This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Fix use of == to compare strings in perlref.pod
[perl5.git] / pod / perlebcdic.pod
CommitLineData
49781f4a
AB
1=encoding utf8
2
d396a558
JH
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
4d2ca8b5 10on EBCDIC based computers.
d396a558 11
4d2ca8b5 12Portions of this document that are still incomplete are marked with XXX.
d396a558 13
4d2ca8b5
KW
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, not all
18the modules found on CPAN but shipped with core Perl work on z/OS.
19
20If you want to use Perl on a non-z/OS EBCDIC machine, please let us know
e1b711da
KW
21by sending mail to perlbug@perl.org
22
4d2ca8b5
KW
23Writing Perl on an EBCDIC platform is really no different than writing
24on an L</ASCII> one, but with different underlying numbers, as we'll see
25shortly. You'll have to know something about those L</ASCII> platforms
26because the documentation is biased and will frequently use example
27numbers that don't apply to EBCDIC. There are also very few CPAN
28modules that are written for EBCDIC and which don't work on ASCII;
29instead the vast majority of CPAN modules are written for ASCII, and
30some may happen to work on EBCDIC, while a few have been designed to
31portably work on both.
32
33If your code just uses the 52 letters A-Z and a-z, plus SPACE, the
34digits 0-9, and the punctuation characters that Perl uses, plus a few
35controls that are denoted by escape sequences like C<\n> and C<\t>, then
36there's nothing special about using Perl, and your code may very well
37work on an ASCII machine without change.
38
39But if you write code that uses C<\005> to mean a TAB or C<\xC1> to mean
40an "A", or C<\xDF> to mean a "E<yuml>" (small C<"y"> with a diaeresis),
41then your code may well work on your EBCDIC platform, but not on an
42ASCII one. That's fine to do if no one will ever want to run your code
43on an ASCII platform; but the bias in this document will be in writing
44code portable between EBCDIC and ASCII systems. Again, if every
45character you care about is easily enterable from your keyboard, you
46don't have to know anything about ASCII, but many keyboards don't easily
47allow you to directly enter, say, the character C<\xDF>, so you have to
48specify it indirectly, such as by using the C<"\xDF"> escape sequence.
49In those cases it's easiest to know something about the ASCII/Unicode
50character sets. If you know that the small "E<yuml>" is C<U+00FF>, then
51you can instead specify it as C<"\N{U+FF}">, and have the computer
52automatically translate it to C<\xDF> on your platform, and leave it as
53C<\xFF> on ASCII ones. Or you could specify it by name, C<\N{LATIN
54SMALL LETTER Y WITH DIAERESIS> and not have to know the numbers.
55Either way works, but require familiarity with Unicode.
56
d396a558
JH
57=head1 COMMON CHARACTER CODE SETS
58
59=head2 ASCII
60
4d2ca8b5
KW
61The American Standard Code for Information Interchange (ASCII or
62US-ASCII) is a set of
63integers running from 0 to 127 (decimal) that have standardized
64interpretations by the computers which use ASCII. For example, 65 means
65the letter "A".
eaf8b9b9
KW
66The range 0..127 can be covered by setting the bits in a 7-bit binary
67digit, hence the set is sometimes referred to as "7-bit ASCII".
68ASCII was described by the American National Standards Institute
69document ANSI X3.4-1986. It was also described by ISO 646:1991
70(with localization for currency symbols). The full ASCII set is
4d2ca8b5
KW
71given in the table L<below|/recipe 3> as the first 128 elements.
72Languages that
eaf8b9b9
KW
73can be written adequately with the characters in ASCII include
74English, Hawaiian, Indonesian, Swahili and some Native American
d396a558
JH
75languages.
76
4d2ca8b5
KW
77Most non-EBCDIC character sets are supersets of ASCII. That is the
78integers 0-127 mean what ASCII says they mean. But integers 128 and
79above are specific to the character set.
80
81Many of these fit entirely into 8 bits, using ASCII as 0-127, while
82specifying what 128-255 mean, and not using anything above 255.
83Thus, these are single-byte (or octet if you prefer) character sets.
84One important one (since Unicode is a superset of it) is the ISO 8859-1
85character set.
51b5cecb 86
d396a558
JH
87=head2 ISO 8859
88
4d2ca8b5
KW
89The ISO 8859-I<B<$n>> are a collection of character code sets from the
90International Organization for Standardization (ISO), each of which adds
91characters to the ASCII set that are typically found in various
5d9fe53c 92languages, many of which are based on the Roman, or Latin, alphabet.
4d2ca8b5
KW
93Most are for European languages, but there are also ones for Arabic,
94Greek, Hebrew, and Thai. There are good references on the web about
95all these.
d396a558
JH
96
97=head2 Latin 1 (ISO 8859-1)
98
eaf8b9b9
KW
99A particular 8-bit extension to ASCII that includes grave and acute
100accented Latin characters. Languages that can employ ISO 8859-1
101include all the languages covered by ASCII as well as Afrikaans,
102Albanian, Basque, Catalan, Danish, Faroese, Finnish, Norwegian,
103Portuguese, Spanish, and Swedish. Dutch is covered albeit without
104the ij ligature. French is covered too but without the oe ligature.
d396a558 105German can use ISO 8859-1 but must do so without German-style
eaf8b9b9 106quotation marks. This set is based on Western European extensions
d396a558 107to ASCII and is commonly encountered in world wide web work.
4d2ca8b5 108In IBM character code set identification terminology, ISO 8859-1 is
51b5cecb 109also known as CCSID 819 (or sometimes 0819 or even 00819).
d396a558
JH
110
111=head2 EBCDIC
112
eaf8b9b9 113The Extended Binary Coded Decimal Interchange Code refers to a
8a50e6a3 114large collection of single- and multi-byte coded character sets that are
4d2ca8b5
KW
115quite different from ASCII and ISO 8859-1, and are all slightly
116different from each other; they typically run on host computers. The
117EBCDIC encodings derive from 8-bit byte extensions of Hollerith punched
118card encodings, which long predate ASCII. The layout on the
119cards was such that high bits were set for the upper and lower case
120alphabetic
121characters C<[a-z]> and C<[A-Z]>, but there were gaps within each Latin
122alphabet range, visible in the table L<below|/recipe 3>. These gaps can
123cause complications.
d396a558 124
eaf8b9b9 125Some IBM EBCDIC character sets may be known by character code set
2c09a866 126identification numbers (CCSID numbers) or code page numbers.
51b5cecb 127
2bbc8d55
SP
128Perl can be compiled on platforms that run any of three commonly used EBCDIC
129character sets, listed below.
130
d5924ca6 131=head3 The 13 variant characters
1e054b24 132
51b5cecb
PP
133Among IBM EBCDIC character code sets there are 13 characters that
134are often mapped to different integer values. Those characters
135are known as the 13 "variant" characters and are:
d396a558 136
eaf8b9b9 137 \ [ ] { } ^ ~ ! # | $ @ `
d396a558 138
6ff677df 139When Perl is compiled for a platform, it looks at all of these characters to
2bbc8d55
SP
140guess which EBCDIC character set the platform uses, and adapts itself
141accordingly to that platform. If the platform uses a character set that is not
142one of the three Perl knows about, Perl will either fail to compile, or
143mistakenly and silently choose one of the three.
4d2ca8b5
KW
144
145=head3 EBCDIC code sets recognized by Perl
2bbc8d55 146
d5924ca6
KW
147=over
148
149=item B<0037>
d396a558 150
eaf8b9b9
KW
151Character code set ID 0037 is a mapping of the ASCII plus Latin-1
152characters (i.e. ISO 8859-1) to an EBCDIC set. 0037 is used
153in North American English locales on the OS/400 operating system
154that runs on AS/400 computers. CCSID 0037 differs from ISO 8859-1
a8f582bb 155in 236 places; in other words they agree on only 20 code point values.
d396a558 156
d5924ca6 157=item B<1047>
d396a558 158
eaf8b9b9
KW
159Character code set ID 1047 is also a mapping of the ASCII plus
160Latin-1 characters (i.e. ISO 8859-1) to an EBCDIC set. 1047 is
161used under Unix System Services for OS/390 or z/OS, and OpenEdition
a8f582bb
KW
162for VM/ESA. CCSID 1047 differs from CCSID 0037 in eight places,
163and from ISO 8859-1 in 236.
d396a558 164
d5924ca6 165=item B<POSIX-BC>
d396a558
JH
166
167The EBCDIC code page in use on Siemens' BS2000 system is distinct from
1681047 and 0037. It is identified below as the POSIX-BC set.
a8f582bb
KW
169Like 0037 and 1047, it is the same as ISO 8859-1 in 20 code point
170values.
d396a558 171
d5924ca6
KW
172=back
173
64c66fb6
JH
174=head2 Unicode code points versus EBCDIC code points
175
176In Unicode terminology a I<code point> is the number assigned to a
177character: for example, in EBCDIC the character "A" is usually assigned
4d2ca8b5
KW
178the number 193. In Unicode, the character "A" is assigned the number 65.
179All the code points in ASCII and Latin-1 (ISO 8859-1) have the same
180meaning in Unicode. All three of the recognized EBCDIC code sets have
181256 code points, and in each code set, all 256 code points are mapped to
182equivalent Latin1 code points. Obviously, "A" will map to "A", "B" =>
183"B", "%" => "%", etc., for all printable characters in Latin1 and these
184code pages.
185
186It also turns out that EBCDIC has nearly precise equivalents for the
187ASCII/Latin1 C0 controls and the DELETE control. (The C0 controls are
188those whose ASCII code points are 0..0x1F; things like TAB, ACK, BEL,
189etc.) A mapping is set up between these ASCII/EBCDIC controls. There
190isn't such a precise mapping between the C1 controls on ASCII platforms
191and the remaining EBCDIC controls. What has been done is to map these
192controls, mostly arbitrarily, to some otherwise unmatched character in
193the other character set. Most of these are very very rarely used
194nowadays in EBCDIC anyway, and their names have been dropped, without
195much complaint. For example the EO (Eight Ones) EBCDIC control
196(consisting of eight one bits = 0xFF) is mapped to the C1 APC control
197(0x9F), and you can't use the name "EO".
198
199The EBCDIC controls provide three possible line terminator characters,
200CR (0x0D), LF (0x25), and NL (0x15). On ASCII platforms, the symbols
201"NL" and "LF" refer to the same character, but in strict EBCDIC
202terminology they are different ones. The EBCDIC NL is mapped to the C1
203control called "NEL" ("Next Line"; here's a case where the mapping makes
204quite a bit of sense, and hence isn't just arbitrary). On some EBCDIC
205platforms, this NL or NEL is the typical line terminator. This is true
206of z/OS and BS2000. In these platforms, the C compilers will swap the
207LF and NEL code points, so that C<"\n"> is 0x15, and refers to NL. Perl
208does that too; you can see it in the code chart L<below|/recipe 3>.
209This makes things generally "just work" without you even having to be
210aware that there is a swap.
dc4af4bb 211
395f5a0c
PK
212=head2 Unicode and UTF
213
4d2ca8b5 214UTF stands for "Unicode Transformation Format".
2bbc8d55
SP
215UTF-8 is an encoding of Unicode into a sequence of 8-bit byte chunks, based on
216ASCII and Latin-1.
217The length of a sequence required to represent a Unicode code point
218depends on the ordinal number of that code point,
219with larger numbers requiring more bytes.
220UTF-EBCDIC is like UTF-8, but based on EBCDIC.
4d2ca8b5
KW
221They are enough alike that often, casual usage will conflate the two
222terms, and use "UTF-8" to mean both the UTF-8 found on ASCII platforms,
223and the UTF-EBCDIC found on EBCDIC ones.
2bbc8d55 224
4d2ca8b5 225You may see the term "invariant" character or code point.
fe749c9a 226This simply means that the character has the same numeric
4d2ca8b5
KW
227value and representation when encoded in UTF-8 (or UTF-EBCDIC) as when
228not. (Note that this is a very different concept from L</The 13 variant
229characters> mentioned above. Careful prose will use the term "UTF-8
230invariant" instead of just "invariant", but most often you'll see just
231"invariant".) For example, the ordinal value of "A" is 193 in most
232EBCDIC code pages, and also is 193 when encoded in UTF-EBCDIC. All
233UTF-8 (or UTF-EBCDIC) variant code points occupy at least two bytes when
234encoded in UTF-8 (or UTF-EBCDIC); by definition, the UTF-8 (or
235UTF-EBCDIC) invariant code points are exactly one byte whether encoded
236in UTF-8 (or UTF-EBCDIC), or not. (By now you see why people typically
237just say "UTF-8" when they also mean "UTF-EBCDIC". For the rest of this
238document, we'll mostly be casual about it too.)
239In ASCII UTF-8, the code points corresponding to the lowest 128
fe749c9a
KW
240ordinal numbers (0 - 127: the ASCII characters) are invariant.
241In UTF-EBCDIC, there are 160 invariant characters.
2bbc8d55 242(If you care, the EBCDIC invariants are those characters
fe749c9a 243which have ASCII equivalents, plus those that correspond to
4d2ca8b5 244the C1 controls (128 - 159 on ASCII platforms).)
fe749c9a 245
c0236afe
KW
246A string encoded in UTF-EBCDIC may be longer (very rarely shorter) than
247one encoded in UTF-8. Perl extends both UTF-8 and UTF-EBCDIC so that
248they can encode code points above the Unicode maximum of U+10FFFF. Both
249extensions are constructed to allow encoding of any code point that fits
250in a 64-bit word.
4d2ca8b5
KW
251
252UTF-EBCDIC is defined by
c0236afe
KW
253L<Unicode Technical Report #16|http://www.unicode.org/reports/tr16>
254(often referred to as just TR16).
4d2ca8b5
KW
255It is defined based on CCSID 1047, not allowing for the differences for
256other code pages. This allows for easy interchange of text between
257computers running different code pages, but makes it unusable, without
258adaptation, for Perl on those other code pages.
259
260The reason for this unusability is that a fundamental assumption of Perl
261is that the characters it cares about for parsing and lexical analysis
262are the same whether or not the text is in UTF-8. For example, Perl
263expects the character C<"["> to have the same representation, no matter
264if the string containing it (or program text) is UTF-8 encoded or not.
265To ensure this, Perl adapts UTF-EBCDIC to the particular code page so
266that all characters it expects to be UTF-8 invariant are in fact UTF-8
267invariant. This means that text generated on a computer running one
268version of Perl's UTF-EBCDIC has to be translated to be intelligible to
269a computer running another.
395f5a0c 270
c0236afe
KW
271TR16 implies a method to extend UTF-EBCDIC to encode points up through
272S<C<2 ** 31 - 1>>. Perl uses this method for code points up through
273S<C<2 ** 30 - 1>>, but uses an incompatible method for larger ones, to
274enable it to handle much larger code points than otherwise.
275
8704cfd1 276=head2 Using Encode
8f94de01 277
4d2ca8b5 278Starting from Perl 5.8 you can use the standard module Encode
2bbc8d55
SP
279to translate from EBCDIC to Latin-1 code points.
280Encode knows about more EBCDIC character sets than Perl can currently
281be compiled to run on.
8f94de01 282
c72e675e 283 use Encode 'from_to';
8f94de01 284
c72e675e 285 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' );
8f94de01 286
c72e675e
KW
287 # $a is in EBCDIC code points
288 from_to($a, $ebcdic{ord '^'}, 'latin1');
289 # $a is ISO 8859-1 code points
8f94de01
JH
290
291and from Latin-1 code points to EBCDIC code points
292
c72e675e 293 use Encode 'from_to';
8f94de01 294
c72e675e 295 my %ebcdic = ( 176 => 'cp37', 95 => 'cp1047', 106 => 'posix-bc' );
8f94de01 296
c72e675e
KW
297 # $a is ISO 8859-1 code points
298 from_to($a, 'latin1', $ebcdic{ord '^'});
299 # $a is in EBCDIC code points
8f94de01
JH
300
301For doing I/O it is suggested that you use the autotranslating features
302of PerlIO, see L<perluniintro>.
303
4d2ca8b5 304Since version 5.8 Perl uses the PerlIO I/O library. This enables
aa2b82fc
JH
305you to use different encodings per IO channel. For example you may use
306
307 use Encode;
308 open($f, ">:encoding(ascii)", "test.ascii");
309 print $f "Hello World!\n";
310 open($f, ">:encoding(cp37)", "test.ebcdic");
311 print $f "Hello World!\n";
312 open($f, ">:encoding(latin1)", "test.latin1");
313 print $f "Hello World!\n";
314 open($f, ">:encoding(utf8)", "test.utf8");
315 print $f "Hello World!\n";
316
2c09a866 317to get four files containing "Hello World!\n" in ASCII, CP 0037 EBCDIC,
2bbc8d55 318ISO 8859-1 (Latin-1) (in this example identical to ASCII since only ASCII
eaf8b9b9 319characters were printed), and
2bbc8d55
SP
320UTF-EBCDIC (in this example identical to normal EBCDIC since only characters
321that don't differ between EBCDIC and UTF-EBCDIC were printed). See the
4d2ca8b5 322documentation of L<Encode::PerlIO> for details.
aa2b82fc
JH
323
324As the PerlIO layer uses raw IO (bytes) internally, all this totally
325ignores things like the type of your filesystem (ASCII or EBCDIC).
326
d396a558
JH
327=head1 SINGLE OCTET TABLES
328
329The following tables list the ASCII and Latin 1 ordered sets including
330the subsets: C0 controls (0..31), ASCII graphics (32..7e), delete (7f),
eaf8b9b9 331C1 controls (80..9f), and Latin-1 (a.k.a. ISO 8859-1) (a0..ff). In the
8d725451 332table names of the Latin 1
eaf8b9b9 333extensions to ASCII have been labelled with character names roughly
8d725451 334corresponding to I<The Unicode Standard, Version 6.1> albeit with
4d2ca8b5
KW
335substitutions such as C<s/LATIN//> and C<s/VULGAR//> in all cases;
336S<C<s/CAPITAL LETTER//>> in some cases; and
337S<C<s/SMALL LETTER ([A-Z])/\l$1/>> in some other
0e56abba 338cases. Controls are listed using their Unicode 6.2 abbreviations.
eaf8b9b9 339The differences between the 0037 and 1047 sets are
4d2ca8b5
KW
340flagged with C<**>. The differences between the 1047 and POSIX-BC sets
341are flagged with C<##.> All C<ord()> numbers listed are decimal. If you
8d725451
KW
342would rather see this table listing octal values, then run the table
343(that is, the pod source text of this document, since this recipe may not
1e054b24 344work with a pod2_other_format translation) through:
d396a558
JH
345
346=over 4
347
348=item recipe 0
349
350=back
351
8d725451
KW
352 perl -ne 'if(/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
353 -e '{printf("%s%-5.03o%-5.03o%-5.03o%.03o\n",$1,$2,$3,$4,$5)}' \
5f26d5fd 354 perlebcdic.pod
395f5a0c
PK
355
356If you want to retain the UTF-x code points then in script form you
357might want to write:
358
359=over 4
360
361=item recipe 1
362
363=back
364
c72e675e
KW
365 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!";
366 while (<FH>) {
f11f9c4c
KW
367 if (/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*)
368 \s+(\d+)\.?(\d*)/x)
5f26d5fd 369 {
c72e675e 370 if ($7 ne '' && $9 ne '') {
5f26d5fd 371 printf(
8d725451 372 "%s%-5.03o%-5.03o%-5.03o%-5.03o%-3o.%-5o%-3o.%.03o\n",
5f26d5fd 373 $1,$2,$3,$4,$5,$6,$7,$8,$9);
c72e675e
KW
374 }
375 elsif ($7 ne '') {
8d725451 376 printf("%s%-5.03o%-5.03o%-5.03o%-5.03o%-3o.%-5o%.03o\n",
c72e675e
KW
377 $1,$2,$3,$4,$5,$6,$7,$8);
378 }
379 else {
8d725451 380 printf("%s%-5.03o%-5.03o%-5.03o%-5.03o%-5.03o%.03o\n",
5f26d5fd 381 $1,$2,$3,$4,$5,$6,$8);
c72e675e
KW
382 }
383 }
384 }
d396a558
JH
385
386If you would rather see this table listing hexadecimal values then
387run the table through:
388
389=over 4
390
395f5a0c 391=item recipe 2
d396a558
JH
392
393=back
394
8d725451
KW
395 perl -ne 'if(/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/)' \
396 -e '{printf("%s%-5.02X%-5.02X%-5.02X%.02X\n",$1,$2,$3,$4,$5)}' \
5f26d5fd 397 perlebcdic.pod
395f5a0c
PK
398
399Or, in order to retain the UTF-x code points in hexadecimal:
400
401=over 4
402
403=item recipe 3
404
405=back
406
c72e675e
KW
407 open(FH,"<perlebcdic.pod") or die "Could not open perlebcdic.pod: $!";
408 while (<FH>) {
f11f9c4c
KW
409 if (/(.{29})(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\.?(\d*)
410 \s+(\d+)\.?(\d*)/x)
5f26d5fd 411 {
c72e675e 412 if ($7 ne '' && $9 ne '') {
5f26d5fd 413 printf(
8d725451 414 "%s%-5.02X%-5.02X%-5.02X%-5.02X%-2X.%-6.02X%02X.%02X\n",
c72e675e
KW
415 $1,$2,$3,$4,$5,$6,$7,$8,$9);
416 }
417 elsif ($7 ne '') {
8d725451 418 printf("%s%-5.02X%-5.02X%-5.02X%-5.02X%-2X.%-6.02X%02X\n",
c72e675e
KW
419 $1,$2,$3,$4,$5,$6,$7,$8);
420 }
421 else {
8d725451 422 printf("%s%-5.02X%-5.02X%-5.02X%-5.02X%-5.02X%02X\n",
5f26d5fd 423 $1,$2,$3,$4,$5,$6,$8);
c72e675e
KW
424 }
425 }
426 }
395f5a0c
PK
427
428
8d725451 429 ISO
f11f9c4c
KW
430 8859-1 POS- CCSID
431 CCSID CCSID CCSID IX- 1047
8d725451
KW
432 chr 0819 0037 1047 BC UTF-8 UTF-EBCDIC
433 ---------------------------------------------------------------------
434 <NUL> 0 0 0 0 0 0
435 <SOH> 1 1 1 1 1 1
436 <STX> 2 2 2 2 2 2
437 <ETX> 3 3 3 3 3 3
438 <EOT> 4 55 55 55 4 55
439 <ENQ> 5 45 45 45 5 45
440 <ACK> 6 46 46 46 6 46
441 <BEL> 7 47 47 47 7 47
442 <BS> 8 22 22 22 8 22
443 <HT> 9 5 5 5 9 5
444 <LF> 10 37 21 21 10 21 **
445 <VT> 11 11 11 11 11 11
446 <FF> 12 12 12 12 12 12
447 <CR> 13 13 13 13 13 13
448 <SO> 14 14 14 14 14 14
449 <SI> 15 15 15 15 15 15
450 <DLE> 16 16 16 16 16 16
451 <DC1> 17 17 17 17 17 17
452 <DC2> 18 18 18 18 18 18
453 <DC3> 19 19 19 19 19 19
454 <DC4> 20 60 60 60 20 60
455 <NAK> 21 61 61 61 21 61
456 <SYN> 22 50 50 50 22 50
457 <ETB> 23 38 38 38 23 38
458 <CAN> 24 24 24 24 24 24
459 <EOM> 25 25 25 25 25 25
460 <SUB> 26 63 63 63 26 63
461 <ESC> 27 39 39 39 27 39
462 <FS> 28 28 28 28 28 28
463 <GS> 29 29 29 29 29 29
464 <RS> 30 30 30 30 30 30
465 <US> 31 31 31 31 31 31
466 <SPACE> 32 64 64 64 32 64
467 ! 33 90 90 90 33 90
468 " 34 127 127 127 34 127
469 # 35 123 123 123 35 123
470 $ 36 91 91 91 36 91
471 % 37 108 108 108 37 108
472 & 38 80 80 80 38 80
473 ' 39 125 125 125 39 125
474 ( 40 77 77 77 40 77
475 ) 41 93 93 93 41 93
476 * 42 92 92 92 42 92
477 + 43 78 78 78 43 78
478 , 44 107 107 107 44 107
479 - 45 96 96 96 45 96
480 . 46 75 75 75 46 75
481 / 47 97 97 97 47 97
482 0 48 240 240 240 48 240
483 1 49 241 241 241 49 241
484 2 50 242 242 242 50 242
485 3 51 243 243 243 51 243
486 4 52 244 244 244 52 244
487 5 53 245 245 245 53 245
488 6 54 246 246 246 54 246
489 7 55 247 247 247 55 247
490 8 56 248 248 248 56 248
491 9 57 249 249 249 57 249
492 : 58 122 122 122 58 122
493 ; 59 94 94 94 59 94
494 < 60 76 76 76 60 76
495 = 61 126 126 126 61 126
496 > 62 110 110 110 62 110
497 ? 63 111 111 111 63 111
498 @ 64 124 124 124 64 124
499 A 65 193 193 193 65 193
500 B 66 194 194 194 66 194
501 C 67 195 195 195 67 195
502 D 68 196 196 196 68 196
503 E 69 197 197 197 69 197
504 F 70 198 198 198 70 198
505 G 71 199 199 199 71 199
506 H 72 200 200 200 72 200
507 I 73 201 201 201 73 201
508 J 74 209 209 209 74 209
509 K 75 210 210 210 75 210
510 L 76 211 211 211 76 211
511 M 77 212 212 212 77 212
512 N 78 213 213 213 78 213
513 O 79 214 214 214 79 214
514 P 80 215 215 215 80 215
515 Q 81 216 216 216 81 216
516 R 82 217 217 217 82 217
517 S 83 226 226 226 83 226
518 T 84 227 227 227 84 227
519 U 85 228 228 228 85 228
520 V 86 229 229 229 86 229
521 W 87 230 230 230 87 230
522 X 88 231 231 231 88 231
523 Y 89 232 232 232 89 232
524 Z 90 233 233 233 90 233
525 [ 91 186 173 187 91 173 ** ##
526 \ 92 224 224 188 92 224 ##
527 ] 93 187 189 189 93 189 **
528 ^ 94 176 95 106 94 95 ** ##
529 _ 95 109 109 109 95 109
530 ` 96 121 121 74 96 121 ##
531 a 97 129 129 129 97 129
532 b 98 130 130 130 98 130
533 c 99 131 131 131 99 131
534 d 100 132 132 132 100 132
535 e 101 133 133 133 101 133
536 f 102 134 134 134 102 134
537 g 103 135 135 135 103 135
538 h 104 136 136 136 104 136
539 i 105 137 137 137 105 137
540 j 106 145 145 145 106 145
541 k 107 146 146 146 107 146
542 l 108 147 147 147 108 147
543 m 109 148 148 148 109 148
544 n 110 149 149 149 110 149
545 o 111 150 150 150 111 150
546 p 112 151 151 151 112 151
547 q 113 152 152 152 113 152
548 r 114 153 153 153 114 153
549 s 115 162 162 162 115 162
550 t 116 163 163 163 116 163
551 u 117 164 164 164 117 164
552 v 118 165 165 165 118 165
553 w 119 166 166 166 119 166
554 x 120 167 167 167 120 167
555 y 121 168 168 168 121 168
556 z 122 169 169 169 122 169
557 { 123 192 192 251 123 192 ##
558 | 124 79 79 79 124 79
559 } 125 208 208 253 125 208 ##
560 ~ 126 161 161 255 126 161 ##
561 <DEL> 127 7 7 7 127 7
562 <PAD> 128 32 32 32 194.128 32
563 <HOP> 129 33 33 33 194.129 33
564 <BPH> 130 34 34 34 194.130 34
565 <NBH> 131 35 35 35 194.131 35
566 <IND> 132 36 36 36 194.132 36
567 <NEL> 133 21 37 37 194.133 37 **
568 <SSA> 134 6 6 6 194.134 6
569 <ESA> 135 23 23 23 194.135 23
570 <HTS> 136 40 40 40 194.136 40
571 <HTJ> 137 41 41 41 194.137 41
572 <VTS> 138 42 42 42 194.138 42
573 <PLD> 139 43 43 43 194.139 43
574 <PLU> 140 44 44 44 194.140 44
575 <RI> 141 9 9 9 194.141 9
576 <SS2> 142 10 10 10 194.142 10
577 <SS3> 143 27 27 27 194.143 27
578 <DCS> 144 48 48 48 194.144 48
579 <PU1> 145 49 49 49 194.145 49
580 <PU2> 146 26 26 26 194.146 26
581 <STS> 147 51 51 51 194.147 51
582 <CCH> 148 52 52 52 194.148 52
583 <MW> 149 53 53 53 194.149 53
584 <SPA> 150 54 54 54 194.150 54
585 <EPA> 151 8 8 8 194.151 8
586 <SOS> 152 56 56 56 194.152 56
587 <SGC> 153 57 57 57 194.153 57
588 <SCI> 154 58 58 58 194.154 58
589 <CSI> 155 59 59 59 194.155 59
590 <ST> 156 4 4 4 194.156 4
591 <OSC> 157 20 20 20 194.157 20
592 <PM> 158 62 62 62 194.158 62
593 <APC> 159 255 255 95 194.159 255 ##
594 <NON-BREAKING SPACE> 160 65 65 65 194.160 128.65
595 <INVERTED "!" > 161 170 170 170 194.161 128.66
596 <CENT SIGN> 162 74 74 176 194.162 128.67 ##
597 <POUND SIGN> 163 177 177 177 194.163 128.68
598 <CURRENCY SIGN> 164 159 159 159 194.164 128.69
599 <YEN SIGN> 165 178 178 178 194.165 128.70
600 <BROKEN BAR> 166 106 106 208 194.166 128.71 ##
601 <SECTION SIGN> 167 181 181 181 194.167 128.72
602 <DIAERESIS> 168 189 187 121 194.168 128.73 ** ##
603 <COPYRIGHT SIGN> 169 180 180 180 194.169 128.74
604 <FEMININE ORDINAL> 170 154 154 154 194.170 128.81
605 <LEFT POINTING GUILLEMET> 171 138 138 138 194.171 128.82
606 <NOT SIGN> 172 95 176 186 194.172 128.83 ** ##
607 <SOFT HYPHEN> 173 202 202 202 194.173 128.84
608 <REGISTERED TRADE MARK> 174 175 175 175 194.174 128.85
609 <MACRON> 175 188 188 161 194.175 128.86 ##
610 <DEGREE SIGN> 176 144 144 144 194.176 128.87
611 <PLUS-OR-MINUS SIGN> 177 143 143 143 194.177 128.88
612 <SUPERSCRIPT TWO> 178 234 234 234 194.178 128.89
613 <SUPERSCRIPT THREE> 179 250 250 250 194.179 128.98
614 <ACUTE ACCENT> 180 190 190 190 194.180 128.99
615 <MICRO SIGN> 181 160 160 160 194.181 128.100
616 <PARAGRAPH SIGN> 182 182 182 182 194.182 128.101
617 <MIDDLE DOT> 183 179 179 179 194.183 128.102
618 <CEDILLA> 184 157 157 157 194.184 128.103
619 <SUPERSCRIPT ONE> 185 218 218 218 194.185 128.104
620 <MASC. ORDINAL INDICATOR> 186 155 155 155 194.186 128.105
621 <RIGHT POINTING GUILLEMET> 187 139 139 139 194.187 128.106
622 <FRACTION ONE QUARTER> 188 183 183 183 194.188 128.112
623 <FRACTION ONE HALF> 189 184 184 184 194.189 128.113
624 <FRACTION THREE QUARTERS> 190 185 185 185 194.190 128.114
625 <INVERTED QUESTION MARK> 191 171 171 171 194.191 128.115
626 <A WITH GRAVE> 192 100 100 100 195.128 138.65
627 <A WITH ACUTE> 193 101 101 101 195.129 138.66
628 <A WITH CIRCUMFLEX> 194 98 98 98 195.130 138.67
629 <A WITH TILDE> 195 102 102 102 195.131 138.68
630 <A WITH DIAERESIS> 196 99 99 99 195.132 138.69
631 <A WITH RING ABOVE> 197 103 103 103 195.133 138.70
632 <CAPITAL LIGATURE AE> 198 158 158 158 195.134 138.71
633 <C WITH CEDILLA> 199 104 104 104 195.135 138.72
634 <E WITH GRAVE> 200 116 116 116 195.136 138.73
635 <E WITH ACUTE> 201 113 113 113 195.137 138.74
636 <E WITH CIRCUMFLEX> 202 114 114 114 195.138 138.81
637 <E WITH DIAERESIS> 203 115 115 115 195.139 138.82
638 <I WITH GRAVE> 204 120 120 120 195.140 138.83
639 <I WITH ACUTE> 205 117 117 117 195.141 138.84
640 <I WITH CIRCUMFLEX> 206 118 118 118 195.142 138.85
641 <I WITH DIAERESIS> 207 119 119 119 195.143 138.86
642 <CAPITAL LETTER ETH> 208 172 172 172 195.144 138.87
643 <N WITH TILDE> 209 105 105 105 195.145 138.88
644 <O WITH GRAVE> 210 237 237 237 195.146 138.89
645 <O WITH ACUTE> 211 238 238 238 195.147 138.98
646 <O WITH CIRCUMFLEX> 212 235 235 235 195.148 138.99
647 <O WITH TILDE> 213 239 239 239 195.149 138.100
648 <O WITH DIAERESIS> 214 236 236 236 195.150 138.101
649 <MULTIPLICATION SIGN> 215 191 191 191 195.151 138.102
650 <O WITH STROKE> 216 128 128 128 195.152 138.103
651 <U WITH GRAVE> 217 253 253 224 195.153 138.104 ##
652 <U WITH ACUTE> 218 254 254 254 195.154 138.105
653 <U WITH CIRCUMFLEX> 219 251 251 221 195.155 138.106 ##
654 <U WITH DIAERESIS> 220 252 252 252 195.156 138.112
655 <Y WITH ACUTE> 221 173 186 173 195.157 138.113 ** ##
656 <CAPITAL LETTER THORN> 222 174 174 174 195.158 138.114
657 <SMALL LETTER SHARP S> 223 89 89 89 195.159 138.115
658 <a WITH GRAVE> 224 68 68 68 195.160 139.65
659 <a WITH ACUTE> 225 69 69 69 195.161 139.66
660 <a WITH CIRCUMFLEX> 226 66 66 66 195.162 139.67
661 <a WITH TILDE> 227 70 70 70 195.163 139.68
662 <a WITH DIAERESIS> 228 67 67 67 195.164 139.69
663 <a WITH RING ABOVE> 229 71 71 71 195.165 139.70
664 <SMALL LIGATURE ae> 230 156 156 156 195.166 139.71
665 <c WITH CEDILLA> 231 72 72 72 195.167 139.72
666 <e WITH GRAVE> 232 84 84 84 195.168 139.73
667 <e WITH ACUTE> 233 81 81 81 195.169 139.74
668 <e WITH CIRCUMFLEX> 234 82 82 82 195.170 139.81
669 <e WITH DIAERESIS> 235 83 83 83 195.171 139.82
670 <i WITH GRAVE> 236 88 88 88 195.172 139.83
671 <i WITH ACUTE> 237 85 85 85 195.173 139.84
672 <i WITH CIRCUMFLEX> 238 86 86 86 195.174 139.85
673 <i WITH DIAERESIS> 239 87 87 87 195.175 139.86
674 <SMALL LETTER eth> 240 140 140 140 195.176 139.87
675 <n WITH TILDE> 241 73 73 73 195.177 139.88
676 <o WITH GRAVE> 242 205 205 205 195.178 139.89
677 <o WITH ACUTE> 243 206 206 206 195.179 139.98
678 <o WITH CIRCUMFLEX> 244 203 203 203 195.180 139.99
679 <o WITH TILDE> 245 207 207 207 195.181 139.100
680 <o WITH DIAERESIS> 246 204 204 204 195.182 139.101
681 <DIVISION SIGN> 247 225 225 225 195.183 139.102
682 <o WITH STROKE> 248 112 112 112 195.184 139.103
683 <u WITH GRAVE> 249 221 221 192 195.185 139.104 ##
684 <u WITH ACUTE> 250 222 222 222 195.186 139.105
685 <u WITH CIRCUMFLEX> 251 219 219 219 195.187 139.106
686 <u WITH DIAERESIS> 252 220 220 220 195.188 139.112
687 <y WITH ACUTE> 253 141 141 141 195.189 139.113
688 <SMALL LETTER thorn> 254 142 142 142 195.190 139.114
689 <y WITH DIAERESIS> 255 223 223 223 195.191 139.115
d396a558
JH
690
691If you would rather see the above table in CCSID 0037 order rather than
692ASCII + Latin-1 order then run the table through:
693
694=over 4
695
395f5a0c 696=item recipe 4
d396a558
JH
697
698=back
699
5f26d5fd 700 perl \
8d725451 701 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\
84f709e7
JH
702 -e '{push(@l,$_)}' \
703 -e 'END{print map{$_->[0]}' \
704 -e ' sort{$a->[1] <=> $b->[1]}' \
8d725451 705 -e ' map{[$_,substr($_,34,3)]}@l;}' perlebcdic.pod
d396a558 706
2c09a866 707If you would rather see it in CCSID 1047 order then change the number
8d725451 70834 in the last line to 39, like this:
d396a558
JH
709
710=over 4
711
395f5a0c 712=item recipe 5
d396a558
JH
713
714=back
715
5f26d5fd 716 perl \
8d725451 717 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\
5f26d5fd
KW
718 -e '{push(@l,$_)}' \
719 -e 'END{print map{$_->[0]}' \
720 -e ' sort{$a->[1] <=> $b->[1]}' \
8d725451 721 -e ' map{[$_,substr($_,39,3)]}@l;}' perlebcdic.pod
d396a558 722
2c09a866 723If you would rather see it in POSIX-BC order then change the number
4d2ca8b5 72434 in the last line to 44, like this:
d396a558
JH
725
726=over 4
727
395f5a0c 728=item recipe 6
d396a558
JH
729
730=back
731
5f26d5fd 732 perl \
8d725451 733 -ne 'if(/.{29}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}\s{2,4}\d{1,3}/)'\
84f709e7
JH
734 -e '{push(@l,$_)}' \
735 -e 'END{print map{$_->[0]}' \
736 -e ' sort{$a->[1] <=> $b->[1]}' \
8d725451 737 -e ' map{[$_,substr($_,44,3)]}@l;}' perlebcdic.pod
d396a558 738
4d2ca8b5 739=head2 Table in hex, sorted in 1047 order
d396a558 740
4d2ca8b5
KW
741Since this document was first written, the convention has become more
742and more to use hexadecimal notation for code points. To do this with
743the recipes and to also sort is a multi-step process, so here, for
744convenience, is the table from above, re-sorted to be in Code Page 1047
745order, and using hex notation.
d396a558 746
4d2ca8b5
KW
747 ISO
748 8859-1 POS- CCSID
749 CCSID CCSID CCSID IX- 1047
750 chr 0819 0037 1047 BC UTF-8 UTF-EBCDIC
751 ---------------------------------------------------------------------
752 <NUL> 00 00 00 00 00 00
753 <SOH> 01 01 01 01 01 01
754 <STX> 02 02 02 02 02 02
755 <ETX> 03 03 03 03 03 03
756 <ST> 9C 04 04 04 C2.9C 04
757 <HT> 09 05 05 05 09 05
758 <SSA> 86 06 06 06 C2.86 06
759 <DEL> 7F 07 07 07 7F 07
760 <EPA> 97 08 08 08 C2.97 08
761 <RI> 8D 09 09 09 C2.8D 09
762 <SS2> 8E 0A 0A 0A C2.8E 0A
763 <VT> 0B 0B 0B 0B 0B 0B
764 <FF> 0C 0C 0C 0C 0C 0C
765 <CR> 0D 0D 0D 0D 0D 0D
766 <SO> 0E 0E 0E 0E 0E 0E
767 <SI> 0F 0F 0F 0F 0F 0F
768 <DLE> 10 10 10 10 10 10
769 <DC1> 11 11 11 11 11 11
770 <DC2> 12 12 12 12 12 12
771 <DC3> 13 13 13 13 13 13
772 <OSC> 9D 14 14 14 C2.9D 14
773 <LF> 0A 25 15 15 0A 15 **
774 <BS> 08 16 16 16 08 16
775 <ESA> 87 17 17 17 C2.87 17
776 <CAN> 18 18 18 18 18 18
777 <EOM> 19 19 19 19 19 19
778 <PU2> 92 1A 1A 1A C2.92 1A
779 <SS3> 8F 1B 1B 1B C2.8F 1B
780 <FS> 1C 1C 1C 1C 1C 1C
781 <GS> 1D 1D 1D 1D 1D 1D
782 <RS> 1E 1E 1E 1E 1E 1E
783 <US> 1F 1F 1F 1F 1F 1F
784 <PAD> 80 20 20 20 C2.80 20
785 <HOP> 81 21 21 21 C2.81 21
786 <BPH> 82 22 22 22 C2.82 22
787 <NBH> 83 23 23 23 C2.83 23
788 <IND> 84 24 24 24 C2.84 24
789 <NEL> 85 15 25 25 C2.85 25 **
790 <ETB> 17 26 26 26 17 26
791 <ESC> 1B 27 27 27 1B 27
792 <HTS> 88 28 28 28 C2.88 28
793 <HTJ> 89 29 29 29 C2.89 29
794 <VTS> 8A 2A 2A 2A C2.8A 2A
795 <PLD> 8B 2B 2B 2B C2.8B 2B
796 <PLU> 8C 2C 2C 2C C2.8C 2C
797 <ENQ> 05 2D 2D 2D 05 2D
798 <ACK> 06 2E 2E 2E 06 2E
799 <BEL> 07 2F 2F 2F 07 2F
800 <DCS> 90 30 30 30 C2.90 30
801 <PU1> 91 31 31 31 C2.91 31
802 <SYN> 16 32 32 32 16 32
803 <STS> 93 33 33 33 C2.93 33
804 <CCH> 94 34 34 34 C2.94 34
805 <MW> 95 35 35 35 C2.95 35
806 <SPA> 96 36 36 36 C2.96 36
807 <EOT> 04 37 37 37 04 37
808 <SOS> 98 38 38 38 C2.98 38
809 <SGC> 99 39 39 39 C2.99 39
810 <SCI> 9A 3A 3A 3A C2.9A 3A
811 <CSI> 9B 3B 3B 3B C2.9B 3B
812 <DC4> 14 3C 3C 3C 14 3C
813 <NAK> 15 3D 3D 3D 15 3D
814 <PM> 9E 3E 3E 3E C2.9E 3E
815 <SUB> 1A 3F 3F 3F 1A 3F
816 <SPACE> 20 40 40 40 20 40
817 <NON-BREAKING SPACE> A0 41 41 41 C2.A0 80.41
818 <a WITH CIRCUMFLEX> E2 42 42 42 C3.A2 8B.43
819 <a WITH DIAERESIS> E4 43 43 43 C3.A4 8B.45
820 <a WITH GRAVE> E0 44 44 44 C3.A0 8B.41
821 <a WITH ACUTE> E1 45 45 45 C3.A1 8B.42
822 <a WITH TILDE> E3 46 46 46 C3.A3 8B.44
823 <a WITH RING ABOVE> E5 47 47 47 C3.A5 8B.46
824 <c WITH CEDILLA> E7 48 48 48 C3.A7 8B.48
825 <n WITH TILDE> F1 49 49 49 C3.B1 8B.58
826 <CENT SIGN> A2 4A 4A B0 C2.A2 80.43 ##
827 . 2E 4B 4B 4B 2E 4B
828 < 3C 4C 4C 4C 3C 4C
829 ( 28 4D 4D 4D 28 4D
830 + 2B 4E 4E 4E 2B 4E
831 | 7C 4F 4F 4F 7C 4F
832 & 26 50 50 50 26 50
833 <e WITH ACUTE> E9 51 51 51 C3.A9 8B.4A
834 <e WITH CIRCUMFLEX> EA 52 52 52 C3.AA 8B.51
835 <e WITH DIAERESIS> EB 53 53 53 C3.AB 8B.52
836 <e WITH GRAVE> E8 54 54 54 C3.A8 8B.49
837 <i WITH ACUTE> ED 55 55 55 C3.AD 8B.54
838 <i WITH CIRCUMFLEX> EE 56 56 56 C3.AE 8B.55
839 <i WITH DIAERESIS> EF 57 57 57 C3.AF 8B.56
840 <i WITH GRAVE> EC 58 58 58 C3.AC 8B.53
841 <SMALL LETTER SHARP S> DF 59 59 59 C3.9F 8A.73
842 ! 21 5A 5A 5A 21 5A
843 $ 24 5B 5B 5B 24 5B
844 * 2A 5C 5C 5C 2A 5C
845 ) 29 5D 5D 5D 29 5D
846 ; 3B 5E 5E 5E 3B 5E
847 ^ 5E B0 5F 6A 5E 5F ** ##
848 - 2D 60 60 60 2D 60
849 / 2F 61 61 61 2F 61
850 <A WITH CIRCUMFLEX> C2 62 62 62 C3.82 8A.43
851 <A WITH DIAERESIS> C4 63 63 63 C3.84 8A.45
852 <A WITH GRAVE> C0 64 64 64 C3.80 8A.41
853 <A WITH ACUTE> C1 65 65 65 C3.81 8A.42
854 <A WITH TILDE> C3 66 66 66 C3.83 8A.44
855 <A WITH RING ABOVE> C5 67 67 67 C3.85 8A.46
856 <C WITH CEDILLA> C7 68 68 68 C3.87 8A.48
857 <N WITH TILDE> D1 69 69 69 C3.91 8A.58
858 <BROKEN BAR> A6 6A 6A D0 C2.A6 80.47 ##
859 , 2C 6B 6B 6B 2C 6B
860 % 25 6C 6C 6C 25 6C
861 _ 5F 6D 6D 6D 5F 6D
862 > 3E 6E 6E 6E 3E 6E
863 ? 3F 6F 6F 6F 3F 6F
864 <o WITH STROKE> F8 70 70 70 C3.B8 8B.67
865 <E WITH ACUTE> C9 71 71 71 C3.89 8A.4A
866 <E WITH CIRCUMFLEX> CA 72 72 72 C3.8A 8A.51
867 <E WITH DIAERESIS> CB 73 73 73 C3.8B 8A.52
868 <E WITH GRAVE> C8 74 74 74 C3.88 8A.49
869 <I WITH ACUTE> CD 75 75 75 C3.8D 8A.54
870 <I WITH CIRCUMFLEX> CE 76 76 76 C3.8E 8A.55
871 <I WITH DIAERESIS> CF 77 77 77 C3.8F 8A.56
872 <I WITH GRAVE> CC 78 78 78 C3.8C 8A.53
873 ` 60 79 79 4A 60 79 ##
874 : 3A 7A 7A 7A 3A 7A
875 # 23 7B 7B 7B 23 7B
876 @ 40 7C 7C 7C 40 7C
877 ' 27 7D 7D 7D 27 7D
878 = 3D 7E 7E 7E 3D 7E
879 " 22 7F 7F 7F 22 7F
880 <O WITH STROKE> D8 80 80 80 C3.98 8A.67
881 a 61 81 81 81 61 81
882 b 62 82 82 82 62 82
883 c 63 83 83 83 63 83
884 d 64 84 84 84 64 84
885 e 65 85 85 85 65 85
886 f 66 86 86 86 66 86
887 g 67 87 87 87 67 87
888 h 68 88 88 88 68 88
889 i 69 89 89 89 69 89
890 <LEFT POINTING GUILLEMET> AB 8A 8A 8A C2.AB 80.52
891 <RIGHT POINTING GUILLEMET> BB 8B 8B 8B C2.BB 80.6A
892 <SMALL LETTER eth> F0 8C 8C 8C C3.B0 8B.57
893 <y WITH ACUTE> FD 8D 8D 8D C3.BD 8B.71
894 <SMALL LETTER thorn> FE 8E 8E 8E C3.BE 8B.72
895 <PLUS-OR-MINUS SIGN> B1 8F 8F 8F C2.B1 80.58
896 <DEGREE SIGN> B0 90 90 90 C2.B0 80.57
897 j 6A 91 91 91 6A 91
898 k 6B 92 92 92 6B 92
899 l 6C 93 93 93 6C 93
900 m 6D 94 94 94 6D 94
901 n 6E 95 95 95 6E 95
902 o 6F 96 96 96 6F 96
903 p 70 97 97 97 70 97
904 q 71 98 98 98 71 98
905 r 72 99 99 99 72 99
906 <FEMININE ORDINAL> AA 9A 9A 9A C2.AA 80.51
907 <MASC. ORDINAL INDICATOR> BA 9B 9B 9B C2.BA 80.69
908 <SMALL LIGATURE ae> E6 9C 9C 9C C3.A6 8B.47
909 <CEDILLA> B8 9D 9D 9D C2.B8 80.67
910 <CAPITAL LIGATURE AE> C6 9E 9E 9E C3.86 8A.47
911 <CURRENCY SIGN> A4 9F 9F 9F C2.A4 80.45
912 <MICRO SIGN> B5 A0 A0 A0 C2.B5 80.64
913 ~ 7E A1 A1 FF 7E A1 ##
914 s 73 A2 A2 A2 73 A2
915 t 74 A3 A3 A3 74 A3
916 u 75 A4 A4 A4 75 A4
917 v 76 A5 A5 A5 76 A5
918 w 77 A6 A6 A6 77 A6
919 x 78 A7 A7 A7 78 A7
920 y 79 A8 A8 A8 79 A8
921 z 7A A9 A9 A9 7A A9
922 <INVERTED "!" > A1 AA AA AA C2.A1 80.42
923 <INVERTED QUESTION MARK> BF AB AB AB C2.BF 80.73
924 <CAPITAL LETTER ETH> D0 AC AC AC C3.90 8A.57
925 [ 5B BA AD BB 5B AD ** ##
926 <CAPITAL LETTER THORN> DE AE AE AE C3.9E 8A.72
927 <REGISTERED TRADE MARK> AE AF AF AF C2.AE 80.55
928 <NOT SIGN> AC 5F B0 BA C2.AC 80.53 ** ##
929 <POUND SIGN> A3 B1 B1 B1 C2.A3 80.44
930 <YEN SIGN> A5 B2 B2 B2 C2.A5 80.46
931 <MIDDLE DOT> B7 B3 B3 B3 C2.B7 80.66
932 <COPYRIGHT SIGN> A9 B4 B4 B4 C2.A9 80.4A
933 <SECTION SIGN> A7 B5 B5 B5 C2.A7 80.48
934 <PARAGRAPH SIGN> B6 B6 B6 B6 C2.B6 80.65
935 <FRACTION ONE QUARTER> BC B7 B7 B7 C2.BC 80.70
936 <FRACTION ONE HALF> BD B8 B8 B8 C2.BD 80.71
937 <FRACTION THREE QUARTERS> BE B9 B9 B9 C2.BE 80.72
938 <Y WITH ACUTE> DD AD BA AD C3.9D 8A.71 ** ##
939 <DIAERESIS> A8 BD BB 79 C2.A8 80.49 ** ##
940 <MACRON> AF BC BC A1 C2.AF 80.56 ##
941 ] 5D BB BD BD 5D BD **
942 <ACUTE ACCENT> B4 BE BE BE C2.B4 80.63
943 <MULTIPLICATION SIGN> D7 BF BF BF C3.97 8A.66
944 { 7B C0 C0 FB 7B C0 ##
945 A 41 C1 C1 C1 41 C1
946 B 42 C2 C2 C2 42 C2
947 C 43 C3 C3 C3 43 C3
948 D 44 C4 C4 C4 44 C4
949 E 45 C5 C5 C5 45 C5
950 F 46 C6 C6 C6 46 C6
951 G 47 C7 C7 C7 47 C7
952 H 48 C8 C8 C8 48 C8
953 I 49 C9 C9 C9 49 C9
954 <SOFT HYPHEN> AD CA CA CA C2.AD 80.54
955 <o WITH CIRCUMFLEX> F4 CB CB CB C3.B4 8B.63
956 <o WITH DIAERESIS> F6 CC CC CC C3.B6 8B.65
957 <o WITH GRAVE> F2 CD CD CD C3.B2 8B.59
958 <o WITH ACUTE> F3 CE CE CE C3.B3 8B.62
959 <o WITH TILDE> F5 CF CF CF C3.B5 8B.64
960 } 7D D0 D0 FD 7D D0 ##
961 J 4A D1 D1 D1 4A D1
962 K 4B D2 D2 D2 4B D2
963 L 4C D3 D3 D3 4C D3
964 M 4D D4 D4 D4 4D D4
965 N 4E D5 D5 D5 4E D5
966 O 4F D6 D6 D6 4F D6
967 P 50 D7 D7 D7 50 D7
968 Q 51 D8 D8 D8 51 D8
969 R 52 D9 D9 D9 52 D9
970 <SUPERSCRIPT ONE> B9 DA DA DA C2.B9 80.68
971 <u WITH CIRCUMFLEX> FB DB DB DB C3.BB 8B.6A
972 <u WITH DIAERESIS> FC DC DC DC C3.BC 8B.70
973 <u WITH GRAVE> F9 DD DD C0 C3.B9 8B.68 ##
974 <u WITH ACUTE> FA DE DE DE C3.BA 8B.69
975 <y WITH DIAERESIS> FF DF DF DF C3.BF 8B.73
976 \ 5C E0 E0 BC 5C E0 ##
977 <DIVISION SIGN> F7 E1 E1 E1 C3.B7 8B.66
978 S 53 E2 E2 E2 53 E2
979 T 54 E3 E3 E3 54 E3
980 U 55 E4 E4 E4 55 E4
981 V 56 E5 E5 E5 56 E5
982 W 57 E6 E6 E6 57 E6
983 X 58 E7 E7 E7 58 E7
984 Y 59 E8 E8 E8 59 E8
985 Z 5A E9 E9 E9 5A E9
986 <SUPERSCRIPT TWO> B2 EA EA EA C2.B2 80.59
987 <O WITH CIRCUMFLEX> D4 EB EB EB C3.94 8A.63
988 <O WITH DIAERESIS> D6 EC EC EC C3.96 8A.65
989 <O WITH GRAVE> D2 ED ED ED C3.92 8A.59
990 <O WITH ACUTE> D3 EE EE EE C3.93 8A.62
991 <O WITH TILDE> D5 EF EF EF C3.95 8A.64
992 0 30 F0 F0 F0 30 F0
993 1 31 F1 F1 F1 31 F1
994 2 32 F2 F2 F2 32 F2
995 3 33 F3 F3 F3 33 F3
996 4 34 F4 F4 F4 34 F4
997 5 35 F5 F5 F5 35 F5
998 6 36 F6 F6 F6 36 F6
999 7 37 F7 F7 F7 37 F7
1000 8 38 F8 F8 F8 38 F8
1001 9 39 F9 F9 F9 39 F9
1002 <SUPERSCRIPT THREE> B3 FA FA FA C2.B3 80.62
1003 <U WITH CIRCUMFLEX> DB FB FB DD C3.9B 8A.6A ##
1004 <U WITH DIAERESIS> DC FC FC FC C3.9C 8A.70
1005 <U WITH GRAVE> D9 FD FD E0 C3.99 8A.68 ##
1006 <U WITH ACUTE> DA FE FE FE C3.9A 8A.69
1007 <APC> 9F FF FF 5F C2.9F FF ##
d396a558 1008
4d2ca8b5 1009=head1 IDENTIFYING CHARACTER CODE SETS
d396a558 1010
4d2ca8b5
KW
1011It is possible to determine which character set you are operating under.
1012But first you need to be really really sure you need to do this. Your
1013code will be simpler and probably just as portable if you don't have
1014to test the character set and do different things, depending. There are
1015actually only very few circumstances where it's not easy to write
1016straight-line code portable to all character sets. See
1017L<perluniintro/Unicode and EBCDIC> for how to portably specify
1018characters.
d396a558 1019
4d2ca8b5
KW
1020But there are some cases where you may want to know which character set
1021you are running under. One possible example is doing
1022L<sorting|/SORTING> in inner loops where performance is critical.
d396a558 1023
4d2ca8b5
KW
1024To determine if you are running under ASCII or EBCDIC, you can use the
1025return value of C<ord()> or C<chr()> to test one or more character
1026values. For example:
d396a558 1027
4d2ca8b5
KW
1028 $is_ascii = "A" eq chr(65);
1029 $is_ebcdic = "A" eq chr(193);
1030 $is_ascii = ord("A") == 65;
1031 $is_ebcdic = ord("A") == 193;
d396a558 1032
4d2ca8b5
KW
1033There's even less need to distinguish between EBCDIC code pages, but to
1034do so try looking at one or more of the characters that differ between
1035them.
d396a558 1036
84f709e7
JH
1037 $is_ascii = ord('[') == 91;
1038 $is_ebcdic_37 = ord('[') == 186;
1039 $is_ebcdic_1047 = ord('[') == 173;
1040 $is_ebcdic_POSIX_BC = ord('[') == 187;
d396a558
JH
1041
1042However, it would be unwise to write tests such as:
1043
84f709e7
JH
1044 $is_ascii = "\r" ne chr(13); # WRONG
1045 $is_ascii = "\n" ne chr(10); # ILL ADVISED
d396a558 1046
4d2ca8b5
KW
1047Obviously the first of these will fail to distinguish most ASCII
1048platforms from either a CCSID 0037, a 1047, or a POSIX-BC EBCDIC
1049platform since S<C<"\r" eq chr(13)>> under all of those coded character
1050sets. But note too that because C<"\n"> is C<chr(13)> and C<"\r"> is
1051C<chr(10)> on old Macintosh (which is an ASCII platform) the second
1052C<$is_ascii> test will lead to trouble there.
d396a558 1053
eaf8b9b9 1054To determine whether or not perl was built under an EBCDIC
d396a558
JH
1055code page you can use the Config module like so:
1056
1057 use Config;
84f709e7 1058 $is_ebcdic = $Config{'ebcdic'} eq 'define';
d396a558
JH
1059
1060=head1 CONVERSIONS
1061
d5924ca6
KW
1062=head2 C<utf8::unicode_to_native()> and C<utf8::native_to_unicode()>
1063
1064These functions take an input numeric code point in one encoding and
1065return what its equivalent value is in the other.
1066
4d2ca8b5
KW
1067See L<utf8>.
1068
1e054b24
PP
1069=head2 tr///
1070
eaf8b9b9 1071In order to convert a string of characters from one character set to
d396a558 1072another a simple list of numbers, such as in the right columns in the
4d2ca8b5 1073above table, along with Perl's C<tr///> operator is all that is needed.
5f26d5fd 1074The data in the table are in ASCII/Latin1 order, hence the EBCDIC columns
eaf8b9b9 1075provide easy-to-use ASCII/Latin1 to EBCDIC operations that are also easily
d396a558
JH
1076reversed.
1077
5f26d5fd 1078For example, to convert ASCII/Latin1 to code page 037 take the output of the
4d2ca8b5
KW
1079second numbers column from the output of recipe 2 (modified to add
1080C<"\"> characters), and use it in C<tr///> like so:
d396a558 1081
eaf8b9b9 1082 $cp_037 =
5f26d5fd
KW
1083 '\x00\x01\x02\x03\x37\x2D\x2E\x2F\x16\x05\x25\x0B\x0C\x0D\x0E\x0F' .
1084 '\x10\x11\x12\x13\x3C\x3D\x32\x26\x18\x19\x3F\x27\x1C\x1D\x1E\x1F' .
1085 '\x40\x5A\x7F\x7B\x5B\x6C\x50\x7D\x4D\x5D\x5C\x4E\x6B\x60\x4B\x61' .
1086 '\xF0\xF1\xF2\xF3\xF4\xF5\xF6\xF7\xF8\xF9\x7A\x5E\x4C\x7E\x6E\x6F' .
1087 '\x7C\xC1\xC2\xC3\xC4\xC5\xC6\xC7\xC8\xC9\xD1\xD2\xD3\xD4\xD5\xD6' .
1088 '\xD7\xD8\xD9\xE2\xE3\xE4\xE5\xE6\xE7\xE8\xE9\xBA\xE0\xBB\xB0\x6D' .
1089 '\x79\x81\x82\x83\x84\x85\x86\x87\x88\x89\x91\x92\x93\x94\x95\x96' .
1090 '\x97\x98\x99\xA2\xA3\xA4\xA5\xA6\xA7\xA8\xA9\xC0\x4F\xD0\xA1\x07' .
1091 '\x20\x21\x22\x23\x24\x15\x06\x17\x28\x29\x2A\x2B\x2C\x09\x0A\x1B' .
1092 '\x30\x31\x1A\x33\x34\x35\x36\x08\x38\x39\x3A\x3B\x04\x14\x3E\xFF' .
1093 '\x41\xAA\x4A\xB1\x9F\xB2\x6A\xB5\xBD\xB4\x9A\x8A\x5F\xCA\xAF\xBC' .
1094 '\x90\x8F\xEA\xFA\xBE\xA0\xB6\xB3\x9D\xDA\x9B\x8B\xB7\xB8\xB9\xAB' .
1095 '\x64\x65\x62\x66\x63\x67\x9E\x68\x74\x71\x72\x73\x78\x75\x76\x77' .
1096 '\xAC\x69\xED\xEE\xEB\xEF\xEC\xBF\x80\xFD\xFE\xFB\xFC\xAD\xAE\x59' .
1097 '\x44\x45\x42\x46\x43\x47\x9C\x48\x54\x51\x52\x53\x58\x55\x56\x57' .
1098 '\x8C\x49\xCD\xCE\xCB\xCF\xCC\xE1\x70\xDD\xDE\xDB\xDC\x8D\x8E\xDF';
d396a558
JH
1099
1100 my $ebcdic_string = $ascii_string;
5f26d5fd 1101 eval '$ebcdic_string =~ tr/\000-\377/' . $cp_037 . '/';
d396a558 1102
0be03469 1103To convert from EBCDIC 037 to ASCII just reverse the order of the tr///
d396a558
JH
1104arguments like so:
1105
1106 my $ascii_string = $ebcdic_string;
5f26d5fd
KW
1107 eval '$ascii_string =~ tr/' . $cp_037 . '/\000-\377/';
1108
1109Similarly one could take the output of the third numbers column from recipe 2
1110to obtain a C<$cp_1047> table. The fourth numbers column of the output from
1111recipe 2 could provide a C<$cp_posix_bc> table suitable for transcoding as
1112well.
d5d9880c 1113
5f26d5fd
KW
1114If you wanted to see the inverse tables, you would first have to sort on the
1115desired numbers column as in recipes 4, 5 or 6, then take the output of the
1116first numbers column.
1e054b24
PP
1117
1118=head2 iconv
d396a558 1119
d5d9880c 1120XPG operability often implies the presence of an I<iconv> utility
d396a558
JH
1121available from the shell or from the C library. Consult your system's
1122documentation for information on iconv.
1123
4d2ca8b5 1124On OS/390 or z/OS see the L<iconv(1)> manpage. One way to invoke the C<iconv>
d396a558
JH
1125shell utility from within perl would be to:
1126
395f5a0c 1127 # OS/390 or z/OS example
84f709e7 1128 $ascii_data = `echo '$ebcdic_data'| iconv -f IBM-1047 -t ISO8859-1`
d396a558
JH
1129
1130or the inverse map:
1131
395f5a0c 1132 # OS/390 or z/OS example
84f709e7 1133 $ebcdic_data = `echo '$ascii_data'| iconv -f ISO8859-1 -t IBM-1047`
d396a558 1134
4d2ca8b5 1135For other Perl-based conversion options see the C<Convert::*> modules on CPAN.
d396a558 1136
1e054b24
PP
1137=head2 C RTL
1138
4d2ca8b5 1139The OS/390 and z/OS C run-time libraries provide C<_atoe()> and C<_etoa()> functions.
1e054b24 1140
d396a558
JH
1141=head1 OPERATOR DIFFERENCES
1142
eaf8b9b9 1143The C<..> range operator treats certain character ranges with
2bbc8d55
SP
1144care on EBCDIC platforms. For example the following array
1145will have twenty six elements on either an EBCDIC platform
1146or an ASCII platform:
d396a558 1147
84f709e7 1148 @alphabet = ('A'..'Z'); # $#alphabet == 25
d396a558
JH
1149
1150The bitwise operators such as & ^ | may return different results
4d2ca8b5 1151when operating on string or character data in a Perl program running
2bbc8d55 1152on an EBCDIC platform than when run on an ASCII platform. Here is
d396a558
JH
1153an example adapted from the one in L<perlop>:
1154
1155 # EBCDIC-based examples
84f709e7 1156 print "j p \n" ^ " a h"; # prints "JAPH\n"
eaf8b9b9 1157 print "JA" | " ph\n"; # prints "japh\n"
84f709e7
JH
1158 print "JAPH\nJunk" & "\277\277\277\277\277"; # prints "japh\n";
1159 print 'p N$' ^ " E<H\n"; # prints "Perl\n";
d396a558
JH
1160
1161An interesting property of the 32 C0 control characters
1162in the ASCII table is that they can "literally" be constructed
4d2ca8b5 1163as control characters in Perl, e.g. C<(chr(0)> eq C<\c@>)>
c72e675e 1164C<(chr(1)> eq C<\cA>)>, and so on. Perl on EBCDIC platforms has been
4d2ca8b5 1165ported to take C<\c@> to C<chr(0)> and C<\cA> to C<chr(1)>, etc. as well, but the
2bd1cbf6 1166characters that result depend on which code page you are
2c09a866
KW
1167using. The table below uses the standard acronyms for the controls.
1168The POSIX-BC and 1047 sets are
eaf8b9b9 1169identical throughout this range and differ from the 0037 set at only
4d2ca8b5 1170one spot (21 decimal). Note that the line terminator character
eaf8b9b9
KW
1171may be generated by C<\cJ> on ASCII platforms but by C<\cU> on 1047 or POSIX-BC
1172platforms and cannot be generated as a C<"\c.letter."> control character on
2c09a866
KW
11730037 platforms. Note also that C<\c\> cannot be the final element in a string
1174or regex, as it will absorb the terminator. But C<\c\I<X>> is a C<FILE
1175SEPARATOR> concatenated with I<X> for all I<X>.
2bd1cbf6
KW
1176The outlier C<\c?> on ASCII, which yields a non-C0 control C<DEL>,
1177yields the outlier control C<APC> on EBCDIC, the one that isn't in the
aae773bb
KW
1178block of contiguous controls. Note that a subtlety of this is that
1179C<\c?> on ASCII platforms is an ASCII character, while it isn't
1180equivalent to any ASCII character in EBCDIC platforms.
2c09a866 1181
eaf8b9b9 1182 chr ord 8859-1 0037 1047 && POSIX-BC
c72e675e 1183 -----------------------------------------------------------------------
2c09a866 1184 \c@ 0 <NUL> <NUL> <NUL>
eaf8b9b9 1185 \cA 1 <SOH> <SOH> <SOH>
2c09a866
KW
1186 \cB 2 <STX> <STX> <STX>
1187 \cC 3 <ETX> <ETX> <ETX>
eaf8b9b9
KW
1188 \cD 4 <EOT> <ST> <ST>
1189 \cE 5 <ENQ> <HT> <HT>
1190 \cF 6 <ACK> <SSA> <SSA>
1191 \cG 7 <BEL> <DEL> <DEL>
1192 \cH 8 <BS> <EPA> <EPA>
1193 \cI 9 <HT> <RI> <RI>
1194 \cJ 10 <LF> <SS2> <SS2>
2c09a866 1195 \cK 11 <VT> <VT> <VT>
eaf8b9b9
KW
1196 \cL 12 <FF> <FF> <FF>
1197 \cM 13 <CR> <CR> <CR>
2c09a866
KW
1198 \cN 14 <SO> <SO> <SO>
1199 \cO 15 <SI> <SI> <SI>
eaf8b9b9 1200 \cP 16 <DLE> <DLE> <DLE>
2c09a866
KW
1201 \cQ 17 <DC1> <DC1> <DC1>
1202 \cR 18 <DC2> <DC2> <DC2>
eaf8b9b9
KW
1203 \cS 19 <DC3> <DC3> <DC3>
1204 \cT 20 <DC4> <OSC> <OSC>
8d725451 1205 \cU 21 <NAK> <NEL> <LF> **
2c09a866 1206 \cV 22 <SYN> <BS> <BS>
eaf8b9b9 1207 \cW 23 <ETB> <ESA> <ESA>
2c09a866
KW
1208 \cX 24 <CAN> <CAN> <CAN>
1209 \cY 25 <EOM> <EOM> <EOM>
eaf8b9b9
KW
1210 \cZ 26 <SUB> <PU2> <PU2>
1211 \c[ 27 <ESC> <SS3> <SS3>
2c09a866
KW
1212 \c\X 28 <FS>X <FS>X <FS>X
1213 \c] 29 <GS> <GS> <GS>
1214 \c^ 30 <RS> <RS> <RS>
1215 \c_ 31 <US> <US> <US>
2bd1cbf6
KW
1216 \c? * <DEL> <APC> <APC>
1217
1218C<*> Note: C<\c?> maps to ordinal 127 (C<DEL>) on ASCII platforms, but
1219since ordinal 127 is a not a control character on EBCDIC machines,
4d2ca8b5
KW
1220C<\c?> instead maps on them to C<APC>, which is 255 in 0037 and 1047,
1221and 95 in POSIX-BC.
d396a558
JH
1222
1223=head1 FUNCTION DIFFERENCES
1224
1225=over 8
1226
4d2ca8b5 1227=item C<chr()>
d396a558 1228
4d2ca8b5 1229C<chr()> must be given an EBCDIC code number argument to yield a desired
2bbc8d55 1230character return value on an EBCDIC platform. For example:
d396a558 1231
84f709e7 1232 $CAPITAL_LETTER_A = chr(193);
d396a558 1233
4d2ca8b5 1234=item C<ord()>
d396a558 1235
4d2ca8b5 1236C<ord()> will return EBCDIC code number values on an EBCDIC platform.
d396a558
JH
1237For example:
1238
84f709e7 1239 $the_number_193 = ord("A");
d396a558 1240
4d2ca8b5 1241=item C<pack()>
d396a558 1242
4d2ca8b5
KW
1243
1244The C<"c"> and C<"C"> templates for C<pack()> are dependent upon character set
d396a558
JH
1245encoding. Examples of usage on EBCDIC include:
1246
1247 $foo = pack("CCCC",193,194,195,196);
1248 # $foo eq "ABCD"
84f709e7 1249 $foo = pack("C4",193,194,195,196);
d396a558
JH
1250 # same thing
1251
1252 $foo = pack("ccxxcc",193,194,195,196);
1253 # $foo eq "AB\0\0CD"
1254
4d2ca8b5
KW
1255The C<"U"> template has been ported to mean "Unicode" on all platforms so
1256that
1257
1258 pack("U", 65) eq 'A'
1259
1260is true on all platforms. If you want native code points for the low
1261256, use the C<"W"> template. This means that the equivalences
1262
1263 pack("W", ord($character)) eq $character
1264 unpack("W", $character) == ord $character
1265
1266will hold.
1267
4d2ca8b5 1268=item C<print()>
d396a558
JH
1269
1270One must be careful with scalars and strings that are passed to
1271print that contain ASCII encodings. One common place
1272for this to occur is in the output of the MIME type header for
4d2ca8b5 1273CGI script writing. For example, many Perl programming guides
d396a558
JH
1274recommend something similar to:
1275
eaf8b9b9 1276 print "Content-type:\ttext/html\015\012\015\012";
d396a558
JH
1277 # this may be wrong on EBCDIC
1278
4d2ca8b5 1279You can instead write
d396a558 1280
5f26d5fd 1281 print "Content-type:\ttext/html\r\n\r\n"; # OK for DGW et al
d396a558 1282
4d2ca8b5
KW
1283and have it work portably.
1284
d396a558 1285That is because the translation from EBCDIC to ASCII is done
4d2ca8b5 1286by the web server in this case. Consult your web server's documentation for
d396a558
JH
1287further details.
1288
4d2ca8b5 1289=item C<printf()>
d396a558
JH
1290
1291The formats that can convert characters to numbers and vice versa
1292will be different from their ASCII counterparts when executed
2bbc8d55 1293on an EBCDIC platform. Examples include:
d396a558
JH
1294
1295 printf("%c%c%c",193,194,195); # prints ABC
1296
4d2ca8b5 1297=item C<sort()>
d396a558 1298
eaf8b9b9 1299EBCDIC sort results may differ from ASCII sort results especially for
4d2ca8b5 1300mixed case strings. This is discussed in more detail L<below|/SORTING>.
d396a558 1301
4d2ca8b5 1302=item C<sprintf()>
d396a558 1303
4d2ca8b5 1304See the discussion of C<L</printf()>> above. An example of the use
d396a558
JH
1305of sprintf would be:
1306
84f709e7 1307 $CAPITAL_LETTER_A = sprintf("%c",193);
d396a558 1308
4d2ca8b5 1309=item C<unpack()>
d396a558 1310
4d2ca8b5 1311See the discussion of C<L</pack()>> above.
d396a558
JH
1312
1313=back
1314
4d2ca8b5
KW
1315Note that it is possible to write portable code for these by specifying
1316things in Unicode numbers, and using a conversion function:
1317
1318 printf("%c",utf8::unicode_to_native(65)); # prints A on all
1319 # platforms
1320 print utf8::native_to_unicode(ord("A")); # Likewise, prints 65
1321
1322See L<perluniintro/Unicode and EBCDIC> and L</CONVERSIONS>
1323for other options.
1324
d396a558
JH
1325=head1 REGULAR EXPRESSION DIFFERENCES
1326
4d2ca8b5
KW
1327You can write your regular expressions just like someone on an ASCII
1328platform would do. But keep in mind that using octal or hex notation to
1329specify a particular code point will give you the character that the
1330EBCDIC code page natively maps to it. (This is also true of all
1331double-quoted strings.) If you want to write portably, just use the
1332C<\N{U+...}> notation everywhere where you would have used C<\x{...}>,
1333and don't use octal notation at all.
1334
1335Starting in Perl v5.22, this applies to ranges in bracketed character
1336classes. If you say, for example, C<qr/[\N{U+20}-\N{U+7F}]/>, it means
1337the characters C<\N{U+20}>, C<\N{U+21}>, ..., C<\N{U+7F}>. This range
1338is all the printable characters that the ASCII character set contains.
1339
1340Prior to v5.22, you couldn't specify any ranges portably, except
1341(starting in Perl v5.5.3) all subsets of the C<[A-Z]> and C<[a-z]>
1342ranges are specially coded to not pick up gap characters. For example,
1343characters such as "E<ocirc>" (C<o WITH CIRCUMFLEX>) that lie between
1344"I" and "J" would not be matched by the regular expression range
1345C</[H-K]/>. But if either of the range end points is explicitly numeric
1346(and neither is specified by C<\N{U+...}>), the gap characters are
1347matched:
1348
1349 /[\x89-\x91]/
1350
1351will match C<\x8e>, even though C<\x89> is "i" and C<\x91 > is "j",
1352and C<\x8e> is a gap character, from the alphabetic viewpoint.
1353
1354Another construct to be wary of is the inappropriate use of hex (unless
1355you use C<\N{U+...}>) or
d396a558
JH
1356octal constants in regular expressions. Consider the following
1357set of subs:
1358
1359 sub is_c0 {
1360 my $char = substr(shift,0,1);
1361 $char =~ /[\000-\037]/;
1362 }
1363
1364 sub is_print_ascii {
1365 my $char = substr(shift,0,1);
1366 $char =~ /[\040-\176]/;
1367 }
1368
1369 sub is_delete {
1370 my $char = substr(shift,0,1);
1371 $char eq "\177";
1372 }
1373
1374 sub is_c1 {
1375 my $char = substr(shift,0,1);
1376 $char =~ /[\200-\237]/;
1377 }
1378
10c526cf 1379 sub is_latin_1 { # But not ASCII; not C1
d396a558
JH
1380 my $char = substr(shift,0,1);
1381 $char =~ /[\240-\377]/;
1382 }
1383
4d2ca8b5
KW
1384These are valid only on ASCII platforms. Starting in Perl v5.22, simply
1385changing the octal constants to equivalent C<\N{U+...}> values makes
1386them portable:
1387
1388 sub is_c0 {
1389 my $char = substr(shift,0,1);
1390 $char =~ /[\N{U+00}-\N{U+1F}]/;
1391 }
1392
1393 sub is_print_ascii {
1394 my $char = substr(shift,0,1);
1395 $char =~ /[\N{U+20}-\N{U+7E}]/;
1396 }
1397
1398 sub is_delete {
1399 my $char = substr(shift,0,1);
1400 $char eq "\N{U+7F}";
1401 }
1402
1403 sub is_c1 {
1404 my $char = substr(shift,0,1);
1405 $char =~ /[\N{U+80}-\N{U+9F}]/;
1406 }
1407
1408 sub is_latin_1 { # But not ASCII; not C1
1409 my $char = substr(shift,0,1);
1410 $char =~ /[\N{U+A0}-\N{U+FF}]/;
1411 }
1412
1413And here are some alternative portable ways to write them:
d396a558
JH
1414
1415 sub Is_c0 {
1416 my $char = substr(shift,0,1);
f11f9c4c
KW
1417 return $char =~ /[[:cntrl:]]/a && ! Is_delete($char);
1418
1419 # Alternatively:
1420 # return $char =~ /[[:cntrl:]]/
1421 # && $char =~ /[[:ascii:]]/
1422 # && ! Is_delete($char);
d396a558
JH
1423 }
1424
1425 sub Is_print_ascii {
1426 my $char = substr(shift,0,1);
10c526cf 1427
f11f9c4c 1428 return $char =~ /[[:print:]]/a;
10c526cf
KW
1429
1430 # Alternatively:
f11f9c4c
KW
1431 # return $char =~ /[[:print:]]/ && $char =~ /[[:ascii:]]/;
1432
1433 # Or
10c526cf
KW
1434 # return $char
1435 # =~ /[ !"\#\$%&'()*+,\-.\/0-9:;<=>?\@A-Z[\\\]^_`a-z{|}~]/;
d396a558
JH
1436 }
1437
1438 sub Is_delete {
1439 my $char = substr(shift,0,1);
10c526cf 1440 return utf8::native_to_unicode(ord $char) == 0x7F;
d396a558
JH
1441 }
1442
1443 sub Is_c1 {
10c526cf 1444 use feature 'unicode_strings';
d396a558 1445 my $char = substr(shift,0,1);
10c526cf 1446 return $char =~ /[[:cntrl:]]/ && $char !~ /[[:ascii:]]/;
d396a558
JH
1447 }
1448
10c526cf
KW
1449 sub Is_latin_1 { # But not ASCII; not C1
1450 use feature 'unicode_strings';
d396a558 1451 my $char = substr(shift,0,1);
10c526cf 1452 return ord($char) < 256
4d2ca8b5
KW
1453 && $char !~ /[[:ascii:]]/
1454 && $char !~ /[[:cntrl:]]/;
d396a558
JH
1455 }
1456
10c526cf 1457Another way to write C<Is_latin_1()> would be
d396a558
JH
1458to use the characters in the range explicitly:
1459
1460 sub Is_latin_1 {
1461 my $char = substr(shift,0,1);
f11f9c4c
KW
1462 $char =~ /[ ¡¢£¤¥¦§¨©ª«¬­®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏ]
1463 [ÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ]/x;
d396a558
JH
1464 }
1465
eaf8b9b9 1466Although that form may run into trouble in network transit (due to the
4d2ca8b5
KW
1467presence of 8 bit characters) or on non ISO-Latin character sets. But
1468it does allow C<Is_c1> to be rewritten so it works on Perls that don't
1469have C<'unicode_strings'> (earlier than v5.14):
1470
1471 sub Is_latin_1 { # But not ASCII; not C1
1472 my $char = substr(shift,0,1);
1473 return ord($char) < 256
1474 && $char !~ /[[:ascii:]]/
1475 && ! Is_latin1($char);
1476 }
d396a558
JH
1477
1478=head1 SOCKETS
1479
1480Most socket programming assumes ASCII character encodings in network
1481byte order. Exceptions can include CGI script writing under a
1482host web server where the server may take care of translation for you.
1483Most host web servers convert EBCDIC data to ISO-8859-1 or Unicode on
1484output.
1485
1486=head1 SORTING
1487
8a50e6a3 1488One big difference between ASCII-based character sets and EBCDIC ones
4d2ca8b5
KW
1489are the relative positions of the characters when sorted in native
1490order. Of most concern are the upper- and lowercase letters, the
1491digits, and the underscore (C<"_">). On ASCII platforms the native sort
1492order has the digits come before the uppercase letters which come before
1493the underscore which comes before the lowercase letters. On EBCDIC, the
1494underscore comes first, then the lowercase letters, then the uppercase
1495ones, and the digits last. If sorted on an ASCII-based platform, the
8a50e6a3
FC
1496two-letter abbreviation for a physician comes before the two letter
1497abbreviation for drive; that is:
d396a558 1498
c72e675e 1499 @sorted = sort(qw(Dr. dr.)); # @sorted holds ('Dr.','dr.') on ASCII,
84f709e7 1500 # but ('dr.','Dr.') on EBCDIC
d396a558 1501
8a50e6a3 1502The property of lowercase before uppercase letters in EBCDIC is
d396a558 1503even carried to the Latin 1 EBCDIC pages such as 0037 and 1047.
4d2ca8b5
KW
1504An example would be that "E<Euml>" (C<E WITH DIAERESIS>, 203) comes
1505before "E<euml>" (C<e WITH DIAERESIS>, 235) on an ASCII platform, but
eaf8b9b9 1506the latter (83) comes before the former (115) on an EBCDIC platform.
4d2ca8b5
KW
1507(Astute readers will note that the uppercase version of "E<szlig>"
1508C<SMALL LETTER SHARP S> is simply "SS" and that the upper case versions
1509of "E<yuml>" (small C<y WITH DIAERESIS>) and "E<micro>" (C<MICRO SIGN>)
1510are not in the 0..255 range but are in Unicode, in a Unicode enabled
1511Perl).
d396a558
JH
1512
1513The sort order will cause differences between results obtained on
2bbc8d55 1514ASCII platforms versus EBCDIC platforms. What follows are some suggestions
d396a558
JH
1515on how to deal with these differences.
1516
51b5cecb 1517=head2 Ignore ASCII vs. EBCDIC sort differences.
d396a558
JH
1518
1519This is the least computationally expensive strategy. It may require
1520some user education.
1521
4d2ca8b5 1522=head2 Use a sort helper function
d396a558 1523
4d2ca8b5
KW
1524This is completely general, but the most computationally expensive
1525strategy. Choose one or the other character set and transform to that
1526for every sort comparision. Here's a complete example that transforms
1527to ASCII sort order:
51b5cecb 1528
4d2ca8b5
KW
1529 sub native_to_uni($) {
1530 my $string = shift;
d396a558 1531
4d2ca8b5
KW
1532 # Saves time on an ASCII platform
1533 return $string if ord 'A' == 65;
d396a558 1534
4d2ca8b5
KW
1535 my $output = "";
1536 for my $i (0 .. length($string) - 1) {
1537 $output
1538 .= chr(utf8::native_to_unicode(ord(substr($string, $i, 1))));
1539 }
1540
1541 # Preserve utf8ness of input onto the output, even if it didn't need
1542 # to be utf8
1543 utf8::upgrade($output) if utf8::is_utf8($string);
51b5cecb 1544
4d2ca8b5
KW
1545 return $output;
1546 }
51b5cecb 1547
4d2ca8b5
KW
1548 sub ascii_order { # Sort helper
1549 return native_to_uni($a) cmp native_to_uni($b);
1550 }
d396a558 1551
4d2ca8b5
KW
1552 sort ascii_order @list;
1553
1554=head2 MONO CASE then sort data (for non-digits, non-underscore)
1555
1556If you don't care about where digits and underscore sort to, you can do
1557something like this
1558
1559 sub case_insensitive_order { # Sort helper
1560 return lc($a) cmp lc($b)
1561 }
1562
1563 sort case_insensitive_order @list;
1564
1565If performance is an issue, and you don't care if the output is in the
1566same case as the input, Use C<tr///> to transform to the case most
1567employed within the data. If the data are primarily UPPERCASE
1568non-Latin1, then apply C<tr/[a-z]/[A-Z]/>, and then C<sort()>. If the
1569data are primarily lowercase non Latin1 then apply C<tr/[A-Z]/[a-z]/>
1570before sorting. If the data are primarily UPPERCASE and include Latin-1
1571characters then apply:
1572
1573 tr/[a-z]/[A-Z]/;
1574 tr/[àáâãäåæçèéêëìíîïðñòóôõöøùúûüýþ]/[ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞ/;
1575 s/ß/SS/g;
1576
1577then C<sort()>. If you have a choice, it's better to lowercase things
1578to avoid the problems of the two Latin-1 characters whose uppercase is
1579outside Latin-1: "E<yuml>" (small C<y WITH DIAERESIS>) and "E<micro>"
1580(C<MICRO SIGN>). If you do need to upppercase, you can; with a
1581Unicode-enabled Perl, do:
1582
1583 tr/ÿ/\x{178}/;
1584 tr/µ/\x{39C}/;
d396a558 1585
2bbc8d55 1586=head2 Perform sorting on one type of platform only.
d396a558
JH
1587
1588This strategy can employ a network connection. As such
1589it would be computationally expensive.
1590
395f5a0c 1591=head1 TRANSFORMATION FORMATS
1e054b24 1592
eaf8b9b9
KW
1593There are a variety of ways of transforming data with an intra character set
1594mapping that serve a variety of purposes. Sorting was discussed in the
1595previous section and a few of the other more popular mapping techniques are
1e054b24
PP
1596discussed next.
1597
1598=head2 URL decoding and encoding
d396a558 1599
51b5cecb 1600Note that some URLs have hexadecimal ASCII code points in them in an
eaf8b9b9 1601attempt to overcome character or protocol limitation issues. For example
1e054b24 1602the tilde character is not on every keyboard hence a URL of the form:
d396a558
JH
1603
1604 http://www.pvhp.com/~pvhp/
1605
1606may also be expressed as either of:
1607
1608 http://www.pvhp.com/%7Epvhp/
1609
1610 http://www.pvhp.com/%7epvhp/
1611
4d2ca8b5 1612where 7E is the hexadecimal ASCII code point for "~". Here is an example
f11f9c4c 1613of decoding such a URL in any EBCDIC code page:
d396a558 1614
84f709e7 1615 $url = 'http://www.pvhp.com/%7Epvhp/';
f11f9c4c
KW
1616 $url =~ s/%([0-9a-fA-F]{2})/
1617 pack("c",utf8::unicode_to_native(hex($1)))/xge;
d396a558 1618
eaf8b9b9 1619Conversely, here is a partial solution for the task of encoding such
f11f9c4c 1620a URL in any EBCDIC code page:
1e054b24 1621
84f709e7 1622 $url = 'http://www.pvhp.com/~pvhp/';
eaf8b9b9
KW
1623 # The following regular expression does not address the
1624 # mappings for: ('.' => '%2E', '/' => '%2F', ':' => '%3A')
10c526cf 1625 $url =~ s/([\t "#%&\(\),;<=>\?\@\[\\\]^`{|}~])/
f11f9c4c 1626 sprintf("%%%02X",utf8::native_to_unicode(ord($1)))/xge;
1e054b24 1627
eaf8b9b9 1628where a more complete solution would split the URL into components
1e054b24
PP
1629and apply a full s/// substitution only to the appropriate parts.
1630
1e054b24
PP
1631=head2 uu encoding and decoding
1632
4d2ca8b5
KW
1633The C<u> template to C<pack()> or C<unpack()> will render EBCDIC data in
1634EBCDIC characters equivalent to their ASCII counterparts. For example,
1635the following will print "Yes indeed\n" on either an ASCII or EBCDIC
1636computer:
1e054b24 1637
84f709e7
JH
1638 $all_byte_chrs = '';
1639 for (0..255) { $all_byte_chrs .= chr($_); }
1640 $uuencode_byte_chrs = pack('u', $all_byte_chrs);
210b36aa 1641 ($uu = <<'ENDOFHEREDOC') =~ s/^\s*//gm;
1e054b24
PP
1642 M``$"`P0%!@<("0H+#`T.#Q`1$A,4%187&!D:&QP='A\@(2(C)"4F)R@I*BLL
1643 M+2XO,#$R,S0U-C<X.3H[/#T^/T!!0D-$149'2$E*2TQ-3D]045)35%565UA9
1644 M6EM<75Y?8&%B8V1E9F=H:6IK;&UN;W!Q<G-T=79W>'EZ>WQ]?G^`@8*#A(6&
1645 MAXB)BHN,C8Z/D)&2DY25EI>8F9J;G)V>GZ"AHJ.DI::GJ*FJJZRMKJ^PL;*S
1646 MM+6VM[BYNKN\O;Z_P,'"P\3%QL?(R<K+S,W.S]#1TM/4U=;7V-G:V]S=WM_@
1647 ?X>+CY.7FY^CIZNOL[>[O\/'R\_3U]O?X^?K[_/W^_P``
1648 ENDOFHEREDOC
84f709e7 1649 if ($uuencode_byte_chrs eq $uu) {
1e054b24
PP
1650 print "Yes ";
1651 }
1652 $uudecode_byte_chrs = unpack('u', $uuencode_byte_chrs);
84f709e7 1653 if ($uudecode_byte_chrs eq $all_byte_chrs) {
1e054b24
PP
1654 print "indeed\n";
1655 }
1656
f11f9c4c 1657Here is a very spartan uudecoder that will work on EBCDIC:
1e054b24 1658
84f709e7 1659 #!/usr/local/bin/perl
84f709e7 1660 $_ = <> until ($mode,$file) = /^begin\s*(\d*)\s*(\S*)/;
1e054b24
PP
1661 open(OUT, "> $file") if $file ne "";
1662 while(<>) {
1663 last if /^end/;
1664 next if /[a-z]/;
f11f9c4c
KW
1665 next unless int((((utf8::native_to_unicode(ord()) - 32 ) & 077)
1666 + 2) / 3)
1667 == int(length() / 4);
1e054b24
PP
1668 print OUT unpack("u", $_);
1669 }
1670 close(OUT);
1671 chmod oct($mode), $file;
1672
1673
1674=head2 Quoted-Printable encoding and decoding
1675
8a50e6a3 1676On ASCII-encoded platforms it is possible to strip characters outside of
1e054b24
PP
1677the printable set using:
1678
1679 # This QP encoder works on ASCII only
4d2ca8b5
KW
1680 $qp_string =~ s/([=\x00-\x1F\x80-\xFF])/
1681 sprintf("=%02X",ord($1))/xge;
1e054b24 1682
4d2ca8b5
KW
1683Starting in Perl v5.22, this is trivially changeable to work portably on
1684both ASCII and EBCDIC platforms.
1685
1686 # This QP encoder works on both ASCII and EBCDIC
1687 $qp_string =~ s/([=\N{U+00}-\N{U+1F}\N{U+80}-\N{U+FF}])/
1688 sprintf("=%02X",ord($1))/xge;
1689
1690For earlier Perls, a QP encoder that works on both ASCII and EBCDIC
1691platforms would look somewhat like the following:
1e054b24 1692
f11f9c4c 1693 $delete = utf8::unicode_to_native(ord("\x7F"));
84f709e7 1694 $qp_string =~
f11f9c4c
KW
1695 s/([^[:print:]$delete])/
1696 sprintf("=%02X",utf8::native_to_unicode(ord($1)))/xage;
1e054b24
PP
1697
1698(although in production code the substitutions might be done
f11f9c4c 1699in the EBCDIC branch with the function call and separately in the
4d2ca8b5
KW
1700ASCII branch without the expense of the identity map; in Perl v5.22, the
1701identity map is optimized out so there is no expense, but the
1702alternative above is simpler and is also available in v5.22).
1e054b24
PP
1703
1704Such QP strings can be decoded with:
1705
1706 # This QP decoder is limited to ASCII only
f11f9c4c 1707 $string =~ s/=([[:xdigit:][[:xdigit:])/chr hex $1/ge;
1e054b24
PP
1708 $string =~ s/=[\n\r]+$//;
1709
eaf8b9b9 1710Whereas a QP decoder that works on both ASCII and EBCDIC platforms
f11f9c4c 1711would look somewhat like the following:
1e054b24 1712
f11f9c4c
KW
1713 $string =~ s/=([[:xdigit:][:xdigit:]])/
1714 chr utf8::native_to_unicode(hex $1)/xge;
1e054b24
PP
1715 $string =~ s/=[\n\r]+$//;
1716
c69ca1d4 1717=head2 Caesarean ciphers
1e054b24
PP
1718
1719The practice of shifting an alphabet one or more characters for encipherment
1720dates back thousands of years and was explicitly detailed by Gaius Julius
eaf8b9b9 1721Caesar in his B<Gallic Wars> text. A single alphabet shift is sometimes
1e054b24 1722referred to as a rotation and the shift amount is given as a number $n after
eaf8b9b9
KW
1723the string 'rot' or "rot$n". Rot0 and rot26 would designate identity maps
1724on the 26-letter English version of the Latin alphabet. Rot13 has the
1725interesting property that alternate subsequent invocations are identity maps
1726(thus rot13 is its own non-trivial inverse in the group of 26 alphabet
1727rotations). Hence the following is a rot13 encoder and decoder that will
2bbc8d55 1728work on ASCII and EBCDIC platforms:
1e054b24
PP
1729
1730 #!/usr/local/bin/perl
1731
84f709e7 1732 while(<>){
1e054b24
PP
1733 tr/n-za-mN-ZA-M/a-zA-Z/;
1734 print;
1735 }
1736
1737In one-liner form:
1738
84f709e7 1739 perl -ne 'tr/n-za-mN-ZA-M/a-zA-Z/;print'
1e054b24
PP
1740
1741
1742=head1 Hashing order and checksums
1743
4d2ca8b5
KW
1744Perl deliberately randomizes hash order for security purposes on both
1745ASCII and EBCDIC platforms.
1746
1747EBCDIC checksums will differ for the same file translated into ASCII
1748and vice versa.
1e054b24 1749
d396a558
JH
1750=head1 I18N AND L10N
1751
eaf8b9b9
KW
1752Internationalization (I18N) and localization (L10N) are supported at least
1753in principle even on EBCDIC platforms. The details are system-dependent
4d2ca8b5 1754and discussed under the L<OS ISSUES> section below.
d396a558 1755
8a50e6a3 1756=head1 MULTI-OCTET CHARACTER SETS
d396a558 1757
4d2ca8b5
KW
1758Perl works with UTF-EBCDIC, a multi-byte encoding. In Perls earlier
1759than v5.22, there may be various bugs in this regard.
395f5a0c
PK
1760
1761Legacy multi byte EBCDIC code pages XXX.
d396a558
JH
1762
1763=head1 OS ISSUES
1764
eaf8b9b9 1765There may be a few system-dependent issues
d396a558
JH
1766of concern to EBCDIC Perl programmers.
1767
522b859a 1768=head2 OS/400
51b5cecb 1769
d396a558
JH
1770=over 8
1771
522b859a
JH
1772=item PASE
1773
8a50e6a3
FC
1774The PASE environment is a runtime environment for OS/400 that can run
1775executables built for PowerPC AIX in OS/400; see L<perlos400>. PASE
522b859a
JH
1776is ASCII-based, not EBCDIC-based as the ILE.
1777
d396a558
JH
1778=item IFS access
1779
1780XXX.
1781
1782=back
1783
395f5a0c 1784=head2 OS/390, z/OS
d396a558 1785
51b5cecb
PP
1786Perl runs under Unix Systems Services or USS.
1787
d396a558
JH
1788=over 8
1789
4d2ca8b5
KW
1790=item C<sigaction>
1791
1792C<SA_SIGINFO> can have segmentation faults.
1793
1794=item C<chcp>
51b5cecb 1795
eaf8b9b9 1796B<chcp> is supported as a shell utility for displaying and changing
75cdcc93 1797one's code page. See also L<chcp(1)>.
51b5cecb 1798
d396a558
JH
1799=item dataset access
1800
1801For sequential data set access try:
1802
1803 my @ds_records = `cat //DSNAME`;
1804
1805or:
1806
1807 my @ds_records = `cat //'HLQ.DSNAME'`;
1808
1809See also the OS390::Stdio module on CPAN.
1810
4d2ca8b5 1811=item C<iconv>
51b5cecb 1812
1e054b24 1813B<iconv> is supported as both a shell utility and a C RTL routine.
4d2ca8b5 1814See also the L<iconv(1)> and L<iconv(3)> manual pages.
51b5cecb 1815
d396a558
JH
1816=item locales
1817
4d2ca8b5
KW
1818Locales are supported. There may be glitches when a locale is another
1819EBCDIC code page which has some of the
1820L<code-page variant characters|/The 13 variant characters> in other
1821positions.
1822
1823There aren't currently any real UTF-8 locales, even though some locale
1824names contain the string "UTF-8".
1825
1826See L<perllocale> for information on locales. The L10N files
1827are in F</usr/nls/locale>. C<$Config{d_setlocale}> is C<'define'> on
1828OS/390 or z/OS.
d396a558
JH
1829
1830=back
1831
d396a558
JH
1832=head2 POSIX-BC?
1833
1834XXX.
1835
51b5cecb
PP
1836=head1 BUGS
1837
4d2ca8b5
KW
1838=over 4
1839
1840=item *
1841
416c06fc
KW
1842The C<cmp> (and hence C<sort>) operators do not necessarily give the
1843correct results when both operands are UTF-EBCDIC encoded strings and
1844there is a mixture of ASCII and/or control characters, along with other
1845characters.
1846
1847=item *
1848
1849Ranges containing C<\N{...}> in the C<tr///> (and C<y///>)
1850transliteration operators are treated differently than the equivalent
f0b1ad2b 1851ranges in regular expression patterns. They should, but don't, cause
416c06fc
KW
1852the values in the ranges to all be treated as Unicode code points, and
1853not native ones. (L<perlre/Version 8 Regular Expressions> gives
1854details as to how it should work.)
1855
1856=item *
1857
51b5cecb 1858Not all shells will allow multiple C<-e> string arguments to perl to
4d2ca8b5
KW
1859be concatenated together properly as recipes in this document
18600, 2, 4, 5, and 6 might
395f5a0c 1861seem to imply.
51b5cecb 1862
4d2ca8b5
KW
1863=item *
1864
1865There are some bugs in the C<pack>/C<unpack> C<"U0"> template
1866
1867=item *
1868
1869There are a significant number of test failures in the CPAN modules
1870shipped with Perl v5.22. These are only in modules not primarily
1871maintained by Perl 5 porters. Some of these are failures in the tests
1872only: they don't realize that it is proper to get different results on
1873EBCDIC platforms. And some of the failures are real bugs. If you
1874compile and do a C<make test> on Perl, all tests on the C</cpan>
1875directory are skipped.
1876
1877In particular, the extensions L<Unicode::Collate> and
1878L<Unicode::Normalize> are not supported under EBCDIC; likewise for the
1879(now deprecated) L<encoding> pragma.
1880
1881L<Encode> partially works.
1882
7a145370
KW
1883=item *
1884
1885In earlier versions, when byte and character data were concatenated,
1886the new string was sometimes created by
1887decoding the byte strings as I<ISO 8859-1 (Latin-1)>, even if the
1888old Unicode string used EBCDIC.
1889
4d2ca8b5
KW
1890=back
1891
b3b6085d
PP
1892=head1 SEE ALSO
1893
395f5a0c 1894L<perllocale>, L<perlfunc>, L<perlunicode>, L<utf8>.
b3b6085d 1895
d396a558
JH
1896=head1 REFERENCES
1897
2bbc8d55 1898L<http://anubis.dkuug.dk/i18n/charmaps>
d396a558 1899
2bbc8d55 1900L<http://www.unicode.org/>
d396a558 1901
2bbc8d55 1902L<http://www.unicode.org/unicode/reports/tr16/>
d396a558 1903
08d7a6b2 1904L<http://www.wps.com/projects/codes/>
51b5cecb
PP
1905B<ASCII: American Standard Code for Information Infiltration> Tom Jennings,
1906September 1999.
1907
eaf8b9b9
KW
1908B<The Unicode Standard, Version 3.0> The Unicode Consortium, Lisa Moore ed.,
1909ISBN 0-201-61633-5, Addison Wesley Developers Press, February 2000.
51b5cecb 1910
eaf8b9b9
KW
1911B<CDRA: IBM - Character Data Representation Architecture -
1912Reference and Registry>, IBM SC09-2190-00, December 1996.
d396a558 1913
eaf8b9b9 1914"Demystifying Character Sets", Andrea Vine, Multilingual Computing
d396a558
JH
1915& Technology, B<#26 Vol. 10 Issue 4>, August/September 1999;
1916ISSN 1523-0309; Multilingual Computing Inc. Sandpoint ID, USA.
1917
1e054b24
PP
1918B<Codes, Ciphers, and Other Cryptic and Clandestine Communication>
1919Fred B. Wrixon, ISBN 1-57912-040-7, Black Dog & Leventhal Publishers,
19201998.
1921
2bbc8d55 1922L<http://www.bobbemer.com/P-BIT.HTM>
395f5a0c
PK
1923B<IBM - EBCDIC and the P-bit; The biggest Computer Goof Ever> Robert Bemer.
1924
1925=head1 HISTORY
1926
192715 April 2001: added UTF-8 and UTF-EBCDIC to main table, pvhp.
1928
d396a558
JH
1929=head1 AUTHOR
1930
eaf8b9b9
KW
1931Peter Prymmer pvhp@best.com wrote this in 1999 and 2000
1932with CCSID 0819 and 0037 help from Chris Leach and
1933AndrE<eacute> Pirard A.Pirard@ulg.ac.be as well as POSIX-BC
b3b6085d 1934help from Thomas Dorner Thomas.Dorner@start.de.
eaf8b9b9
KW
1935Thanks also to Vickie Cooper, Philip Newton, William Raffloer, and
1936Joe Smith. Trademarks, registered trademarks, service marks and
1937registered service marks used in this document are the property of
1e054b24 1938their respective owners.
4d2ca8b5
KW
1939
1940Now maintained by Perl5 Porters.