This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
utf8.h, utfebcdic.h: Add, fix comments
[perl5.git] / utfebcdic.h
CommitLineData
1d72bdf6
NIS
1/* utfebcdic.h
2 *
2eee27d7
SS
3 * Copyright (C) 2001, 2002, 2003, 2005, 2006, 2007, 2009,
4 * 2010, 2011 by Larry Wall, Nick Ing-Simmons, and others
1d72bdf6
NIS
5 *
6 * You may distribute under the terms of either the GNU General Public
7 * License or the Artistic License, as specified in the README file.
8 *
9 * Macros to implement UTF-EBCDIC as perl's internal encoding
486ec47a 10 * Taken from version 7.1 of Unicode Technical Report #16:
1d72bdf6 11 * http://www.unicode.org/unicode/reports/tr16
fe749c9a
KW
12 *
13 * To summarize, the way it works is:
14 * To convert an EBCDIC character to UTF-EBCDIC:
d06134e5 15 * 1) convert to Unicode. The table in this file that does this for
fe749c9a
KW
16 * EBCDIC bytes is PL_e2a (with inverse PLa2e). The 'a' stands for
17 * ASCIIish, meaning latin1.
d06134e5
KW
18 * 2) convert that to a utf8-like string called I8 (I stands for
19 * intermediate) with variant characters occupying multiple bytes. This
20 * step is similar to the utf8-creating step from Unicode, but the details
21 * are different. This transformation is called UTF8-Mod. There is a
22 * chart about the bit patterns in a comment later in this file. But
fe749c9a
KW
23 * essentially here are the differences:
24 * UTF8 I8
25 * invariant byte starts with 0 starts with 0 or 100
26 * continuation byte starts with 10 starts with 101
27 * start byte same in both: if the code point requires N bytes,
28 * then the leading N bits are 1, followed by a 0. (No
29 * trailing 0 for the very largest possible allocation
30 * in I8, far beyond the current Unicode standard's
31 * max, as shown in the comment later in this file.)
32 * 3) Use the table published in tr16 to convert each byte from step 2 into
d06134e5
KW
33 * final UTF-EBCDIC. That table is reproduced in this file as PL_utf2e,
34 * and its inverse is PL_e2utf. They are constructed so that all EBCDIC
35 * invariants remain invariant, but no others do. For example, the
36 * ordinal value of 'A' is 193 in EBCDIC, and also is 193 in UTF-EBCDIC.
37 * Step 1) converts it to 65, Step 2 leaves it at 65, and Step 3 converts
38 * it back to 193. As an example of how a variant character works, take
486ec47a 39 * LATIN SMALL LETTER Y WITH DIAERESIS, which is typically 0xDF in
d06134e5
KW
40 * EBCDIC. Step 1 converts it to the Unicode value, 0xFF. Step 2
41 * converts that to two bytes = 11000111 10111111 = C7 BF, and Step 3
42 * converts those to 0x8B 0x73. The table is constructed so that the
43 * first byte of the final form of a variant will always have its upper
44 * bit set (at least in the encodings that Perl recognizes, and probably
45 * all). But note that the upper bit of some invariants is also 1.
fe749c9a
KW
46 *
47 * If you're starting from Unicode, skip step 1. For UTF-EBCDIC to straight
48 * EBCDIC, reverse the steps.
49 *
50 * The EBCDIC invariants have been chosen to be those characters whose Unicode
51 * equivalents have ordinal numbers less than 160, that is the same characters
52 * that are expressible in ASCII, plus the C1 controls. So there are 160
53 * invariants instead of the 128 in UTF-8. (My guess is that this is because
54 * the C1 control NEL (and maybe others) is important in IBM.)
55 *
56 * The purpose of Step 3 is to make the encoding be invariant for the chosen
57 * characters. This messes up the convenient patterns found in step 2, so
58 * generally, one has to undo step 3 into a temporary to use them. However,
59 * a "shadow", or parallel table, PL_utf8skip, has been constructed so that for
60 * each byte, it says how long the sequence is if that byte were to begin it
61 *
62 * There are actually 3 slightly different UTF-EBCDIC encodings in this file,
63 * one for each of the code pages recognized by Perl. That means that there
64 * are actually three different sets of tables, one for each code page. (If
d06134e5
KW
65 * Perl is compiled on platforms using another EBCDIC code page, it may not
66 * compile, or Perl may silently mistake it for one of the three.)
fe749c9a
KW
67 *
68 * EBCDIC characters above 0xFF are the same as Unicode in Perl's
69 * implementation of all 3 encodings, so for those Step 1 is trivial.
70 *
71 * (Note that the entries for invariant characters are necessarily the same in
72 * PL_e2a and PLe2f, and the same for their inverses.)
73 *
74 * UTF-EBCDIC strings are the same length or longer than UTF-8 representations
75 * of the same string. The maximum code point representable as 2 bytes in
76 * UTF-EBCDIC is 0x3FFF, instead of 0x7FFF in UTF-8.
1d72bdf6
NIS
77 */
78
79START_EXTERN_C
80
81#ifdef DOINIT
82/* Indexed by encoded byte this table gives the length of the sequence.
83 Adapted from the shadow flags table in tr16.
9df205e2 84 The entries marked 9 in tr16 are continuation bytes and are marked
c4d5f83a 85 as length 1 here so that we can recover.
1d72bdf6 86*/
f5e1abaf 87#if '^' == 95 /* if defined(__MVS__) || defined(??) (VM/ESA?) 1047 */
1d72bdf6
NIS
88EXTCONST unsigned char PL_utf8skip[] = {
891,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
901,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
911,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
921,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
c4d5f83a
NIS
931,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
941,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
951,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
961,1,1,1,2,2,2,2,2,1,1,1,1,1,1,1,
1d72bdf6
NIS
972,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
982,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
992,1,1,1,1,1,1,1,1,1,2,2,2,1,2,2,
1002,2,2,2,2,2,2,3,3,3,3,3,3,1,3,3,
1011,1,1,1,1,1,1,1,1,1,3,3,3,3,3,3,
1021,1,1,1,1,1,1,1,1,1,3,3,4,4,4,4,
1031,4,1,1,1,1,1,1,1,1,4,4,4,5,5,5,
1041,1,1,1,1,1,1,1,1,1,5,6,6,7,7,1
105};
f5e1abaf
JH
106#endif
107
108#if '^' == 106 /* if defined(_OSD_POSIX) POSIX-BC */
109unsigned char PL_utf8skip[] = {
1101,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1111,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1121,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1131,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1141,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1151,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1161,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1171,1,1,1,2,2,2,2,2,3,1,1,1,1,1,1,
1182,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
1192,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
1202,3,1,1,1,1,1,1,1,1,2,2,2,3,2,2,
1211,2,2,2,2,2,2,3,3,3,2,1,1,1,3,3,
1224,1,1,1,1,1,1,1,1,1,3,3,3,3,3,3,
1231,1,1,1,1,1,1,1,1,1,3,3,4,6,4,4,
1247,4,1,1,1,1,1,1,1,1,4,4,4,5,5,5,
1251,1,1,1,1,1,1,1,1,1,5,1,6,1,7,1
126};
127#endif
128
129#if '^' == 176 /* if defined(??) (OS/400?) 037 */
130unsigned char PL_utf8skip[] = {
1311,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1321,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1331,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1341,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1351,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1361,1,1,1,1,1,1,1,1,1,1,1,1,1,1,2,
1371,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1381,1,1,1,2,2,2,2,2,1,1,1,1,1,1,1,
1392,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
1402,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
1412,1,1,1,1,1,1,1,1,1,2,2,2,3,2,2,
1421,2,2,2,2,2,2,3,3,3,1,1,3,3,3,3,
1431,1,1,1,1,1,1,1,1,1,3,3,3,3,3,3,
1441,1,1,1,1,1,1,1,1,1,3,3,4,4,4,4,
1451,4,1,1,1,1,1,1,1,1,4,4,4,5,5,5,
1461,1,1,1,1,1,1,1,1,1,5,6,6,7,7,1
147};
148#endif
1d72bdf6 149
fe749c9a
KW
150/* Transform tables from tr16 applied after encoding to render encoding EBCDIC
151 * like, meaning that all the invariants are actually invariant, eg, that 'A'
152 * remains 'A' */
1d72bdf6 153
f5e1abaf 154#if '^' == 95 /* if defined(__MVS__) || defined(??) (VM/ESA?) 1047 */
d06134e5 155EXTCONST unsigned char PL_utf2e[] = { /* I8 to EBCDIC (IBM-1047) */
1d72bdf6
NIS
156 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, 0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
157 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, 0x18, 0x19, 0x3F, 0x27, 0x1C, 0x1D, 0x1E, 0x1F,
158 0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
159 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
160 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
161 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xAD, 0xE0, 0xBD, 0x5F, 0x6D,
162 0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
163 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xC0, 0x4F, 0xD0, 0xA1, 0x07,
164 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x06, 0x17, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x09, 0x0A, 0x1B,
165 0x30, 0x31, 0x1A, 0x33, 0x34, 0x35, 0x36, 0x08, 0x38, 0x39, 0x3A, 0x3B, 0x04, 0x14, 0x3E, 0xFF,
166 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4A, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56,
167 0x57, 0x58, 0x59, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x70, 0x71, 0x72, 0x73,
168 0x74, 0x75, 0x76, 0x77, 0x78, 0x80, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F, 0x90, 0x9A, 0x9B, 0x9C,
169 0x9D, 0x9E, 0x9F, 0xA0, 0xAA, 0xAB, 0xAC, 0xAE, 0xAF, 0xB0, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6,
170 0xB7, 0xB8, 0xB9, 0xBA, 0xBB, 0xBC, 0xBE, 0xBF, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF, 0xDA, 0xDB,
cbbb00c6 171 0xDC, 0xDD, 0xDE, 0xDF, 0xE1, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE
1d72bdf6
NIS
172};
173
d06134e5 174EXTCONST unsigned char PL_e2utf[] = { /* EBCDIC (IBM-1047) to I8 */
1d72bdf6
NIS
175 0x00, 0x01, 0x02, 0x03, 0x9C, 0x09, 0x86, 0x7F, 0x97, 0x8D, 0x8E, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
176 0x10, 0x11, 0x12, 0x13, 0x9D, 0x0A, 0x08, 0x87, 0x18, 0x19, 0x92, 0x8F, 0x1C, 0x1D, 0x1E, 0x1F,
177 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x17, 0x1B, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x05, 0x06, 0x07,
178 0x90, 0x91, 0x16, 0x93, 0x94, 0x95, 0x96, 0x04, 0x98, 0x99, 0x9A, 0x9B, 0x14, 0x15, 0x9E, 0x1A,
179 0x20, 0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0x2E, 0x3C, 0x28, 0x2B, 0x7C,
180 0x26, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF, 0xB0, 0xB1, 0xB2, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0x5E,
181 0x2D, 0x2F, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7, 0xB8, 0xB9, 0xBA, 0xBB, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
182 0xBC, 0xBD, 0xBE, 0xBF, 0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
183 0xC5, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xC6, 0xC7, 0xC8, 0xC9, 0xCA, 0xCB,
184 0xCC, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0xCD, 0xCE, 0xCF, 0xD0, 0xD1, 0xD2,
185 0xD3, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0xD4, 0xD5, 0xD6, 0x5B, 0xD7, 0xD8,
186 0xD9, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF, 0xE0, 0xE1, 0xE2, 0xE3, 0xE4, 0xE5, 0x5D, 0xE6, 0xE7,
187 0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED,
188 0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0xEE, 0xEF, 0xF0, 0xF1, 0xF2, 0xF3,
189 0x5C, 0xF4, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0xFA,
cbbb00c6 190 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF, 0x9F
1d72bdf6 191};
f5e1abaf
JH
192#endif /* 1047 */
193
194#if '^' == 106 /* if defined(_OSD_POSIX) POSIX-BC */
d06134e5 195unsigned char PL_utf2e[] = { /* I8 to EBCDIC (POSIX-BC) */
f5e1abaf
JH
196 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, 0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
197 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, 0x18, 0x19, 0x3F, 0x27, 0x1C, 0x1D, 0x1E, 0x1F,
198 0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
199 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
200 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
201 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xBB, 0xBC, 0xBD, 0x6A, 0x6D,
202 0x4A, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
203 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xFB, 0x4F, 0xFD, 0xFF, 0x07,
204 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x06, 0x17, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x09, 0x0A, 0x1B,
205 0x30, 0x31, 0x1A, 0x33, 0x34, 0x35, 0x36, 0x08, 0x38, 0x39, 0x3A, 0x3B, 0x04, 0x14, 0x3E, 0x5F,
206 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0xB0, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56,
207 0x57, 0x58, 0x59, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xD0, 0x70, 0x71, 0x72, 0x73,
208 0x74, 0x75, 0x76, 0x77, 0x78, 0x80, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F, 0x90, 0x9A, 0x9B, 0x9C,
209 0x9D, 0x9E, 0x9F, 0xA0, 0xAA, 0xAB, 0xAC, 0xAE, 0xAF, 0xBA, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6,
210 0xB7, 0xB8, 0xB9, 0xAD, 0x79, 0xA1, 0xBE, 0xBF, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF, 0xDA, 0xDB,
cbbb00c6 211 0xDC, 0xC0, 0xDE, 0xDF, 0xE1, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF, 0xFA, 0xDD, 0xFC, 0xE0, 0xFE
f5e1abaf
JH
212};
213
d06134e5 214unsigned char PL_e2utf[] = { /* EBCDIC (POSIX-BC) to I8 */
f5e1abaf
JH
215 0x00, 0x01, 0x02, 0x03, 0x9C, 0x09, 0x86, 0x7F, 0x97, 0x8D, 0x8E, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
216 0x10, 0x11, 0x12, 0x13, 0x9D, 0x0A, 0x08, 0x87, 0x18, 0x19, 0x92, 0x8F, 0x1C, 0x1D, 0x1E, 0x1F,
217 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x17, 0x1B, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x05, 0x06, 0x07,
218 0x90, 0x91, 0x16, 0x93, 0x94, 0x95, 0x96, 0x04, 0x98, 0x99, 0x9A, 0x9B, 0x14, 0x15, 0x9E, 0x1A,
219 0x20, 0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0x60, 0x2E, 0x3C, 0x28, 0x2B, 0x7C,
220 0x26, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF, 0xB0, 0xB1, 0xB2, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0x9F,
221 0x2D, 0x2F, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7, 0xB8, 0xB9, 0xBA, 0x5E, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
222 0xBC, 0xBD, 0xBE, 0xBF, 0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0xE4, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
223 0xC5, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xC6, 0xC7, 0xC8, 0xC9, 0xCA, 0xCB,
224 0xCC, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0xCD, 0xCE, 0xCF, 0xD0, 0xD1, 0xD2,
225 0xD3, 0xE5, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0xD4, 0xD5, 0xD6, 0xE3, 0xD7, 0xD8,
226 0xA9, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF, 0xE0, 0xE1, 0xE2, 0xD9, 0x5B, 0x5C, 0x5D, 0xE6, 0xE7,
227 0xF1, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED,
228 0xBB, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0xEE, 0xEF, 0xF0, 0xFC, 0xF2, 0xF3,
229 0xFE, 0xF4, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0xFA,
cbbb00c6 230 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0xFB, 0x7B, 0xFD, 0x7D, 0xFF, 0x7E
f5e1abaf
JH
231};
232#endif /* POSIX-BC */
233
234#if '^' == 176 /* if defined(??) (OS/400?) 037 */
d06134e5 235unsigned char PL_utf2e[] = { /* I8 to EBCDIC (IBM-037) */
f5e1abaf
JH
236 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, 0x16, 0x05, 0x25, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
237 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, 0x18, 0x19, 0x3F, 0x27, 0x1C, 0x1D, 0x1E, 0x1F,
238 0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
239 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
240 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
241 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xBA, 0xE0, 0xBB, 0xB0, 0x6D,
242 0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
243 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xC0, 0x4F, 0xD0, 0xA1, 0x07,
244 0x20, 0x21, 0x22, 0x23, 0x24, 0x15, 0x06, 0x17, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x09, 0x0A, 0x1B,
245 0x30, 0x31, 0x1A, 0x33, 0x34, 0x35, 0x36, 0x08, 0x38, 0x39, 0x3A, 0x3B, 0x04, 0x14, 0x3E, 0xFF,
246 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4A, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56,
247 0x57, 0x58, 0x59, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x70, 0x71, 0x72, 0x73,
248 0x74, 0x75, 0x76, 0x77, 0x78, 0x80, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F, 0x90, 0x9A, 0x9B, 0x9C,
249 0x9D, 0x9E, 0x9F, 0xA0, 0xAA, 0xAB, 0xAC, 0xAE, 0xAF, 0x5F, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6,
250 0xB7, 0xB8, 0xB9, 0xAD, 0xBD, 0xBC, 0xBE, 0xBF, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF, 0xDA, 0xDB,
cbbb00c6 251 0xDC, 0xDD, 0xDE, 0xDF, 0xE1, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE
f5e1abaf
JH
252};
253
d06134e5 254unsigned char PL_e2utf[] = { /* EBCDIC (IBM-037) to I8 */
f5e1abaf
JH
255 0x00, 0x01, 0x02, 0x03, 0x9C, 0x09, 0x86, 0x7F, 0x97, 0x8D, 0x8E, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
256 0x10, 0x11, 0x12, 0x13, 0x9D, 0x85, 0x08, 0x87, 0x18, 0x19, 0x92, 0x8F, 0x1C, 0x1D, 0x1E, 0x1F,
257 0x80, 0x81, 0x82, 0x83, 0x84, 0x0A, 0x17, 0x1B, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x05, 0x06, 0x07,
258 0x90, 0x91, 0x16, 0x93, 0x94, 0x95, 0x96, 0x04, 0x98, 0x99, 0x9A, 0x9B, 0x14, 0x15, 0x9E, 0x1A,
259 0x20, 0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0x2E, 0x3C, 0x28, 0x2B, 0x7C,
260 0x26, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF, 0xB0, 0xB1, 0xB2, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0xD9,
261 0x2D, 0x2F, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7, 0xB8, 0xB9, 0xBA, 0xBB, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
262 0xBC, 0xBD, 0xBE, 0xBF, 0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
263 0xC5, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xC6, 0xC7, 0xC8, 0xC9, 0xCA, 0xCB,
264 0xCC, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0xCD, 0xCE, 0xCF, 0xD0, 0xD1, 0xD2,
265 0xD3, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0xD4, 0xD5, 0xD6, 0xE3, 0xD7, 0xD8,
266 0x5E, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF, 0xE0, 0xE1, 0xE2, 0x5B, 0x5D, 0xE5, 0xE4, 0xE6, 0xE7,
267 0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED,
268 0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0xEE, 0xEF, 0xF0, 0xF1, 0xF2, 0xF3,
269 0x5C, 0xF4, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0xFA,
cbbb00c6 270 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF, 0x9F
f5e1abaf
JH
271};
272#endif /* 037 */
1d72bdf6
NIS
273
274/* These tables moved from perl.h and converted to hex.
486ec47a 275 They map platform code page from/to bottom 256 codes of Unicode (i.e. iso-8859-1).
1d72bdf6
NIS
276*/
277
278#if '^' == 95 /* if defined(__MVS__) || defined(??) (VM/ESA?) 1047 */
279EXTCONST unsigned char PL_a2e[] = { /* ASCII (iso-8859-1) to EBCDIC (IBM-1047) */
280 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, 0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
281 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, 0x18, 0x19, 0x3F, 0x27, 0x1C, 0x1D, 0x1E, 0x1F,
282 0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
283 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
284 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
285 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xAD, 0xE0, 0xBD, 0x5F, 0x6D,
286 0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
287 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xC0, 0x4F, 0xD0, 0xA1, 0x07,
288 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x06, 0x17, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x09, 0x0A, 0x1B,
289 0x30, 0x31, 0x1A, 0x33, 0x34, 0x35, 0x36, 0x08, 0x38, 0x39, 0x3A, 0x3B, 0x04, 0x14, 0x3E, 0xFF,
290 0x41, 0xAA, 0x4A, 0xB1, 0x9F, 0xB2, 0x6A, 0xB5, 0xBB, 0xB4, 0x9A, 0x8A, 0xB0, 0xCA, 0xAF, 0xBC,
291 0x90, 0x8F, 0xEA, 0xFA, 0xBE, 0xA0, 0xB6, 0xB3, 0x9D, 0xDA, 0x9B, 0x8B, 0xB7, 0xB8, 0xB9, 0xAB,
292 0x64, 0x65, 0x62, 0x66, 0x63, 0x67, 0x9E, 0x68, 0x74, 0x71, 0x72, 0x73, 0x78, 0x75, 0x76, 0x77,
293 0xAC, 0x69, 0xED, 0xEE, 0xEB, 0xEF, 0xEC, 0xBF, 0x80, 0xFD, 0xFE, 0xFB, 0xFC, 0xBA, 0xAE, 0x59,
294 0x44, 0x45, 0x42, 0x46, 0x43, 0x47, 0x9C, 0x48, 0x54, 0x51, 0x52, 0x53, 0x58, 0x55, 0x56, 0x57,
295 0x8C, 0x49, 0xCD, 0xCE, 0xCB, 0xCF, 0xCC, 0xE1, 0x70, 0xDD, 0xDE, 0xDB, 0xDC, 0x8D, 0x8E, 0xDF
296};
297
431fc773 298EXTCONST unsigned char PL_e2a[] = { /* EBCDIC (IBM-1047) to ASCII (iso-8859-1) */
1d72bdf6
NIS
299 0x00, 0x01, 0x02, 0x03, 0x9C, 0x09, 0x86, 0x7F, 0x97, 0x8D, 0x8E, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
300 0x10, 0x11, 0x12, 0x13, 0x9D, 0x0A, 0x08, 0x87, 0x18, 0x19, 0x92, 0x8F, 0x1C, 0x1D, 0x1E, 0x1F,
301 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x17, 0x1B, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x05, 0x06, 0x07,
302 0x90, 0x91, 0x16, 0x93, 0x94, 0x95, 0x96, 0x04, 0x98, 0x99, 0x9A, 0x9B, 0x14, 0x15, 0x9E, 0x1A,
303 0x20, 0xA0, 0xE2, 0xE4, 0xE0, 0xE1, 0xE3, 0xE5, 0xE7, 0xF1, 0xA2, 0x2E, 0x3C, 0x28, 0x2B, 0x7C,
304 0x26, 0xE9, 0xEA, 0xEB, 0xE8, 0xED, 0xEE, 0xEF, 0xEC, 0xDF, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0x5E,
305 0x2D, 0x2F, 0xC2, 0xC4, 0xC0, 0xC1, 0xC3, 0xC5, 0xC7, 0xD1, 0xA6, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
306 0xF8, 0xC9, 0xCA, 0xCB, 0xC8, 0xCD, 0xCE, 0xCF, 0xCC, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
307 0xD8, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xAB, 0xBB, 0xF0, 0xFD, 0xFE, 0xB1,
308 0xB0, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0xAA, 0xBA, 0xE6, 0xB8, 0xC6, 0xA4,
309 0xB5, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0xA1, 0xBF, 0xD0, 0x5B, 0xDE, 0xAE,
310 0xAC, 0xA3, 0xA5, 0xB7, 0xA9, 0xA7, 0xB6, 0xBC, 0xBD, 0xBE, 0xDD, 0xA8, 0xAF, 0x5D, 0xB4, 0xD7,
311 0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0xAD, 0xF4, 0xF6, 0xF2, 0xF3, 0xF5,
312 0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0xB9, 0xFB, 0xFC, 0xF9, 0xFA, 0xFF,
313 0x5C, 0xF7, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0xB2, 0xD4, 0xD6, 0xD2, 0xD3, 0xD5,
314 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0xB3, 0xDB, 0xDC, 0xD9, 0xDA, 0x9F
315};
d02f4dad
KW
316
317EXTCONST unsigned char PL_fold[] = { /* fast EBCDIC case folding table, 'A' =>
318 'a'; 'a' => 'A' */
319 0, 1, 2, 3, 4, 5, 6, 7,
320 8, 9, 10, 11, 12, 13, 14, 15,
321 16, 17, 18, 19, 20, 21, 22, 23,
322 24, 25, 26, 27, 28, 29, 30, 31,
323 32, 33, 34, 35, 36, 37, 38, 39,
324 40, 41, 42, 43, 44, 45, 46, 47,
325 48, 49, 50, 51, 52, 53, 54, 55,
326 56, 57, 58, 59, 60, 61, 62, 63,
327 64, 65, 98, 99, 100, 101, 102, 103,
328 104, 105, 74, 75, 76, 77, 78, 79,
329 80, 113, 114, 115, 116, 117, 118, 119,
330 120, 89, 90, 91, 92, 93, 94, 95,
331 96, 97, 66, 67, 68, 69, 70, 71,
332 72, 73, 106, 107, 108, 109, 110, 111,
333 128, 81, 82, 83, 84, 85, 86, 87,
334 88, 121, 122, 123, 124, 125, 126, 127,
335 112, 'A', 'B', 'C', 'D', 'E', 'F', 'G',
336 'H', 'I', 138, 139, 172, 186, 174, 143,
337 144, 'J', 'K', 'L', 'M', 'N', 'O', 'P',
338 'Q', 'R', 154, 155, 158, 157, 156, 159,
339 160, 161, 'S', 'T', 'U', 'V', 'W', 'X',
340 'Y', 'Z', 170, 171, 140, 173, 142, 175,
341 176, 177, 178, 179, 180, 181, 182, 183,
342 184, 185, 141, 187, 188, 189, 190, 191,
343 192, 'a', 'b', 'c', 'd', 'e', 'f', 'g',
344 'h', 'i', 202, 235, 236, 237, 238, 239,
345 208, 'j', 'k', 'l', 'm', 'n', 'o', 'p',
346 'q', 'r', 218, 251, 252, 253, 254, 223,
347 224, 225, 's', 't', 'u', 'v', 'w', 'x',
348 'y', 'z', 234, 203, 204, 205, 206, 207,
349 240, 241, 242, 243, 244, 245, 246, 247,
350 248, 249, 250, 219, 220, 221, 222, 255
351};
1d72bdf6
NIS
352#endif /* 1047 */
353
354#if '^' == 106 /* if defined(_OSD_POSIX) POSIX-BC */
355EXTCONST unsigned char PL_a2e[] = { /* ASCII (ISO8859-1) to EBCDIC (POSIX-BC) */
356 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, 0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
357 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, 0x18, 0x19, 0x3F, 0x27, 0x1C, 0x1D, 0x1E, 0x1F,
358 0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
359 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
360 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
361 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xBB, 0xBC, 0xBD, 0x6A, 0x6D,
362 0x4A, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
363 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xFB, 0x4F, 0xFD, 0xFF, 0x07,
364 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x06, 0x17, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x09, 0x0A, 0x1B,
365 0x30, 0x31, 0x1A, 0x33, 0x34, 0x35, 0x36, 0x08, 0x38, 0x39, 0x3A, 0x3B, 0x04, 0x14, 0x3E, 0x5F,
366 0x41, 0xAA, 0xB0, 0xB1, 0x9F, 0xB2, 0xD0, 0xB5, 0x79, 0xB4, 0x9A, 0x8A, 0xBA, 0xCA, 0xAF, 0xA1,
367 0x90, 0x8F, 0xEA, 0xFA, 0xBE, 0xA0, 0xB6, 0xB3, 0x9D, 0xDA, 0x9B, 0x8B, 0xB7, 0xB8, 0xB9, 0xAB,
368 0x64, 0x65, 0x62, 0x66, 0x63, 0x67, 0x9E, 0x68, 0x74, 0x71, 0x72, 0x73, 0x78, 0x75, 0x76, 0x77,
369 0xAC, 0x69, 0xED, 0xEE, 0xEB, 0xEF, 0xEC, 0xBF, 0x80, 0xE0, 0xFE, 0xDD, 0xFC, 0xAD, 0xAE, 0x59,
370 0x44, 0x45, 0x42, 0x46, 0x43, 0x47, 0x9C, 0x48, 0x54, 0x51, 0x52, 0x53, 0x58, 0x55, 0x56, 0x57,
371 0x8C, 0x49, 0xCD, 0xCE, 0xCB, 0xCF, 0xCC, 0xE1, 0x70, 0xC0, 0xDE, 0xDB, 0xDC, 0x8D, 0x8E, 0xDF
372};
373
374EXTCONST unsigned char PL_e2a[] = { /* EBCDIC (POSIX-BC) to ASCII (ISO8859-1) */
375 0x00, 0x01, 0x02, 0x03, 0x9C, 0x09, 0x86, 0x7F, 0x97, 0x8D, 0x8E, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
376 0x10, 0x11, 0x12, 0x13, 0x9D, 0x0A, 0x08, 0x87, 0x18, 0x19, 0x92, 0x8F, 0x1C, 0x1D, 0x1E, 0x1F,
377 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x17, 0x1B, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x05, 0x06, 0x07,
378 0x90, 0x91, 0x16, 0x93, 0x94, 0x95, 0x96, 0x04, 0x98, 0x99, 0x9A, 0x9B, 0x14, 0x15, 0x9E, 0x1A,
379 0x20, 0xA0, 0xE2, 0xE4, 0xE0, 0xE1, 0xE3, 0xE5, 0xE7, 0xF1, 0x60, 0x2E, 0x3C, 0x28, 0x2B, 0x7C,
380 0x26, 0xE9, 0xEA, 0xEB, 0xE8, 0xED, 0xEE, 0xEF, 0xEC, 0xDF, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0x9F,
381 0x2D, 0x2F, 0xC2, 0xC4, 0xC0, 0xC1, 0xC3, 0xC5, 0xC7, 0xD1, 0x5E, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
382 0xF8, 0xC9, 0xCA, 0xCB, 0xC8, 0xCD, 0xCE, 0xCF, 0xCC, 0xA8, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
383 0xD8, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xAB, 0xBB, 0xF0, 0xFD, 0xFE, 0xB1,
384 0xB0, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0xAA, 0xBA, 0xE6, 0xB8, 0xC6, 0xA4,
385 0xB5, 0xAF, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0xA1, 0xBF, 0xD0, 0xDD, 0xDE, 0xAE,
386 0xA2, 0xA3, 0xA5, 0xB7, 0xA9, 0xA7, 0xB6, 0xBC, 0xBD, 0xBE, 0xAC, 0x5B, 0x5C, 0x5D, 0xB4, 0xD7,
387 0xF9, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0xAD, 0xF4, 0xF6, 0xF2, 0xF3, 0xF5,
388 0xA6, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0xB9, 0xFB, 0xFC, 0xDB, 0xFA, 0xFF,
389 0xD9, 0xF7, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0xB2, 0xD4, 0xD6, 0xD2, 0xD3, 0xD5,
390 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0xB3, 0x7B, 0xDC, 0x7D, 0xDA, 0x7E
391};
d02f4dad
KW
392
393EXTCONST unsigned char PL_fold[] = { /* fast EBCDIC case folding table, 'A' =>
394 'a'; 'a' => 'A' */
395 0, 1, 2, 3, 4, 5, 6, 7,
396 8, 9, 10, 11, 12, 13, 14, 15,
397 16, 17, 18, 19, 20, 21, 22, 23,
398 24, 25, 26, 27, 28, 29, 30, 31,
399 32, 33, 34, 35, 36, 37, 38, 39,
400 40, 41, 42, 43, 44, 45, 46, 47,
401 48, 49, 50, 51, 52, 53, 54, 55,
402 56, 57, 58, 59, 60, 61, 62, 63,
403 64, 65, 98, 99, 100, 101, 102, 103,
404 104, 105, 74, 75, 76, 77, 78, 79,
405 80, 113, 114, 115, 116, 117, 118, 119,
406 120, 89, 90, 91, 92, 93, 94, 95,
407 96, 97, 66, 67, 68, 69, 70, 71,
408 72, 73, 106, 107, 108, 109, 110, 111,
409 128, 81, 82, 83, 84, 85, 86, 87,
410 88, 121, 122, 123, 124, 125, 126, 127,
411 112, 'A', 'B', 'C', 'D', 'E', 'F', 'G',
412 'H', 'I', 138, 139, 172, 173, 174, 143,
413 144, 'J', 'K', 'L', 'M', 'N', 'O', 'P',
414 'Q', 'R', 154, 155, 158, 157, 156, 159,
415 160, 161, 'S', 'T', 'U', 'V', 'W', 'X',
416 'Y', 'Z', 170, 171, 140, 141, 142, 175,
417 176, 177, 178, 179, 180, 181, 182, 183,
418 184, 185, 186, 187, 188, 189, 190, 191,
419 224, 'a', 'b', 'c', 'd', 'e', 'f', 'g',
420 'h', 'i', 202, 235, 236, 237, 238, 239,
421 208, 'j', 'k', 'l', 'm', 'n', 'o', 'p',
422 'q', 'r', 218, 221, 252, 219, 254, 223,
423 192, 225, 's', 't', 'u', 'v', 'w', 'x',
424 'y', 'z', 234, 203, 204, 205, 206, 207,
425 240, 241, 242, 243, 244, 245, 246, 247,
426 248, 249, 250, 251, 220, 253, 222, 255
427};
1d72bdf6 428#endif /* POSIX-BC */
f5e1abaf 429
1d72bdf6
NIS
430#if '^' == 176 /* if defined(??) (OS/400?) 037 */
431EXTCONST unsigned char PL_a2e[] = { /* ASCII (ISO8859-1) to EBCDIC (IBM-037) */
432 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, 0x16, 0x05, 0x25, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
433 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, 0x18, 0x19, 0x3F, 0x27, 0x1C, 0x1D, 0x1E, 0x1F,
434 0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
435 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
436 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
437 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xBA, 0xE0, 0xBB, 0xB0, 0x6D,
438 0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
439 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xC0, 0x4F, 0xD0, 0xA1, 0x07,
440 0x20, 0x21, 0x22, 0x23, 0x24, 0x15, 0x06, 0x17, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x09, 0x0A, 0x1B,
441 0x30, 0x31, 0x1A, 0x33, 0x34, 0x35, 0x36, 0x08, 0x38, 0x39, 0x3A, 0x3B, 0x04, 0x14, 0x3E, 0xFF,
442 0x41, 0xAA, 0x4A, 0xB1, 0x9F, 0xB2, 0x6A, 0xB5, 0xBD, 0xB4, 0x9A, 0x8A, 0x5F, 0xCA, 0xAF, 0xBC,
443 0x90, 0x8F, 0xEA, 0xFA, 0xBE, 0xA0, 0xB6, 0xB3, 0x9D, 0xDA, 0x9B, 0x8B, 0xB7, 0xB8, 0xB9, 0xAB,
444 0x64, 0x65, 0x62, 0x66, 0x63, 0x67, 0x9E, 0x68, 0x74, 0x71, 0x72, 0x73, 0x78, 0x75, 0x76, 0x77,
445 0xAC, 0x69, 0xED, 0xEE, 0xEB, 0xEF, 0xEC, 0xBF, 0x80, 0xFD, 0xFE, 0xFB, 0xFC, 0xAD, 0xAE, 0x59,
446 0x44, 0x45, 0x42, 0x46, 0x43, 0x47, 0x9C, 0x48, 0x54, 0x51, 0x52, 0x53, 0x58, 0x55, 0x56, 0x57,
447 0x8C, 0x49, 0xCD, 0xCE, 0xCB, 0xCF, 0xCC, 0xE1, 0x70, 0xDD, 0xDE, 0xDB, 0xDC, 0x8D, 0x8E, 0xDF
448};
449
450EXTCONST unsigned char PL_e2a[] = { /* EBCDIC (IBM-037) to ASCII (ISO8859-1) */
451 0x00, 0x01, 0x02, 0x03, 0x9C, 0x09, 0x86, 0x7F, 0x97, 0x8D, 0x8E, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
452 0x10, 0x11, 0x12, 0x13, 0x9D, 0x85, 0x08, 0x87, 0x18, 0x19, 0x92, 0x8F, 0x1C, 0x1D, 0x1E, 0x1F,
453 0x80, 0x81, 0x82, 0x83, 0x84, 0x0A, 0x17, 0x1B, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x05, 0x06, 0x07,
454 0x90, 0x91, 0x16, 0x93, 0x94, 0x95, 0x96, 0x04, 0x98, 0x99, 0x9A, 0x9B, 0x14, 0x15, 0x9E, 0x1A,
455 0x20, 0xA0, 0xE2, 0xE4, 0xE0, 0xE1, 0xE3, 0xE5, 0xE7, 0xF1, 0xA2, 0x2E, 0x3C, 0x28, 0x2B, 0x7C,
456 0x26, 0xE9, 0xEA, 0xEB, 0xE8, 0xED, 0xEE, 0xEF, 0xEC, 0xDF, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0xAC,
457 0x2D, 0x2F, 0xC2, 0xC4, 0xC0, 0xC1, 0xC3, 0xC5, 0xC7, 0xD1, 0xA6, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
458 0xF8, 0xC9, 0xCA, 0xCB, 0xC8, 0xCD, 0xCE, 0xCF, 0xCC, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
459 0xD8, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xAB, 0xBB, 0xF0, 0xFD, 0xFE, 0xB1,
460 0xB0, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0xAA, 0xBA, 0xE6, 0xB8, 0xC6, 0xA4,
461 0xB5, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0xA1, 0xBF, 0xD0, 0xDD, 0xDE, 0xAE,
462 0x5E, 0xA3, 0xA5, 0xB7, 0xA9, 0xA7, 0xB6, 0xBC, 0xBD, 0xBE, 0x5B, 0x5D, 0xAF, 0xA8, 0xB4, 0xD7,
463 0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0xAD, 0xF4, 0xF6, 0xF2, 0xF3, 0xF5,
464 0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0xB9, 0xFB, 0xFC, 0xF9, 0xFA, 0xFF,
465 0x5C, 0xF7, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0xB2, 0xD4, 0xD6, 0xD2, 0xD3, 0xD5,
466 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0xB3, 0xDB, 0xDC, 0xD9, 0xDA, 0x9F
467};
d02f4dad
KW
468
469EXTCONST unsigned char PL_fold[] = { /* fast EBCDIC case folding table, 'A' =>
470 'a'; 'a' => 'A' */
471 0, 1, 2, 3, 4, 5, 6, 7,
472 8, 9, 10, 11, 12, 13, 14, 15,
473 16, 17, 18, 19, 20, 21, 22, 23,
474 24, 25, 26, 27, 28, 29, 30, 31,
475 32, 33, 34, 35, 36, 37, 38, 39,
476 40, 41, 42, 43, 44, 45, 46, 47,
477 48, 49, 50, 51, 52, 53, 54, 55,
478 56, 57, 58, 59, 60, 61, 62, 63,
479 64, 65, 98, 99, 100, 101, 102, 103,
480 104, 105, 74, 75, 76, 77, 78, 79,
481 80, 113, 114, 115, 116, 117, 118, 119,
482 120, 89, 90, 91, 92, 93, 94, 95,
483 96, 97, 66, 67, 68, 69, 70, 71,
484 72, 73, 106, 107, 108, 109, 110, 111,
485 128, 81, 82, 83, 84, 85, 86, 87,
486 88, 121, 122, 123, 124, 125, 126, 127,
487 112, 'A', 'B', 'C', 'D', 'E', 'F', 'G',
488 'H', 'I', 138, 139, 172, 173, 174, 143,
489 144, 'J', 'K', 'L', 'M', 'N', 'O', 'P',
490 'Q', 'R', 154, 155, 158, 157, 156, 159,
491 160, 161, 'S', 'T', 'U', 'V', 'W', 'X',
492 'Y', 'Z', 170, 171, 140, 141, 142, 175,
493 176, 177, 178, 179, 180, 181, 182, 183,
494 184, 185, 186, 187, 188, 189, 190, 191,
495 192, 'a', 'b', 'c', 'd', 'e', 'f', 'g',
496 'h', 'i', 202, 235, 236, 237, 238, 239,
497 208, 'j', 'k', 'l', 'm', 'n', 'o', 'p',
498 'q', 'r', 218, 251, 252, 253, 254, 223,
499 224, 225, 's', 't', 'u', 'v', 'w', 'x',
500 'y', 'z', 234, 203, 204, 205, 206, 207,
501 240, 241, 242, 243, 244, 245, 246, 247,
502 248, 249, 250, 219, 220, 221, 222, 255
503};
1d72bdf6
NIS
504#endif /* 037 */
505
506#else
507EXTCONST unsigned char PL_utf8skip[];
508EXTCONST unsigned char PL_e2utf[];
509EXTCONST unsigned char PL_utf2e[];
510EXTCONST unsigned char PL_e2a[];
511EXTCONST unsigned char PL_a2e[];
d02f4dad 512EXTCONST unsigned char PL_fold[];
1d72bdf6
NIS
513#endif
514
d02f4dad
KW
515/* Since the EBCDIC code pages are isomorphic to Latin1, that table is merely a
516 * duplicate */
517EXTCONST unsigned char * PL_fold_latin1 = PL_fold;
518
1d72bdf6
NIS
519END_EXTERN_C
520
1e54db1a 521/* EBCDIC-happy ways of converting native code to UTF-8 */
1d72bdf6
NIS
522
523/* Native to iso-8859-1 */
524#define NATIVE_TO_ASCII(ch) PL_e2a[(U8)(ch)]
525#define ASCII_TO_NATIVE(ch) PL_a2e[(U8)(ch)]
9df205e2
KW
526/* Transform after encoding, essentially converts to/from I8 */
527#define NATIVE_TO_UTF(ch) PL_e2utf[(U8)(ch)] /* to I8 */
cdebb8b0 528#define NATIVE_TO_I8(ch) NATIVE_TO_UTF(ch) /* synonym */
9df205e2 529#define UTF_TO_NATIVE(ch) PL_utf2e[(U8)(ch)] /* from I8 */
cdebb8b0 530#define I8_TO_NATIVE(ch) UTF_TO_NATIVE(ch) /* synonym */
1d72bdf6
NIS
531/* Transform in wide UV char space */
532#define NATIVE_TO_UNI(ch) (((ch) > 255) ? (ch) : NATIVE_TO_ASCII(ch))
c4d5f83a 533#define UNI_TO_NATIVE(ch) (((ch) > 255) ? (ch) : ASCII_TO_NATIVE(ch))
1d72bdf6
NIS
534/* Transform in invariant..byte space */
535#define NATIVE_TO_NEED(enc,ch) ((enc) ? UTF_TO_NATIVE(NATIVE_TO_ASCII(ch)) : (ch))
536#define ASCII_TO_NEED(enc,ch) ((enc) ? UTF_TO_NATIVE(ch) : ASCII_TO_NATIVE(ch))
537
538/*
d06134e5 539 The following table is adapted from tr16, it shows I8 encoding of Unicode code points.
1d72bdf6
NIS
540
541 Unicode Bit pattern 1st Byte 2nd Byte 3rd Byte 4th Byte 5th Byte 6th Byte 7th byte
542 U+0000..U+007F 000000000xxxxxxx 0xxxxxxx
543 U+0080..U+009F 00000000100xxxxx 100xxxxx
1d72bdf6
NIS
544 U+00A0..U+03FF 000000yyyyyxxxxx 110yyyyy 101xxxxx
545 U+0400..U+3FFF 00zzzzyyyyyxxxxx 1110zzzz 101yyyyy 101xxxxx
546 U+4000..U+3FFFF 0wwwzzzzzyyyyyxxxxx 11110www 101zzzzz 101yyyyy 101xxxxx
547 U+40000..U+3FFFFF 0vvwwwwwzzzzzyyyyyxxxxx 111110vv 101wwwww 101zzzzz 101yyyyy 101xxxxx
548 U+400000..U+3FFFFFF 0uvvvvvwwwwwzzzzzyyyyyxxxxx 1111110u 101vvvvv 101wwwww 101zzzzz 101yyyyy 101xxxxx
549 U+4000000..U+7FFFFFFF 0tuuuuuvvvvvwwwwwzzzzzyyyyyxxxxx 1111111t 101uuuuu 101vvvvv 101wwwww 101zzzzz 101yyyyy 101xxxxx
550
d06134e5 551 Note: The I8 transformation is valid for UCS-4 values X'0' to
1d72bdf6
NIS
552 X'7FFFFFFF' (the full extent of ISO/IEC 10646 coding space).
553
554 */
555
556#define UNISKIP(uv) ( (uv) < 0xA0 ? 1 : \
557 (uv) < 0x400 ? 2 : \
558 (uv) < 0x4000 ? 3 : \
559 (uv) < 0x40000 ? 4 : \
560 (uv) < 0x400000 ? 5 : \
561 (uv) < 0x4000000 ? 6 : 7 )
562
c4d5f83a 563#define UNI_IS_INVARIANT(c) ((c) < 0xA0)
530495eb 564
15824458
KW
565/* UTF-EBCDIC semantic macros - transform back into I8 and then compare
566 * Comments as to the meaning of each are given at their corresponding utf8.h
567 * definitions */
0447e8df
KW
568
569#define UTF8_IS_START(c) (NATIVE_TO_UTF(c) >= 0xC5 && NATIVE_TO_UTF(c) != 0xE0)
2b9d42f0 570#define UTF8_IS_CONTINUATION(c) ((NATIVE_TO_UTF(c) & 0xE0) == 0xA0)
1d72bdf6 571#define UTF8_IS_CONTINUED(c) (NATIVE_TO_UTF(c) >= 0xA0)
0ae1fa71 572#define UTF8_IS_DOWNGRADEABLE_START(c) (NATIVE_TO_UTF(c) >= 0xC5 && NATIVE_TO_UTF(c) <= 0xC7)
7b4252f4 573#define UTF8_IS_ABOVE_LATIN1(c) (NATIVE_TO_I8(c) >= 0xC8)
1d72bdf6 574
22901f30
RGS
575#define UTF_START_MARK(len) (((len) > 7) ? 0xFF : ((U8)(0xFE << (7-(len)))))
576#define UTF_START_MASK(len) (((len) >= 6) ? 0x01 : (0x1F >> ((len)-2)))
1d72bdf6
NIS
577#define UTF_CONTINUATION_MARK 0xA0
578#define UTF_CONTINUATION_MASK ((U8)0x1f)
579#define UTF_ACCUMULATION_SHIFT 5
580
03c76984
KW
581/* How wide can a single UTF-8 encoded character become in bytes. */
582/* NOTE: Strictly speaking Perl's UTF-8 should not be called UTF-8 since UTF-8
583 * is an encoding of Unicode, and Unicode's upper limit, 0x10FFFF, can be
584 * expressed with 5 bytes. However, Perl thinks of UTF-8 as a way to encode
585 * non-negative integers in a binary format, even those above Unicode */
586#define UTF8_MAXBYTES 7
587
588/* The maximum number of UTF-8 bytes a single Unicode character can
589 * uppercase/lowercase/fold into. Unicode guarantees that the maximum
590 * expansion is 3 characters. On EBCDIC platforms, the highest Unicode
591 * character occupies 5 bytes, therefore this number is 15 */
592#define UTF8_MAXBYTES_CASE 15
593
e9a8c099
MHM
594/*
595 * Local variables:
596 * c-indentation-style: bsd
597 * c-basic-offset: 4
14d04a33 598 * indent-tabs-mode: nil
e9a8c099
MHM
599 * End:
600 *
14d04a33 601 * ex: set ts=8 sts=4 sw=4 et:
e9a8c099 602 */