cp37)", "test.ebcdic"); 315 print $f "Hello World!\n"; 316 open($f, ">:encoding(latin1)", "test.latin1"); 317 print $f "Hello World!\n"; 318 open($f, ">:encoding(utf8)", "test.utf8"); 319 print $f "Hello World!\n"; 320 321to get four files containing "Hello World!\n" in ASCII, CP 0037 EBCDIC, 322ISO 8859-1 (Latin-1) (in this example identical to ASCII since only ASCII 323characters were printed), and 324UTF-EBCDIC (in this example identical to normal EBCDIC since only characters 325that don't differ between EBCDIC and UTF-EBCDIC were printed). See the 326documentation of L<Encode::PerlIO> for details. 327 328As the PerlIO layer uses raw IO (bytes) internally, all this totally 329ignores things like the type of your filesystem (ASCII or EBCDIC). 330 331=head1 SINGLE OCTET TABLES 332 333The following tables list the ASCII and Latin 1 ordered sets including 334the subsets: C0 controls (0..31), ASCII graphics (32..7e), delete (7f), 335C1 controls (80..9f), and Latin-1 (a.k.a. ISO 8859-1) (a0..ff). In the 336table names of the Latin 1 337extensions to ASCII have been labelled with character names roughly 338corresponding to I<The Unicode Standard, Version 6.1> albeit with 339substitutions such as C<s/LATIN//> and C<s/VULGAR//> in all cases; 340S<C<s/CAPITAL LETTER//>> in some cases; and 341S<C<s/SMALL LETTER ([A-Z])/\l$1/>> in some other 342cases. Controls are listed using their Unicode 6.2 abbreviations. 343The differences between the 0037 and 1047 sets are 344flagged with C<**>. The differences between the 1047 and POSIX-BC sets 345are flagged with C<##.> All C<ord()> numbers listed are decimal. If you 346would rather see this table listing octal values, then run the table 347(that is, the pod source text of this document, since this recipe may not 348work with a pod2_other_format translation) through: 349 350=over 4 351 352=item recipe 0 353 354=back 355 356 perl -ne 'if(/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \ 357 -e '{printf("%s%-5.03o%-5.03o%-5.03o%.03o\n",$1,$2,$3,$4,$5)}' \ 358 perlebcdic.pod 359 360If you want to retain the UTF-x code points then in script form you 361might want to write: 362 363=over 4 364 365=item recipe 1 366 367=back 368 369 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!"; 370 while (<FH>) { 371 if (/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*) 372 \s+(\d+)\.?(\d*)/x) 373 { 374 if ($7 ne '' && $9 ne '') { 375 printf( 376 "%s%-5.03o%-5.03o%-5.03o%-5.03o%-3o.%-5o%-3o.%.03o\n", 377 $1,$2,$3,$4,$5,$6,$7,$8,$9); 378 } 379 elsif ($7 ne '') { 380 printf("%s%-5.03o%-5.03o%-5.03o%-5.03o%-3o.%-5o%.03o\n", 381 $1,$2,$3,$4,$5,$6,$7,$8); 382 } 383 else { 384 printf("%s%-5.03o%-5.03o%-5.03o%-5.03o%-5.03o%.03o\n", 385 $1,$2,$3,$4,$5,$6,$8); 386 } 387 } 388 } 389 390If you would rather see this table listing hexadecimal values then 391run the table through: 392 393=over 4 394 395=item recipe 2 396 397=back 398 399 perl -ne 'if(/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \ 400 -e '{printf("%s%-5.02X%-5.02X%-5.02X%.02X\n",$1,$2,$3,$4,$5)}' \ 401 perlebcdic.pod 402 403Or, in order to retain the UTF-x code points in hexadecimal: 404 405=over 4 406 407=item recipe 3 408 409=back 410 411 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!"; 412 while (<FH>) { 413 if (/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*) 414 \s+(\d+)\.?(\d*)/x) 415 { 416 if ($7 ne '' && $9 ne '') { 417 printf( 418 "%s%-5.02X%-5.02X%-5.02X%-5.02X%-2X.%-6.02X%02X.%02X\n", 419 $1,$2,$3,$4,$5,$6,$7,$8,$9); 420 } 421 elsif ($7 ne '') { 422 printf("%s%-5.02X%-5.02X%-5.02X%-5.02X%-2X.%-6.02X%02X\n", 423 $1,$2,$3,$4,$5,$6,$7,$8); 424 } 425 else { 426 printf("%s%-5.02X%-5.02X%-5.02X%-5.02X%-5.02X%02X\n", 427 $1,$2,$3,$4,$5,$6,$8); 428 } 429 } 430 } 431 432 433 ISO 434 8859-1 POS- CCSID 435 CCSID CCSID CCSID IX- 1047 436 chr 0819 0037 1047 BC UTF-8 UTF-EBCDIC 437 --------------------------------------------------------------------- 438 <NUL> 0 0 0 0 0 0 439 <SOH> 1 1 1 1 1 1 440 <STX> 2 2 2 2 2 2 441 <ETX> 3 3 3 3 3 3 442 <EOT> 4 55 55 55 4 55 443 <ENQ> 5 45 45 45 5 45 444 <ACK> 6 46 46 46 6 46 445 <BEL> 7 47 47 47 7 47 446 <BS> 8 22 22 22 8 22 447 <HT> 9 5 5 5 9 5 448 <LF> 10 37 21 21 10 21 ** 449 <VT> 11 11 11 11 11 11 450 <FF> 12 12 12 12 12 12 451 <CR> 13 13 13 13 13 13 452 <SO> 14 14 14 14 14 14 453 <SI> 15 15 15 15 15 15 454 <DLE> 16 16 16 16 16 16 455 <DC1> 17 17 17 17 17 17 456 <DC2> 18 18 18 18 18 18 457 <DC3> 19 19 19 19 19 19 458 <DC4> 20 60 60 60 20 60 459 <NAK> 21 61 61 61 21 61 460 <SYN> 22 50 50 50 22 50 461 <ETB> 23 38 38 38 23 38 462 <CAN> 24 24 24 24 24 24 463 <EOM> 25 25 25 25 25 25 464 <SUB> 26 63 63 63 26 63 465 <ESC> 27 39 39 39 27 39 466 <FS> 28 28 28 28 28 28 467 <GS> 29 29 29 29 29 29 468 <RS> 30 30 30 30 30 30 469 <US> 31 31 31 31 31 31 470 <SPACE> 32 64 64 64 32 64 471 ! 33 90 90 90 33 90 472 " 34 127 127 127 34 127 473 # 35 123 123 123 35 123 474 $ 36 91 91 91 36 91 475 % 37 108 108 108 37 108 476 & 38 80 80 80 38 80 477 ' 39 125 125 125 39 125 478 ( 40 77 77 77 40 77 479 ) 41 93 93 93 41 93 480 * 42 92 92 92 42 92 481 + 43 78 78 78 43 78 482 , 44 107 107 107 44 107 483 - 45 96 96 96 45 96 484 . 46 75 75 75 46 75 485 / 47 97 97 97 47 97 486 0 48 240 240 240 48 240 487 1 49 241 241 241 49 241 488 2 50 242 242 242 50 242 489 3 51 243 243 243 51 243 490 4 52 244 244 244 52 244 491 5 53 245 245 245 53 245 492 6 54 246 246 246 54 246 493 7 55 247 247 247 55 247 494 8 56 248 248 248 56 248 495 9 57 249 249 249 57 249 496 : 58 122 122 122 58 122 497 ; 59 94 94 94 59 94 498 < 60 76 76 76 60 76 499 = 61 126 126 126 61 126 500 > 62 110 110 110 62 110 501 ? 63 111 111 111 63 111 502 @ 64 124 124 124 64 124 503 A 65 193 193 193 65 193 504 B 66 194 194 194 66 194 505 C 67 195 195 195 67 195 506 D 68 196 196 196 68 196 507 E 69 197 197 197 69 197 508 F 70 198 198 198 70 198 509 G 71 199 199 199 71 199 510 H 72 200 200 200 72 200 511 I 73 201 201 201 73 201 512 J 74 209 209 209 74 209 513 K 75 210 210 210 75 210 514 L 76 211 211 211 76 211 515 M 77 212 212 212 77 212 516 N 78 213 213 213 78 213 517 O 79 214 214 214 79 214 518 P 80 215 215 215 80 215 519 Q 81 216 216 216 81 216 520 R 82 217 217 217 82 217 521 S 83 226 226 226 83 226 522 T 84 227 227 227 84 227 523 U 85 228 228 228 85 228 524 V 86 229 229 229 86 229 525 W 87 230 230 230 87 230 526 X 88 231 231 231 88 231 527 Y 89 232 232 232 89 232 528 Z 90 233 233 233 90 233 529 [ 91 186 173 187 91 173 ** ## 530 \ 92 224 224 188 92 224 ## 531 ] 93 187 189 189 93 189 ** 532 ^ 94 176 95 106 94 95 ** ## 533 _ 95 109 109 109 95 109 534 ` 96 121 121 74 96 121 ## 535 a 97 129 129 129 97 129 536 b 98 130 130 130 98 130 537 c 99 131 131 131 99 131 538 d 100 132 132 132 100 132 539 e 101 133 133 133 101 133 540 f 102 134 134 134 102 134 541 g 103 135 135 135 103 135 542 h 104 136 136 136 104 136 543 i 105 137 137 137 105 137 544 j 106 145 145 145 106 145 545 k 107 146 146 146 107 146 546 l 108 147 147 147 108 147 547 m 109 148 148 148 109 148 548 n 110 149 149 149 110 149 549 o 111 150 150 150 111 150 550 p 112 151 151 151 112 151 551 q 113 152 152 152 113 152 552 r 114 153 153 153 114 153 553 s 115 162 162 162 115 162 554 t 116 163 163 163 116 163 555 u 117 164 164 164 117 164 556 v 118 165 165 165 118 165 557 w 119 166 166 166 119 166 558 x 120 167 167 167 120 167 559 y 121 168 168 168 121 168 560 z 122 169 169 169 122 169 561 { 123 192 192 251 123 192 ## 562 | 124 79 79 79 124 79 563 } 125 208 208 253 125 208 ## 564 ~ 126 161 161 255 126 161 ## 565 <DEL> 127 7 7 7 127 7 566 <PAD> 128 32 32 32 194.128 32 567 <HOP> 129 33 33 33 194.129 33 568 <BPH> 130 34 34 34 194.130 34 569 <NBH> 131 35 35 35 194.131 35 570 <IND> 132 36 36 36 194.132 36 571 <NEL> 133 21 37 37 194.133 37 ** 572 <SSA> 134 6 6 6 194.134 6 573 <ESA> 135 23 23 23 194.135 23 574 <HTS> 136 40 40 40 194.136 40 575 <HTJ> 137 41 41 41 194.137 41 576 <VTS> 138 42 42 42 194.138 42 577 <PLD> 139 43 43 43 194.139 43 578 <PLU> 140 44 44 44 194.140 44 579 <RI> 141 9 9 9 194.141 9 580 <SS2> 142 10 10 10 194.142 10 581 <SS3> 143 27 27 27 194.143 27 582 <DCS> 144 48 48 48 194.144 48 583 <PU1> 145 49 49 49 194.145 49 584 <PU2> 146 26 26 26 194.146 26 585 <STS> 147 51 51 51 194.147 51 586 <CCH> 148 52 52 52 194.148 52 587 <MW> 149 53 53 53 194.149 53 588 <SPA> 150 54 54 54 194.150 54 589 <EPA> 151 8 8 8 194.151 8 590 <SOS> 152 56 56 56 194.152 56 591 <SGC> 153 57 57 57 194.153 57 592 <SCI> 154 58 58 58 194.154 58 593 <CSI> 155 59 59 59 194.155 59 594 <ST> 156 4 4 4 194.156 4 595 <OSC> 157 20 20 20 194.157 20 596 <PM> 158 62 62 62 194.158 62 597 <APC> 159 255 255 95 194.159 255 ## 598 <NON-BREAKING SPACE> 160 65 65 65 194.160 128.65 599 <INVERTED "!" > 161 170 170 170 194.161 128.66 600 <CENT SIGN> 162 74 74 176 194.162 128.67 ## 601 <POUND SIGN> 163 177 177 177 194.163 128.68 602 <CURRENCY SIGN> 164 159 159 159 194.164 128.69 603 <YEN SIGN> 165 178 178 178 194.165 128.70 604 <BROKEN BAR> 166 106 106 208 194.166 128.71 ## 605 <SECTION SIGN> 167 181 181 181 194.167 128.72 606 <DIAERESIS> 168 189 187 121 194.168 128.73 ** ## 607 <COPYRIGHT SIGN> 169 180 180 180 194.169 128.74 608 <FEMININE ORDINAL> 170 154 154 154 194.170 128.81 609 <LEFT POINTING GUILLEMET> 171 138 138 138 194.171 128.82 610 <NOT SIGN> 172 95 176 186 194.172 128.83 ** ## 611 <SOFT HYPHEN> 173 202 202 202 194.173 128.84 612 <REGISTERED TRADE MARK> 174 175 175 175 194.174 128.85 613 <MACRON> 175 188 188 161 194.175 128.86 ## 614 <DEGREE SIGN> 176 144 144 144 194.176 128.87 615 <PLUS-OR-MINUS SIGN> 177 143 143 143 194.177 128.88 616 <SUPERSCRIPT TWO> 178 234 234 234 194.178 128.89 617 <SUPERSCRIPT THREE> 179 250 250 250 194.179 128.98 618 <ACUTE ACCENT> 180 190 190 190 194.180 128.99 619 <MICRO SIGN> 181 160 160 160 194.181 128.100 620 <PARAGRAPH SIGN> 182 182 182 182 194.182 128.101 621 <MIDDLE DOT> 183 179 179 179 194.183 128.102 622 <CEDILLA> 184 157 157 157 194.184 128.103 623 <SUPERSCRIPT ONE> 185 218 218 218 194.185 128.104 624 <MASC. ORDINAL INDICATOR> 186 155 155 155 194.186 128.105 625 <RIGHT POINTING GUILLEMET> 187 139 139 139 194.187 128.106 626 <FRACTION ONE QUARTER> 188 183 183 183 194.188 128.112 627 <FRACTION ONE HALF> 189 184 184 184 194.189 128.113 628 <FRACTION THREE QUARTERS> 190 185 185 185 194.190 128.114 629 <INVERTED QUESTION MARK> 191 171 171 171 194.191 128.115 630 <A WITH GRAVE> 192 100 100 100 195.128 138.65 631 <A WITH ACUTE> 193 101 101 101 195.129 138.66 632 <A WITH CIRCUMFLEX> 194 98 98 98 195.130 138.67 633 <A WITH TILDE> 195 102 102 102 195.131 138.68 634 <A WITH DIAERESIS> 196 99 99 99 195.132 138.69 635 <A WITH RING ABOVE> 197 103 103 103 195.133 138.70 636 <CAPITAL LIGATURE AE> 198 158 158 158 195.134 138.71 637 <C WITH CEDILLA> 199 104 104 104 195.135 138.72 638 <E WITH GRAVE> 200 116 116 116 195.136 138.73 639 <E WITH ACUTE> 201 113 113 113 195.137 138.74 640 <E WITH CIRCUMFLEX> 202 114 114 114 195.138 138.81 641 <E WITH DIAERESIS> 203 115 115 115 195.139 138.82 642 <I WITH GRAVE> 204 120 120 120 195.140 138.83 643 <I WITH ACUTE> 205 117 117 117 195.141 138.84 644 <I WITH CIRCUMFLEX> 206 118 118 118 195.142 138.85 645 <I WITH DIAERESIS> 207 119 119 119 195.143 138.86 646 <CAPITAL LETTER ETH> 208 172 172 172 195.144 138.87 647 <N WITH TILDE> 209 105 105 105 195.145 138.88 648 <O WITH GRAVE> 210 237 237 237 195.146 138.89 649 <O WITH ACUTE> 211 238 238 238 195.147 138.98 650 <O WITH CIRCUMFLEX> 212 235 235 235 195.148 138.99 651 <O WITH TILDE> 213 239 239 239 195.149 138.100 652 <O WITH DIAERESIS> 214 236 236 236 195.150 138.101 653 <MULTIPLICATION SIGN> 215 191 191 191 195.151 138.102 654 <O WITH STROKE> 216 128 128 128 195.152 138.103 655 <U WITH GRAVE> 217 253 253 224 195.153 138.104 ## 656 <U WITH ACUTE> 218 254 254 254 195.154 138.105 657 <U WITH CIRCUMFLEX> 219 251 251 221 195.155 138.106 ## 658 <U WITH DIAERESIS> 220 252 252 252 195.156 138.112 659 <Y WITH ACUTE> 221 173 186 173 195.157 138.113 ** ## 660 <CAPITAL LETTER THORN> 222 174 174 174 195.158 138.114 661 <SMALL LETTER SHARP S> 223 89 89 89 195.159 138.115 662 <a WITH GRAVE> 224 68 68 68 195.160 139.65 663 <a WITH ACUTE> 225 69 69 69 195.161 139.66 664 <a WITH CIRCUMFLEX> 226 66 66 66 195.162 139.67 665 <a WITH TILDE> 227 70 70 70 195.163 139.68 666 <a WITH DIAERESIS> 228 67 67 67 195.164 139.69 667 <a WITH RING ABOVE> 229 71 71 71 195.165 139.70 668 <SMALL LIGATURE ae> 230 156 156 156 195.166 139.71 669 <c WITH CEDILLA> 231 72 72 72 195.167 139.72 670 <e WITH GRAVE> 232 84 84 84 195.168 139.73 671 <e WITH ACUTE> 233 81 81 81 195.169 139.74 672 <e WITH CIRCUMFLEX> 234 82 82 82 195.170 139.81 673 <e WITH DIAERESIS> 235 83 83 83 195.171 139.82 674 <i WITH GRAVE> 236 88 88 88 195.172 139.83 675 <i WITH ACUTE> 237 85 85 85 195.173 139.84 676 <i WITH CIRCUMFLEX> 238 86 86 86 195.174 139.85 677 <i WITH DIAERESIS> 239 87 87 87 195.175 139.86 678 <SMALL LETTER eth> 240 140 140 140 195.176 139.87 679 <n WITH TILDE> 241 73 73 73 195.177 139.88 680 <o WITH GRAVE> 242 205 205 205 195.178 139.89 681 <o WITH ACUTE> 243 206 206 206 195.179 139.98 682 <o WITH CIRCUMFLEX> 244 203 203 203 195.180 139.99 683 <o WITH TILDE> 245 207 207 207 195.181 139.100 684 <o WITH DIAERESIS> 246 204 204 204 195.182 139.101 685 <DIVISION SIGN> 247 225 225 225 195.183 139.102 686 <o WITH STROKE> 248 112 112 112 195.184 139.103 687 <u WITH GRAVE> 249 221 221 192 195.185 139.104 ## 688 <u WITH ACUTE> 250 222 222 222 195.186 139.105 689 <u WITH CIRCUMFLEX> 251 219 219 219 195.187 139.106 690 <u WITH DIAERESIS> 252 220 220 220 195.188 139.112 691 <y WITH ACUTE> 253 141 141 141 195.189 139.113 692 <SMALL LETTER thorn> 254 142 142 142 195.190 139.114 693 <y WITH DIAERESIS> 255 223 223 223 195.191 139.115 694 695If you would rather see the above table in CCSID 0037 order rather than 696ASCII + Latin-1 order then run the table through: 697 698=over 4 699 700=item recipe 4 701 702=back 703 704 perl \ 705 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\ 706 -e '{push(@l,$_)}' \ 707 -e 'END{print map{$_->[0]}' \ 708 -e ' sort{$a->[1] <=> $b->[1]}' \ 709 -e ' map{[$_,substr($_,34,3)]}@l;}' perlebcdic.pod 710 711If you would rather see it in CCSID 1047 order then change the number 71234 in the last line to 39, like this: 713 714=over 4 715 716=item recipe 5 717 718=back 719 720 perl \ 721 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\ 722 -e '{push(@l,$_)}' \ 723 -e 'END{print map{$_->[0]}' \ 724 -e ' sort{$a->[1] <=> $b->[1]}' \ 725 -e ' map{[$_,substr($_,39,3)]}@l;}' perlebcdic.pod 726 727If you would rather see it in POSIX-BC order then change the number 72834 in the last line to 44, like this: 729 730=over 4 731 732=item recipe 6 733 734=back 735 736 perl \ 737 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\ 738 -e '{push(@l,$_)}' \ 739 -e 'END{print map{$_->[0]}' \ 740 -e ' sort{$a->[1] <=> $b->[1]}' \ 741 -e ' map{[$_,substr($_,44,3)]}@l;}' perlebcdic.pod 742 743=head2 Table in hex, sorted in 1047 order 744 745Since this document was first written, the convention has become more 746and more to use hexadecimal notation for code points. To do this with 747the recipes and to also sort is a multi-step process, so here, for 748convenience, is the table from above, re-sorted to be in Code Page 1047 749order, and using hex notation. 750 751 ISO 752 8859-1 POS- CCSID 753 CCSID CCSID CCSID IX- 1047 754 chr 0819 0037 1047 BC UTF-8 UTF-EBCDIC 755 --------------------------------------------------------------------- 756 <NUL> 00 00 00 00 00 00 757 <SOH> 01 01 01 01 01 01 758 <STX> 02 02 02 02 02 02 759 <ETX> 03 03 03 03 03 03 760 <ST> 9C 04 04 04 C2.9C 04 761 <HT> 09 05 05 05 09 05 762 <SSA> 86 06 06 06 C2.86 06 763 <DEL> 7F 07 07 07 7F 07 764 <EPA> 97 08 08 08 C2.97 08 765 <RI> 8D 09 09 09 C2.8D 09 766 <SS2> 8E 0A 0A 0A C2.8E 0A 767 <VT> 0B 0B 0B 0B 0B 0B 768 <FF> 0C 0C 0C 0C 0C 0C 769 <CR> 0D 0D 0D 0D 0D 0D 770 <SO> 0E 0E 0E 0E 0E 0E 771 <SI> 0F 0F 0F 0F 0F 0F 772 <DLE> 10 10 10 10 10 10 773 <DC1> 11 11 11 11 11 11 774 <DC2> 12 12 12 12 12 12 775 <DC3> 13 13 13 13 13 13 776 <OSC> 9D 14 14 14 C2.9D 14 777 <LF> 0A 25 15 15 0A 15 ** 778 <BS> 08 16 16 16 08 16 779 <ESA> 87 17 17 17 C2.87 17 780 <CAN> 18 18 18 18 18 18 781 <EOM> 19 19 19 19 19 19 782 <PU2> 92 1A 1A 1A C2.92 1A 783 <SS3> 8F 1B 1B 1B C2.8F 1B 784 <FS> 1C 1C 1C 1C 1C 1C 785 <GS> 1D 1D 1D 1D 1D 1D 786 <RS> 1E 1E 1E 1E 1E 1E 787 <US> 1F 1F 1F 1F 1F 1F 788 <PAD> 80 20 20 20 C2.80 20 789 <HOP> 81 21 21 21 C2.81 21 790 <BPH> 82 22 22 22 C2.82 22 791 <NBH> 83 23 23 23 C2.83 23 792 <IND> 84 24 24 24 C2.84 24 793 <NEL> 85 15 25 25 C2.85 25 ** 794 <ETB> 17 26 26 26 17 26 795 <ESC> 1B 27 27 27 1B 27 796 <HTS> 88 28 28 28 C2.88 28 797 <HTJ> 89 29 29 29 C2.89 29 798 <VTS> 8A 2A 2A 2A C2.8A 2A 799 <PLD> 8B 2B 2B 2B C2.8B 2B 800 <PLU> 8C 2C 2C 2C C2.8C 2C 801 <ENQ> 05 2D 2D 2D 05 2D 802 <ACK> 06 2E 2E 2E 06 2E 803 <BEL> 07 2F 2F 2F 07 2F 804 <DCS> 90 30 30 30 C2.90 30 805 <PU1> 91 31 31 31 C2.91 31 806 <SYN> 16 32 32 32 16 32 807 <STS> 93 33 33 33 C2.93 33 808 <CCH> 94 34 34 34 C2.94 34 809 <MW> 95 35 35 35 C2.95 35 810 <SPA> 96 36 36 36 C2.96 36 811 <EOT> 04 37 37 37 04 37 812 <SOS> 98 38 38 38 C2.98 38 813 <SGC> 99 39 39 39 C2.99 39 814 <SCI> 9A 3A 3A 3A C2.9A 3A 815 <CSI> 9B 3B 3B 3B C2.9B 3B 816 <DC4> 14 3C 3C 3C 14 3C 817 <NAK> 15 3D 3D 3D 15 3D 818 <PM> 9E 3E 3E 3E C2.9E 3E 819 <SUB> 1A 3F 3F 3F 1A 3F 820 <SPACE> 20 40 40 40 20 40 821 <NON-BREAKING SPACE> A0 41 41 41 C2.A0 80.41 822 <a WITH CIRCUMFLEX> E2 42 42 42 C3.A2 8B.43 823 <a WITH DIAERESIS> E4 43 43 43 C3.A4 8B.45 824 <a WITH GRAVE> E0 44 44 44 C3.A0 8B.41 825 <a WITH ACUTE> E1 45 45 45 C3.A1 8B.42 826 <a WITH TILDE> E3 46 46 46 C3.A3 8B.44 827 <a WITH RING ABOVE> E5 47 47 47 C3.A5 8B.46 828 <c WITH CEDILLA> E7 48 48 48 C3.A7 8B.48 829 <n WITH TILDE> F1 49 49 49 C3.B1 8B.58 830 <CENT SIGN> A2 4A 4A B0 C2.A2 80.43 ## 831 . 2E 4B 4B 4B 2E 4B 832 < 3C 4C 4C 4C 3C 4C 833 ( 28 4D 4D 4D 28 4D 834 + 2B 4E 4E 4E 2B 4E 835 | 7C 4F 4F 4F 7C 4F 836 & 26 50 50 50 26 50 837 <e WITH ACUTE> E9 51 51 51 C3.A9 8B.4A 838 <e WITH CIRCUMFLEX> EA 52 52 52 C3.AA 8B.51 839 <e WITH DIAERESIS> EB 53 53 53 C3.AB 8B.52 840 <e WITH GRAVE> E8 54 54 54 C3.A8 8B.49 841 <i WITH ACUTE> ED 55 55 55 C3.AD 8B.54 842 <i WITH CIRCUMFLEX> EE 56 56 56 C3.AE 8B.55 843 <i WITH DIAERESIS> EF 57 57 57 C3.AF 8B.56 844 <i WITH GRAVE> EC 58 58 58 C3.AC 8B.53 845 <SMALL LETTER SHARP S> DF 59 59 59 C3.9F 8A.73 846 ! 21 5A 5A 5A 21 5A 847 $ 24 5B 5B 5B 24 5B 848 * 2A 5C 5C 5C 2A 5C 849 ) 29 5D 5D 5D 29 5D 850 ; 3B 5E 5E 5E 3B 5E 851 ^ 5E B0 5F 6A 5E 5F ** ## 852 - 2D 60 60 60 2D 60 853 / 2F 61 61 61 2F 61 854 <A WITH CIRCUMFLEX> C2 62 62 62 C3.82 8A.43 855 <A WITH DIAERESIS> C4 63 63 63 C3.84 8A.45 856 <A WITH GRAVE> C0 64 64 64 C3.80 8A.41 857 <A WITH ACUTE> C1 65 65 65 C3.81 8A.42 858 <A WITH TILDE> C3 66 66 66 C3.83 8A.44 859 <A WITH RING ABOVE> C5 67 67 67 C3.85 8A.46 860 <C WITH CEDILLA> C7 68 68 68 C3.87 8A.48 861 <N WITH TILDE> D1 69 69 69 C3.91 8A.58 862 <BROKEN BAR> A6 6A 6A D0 C2.A6 80.47 ## 863 , 2C 6B 6B 6B 2C 6B 864 % 25 6C 6C 6C 25 6C 865 _ 5F 6D 6D 6D 5F 6D 866 > 3E 6E 6E 6E 3E 6E 867 ? 3F 6F 6F 6F 3F 6F 868 <o WITH STROKE> F8 70 70 70 C3.B8 8B.67 869 <E WITH ACUTE> C9 71 71 71 C3.89 8A.4A 870 <E WITH CIRCUMFLEX> CA 72 72 72 C3.8A 8A.51 871 <E WITH DIAERESIS> CB 73 73 73 C3.8B 8A.52 872 <E WITH GRAVE> C8 74 74 74 C3.88 8A.49 873 <I WITH ACUTE> CD 75 75 75 C3.8D 8A.54 874 <I WITH CIRCUMFLEX> CE 76 76 76 C3.8E 8A.55 875 <I WITH DIAERESIS> CF 77 77 77 C3.8F 8A.56 876 <I WITH GRAVE> CC 78 78 78 C3.8C 8A.53 877 ` 60 79 79 4A 60 79 ## 878 : 3A 7A 7A 7A 3A 7A 879 # 23 7B 7B 7B 23 7B 880 @ 40 7C 7C 7C 40 7C 881 ' 27 7D 7D 7D 27 7D 882 = 3D 7E 7E 7E 3D 7E 883 " 22 7F 7F 7F 22 7F 884 <O WITH STROKE> D8 80 80 80 C3.98 8A.67 885 a 61 81 81 81 61 81 886 b 62 82 82 82 62 82 887 c 63 83 83 83 63 83 888 d 64 84 84 84 64 84 889 e 65 85 85 85 65 85 890 f 66 86 86 86 66 86 891 g 67 87 87 87 67 87 892 h 68 88 88 88 68 88 893 i 69 89 89 89 69 89 894 <LEFT POINTING GUILLEMET> AB 8A 8A 8A C2.AB 80.52 895 <RIGHT POINTING GUILLEMET> BB 8B 8B 8B C2.BB 80.6A 896 <SMALL LETTER eth> F0 8C 8C 8C C3.B0 8B.57 897 <y WITH ACUTE> FD 8D 8D 8D C3.BD 8B.71 898 <SMALL LETTER thorn> FE 8E 8E 8E C3.BE 8B.72 899 <PLUS-OR-MINUS SIGN> B1 8F 8F 8F C2.B1 80.58 900 <DEGREE SIGN> B0 90 90 90 C2.B0 80.57 901 j 6A 91 91 91 6A 91 902 k 6B 92 92 92 6B 92 903 l 6C 93 93 93 6C 93 904 m 6D 94 94 94 6D 94 905 n 6E 95 95 95 6E 95 906 o 6F 96 96 96 6F 96 907 p 70 97 97 97 70 97 908 q 71 98 98 98 71 98 909 r 72 99 99 99 72 99 910 <FEMININE ORDINAL> AA 9A 9A 9A C2.AA 80.51 911 <MASC. ORDINAL INDICATOR> BA 9B 9B 9B C2.BA 80.69 912 <SMALL LIGATURE ae> E6 9C 9C 9C C3.A6 8B.47 913 <CEDILLA> B8 9D 9D 9D C2.B8 80.67 914 <CAPITAL LIGATURE AE> C6 9E 9E 9E C3.86 8A.47 915 <CURRENCY SIGN> A4 9F 9F 9F C2.A4 80.45 916 <MICRO SIGN> B5 A0 A0 A0 C2.B5 80.64 917 ~ 7E A1 A1 FF 7E A1 ## 918 s 73 A2 A2 A2 73 A2 919 t 74 A3 A3 A3 74 A3 920 u 75 A4 A4 A4 75 A4 921 v 76 A5 A5 A5 76 A5 922 w 77 A6 A6 A6 77 A6 923 x 78 A7 A7 A7 78 A7 924 y 79 A8 A8 A8 79 A8 925 z 7A A9 A9 A9 7A A9 926 <INVERTED "!" > A1 AA AA AA C2.A1 80.42 927 <INVERTED QUESTION MARK> BF AB AB AB C2.BF 80.73 928 <CAPITAL LETTER ETH> D0 AC AC AC C3.90 8A.57 929 [ 5B BA AD BB 5B AD ** ## 930 <CAPITAL LETTER THORN> DE AE AE AE C3.9E 8A.72 931 <REGISTERED TRADE MARK> AE AF AF AF C2.AE 80.55 932 <NOT SIGN> AC 5F B0 BA C2.AC 80.53 ** ## 933 <POUND SIGN> A3 B1 B1 B1 C2.A3 80.44 934 <YEN SIGN> A5 B2 B2 B2 C2.A5 80.46 935 <MIDDLE DOT> B7 B3 B3 B3 C2.B7 80.66 936 <COPYRIGHT SIGN> A9 B4 B4 B4 C2.A9 80.4A 937 <SECTION SIGN> A7 B5 B5 B5 C2.A7 80.48 938 <PARAGRAPH SIGN> B6 B6 B6 B6 C2.B6 80.65 939 <FRACTION ONE QUARTER> BC B7 B7 B7 C2.BC 80.70 940 <FRACTION ONE HALF> BD B8 B8 B8 C2.BD 80.71 941 <FRACTION THREE QUARTERS> BE B9 B9 B9 C2.BE 80.72 942 <Y WITH ACUTE> DD AD BA AD C3.9D 8A.71 ** ## 943 <DIAERESIS> A8 BD BB 79 C2.A8 80.49 ** ## 944 <MACRON> AF BC BC A1 C2.AF 80.56 ## 945 ] 5D BB BD BD 5D BD ** 946 <ACUTE ACCENT> B4 BE BE BE C2.B4 80.63 947 <MULTIPLICATION SIGN> D7 BF BF BF C3.97 8A.66 948 { 7B C0 C0 FB 7B C0 ## 949 A 41 C1 C1 C1 41 C1 950 B 42 C2 C2 C2 42 C2 951 C 43 C3 C3 C3 43 C3 952 D 44 C4 C4 C4 44 C4 953 E 45 C5 C5 C5 45 C5 954 F 46 C6 C6 C6 46 C6 955 G 47 C7 C7 C7 47 C7 956 H 48 C8 C8 C8 48 C8 957 I 49 C9 C9 C9 49 C9 958 <SOFT HYPHEN> AD CA CA CA C2.AD 80.54 959 <o WITH CIRCUMFLEX> F4 CB CB CB C3.B4 8B.63 960 <o WITH DIAERESIS> F6 CC CC CC C3.B6 8B.65 961 <o WITH GRAVE> F2 CD CD CD C3.B2 8B.59 962 <o WITH ACUTE> F3 CE CE CE C3.B3 8B.62 963 <o WITH TILDE> F5 CF CF CF C3.B5 8B.64 964 } 7D D0 D0 FD 7D D0 ## 965 J 4A D1 D1 D1 4A D1 966 K 4B D2 D2 D2 4B D2 967 L 4C D3 D3 D3 4C D3 968 M 4D D4 D4 D4 4D D4 969 N 4E D5 D5 D5 4E D5 970 O 4F D6 D6 D6 4F D6 971 P 50 D7 D7 D7 50 D7 972 Q 51 D8 D8 D8 51 D8 973 R 52 D9 D9 D9 52 D9 974 <SUPERSCRIPT ONE> B9 DA DA DA C2.B9 80.68 975 <u WITH CIRCUMFLEX> FB DB DB DB C3.BB 8B.6A 976 <u WITH DIAERESIS> FC DC DC DC C3.BC 8B.70 977 <u WITH GRAVE> F9 DD DD C0 C3.B9 8B.68 ## 978 <u WITH ACUTE> FA DE DE DE C3.BA 8B.69 979 <y WITH DIAERESIS> FF DF DF DF C3.BF 8B.73 980 \ 5C E0 E0 BC 5C E0 ## 981 <DIVISION SIGN> F7 E1 E1 E1 C3.B7 8B.66 982 S 53 E2 E2 E2 53 E2 983 T 54 E3 E3 E3 54 E3 984 U 55 E4 E4 E4 55 E4 985 V 56 E5 E5 E5 56 E5 986 W 57 E6 E6 E6 57 E6 987 X 58 E7 E7 E7 58 E7 988 Y 59 E8 E8 E8 59 E8 989 Z 5A E9 E9 E9 5A E9 990 <SUPERSCRIPT TWO> B2 EA EA EA C2.B2 80.59 991 <O WITH CIRCUMFLEX> D4 EB EB EB C3.94 8A.63 992 <O WITH DIAERESIS> D6 EC EC EC C3.96 8A.65 993 <O WITH GRAVE> D2 ED ED ED C3.92 8A.59 994 <O WITH ACUTE> D3 EE EE EE C3.93 8A.62 995 <O WITH TILDE> D5 EF EF EF C3.95 8A.64 996 0 30 F0 F0 F0 30 F0 997 1 31 F1 F1 F1 31 F1 998 2 32 F2 F2 F2 32 F2 999 3 33 F3 F3 F3 33 F3 1000 4 34 F4 F4 F4 34 F4 1001 5 35 F5 F5 F5 35 F5 1002 6 36 F6 F6 F6 36 F6 1003 7 37 F7 F7 F7 37 F7 1004 8 38 F8 F8 F8 38 F8 1005 9 39 F9 F9 F9 39 F9 1006 <SUPERSCRIPT THREE> B3 FA FA FA C2.B3 80.62 1007 <U WITH CIRCUMFLEX> DB FB FB DD C3.9B 8A.6A ## 1008 <U WITH DIAERESIS> DC FC FC FC C3.9C 8A.70 1009 <U WITH GRAVE> D9 FD FD E0 C3.99 8A.68 ## 1010 <U WITH ACUTE> DA FE FE FE C3.9A 8A.69 1011 <APC> 9F FF FF 5F C2.9F FF ## 1012 1013=head1 IDENTIFYING CHARACTER CODE SETS 1014 1015It is possible to determine which character set you are operating under. 1016But first you need to be really really sure you need to do this. Your 1017code will be simpler and probably just as portable if you don't have 1018to test the character set and do different things, depending. There are 1019actually only very few circumstances where it's not easy to write 1020straight-line code portable to all character sets. See 1021L<perluniintro/Unicode and EBCDIC> for how to portably specify 1022characters. 1023 1024But there are some cases where you may want to know which character set 1025you are running under. One possible example is doing 1026L<sorting|/SORTING> in inner loops where performance is critical. 1027 1028To determine if you are running under ASCII or EBCDIC, you can use the 1029return value of C<ord()> or C<chr()> to test one or more character 1030values. For example: 1031 1032 $is_ascii = "A" eq chr(65); 1033 $is_ebcdic = "A" eq chr(193); 1034 $is_ascii = ord("A") == 65; 1035 $is_ebcdic = ord("A") == 193; 1036 1037There's even less need to distinguish between EBCDIC code pages, but to 1038do so try looking at one or more of the characters that differ between 1039them. 1040 1041 $is_ascii = ord('[') == 91; 1042 $is_ebcdic_37 = ord('[') == 186; 1043 $is_ebcdic_1047 = ord('[') == 173; 1044 $is_ebcdic_POSIX_BC = ord('[') == 187; 1045 1046However, it would be unwise to write tests such as: 1047 1048 $is_ascii = "\r" ne chr(13); # WRONG 1049 $is_ascii = "\n" ne chr(10); # ILL ADVISED 1050 1051Obviously the first of these will fail to distinguish most ASCII 1052platforms from either a CCSID 0037, a 1047, or a POSIX-BC EBCDIC 1053platform since S<C<"\r" eq chr(13)>> under all of those coded character 1054sets. But note too that because C<"\n"> is C<chr(13)> and C<"\r"> is 1055C<chr(10)> on old Macintosh (which is an ASCII platform) the second 1056C<$is_ascii> test will lead to trouble there. 1057 1058To determine whether or not perl was built under an EBCDIC 1059code page you can use the Config module like so: 1060 1061 use Config; 1062 $is_ebcdic = $Config{'ebcdic'} eq 'define'; 1063 1064=head1 CONVERSIONS 1065 1066=head2 C<utf8::unicode_to_native()> and C<utf8::native_to_unicode()> 1067 1068These functions take an input numeric code point in one encoding and 1069return what its equivalent value is in the other. 1070 1071See L<utf8>. 1072 1073=head2 tr/// 1074 1075In order to convert a string of characters from one character set to 1076another a simple list of numbers, such as in the right columns in the 1077above table, along with Perl's C<tr///> operator is all that is needed. 1078The data in the table are in ASCII/Latin1 order, hence the EBCDIC columns 1079provide easy-to-use ASCII/Latin1 to EBCDIC operations that are also easily 1080reversed. 1081 1082For example, to convert ASCII/Latin1 to code page 037 take the output of the 1083second numbers column from the output of recipe 2 (modified to add 1084C<"\"> characters), and use it in C<tr///> like so: 1085 1086 $cp_037 = 1087 '\x00\x01\x02\x03\x37\x2D\x2E\x2F\x16\x05\x25\x0B\x0C\x0D\x0E\x0F' . 1088 '\x10\x11\x12\x13\x3C\x3D\x32\x26\x18\x19\x3F\x27\x1C\x1D\x1E\x1F' . 1089 '\x40\x5A\x7F\x7B\x5B\x6C\x50\x7D\x4D\x5D\x5C\x4E\x6B\x60\x4B\x61' . 1090 '\xF0\xF1\xF2\xF3\xF4\xF5\xF6\xF7\xF8\xF9\x7A\x5E\x4C\x7E\x6E\x6F' . 1091 '\x7C\xC1\xC2\xC3\xC4\xC5\xC6\xC7\xC8\xC9\xD1\xD2\xD3\xD4\xD5\xD6' . 1092 '\xD7\xD8\xD9\xE2\xE3\xE4\xE5\xE6\xE7\xE8\xE9\xBA\xE0\xBB\xB0\x6D' . 1093 '\x79\x81\x82\x83\x84\x85\x86\x87\x88\x89\x91\x92\x93\x94\x95\x96' . 1094 '\x97\x98\x99\xA2\xA3\xA4\xA5\xA6\xA7\xA8\xA9\xC0\x4F\xD0\xA1\x07' . 1095 '\x20\x21\x22\x23\x24\x15\x06\x17\x28\x29\x2A\x2B\x2C\x09\x0A\x1B' . 1096 '\x30\x31\x1A\x33\x34\x35\x36\x08\x38\x39\x3A\x3B\x04\x14\x3E\xFF' . 1097 '\x41\xAA\x4A\xB1\x9F\xB2\x6A\xB5\xBD\xB4\x9A\x8A\x5F\xCA\xAF\xBC' . 1098 '\x90\x8F\xEA\xFA\xBE\xA0\xB6\xB3\x9D\xDA\x9B\x8B\xB7\xB8\xB9\xAB' . 1099 '\x64\x65\x62\x66\x63\x67\x9E\x68\x74\x71\x72\x73\x78\x75\x76\x77' . 1100 '\xAC\x69\xED\xEE\xEB\xEF\xEC\xBF\x80\xFD\xFE\xFB\xFC\xAD\xAE\x59' . 1101 '\x44\x45\x42\x46\x43\x47\x9C\x48\x54\x51\x52\x53\x58\x55\x56\x57' . 1102 '\x8C\x49\xCD\xCE\xCB\xCF\xCC\xE1\x70\xDD\xDE\xDB\xDC\x8D\x8E\xDF'; 1103 1104 my $ebcdic_string = $ascii_string; 1105 eval '$ebcdic_string =~ tr/\000-\377/' . $cp_037 . '/'; 1106 1107To convert from EBCDIC 037 to ASCII just reverse the order of the tr/// 1108arguments like so: 1109 1110 my $ascii_string = $ebcdic_string; 1111 eval '$ascii_string =~ tr/' . $cp_037 . '/\000-\377/'; 1112 1113Similarly one could take the output of the third numbers column from recipe 2 1114to obtain a C<$cp_1047> table. The fourth numbers column of the output from 1115recipe 2 could provide a C<$cp_posix_bc> table suitable for transcoding as 1116well. 1117 1118If you wanted to see the inverse tables, you would first have to sort on the 1119desired numbers column as in recipes 4, 5 or 6, then take the output of the 1120first numbers column. 1121 1122=head2 iconv 1123 1124XPG operability often implies the presence of an I<iconv> utility 1125available from the shell or from the C library. Consult your system's 1126documentation for information on iconv. 1127 1128On OS/390 or z/OS see the L<iconv(1)> manpage. One way to invoke the C<iconv> 1129shell utility from within perl would be to: 1130 1131 # OS/390 or z/OS example 1132 $ascii_data = `echo '$ebcdic_data'| iconv -f IBM-1047 -t ISO8859-1` 1133 1134or the inverse map: 1135 1136 # OS/390 or z/OS example 1137 $ebcdic_data = `echo '$ascii_data'| iconv -f ISO8859-1 -t IBM-1047` 1138 1139For other Perl-based conversion options see the C<Convert::*> modules on CPAN. 1140 1141=head2 C RTL 1142 1143The OS/390 and z/OS C run-time libraries provide C<_atoe()> and C<_etoa()> functions. 1144 1145=head1 OPERATOR DIFFERENCES 1146 1147The C<..> range operator treats certain character ranges with 1148care on EBCDIC platforms. For example the following array 1149will have twenty six elements on either an EBCDIC platform 1150or an ASCII platform: 1151 1152 @alphabet = ('A'..'Z'); # $#alphabet == 25 1153 1154The bitwise operators such as & ^ | may return different results 1155when operating on string or character data in a Perl program running 1156on an EBCDIC platform than when run on an ASCII platform. Here is 1157an example adapted from the one in L<perlop>: 1158 1159 # EBCDIC-based examples 1160 print "j p \n" ^ " a h"; # prints "JAPH\n" 1161 print "JA" | " ph\n"; # prints "japh\n" 1162 print "JAPH\nJunk" & "\277\277\277\277\277"; # prints "japh\n"; 1163 print 'p N$' ^ " E<H\n"; # prints "Perl\n"; 1164 1165An interesting property of the 32 C0 control characters 1166in the ASCII table is that they can "literally" be constructed 1167as control characters in Perl, e.g. C<(chr(0)> eq C<\c@>)> 1168C<(chr(1)> eq C<\cA>)>, and so on. Perl on EBCDIC platforms has been 1169ported to take C<\c@> to C<chr(0)> and C<\cA> to C<chr(1)>, etc. as well, but the 1170characters that result depend on which code page you are 1171using. The table below uses the standard acronyms for the controls. 1172The POSIX-BC and 1047 sets are 1173identical throughout this range and differ from the 0037 set at only 1174one spot (21 decimal). Note that the line terminator character 1175may be generated by C<\cJ> on ASCII platforms but by C<\cU> on 1047 or POSIX-BC 1176platforms and cannot be generated as a C<"\c.letter."> control character on 11770037 platforms. Note also that C<\c\> cannot be the final element in a string 1178or regex, as it will absorb the terminator. But C<\c\I<X>> is a C<FILE 1179SEPARATOR> concatenated with I<X> for all I<X>. 1180The outlier C<\c?> on ASCII, which yields a non-C0 control C<DEL>, 1181yields the outlier control C<APC> on EBCDIC, the one that isn't in the 1182block of contiguous controls. Note that a subtlety of this is that 1183C<\c?> on ASCII platforms is an ASCII character, while it isn't 1184equivalent to any ASCII character in EBCDIC platforms. 1185 1186 chr ord 8859-1 0037 1047 && POSIX-BC 1187 ----------------------------------------------------------------------- 1188 \c@ 0 <NUL> <NUL> <NUL> 1189 \cA 1 <SOH> <SOH> <SOH> 1190 \cB 2 <STX> <STX> <STX> 1191 \cC 3 <ETX> <ETX> <ETX> 1192 \cD 4 <EOT> <ST> <ST> 1193 \cE 5 <ENQ> <HT> <HT> 1194 \cF 6 <ACK> <SSA> <SSA> 1195 \cG 7 <BEL> <DEL> <DEL> 1196 \cH 8 <BS> <EPA> <EPA> 1197 \cI 9 <HT> <RI> <RI> 1198 \cJ 10 <LF> <SS2> <SS2> 1199 \cK 11 <VT> <VT> <VT> 1200 \cL 12 <FF> <FF> <FF> 1201 \cM 13 <CR> <CR> <CR> 1202 \cN 14 <SO> <SO> <SO> 1203 \cO 15 <SI> <SI> <SI> 1204 \cP 16 <DLE> <DLE> <DLE> 1205 \cQ 17 <DC1> <DC1> <DC1> 1206 \cR 18 <DC2> <DC2> <DC2> 1207 \cS 19 <DC3> <DC3> <DC3> 1208 \cT 20 <DC4> <OSC> <OSC> 1209 \cU 21 <NAK> <NEL> <LF> ** 1210 \cV 22 <SYN> <BS> <BS> 1211 \cW 23 <ETB> <ESA> <ESA> 1212 \cX 24 <CAN> <CAN> <CAN> 1213 \cY 25 <EOM> <EOM> <EOM> 1214 \cZ 26 <SUB> <PU2> <PU2> 1215 \c[ 27 <ESC> <SS3> <SS3> 1216 \c\X 28 <FS>X <FS>X <FS>X 1217 \c] 29 <GS> <GS> <GS> 1218 \c^ 30 <RS> <RS> <RS> 1219 \c_ 31 <US> <US> <US> 1220 \c? * <DEL> <APC> <APC> 1221 1222C<*> Note: C<\c?> maps to ordinal 127 (C<DEL>) on ASCII platforms, but 1223since ordinal 127 is a not a control character on EBCDIC machines, 1224C<\c?> instead maps on them to C<APC>, which is 255 in 0037 and 1047, 1225and 95 in POSIX-BC. 1226 1227=head1 FUNCTION DIFFERENCES 1228 1229=over 8 1230 1231=item C<chr()> 1232 1233C<chr()> must be given an EBCDIC code number argument to yield a desired 1234character return value on an EBCDIC platform. For example: 1235 1236 $CAPITAL_LETTER_A = chr(193); 1237 1238=item C<ord()> 1239 1240C<ord()> will return EBCDIC code number values on an EBCDIC platform. 1241For example: 1242 1243 $the_number_193 = ord("A"); 1244 1245=item C<pack()> 1246 1247 1248The C<"c"> and C<"C"> templates for C<pack()> are dependent upon character set 1249encoding. Examples of usage on EBCDIC include: 1250 1251 $foo = pack("CCCC",193,194,195,196); 1252 # $foo eq "ABCD" 1253 $foo = pack("C4",193,194,195,196); 1254 # same thing 1255 1256 $foo = pack("ccxxcc",193,194,195,196); 1257 # $foo eq "AB\0\0CD" 1258 1259The C<"U"> template has been ported to mean "Unicode" on all platforms so 1260that 1261 1262 pack("U", 65) eq 'A' 1263 1264is true on all platforms. If you want native code points for the low 1265256, use the C<"W"> template. This means that the equivalences 1266 1267 pack("W", ord($character)) eq $character 1268 unpack("W", $character) == ord $character 1269 1270will hold. 1271 1272=item C<print()> 1273 1274One must be careful with scalars and strings that are passed to 1275print that contain ASCII encodings. One common place 1276for this to occur is in the output of the MIME type header for 1277CGI script writing. For example, many Perl programming guides 1278recommend something similar to: 1279 1280 print "Content-type:\ttext/html\015\012\015\012"; 1281 # this may be wrong on EBCDIC 1282 1283You can instead write 1284 1285 print "Content-type:\ttext/html\r\n\r\n"; # OK for DGW et al 1286 1287and have it work portably. 1288 1289That is because the translation from EBCDIC to ASCII is done 1290by the web server in this case. Consult your web server's documentation for 1291further details. 1292 1293=item C<printf()> 1294 1295The formats that can convert characters to numbers and vice versa 1296will be different from their ASCII counterparts when executed 1297on an EBCDIC platform. Examples include: 1298 1299 printf("%c%c%c",193,194,195); # prints ABC 1300 1301=item C<sort()> 1302 1303EBCDIC sort results may differ from ASCII sort results especially for 1304mixed case strings. This is discussed in more detail L<below|/SORTING>. 1305 1306=item C<sprintf()> 1307 1308See the discussion of C<L</printf()>> above. An example of the use 1309of sprintf would be: 1310 1311 $CAPITAL_LETTER_A = sprintf("%c",193); 1312 1313=item C<unpack()> 1314 1315See the discussion of C<L</pack()>> above. 1316 1317=back 1318 1319Note that it is possible to write portable code for these by specifying 1320things in Unicode numbers, and using a conversion function: 1321 1322 printf("%c",utf8::unicode_to_native(65)); # prints A on all 1323 # platforms 1324 print utf8::native_to_unicode(ord("A")); # Likewise, prints 65 1325 1326See L<perluniintro/Unicode and EBCDIC> and L</CONVERSIONS> 1327for other options. 1328 1329=head1 REGULAR EXPRESSION DIFFERENCES 1330 1331You can write your regular expressions just like someone on an ASCII 1332platform would do. But keep in mind that using octal or hex notation to 1333specify a particular code point will give you the character that the 1334EBCDIC code page natively maps to it. (This is also true of all 1335double-quoted strings.) If you want to write portably, just use the 1336C<\N{U+...}> notation everywhere where you would have used C<\x{...}>, 1337and don't use octal notation at all. 1338 1339Starting in Perl v5.22, this applies to ranges in bracketed character 1340classes. If you say, for example, C<qr/[\N{U+20}-\N{U+7F}]/>, it means 1341the characters C<\N{U+20}>, C<\N{U+21}>, ..., C<\N{U+7F}>. This range 1342is all the printable characters that the ASCII character set contains. 1343 1344Prior to v5.22, you couldn't specify any ranges portably, except 1345(starting in Perl v5.5.3) all subsets of the C<[A-Z]> and C<[a-z]> 1346ranges are specially coded to not pick up gap characters. For example, 1347characters such as "E<ocirc>" (C<o WITH CIRCUMFLEX>) that lie between 1348"I" and "J" would not be matched by the regular expression range 1349C</[H-K]/>. But if either of the range end points is explicitly numeric 1350(and neither is specified by C<\N{U+...}>), the gap characters are 1351matched: 1352 1353 /[\x89-\x91]/ 1354 1355will match C<\x8e>, even though C<\x89> is "i" and C<\x91 > is "j", 1356and C<\x8e> is a gap character, from the alphabetic viewpoint. 1357 1358Another construct to be wary of is the inappropriate use of hex (unless 1359you use C<\N{U+...}>) or 1360octal constants in regular expressions. Consider the following 1361set of subs: 1362 1363 sub is_c0 { 1364 my $char = substr(shift,0,1); 1365 $char =~ /[\000-\037]/; 1366 } 1367 1368 sub is_print_ascii { 1369 my $char = substr(shift,0,1); 1370 $char =~ /[\040-\176]/; 1371 } 1372 1373 sub is_delete { 1374 my $char = substr(shift,0,1); 1375 $char eq "\177"; 1376 } 1377 1378 sub is_c1 { 1379 my $char = substr(shift,0,1); 1380 $char =~ /[\200-\237]/; 1381 } 1382 1383 sub is_latin_1 { # But not ASCII; not C1 1384 my $char = substr(shift,0,1); 1385 $char =~ /[\240-\377]/; 1386 } 1387 1388These are valid only on ASCII platforms. Starting in Perl v5.22, simply 1389changing the octal constants to equivalent C<\N{U+...}> values makes 1390them portable: 1391 1392 sub is_c0 { 1393 my $char = substr(shift,0,1); 1394 $char =~ /[\N{U+00}-\N{U+1F}]/; 1395 } 1396 1397 sub is_print_ascii { 1398 my $char = substr(shift,0,1); 1399 $char =~ /[\N{U+20}-\N{U+7E}]/; 1400 } 1401 1402 sub is_delete { 1403 my $char = substr(shift,0,1); 1404 $char eq "\N{U+7F}"; 1405 } 1406 1407 sub is_c1 { 1408 my $char = substr(shift,0,1); 1409 $char =~ /[\N{U+80}-\N{U+9F}]/; 1410 } 1411 1412 sub is_latin_1 { # But not ASCII; not C1 1413 my $char = substr(shift,0,1); 1414 $char =~ /[\N{U+A0}-\N{U+FF}]/; 1415 } 1416 1417And here are some alternative portable ways to write them: 1418 1419 sub Is_c0 { 1420 my $char = substr(shift,0,1); 1421 return $char =~ /[[:cntrl:]]/a && ! Is_delete($char); 1422 1423 # Alternatively: 1424 # return $char =~ /[[:cntrl:]]/ 1425 # && $char =~ /[[:ascii:]]/ 1426 # && ! Is_delete($char); 1427 } 1428 1429 sub Is_print_ascii { 1430 my $char = substr(shift,0,1); 1431 1432 return $char =~ /[[:print:]]/a; 1433 1434 # Alternatively: 1435 # return $char =~ /[[:print:]]/ && $char =~ /[[:ascii:]]/; 1436 1437 # Or 1438 # return $char 1439 # =~ /[ !"\#\$%&'()*+,\-.\/0-9:;<=>?\@A-Z[\\\]^_`a-z{|}~]/; 1440 } 1441 1442 sub Is_delete { 1443 my $char = substr(shift,0,1); 1444 return utf8::native_to_unicode(ord $char) == 0x7F; 1445 } 1446 1447 sub Is_c1 { 1448 use feature 'unicode_strings'; 1449 my $char = substr(shift,0,1); 1450 return $char =~ /[[:cntrl:]]/ && $char !~ /[[:ascii:]]/; 1451 } 1452 1453 sub Is_latin_1 { # But not ASCII; not C1 1454 use feature 'unicode_strings'; 1455 my $char = substr(shift,0,1); 1456 return ord($char) < 256 1457 && $char !~ /[[:ascii:]]/ 1458 && $char !~ /[[:cntrl:]]/; 1459 } 1460 1461Another way to write C<Is_latin_1()> would be 1462to use the characters in the range explicitly: 1463 1464 sub Is_latin_1 { 1465 my $char = substr(shift,0,1); 1466 $char =~ /[ ¡¢£¤¥¦§¨©ª«¬®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏ] 1467 [ÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ]/x; 1468 } 1469 1470Although that form may run into trouble in network transit (due to the 1471presence of 8 bit characters) or on non ISO-Latin character sets. But 1472it does allow C<Is_c1> to be rewritten so it works on Perls that don't 1473have C<'unicode_strings'> (earlier than v5.14): 1474 1475 sub Is_latin_1 { # But not ASCII; not C1 1476 my $char = substr(shift,0,1); 1477 return ord($char) < 256 1478 && $char !~ /[[:ascii:]]/ 1479 && ! Is_latin1($char); 1480 } 1481 1482=head1 SOCKETS 1483 1484Most socket programming assumes ASCII character encodings in network 1485byte order. Exceptions can include CGI script writing under a 1486host web server where the server may take care of translation for you. 1487Most host web servers convert EBCDIC data to ISO-8859-1 or Unicode on 1488output. 1489 1490=head1 SORTING 1491 1492One big difference between ASCII-based character sets and EBCDIC ones 1493are the relative positions of the characters when sorted in native 1494order. Of most concern are the upper- and lowercase letters, the 1495digits, and the underscore (C<"_">). On ASCII platforms the native sort 1496order has the digits come before the uppercase letters which come before 1497the underscore which comes before the lowercase letters. On EBCDIC, the 1498underscore comes first, then the lowercase letters, then the uppercase 1499ones, and the digits last. If sorted on an ASCII-based platform, the 1500two-letter abbreviation for a physician comes before the two letter 1501abbreviation for drive; that is: 1502 1503 @sorted = sort(qw(Dr. dr.)); # @sorted holds ('Dr.','dr.') on ASCII, 1504 # but ('dr.','Dr.') on EBCDIC 1505 1506The property of lowercase before uppercase letters in EBCDIC is 1507even carried to the Latin 1 EBCDIC pages such as 0037 and 1047. 1508An example would be that "E<Euml>" (C<E WITH DIAERESIS>, 203) comes 1509before "E<euml>" (C<e WITH DIAERESIS>, 235) on an ASCII platform, but 1510the latter (83) comes before the former (115) on an EBCDIC platform. 1511(Astute readers will note that the uppercase version of "E<szlig>" 1512C<SMALL LETTER SHARP S> is simply "SS" and that the upper case versions 1513of "E<yuml>" (small C<y WITH DIAERESIS>) and "E<micro>" (C<MICRO SIGN>) 1514are not in the 0..255 range but are in Unicode, in a Unicode enabled 1515Perl). 1516 1517The sort order will cause differences between results obtained on 1518ASCII platforms versus EBCDIC platforms. What follows are some suggestions 1519on how to deal with these differences. 1520 1521=head2 Ignore ASCII vs. EBCDIC sort differences. 1522 1523This is the least computationally expensive strategy. It may require 1524some user education. 1525 1526=head2 Use a sort helper function 1527 1528This is completely general, but the most computationally expensive 1529strategy. Choose one or the other character set and transform to that 1530for every sort comparison. Here's a complete example that transforms 1531to ASCII sort order: 1532 1533 sub native_to_uni($) { 1534 my $string = shift; 1535 1536 # Saves time on an ASCII platform 1537 return $string if ord 'A' == 65; 1538 1539 my $output = ""; 1540 for my $i (0 .. length($string) - 1) { 1541 $output 1542 .= chr(utf8::native_to_unicode(ord(substr($string, $i, 1)))); 1543 } 1544 1545 # Preserve utf8ness of input onto the output, even if it didn't need 1546 # to be utf8 1547 utf8::upgrade($output) if utf8::is_utf8($string); 1548 1549 return $output; 1550 } 1551 1552 sub ascii_order { # Sort helper 1553 return native_to_uni($a) cmp native_to_uni($b); 1554 } 1555 1556 sort ascii_order @list; 1557 1558=head2 MONO CASE then sort data (for non-digits, non-underscore) 1559 1560If you don't care about where digits and underscore sort to, you can do 1561something like this 1562 1563 sub case_insensitive_order { # Sort helper 1564 return lc($a) cmp lc($b) 1565 } 1566 1567 sort case_insensitive_order @list; 1568 1569If performance is an issue, and you don't care if the output is in the 1570same case as the input, Use C<tr///> to transform to the case most 1571employed within the data. If the data are primarily UPPERCASE 1572non-Latin1, then apply C<tr/[a-z]/[A-Z]/>, and then C<sort()>. If the 1573data are primarily lowercase non Latin1 then apply C<tr/[A-Z]/[a-z]/> 1574before sorting. If the data are primarily UPPERCASE and include Latin-1 1575characters then apply: 1576 1577 tr/[a-z]/[A-Z]/; 1578 tr/[àáâãäåæçèéêëìíîïðñòóôõöøùúûüýþ]/[ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞ/; 1579 s/ß/SS/g; 1580 1581then C<sort()>. If you have a choice, it's better to lowercase things 1582to avoid the problems of the two Latin-1 characters whose uppercase is 1583outside Latin-1: "E<yuml>" (small C<y WITH DIAERESIS>) and "E<micro>" 1584(C<MICRO SIGN>). If you do need to upppercase, you can; with a 1585Unicode-enabled Perl, do: 1586 1587 tr/ÿ/\x{178}/; 1588 tr/µ/\x{39C}/; 1589 1590=head2 Perform sorting on one type of platform only. 1591 1592This strategy can employ a network connection. As such 1593it would be computationally expensive. 1594 1595=head1 TRANSFORMATION FORMATS 1596 1597There are a variety of ways of transforming data with an intra character set 1598mapping that serve a variety of purposes. Sorting was discussed in the 1599previous section and a few of the other more popular mapping techniques are 1600discussed next. 1601 1602=head2 URL decoding and encoding 1603 1604Note that some URLs have hexadecimal ASCII code points in them in an 1605attempt to overcome character or protocol limitation issues. For example 1606the tilde character is not on every keyboard hence a URL of the form: 1607 1608 http://www.pvhp.com/~pvhp/ 1609 1610may also be expressed as either of: 1611 1612 http://www.pvhp.com/%7Epvhp/ 1613 1614 http://www.pvhp.com/%7epvhp/ 1615 1616where 7E is the hexadecimal ASCII code point for "~". Here is an example 1617of decoding such a URL in any EBCDIC code page: 1618 1619 $url = 'http://www.pvhp.com/%7Epvhp/'; 1620 $url =~ s/%([0-9a-fA-F]{2})/ 1621 pack("c",utf8::unicode_to_native(hex($1)))/xge; 1622 1623Conversely, here is a partial solution for the task of encoding such 1624a URL in any EBCDIC code page: 1625 1626 $url = 'http://www.pvhp.com/~pvhp/'; 1627 # The following regular expression does not address the 1628 # mappings for: ('.' => '%2E', '/' => '%2F', ':' => '%3A') 1629 $url =~ s/([\t "#%&\(\),;<=>\?\@\[\\\]^`{|}~])/ 1630 sprintf("%%%02X",utf8::native_to_unicode(ord($1)))/xge; 1631 1632where a more complete solution would split the URL into components 1633and apply a full s/// substitution only to the appropriate parts. 1634 1635=head2 uu encoding and decoding 1636 1637The C<u> template to C<pack()> or C<unpack()> will render EBCDIC data in 1638EBCDIC characters equivalent to their ASCII counterparts. For example, 1639the following will print "Yes indeed\n" on either an ASCII or EBCDIC 1640computer: 1641 1642 $all_byte_chrs = ''; 1643 for (0..255) { $all_byte_chrs .= chr($_); } 1644 $uuencode_byte_chrs = pack('u', $all_byte_chrs); 1645 ($uu = <<'ENDOFHEREDOC') =~ s/^\s*//gm; 1646 M``$"`P0%!@<("0H+#`T.#Q`1$A,4%187&!D:&QP='A\@(2(C)"4F)R@I*BLL 1647 M+2XO,#$R,S0U-C<X.3H[/#T^/T!!0D-$149'2$E*2TQ-3D]045)35%565UA9 1648 M6EM<75Y?8&%B8V1E9F=H:6IK;&UN;W!Q<G-T=79W>'EZ>WQ]?G^`@8*#A(6& 1649 MAXB)BHN,C8Z/D)&2DY25EI>8F9J;G)V>GZ"AHJ.DI::GJ*FJJZRMKJ^PL;*S 1650 MM+6VM[BYNKN\O;Z_P,'"P\3%QL?(R<K+S,W.S]#1TM/4U=;7V-G:V]S=WM_@ 1651 ?X>+CY.7FY^CIZNOL[>[O\/'R\_3U]O?X^?K[_/W^_P`` 1652 ENDOFHEREDOC 1653 if ($uuencode_byte_chrs eq $uu) { 1654 print "Yes "; 1655 } 1656 $uudecode_byte_chrs = unpack('u', $uuencode_byte_chrs); 1657 if ($uudecode_byte_chrs eq $all_byte_chrs) { 1658 print "indeed\n"; 1659 } 1660 1661Here is a very spartan uudecoder that will work on EBCDIC: 1662 1663 #!/usr/local/bin/perl 1664 $_ = <> until ($mode,$file) = /^begin\s*(\d*)\s*(\S*)/; 1665 open(OUT, "> $file") if $file ne ""; 1666 while(<>) { 1667 last if /^end/; 1668 next if /[a-z]/; 1669 next unless int((((utf8::native_to_unicode(ord()) - 32 ) & 077) 1670 + 2) / 3) 1671 == int(length() / 4); 1672 print OUT unpack("u", $_); 1673 } 1674 close(OUT); 1675 chmod oct($mode), $file; 1676 1677 1678=head2 Quoted-Printable encoding and decoding 1679 1680On ASCII-encoded platforms it is possible to strip characters outside of 1681the printable set using: 1682 1683 # This QP encoder works on ASCII only 1684 $qp_string =~ s/([=\x00-\x1F\x80-\xFF])/ 1685 sprintf("=%02X",ord($1))/xge; 1686 1687Starting in Perl v5.22, this is trivially changeable to work portably on 1688both ASCII and EBCDIC platforms. 1689 1690 # This QP encoder works on both ASCII and EBCDIC 1691 $qp_string =~ s/([=\N{U+00}-\N{U+1F}\N{U+80}-\N{U+FF}])/ 1692 sprintf("=%02X",ord($1))/xge; 1693 1694For earlier Perls, a QP encoder that works on both ASCII and EBCDIC 1695platforms would look somewhat like the following: 1696 1697 $delete = utf8::unicode_to_native(ord("\x7F")); 1698 $qp_string =~ 1699 s/([^[:print:]$delete])/ 1700 sprintf("=%02X",utf8::native_to_unicode(ord($1)))/xage; 1701 1702(although in production code the substitutions might be done 1703in the EBCDIC branch with the function call and separately in the 1704ASCII branch without the expense of the identity map; in Perl v5.22, the 1705identity map is optimized out so there is no expense, but the 1706alternative above is simpler and is also available in v5.22). 1707 1708Such QP strings can be decoded with: 1709 1710 # This QP decoder is limited to ASCII only 1711 $string =~ s/=([[:xdigit:][[:xdigit:])/chr hex $1/ge; 1712 $string =~ s/=[\n\r]+$//; 1713 1714Whereas a QP decoder that works on both ASCII and EBCDIC platforms 1715would look somewhat like the following: 1716 1717 $string =~ s/=([[:xdigit:][:xdigit:]])/ 1718 chr utf8::native_to_unicode(hex $1)/xge; 1719 $string =~ s/=[\n\r]+$//; 1720 1721=head2 Caesarean ciphers 1722 1723The practice of shifting an alphabet one or more characters for encipherment 1724dates back thousands of years and was explicitly detailed by Gaius Julius 1725Caesar in his B<Gallic Wars> text. A single alphabet shift is sometimes 1726referred to as a rotation and the shift amount is given as a number $n after 1727the string 'rot' or "rot$n". Rot0 and rot26 would designate identity maps 1728on the 26-letter English version of the Latin alphabet. Rot13 has the 1729interesting property that alternate subsequent invocations are identity maps 1730(thus rot13 is its own non-trivial inverse in the group of 26 alphabet 1731rotations). Hence the following is a rot13 encoder and decoder that will 1732work on ASCII and EBCDIC platforms: 1733 1734 #!/usr/local/bin/perl 1735 1736 while(<>){ 1737 tr/n-za-mN-ZA-M/a-zA-Z/; 1738 print; 1739 } 1740 1741In one-liner form: 1742 1743 perl -ne 'tr/n-za-mN-ZA-M/a-zA-Z/;print' 1744 1745 1746=head1 Hashing order and checksums 1747 1748Perl deliberately randomizes hash order for security purposes on both 1749ASCII and EBCDIC platforms. 1750 1751EBCDIC checksums will differ for the same file translated into ASCII 1752and vice versa. 1753 1754=head1 I18N AND L10N 1755 1756Internationalization (I18N) and localization (L10N) are supported at least 1757in principle even on EBCDIC platforms. The details are system-dependent 1758and discussed under the L</OS ISSUES> section below. 1759 1760=head1 MULTI-OCTET CHARACTER SETS 1761 1762Perl works with UTF-EBCDIC, a multi-byte encoding. In Perls earlier 1763than v5.22, there may be various bugs in this regard. 1764 1765Legacy multi byte EBCDIC code pages XXX. 1766 1767=head1 OS ISSUES 1768 1769There may be a few system-dependent issues 1770of concern to EBCDIC Perl programmers. 1771 1772=head2 OS/400 1773 1774=over 8 1775 1776=item PASE 1777 1778The PASE environment is a runtime environment for OS/400 that can run 1779executables built for PowerPC AIX in OS/400; see L<perlos400>. PASE 1780is ASCII-based, not EBCDIC-based as the ILE. 1781 1782=item IFS access 1783 1784XXX. 1785 1786=back 1787 1788=head2 OS/390, z/OS 1789 1790Perl runs under Unix Systems Services or USS. 1791 1792=over 8 1793 1794=item C<sigaction> 1795 1796C<SA_SIGINFO> can have segmentation faults. 1797 1798=item C<chcp> 1799 1800B<chcp> is supported as a shell utility for displaying and changing 1801one's code page. See also L<chcp(1)>. 1802 1803=item dataset access 1804 1805For sequential data set access try: 1806 1807 my @ds_records = `cat //DSNAME`; 1808 1809or: 1810 1811 my @ds_records = `cat //'HLQ.DSNAME'`; 1812 1813See also the OS390::Stdio module on CPAN. 1814 1815=item C<iconv> 1816 1817B<iconv> is supported as both a shell utility and a C RTL routine. 1818See also the L<iconv(1)> and L<iconv(3)> manual pages. 1819 1820=item locales 1821 1822Locales are supported. There may be glitches when a locale is another 1823EBCDIC code page which has some of the 1824L<code-page variant characters|/The 13 variant characters> in other 1825positions. 1826 1827There aren't currently any real UTF-8 locales, even though some locale 1828names contain the string "UTF-8". 1829 1830See L<perllocale> for information on locales. The L10N files 1831are in F</usr/nls/locale>. C<$Config{d_setlocale}> is C<'define'> on 1832OS/390 or z/OS. 1833 1834=back 1835 1836=head2 POSIX-BC? 1837 1838XXX. 1839 1840=head1 BUGS 1841 1842=over 4 1843 1844=item * 1845 1846Not all shells will allow multiple C<-e> string arguments to perl to 1847be concatenated together properly as recipes in this document 18480, 2, 4, 5, and 6 might 1849seem to imply. 1850 1851=item * 1852 1853There are a significant number of test failures in the CPAN modules 1854shipped with Perl v5.22 and 5.24. These are only in modules not primarily 1855maintained by Perl 5 porters. Some of these are failures in the tests 1856only: they don't realize that it is proper to get different results on 1857EBCDIC platforms. And some of the failures are real bugs. If you 1858compile and do a C<make test> on Perl, all tests on the C</cpan> 1859directory are skipped. 1860 1861L<Encode> partially works. 1862 1863=item * 1864 1865In earlier Perl versions, when byte and character data were 1866concatenated, the new string was sometimes created by 1867decoding the byte strings as I<ISO 8859-1 (Latin-1)>, even if the 1868old Unicode string used EBCDIC. 1869 1870=back 1871 1872=head1 SEE ALSO 1873 1874L<perllocale>, L<perlfunc>, L<perlunicode>, L<utf8>. 1875 1876=head1 REFERENCES 1877 1878L<http://std.dkuug.dk/i18n/charmaps> 1879 1880L<https://www.unicode.org/> 1881 1882L<https://www.unicode.org/reports/tr16/> 1883 1884L<https://www.sr-ix.com/Archive/CharCodeHist/index.html> 1885B<ASCII: American Standard Code for Information Infiltration> Tom Jennings, 1886September 1999. 1887 1888B<The Unicode Standard, Version 3.0> The Unicode Consortium, Lisa Moore ed., 1889ISBN 0-201-61633-5, Addison Wesley Developers Press, February 2000. 1890 1891B<CDRA: IBM - Character Data Representation Architecture - 1892Reference and Registry>, IBM SC09-2190-00, December 1996. 1893 1894"Demystifying Character Sets", Andrea Vine, Multilingual Computing 1895& Technology, B<#26 Vol. 10 Issue 4>, August/September 1999; 1896ISSN 1523-0309; Multilingual Computing Inc. Sandpoint ID, USA. 1897 1898B<Codes, Ciphers, and Other Cryptic and Clandestine Communication> 1899Fred B. Wrixon, ISBN 1-57912-040-7, Black Dog & Leventhal Publishers, 19001998. 1901 1902L<http://www.bobbemer.com/P-BIT.HTM> 1903B<IBM - EBCDIC and the P-bit; The biggest Computer Goof Ever> Robert Bemer. 1904 1905=head1 HISTORY 1906 190715 April 2001: added UTF-8 and UTF-EBCDIC to main table, pvhp. 1908 1909=head1 AUTHOR 1910 1911Peter Prymmer pvhp@best.com wrote this in 1999 and 2000 1912with CCSID 0819 and 0037 help from Chris Leach and 1913AndrE<eacute> Pirard A.Pirard@ulg.ac.be as well as POSIX-BC 1914help from Thomas Dorner Thomas.Dorner@start.de. 1915Thanks also to Vickie Cooper, Philip Newton, William Raffloer, and 1916Joe Smith. Trademarks, registered trademarks, service marks and 1917registered service marks used in this document are the property of 1918their respective owners. 1919 1920Now maintained by Perl5 Porters. 1921