This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Revert "Fix handy.t for systems without $Config{d_isblank}."
[perl5.git] / locale.c
CommitLineData
98994639
HS
1/* locale.c
2 *
1129b882
NC
3 * Copyright (C) 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001,
4 * 2002, 2003, 2005, 2006, 2007, 2008 by Larry Wall and others
98994639
HS
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 */
10
11/*
4ac71550 12 * A Elbereth Gilthoniel,
cdad3b53 13 * silivren penna míriel
4ac71550 14 * o menel aglar elenath!
cdad3b53 15 * Na-chaered palan-díriel
4ac71550
TC
16 * o galadhremmin ennorath,
17 * Fanuilos, le linnathon
18 * nef aear, si nef aearon!
19 *
20 * [p.238 of _The Lord of the Rings_, II/i: "Many Meetings"]
98994639
HS
21 */
22
166f8a29
DM
23/* utility functions for handling locale-specific stuff like what
24 * character represents the decimal point.
0d071d52
KW
25 *
26 * All C programs have an underlying locale. Perl generally doesn't pay any
27 * attention to it except within the scope of a 'use locale'. For most
28 * categories, it accomplishes this by just using different operations if it is
29 * in such scope than if not. However, various libc functions called by Perl
30 * are affected by the LC_NUMERIC category, so there are macros in perl.h that
31 * are used to toggle between the current locale and the C locale depending on
32 * the desired behavior of those functions at the moment.
166f8a29
DM
33 */
34
98994639
HS
35#include "EXTERN.h"
36#define PERL_IN_LOCALE_C
37#include "perl.h"
38
b310b053
JH
39#ifdef I_LANGINFO
40# include <langinfo.h>
41#endif
42
a4af207c
JH
43#include "reentr.h"
44
8ef6e574
KW
45#ifdef USE_LOCALE
46
98994639 47/*
0d071d52
KW
48 * Standardize the locale name from a string returned by 'setlocale', possibly
49 * modifying that string.
98994639 50 *
0ef2a2b2 51 * The typical return value of setlocale() is either
98994639
HS
52 * (1) "xx_YY" if the first argument of setlocale() is not LC_ALL
53 * (2) "xa_YY xb_YY ..." if the first argument of setlocale() is LC_ALL
54 * (the space-separated values represent the various sublocales,
0ef2a2b2 55 * in some unspecified order). This is not handled by this function.
98994639
HS
56 *
57 * In some platforms it has a form like "LC_SOMETHING=Lang_Country.866\n",
0ef2a2b2
KW
58 * which is harmful for further use of the string in setlocale(). This
59 * function removes the trailing new line and everything up through the '='
98994639
HS
60 *
61 */
62STATIC char *
63S_stdize_locale(pTHX_ char *locs)
64{
7452cf6a 65 const char * const s = strchr(locs, '=');
98994639
HS
66 bool okay = TRUE;
67
7918f24d
NC
68 PERL_ARGS_ASSERT_STDIZE_LOCALE;
69
8772537c
AL
70 if (s) {
71 const char * const t = strchr(s, '.');
98994639 72 okay = FALSE;
8772537c
AL
73 if (t) {
74 const char * const u = strchr(t, '\n');
75 if (u && (u[1] == 0)) {
76 const STRLEN len = u - s;
77 Move(s + 1, locs, len, char);
78 locs[len] = 0;
79 okay = TRUE;
98994639
HS
80 }
81 }
82 }
83
84 if (!okay)
85 Perl_croak(aTHX_ "Can't fix broken locale name \"%s\"", locs);
86
87 return locs;
88}
89
8ef6e574
KW
90#endif
91
98994639
HS
92void
93Perl_set_numeric_radix(pTHX)
94{
95#ifdef USE_LOCALE_NUMERIC
97aff369 96 dVAR;
98994639 97# ifdef HAS_LOCALECONV
7452cf6a 98 const struct lconv* const lc = localeconv();
98994639 99
98994639
HS
100 if (lc && lc->decimal_point) {
101 if (lc->decimal_point[0] == '.' && lc->decimal_point[1] == 0) {
102 SvREFCNT_dec(PL_numeric_radix_sv);
a0714e2c 103 PL_numeric_radix_sv = NULL;
98994639
HS
104 }
105 else {
106 if (PL_numeric_radix_sv)
107 sv_setpv(PL_numeric_radix_sv, lc->decimal_point);
108 else
109 PL_numeric_radix_sv = newSVpv(lc->decimal_point, 0);
28acfe03
KW
110 if (! is_ascii_string((U8 *) lc->decimal_point, 0)
111 && is_utf8_string((U8 *) lc->decimal_point, 0)
112 && is_cur_LC_category_utf8(LC_NUMERIC))
113 {
114 SvUTF8_on(PL_numeric_radix_sv);
115 }
98994639
HS
116 }
117 }
118 else
a0714e2c 119 PL_numeric_radix_sv = NULL;
98994639
HS
120# endif /* HAS_LOCALECONV */
121#endif /* USE_LOCALE_NUMERIC */
122}
123
98994639 124void
8772537c 125Perl_new_numeric(pTHX_ const char *newnum)
98994639
HS
126{
127#ifdef USE_LOCALE_NUMERIC
0d071d52
KW
128
129 /* Called after all libc setlocale() calls affecting LC_NUMERIC, to tell
130 * core Perl this and that 'newnum' is the name of the new locale.
131 * It installs this locale as the current underlying default.
132 *
133 * The default locale and the C locale can be toggled between by use of the
134 * set_numeric_local() and set_numeric_standard() functions, which should
135 * probably not be called directly, but only via macros like
136 * SET_NUMERIC_STANDARD() in perl.h.
137 *
138 * The toggling is necessary mainly so that a non-dot radix decimal point
139 * character can be output, while allowing internal calculations to use a
140 * dot.
141 *
142 * This sets several interpreter-level variables:
143 * PL_numeric_name The default locale's name: a copy of 'newnum'
144 * PL_numeric_local A boolean indicating if the toggled state is such
145 * that the current locale is the default locale
146 * PL_numeric_standard A boolean indicating if the toggled state is such
147 * that the current locale is the C locale
148 * Note that both of the last two variables can be true at the same time,
149 * if the underlying locale is C. (Toggling is a no-op under these
150 * circumstances.)
151 *
152 * Any code changing the locale (outside this file) should use
153 * POSIX::setlocale, which calls this function. Therefore this function
154 * should be called directly only from this file and from
155 * POSIX::setlocale() */
156
b03f34cf 157 char *save_newnum;
97aff369 158 dVAR;
98994639
HS
159
160 if (! newnum) {
43c5f42d
NC
161 Safefree(PL_numeric_name);
162 PL_numeric_name = NULL;
98994639
HS
163 PL_numeric_standard = TRUE;
164 PL_numeric_local = TRUE;
165 return;
166 }
167
b03f34cf
KW
168 save_newnum = stdize_locale(savepv(newnum));
169 if (! PL_numeric_name || strNE(PL_numeric_name, save_newnum)) {
98994639 170 Safefree(PL_numeric_name);
b03f34cf 171 PL_numeric_name = save_newnum;
b03f34cf 172 }
98994639 173
e19f01cb
KW
174 PL_numeric_standard = ((*save_newnum == 'C' && save_newnum[1] == '\0')
175 || strEQ(save_newnum, "POSIX"));
176 PL_numeric_local = TRUE;
177 set_numeric_radix();
6959d69d 178
98994639
HS
179#endif /* USE_LOCALE_NUMERIC */
180}
181
182void
183Perl_set_numeric_standard(pTHX)
184{
185#ifdef USE_LOCALE_NUMERIC
97aff369 186 dVAR;
98994639 187
0d071d52
KW
188 /* Toggle the LC_NUMERIC locale to C, if not already there. Probably
189 * should use the macros like SET_NUMERIC_STANDARD() in perl.h instead of
190 * calling this directly. */
191
98994639
HS
192 if (! PL_numeric_standard) {
193 setlocale(LC_NUMERIC, "C");
194 PL_numeric_standard = TRUE;
195 PL_numeric_local = FALSE;
196 set_numeric_radix();
197 }
198
199#endif /* USE_LOCALE_NUMERIC */
200}
201
202void
203Perl_set_numeric_local(pTHX)
204{
205#ifdef USE_LOCALE_NUMERIC
97aff369 206 dVAR;
98994639 207
0d071d52
KW
208 /* Toggle the LC_NUMERIC locale to the current underlying default, if not
209 * already there. Probably should use the macros like SET_NUMERIC_LOCAL()
210 * in perl.h instead of calling this directly. */
211
98994639
HS
212 if (! PL_numeric_local) {
213 setlocale(LC_NUMERIC, PL_numeric_name);
214 PL_numeric_standard = FALSE;
215 PL_numeric_local = TRUE;
216 set_numeric_radix();
217 }
218
219#endif /* USE_LOCALE_NUMERIC */
220}
221
222/*
223 * Set up for a new ctype locale.
224 */
225void
8772537c 226Perl_new_ctype(pTHX_ const char *newctype)
98994639
HS
227{
228#ifdef USE_LOCALE_CTYPE
0d071d52
KW
229
230 /* Called after all libc setlocale() calls affecting LC_CTYPE, to tell
231 * core Perl this and that 'newctype' is the name of the new locale.
232 *
233 * This function sets up the folding arrays for all 256 bytes, assuming
234 * that tofold() is tolc() since fold case is not a concept in POSIX,
235 *
236 * Any code changing the locale (outside this file) should use
237 * POSIX::setlocale, which calls this function. Therefore this function
238 * should be called directly only from this file and from
239 * POSIX::setlocale() */
240
27da23d5 241 dVAR;
68067e4e 242 UV i;
98994639 243
7918f24d
NC
244 PERL_ARGS_ASSERT_NEW_CTYPE;
245
31f05a37
KW
246 PL_in_utf8_CTYPE_locale = is_cur_LC_category_utf8(LC_CTYPE);
247
248 /* A UTF-8 locale gets standard rules. But note that code still has to
249 * handle this specially because of the three problematic code points */
250 if (PL_in_utf8_CTYPE_locale) {
251 Copy(PL_fold_latin1, PL_fold_locale, 256, U8);
252 }
253 else {
baa60164
KW
254 for (i = 0; i < 256; i++) {
255 if (isUPPER_LC((U8) i))
256 PL_fold_locale[i] = (U8) toLOWER_LC((U8) i);
257 else if (isLOWER_LC((U8) i))
258 PL_fold_locale[i] = (U8) toUPPER_LC((U8) i);
259 else
260 PL_fold_locale[i] = (U8) i;
261 }
31f05a37 262 }
98994639
HS
263
264#endif /* USE_LOCALE_CTYPE */
7918f24d 265 PERL_ARGS_ASSERT_NEW_CTYPE;
8772537c 266 PERL_UNUSED_ARG(newctype);
96a5add6 267 PERL_UNUSED_CONTEXT;
98994639
HS
268}
269
98994639 270void
8772537c 271Perl_new_collate(pTHX_ const char *newcoll)
98994639
HS
272{
273#ifdef USE_LOCALE_COLLATE
0d071d52
KW
274
275 /* Called after all libc setlocale() calls affecting LC_COLLATE, to tell
276 * core Perl this and that 'newcoll' is the name of the new locale.
277 *
278 * Any code changing the locale (outside this file) should use
279 * POSIX::setlocale, which calls this function. Therefore this function
280 * should be called directly only from this file and from
281 * POSIX::setlocale() */
282
97aff369 283 dVAR;
98994639
HS
284
285 if (! newcoll) {
286 if (PL_collation_name) {
287 ++PL_collation_ix;
288 Safefree(PL_collation_name);
289 PL_collation_name = NULL;
290 }
291 PL_collation_standard = TRUE;
292 PL_collxfrm_base = 0;
293 PL_collxfrm_mult = 2;
294 return;
295 }
296
297 if (! PL_collation_name || strNE(PL_collation_name, newcoll)) {
298 ++PL_collation_ix;
299 Safefree(PL_collation_name);
300 PL_collation_name = stdize_locale(savepv(newcoll));
770526c1
NC
301 PL_collation_standard = ((*newcoll == 'C' && newcoll[1] == '\0')
302 || strEQ(newcoll, "POSIX"));
98994639
HS
303
304 {
305 /* 2: at most so many chars ('a', 'b'). */
306 /* 50: surely no system expands a char more. */
307#define XFRMBUFSIZE (2 * 50)
308 char xbuf[XFRMBUFSIZE];
8772537c
AL
309 const Size_t fa = strxfrm(xbuf, "a", XFRMBUFSIZE);
310 const Size_t fb = strxfrm(xbuf, "ab", XFRMBUFSIZE);
311 const SSize_t mult = fb - fa;
e0601d3c 312 if (mult < 1 && !(fa == 0 && fb == 0))
ad49ad39
NC
313 Perl_croak(aTHX_ "panic: strxfrm() gets absurd - a => %"UVuf", ab => %"UVuf,
314 (UV) fa, (UV) fb);
eb160463 315 PL_collxfrm_base = (fa > (Size_t)mult) ? (fa - mult) : 0;
98994639
HS
316 PL_collxfrm_mult = mult;
317 }
318 }
319
320#endif /* USE_LOCALE_COLLATE */
321}
322
323/*
324 * Initialize locale awareness.
325 */
326int
327Perl_init_i18nl10n(pTHX_ int printwarn)
328{
329 int ok = 1;
330 /* returns
331 * 1 = set ok or not applicable,
332 * 0 = fallback to C locale,
333 * -1 = fallback to C locale failed
334 */
335
336#if defined(USE_LOCALE)
97aff369 337 dVAR;
98994639
HS
338
339#ifdef USE_LOCALE_CTYPE
340 char *curctype = NULL;
341#endif /* USE_LOCALE_CTYPE */
342#ifdef USE_LOCALE_COLLATE
343 char *curcoll = NULL;
344#endif /* USE_LOCALE_COLLATE */
345#ifdef USE_LOCALE_NUMERIC
346 char *curnum = NULL;
347#endif /* USE_LOCALE_NUMERIC */
348#ifdef __GLIBC__
7452cf6a 349 char * const language = PerlEnv_getenv("LANGUAGE");
98994639 350#endif
ccd65d51
KW
351 /* NULL uses the existing already set up locale */
352 const char * const setlocale_init = (PerlEnv_getenv("PERL_SKIP_LOCALE_INIT"))
353 ? NULL
354 : "";
7452cf6a
AL
355 char * const lc_all = PerlEnv_getenv("LC_ALL");
356 char * const lang = PerlEnv_getenv("LANG");
98994639
HS
357 bool setlocale_failure = FALSE;
358
359#ifdef LOCALE_ENVIRON_REQUIRED
360
361 /*
362 * Ultrix setlocale(..., "") fails if there are no environment
363 * variables from which to get a locale name.
364 */
365
366 bool done = FALSE;
367
b3e384bf 368# ifdef LC_ALL
98994639 369 if (lang) {
ccd65d51 370 if (setlocale(LC_ALL, setlocale_init))
98994639
HS
371 done = TRUE;
372 else
373 setlocale_failure = TRUE;
374 }
375 if (!setlocale_failure) {
b3e384bf 376# ifdef USE_LOCALE_CTYPE
56279a21 377 Safefree(curctype);
98994639
HS
378 if (! (curctype =
379 setlocale(LC_CTYPE,
380 (!done && (lang || PerlEnv_getenv("LC_CTYPE")))
ccd65d51 381 ? setlocale_init : NULL)))
98994639
HS
382 setlocale_failure = TRUE;
383 else
384 curctype = savepv(curctype);
b3e384bf
KW
385# endif /* USE_LOCALE_CTYPE */
386# ifdef USE_LOCALE_COLLATE
56279a21 387 Safefree(curcoll);
98994639
HS
388 if (! (curcoll =
389 setlocale(LC_COLLATE,
390 (!done && (lang || PerlEnv_getenv("LC_COLLATE")))
ccd65d51 391 ? setlocale_init : NULL)))
98994639
HS
392 setlocale_failure = TRUE;
393 else
394 curcoll = savepv(curcoll);
b3e384bf
KW
395# endif /* USE_LOCALE_COLLATE */
396# ifdef USE_LOCALE_NUMERIC
56279a21 397 Safefree(curnum);
98994639
HS
398 if (! (curnum =
399 setlocale(LC_NUMERIC,
400 (!done && (lang || PerlEnv_getenv("LC_NUMERIC")))
ccd65d51 401 ? setlocale_init : NULL)))
98994639
HS
402 setlocale_failure = TRUE;
403 else
404 curnum = savepv(curnum);
b3e384bf 405# endif /* USE_LOCALE_NUMERIC */
98994639
HS
406 }
407
b3e384bf 408# endif /* LC_ALL */
98994639
HS
409
410#endif /* !LOCALE_ENVIRON_REQUIRED */
411
412#ifdef LC_ALL
ccd65d51 413 if (! setlocale(LC_ALL, setlocale_init))
98994639
HS
414 setlocale_failure = TRUE;
415#endif /* LC_ALL */
416
417 if (!setlocale_failure) {
418#ifdef USE_LOCALE_CTYPE
6cb43dbf 419 Safefree(curctype);
ccd65d51 420 if (! (curctype = setlocale(LC_CTYPE, setlocale_init)))
98994639
HS
421 setlocale_failure = TRUE;
422 else
423 curctype = savepv(curctype);
424#endif /* USE_LOCALE_CTYPE */
425#ifdef USE_LOCALE_COLLATE
6cb43dbf 426 Safefree(curcoll);
ccd65d51 427 if (! (curcoll = setlocale(LC_COLLATE, setlocale_init)))
98994639
HS
428 setlocale_failure = TRUE;
429 else
430 curcoll = savepv(curcoll);
431#endif /* USE_LOCALE_COLLATE */
432#ifdef USE_LOCALE_NUMERIC
6cb43dbf 433 Safefree(curnum);
ccd65d51 434 if (! (curnum = setlocale(LC_NUMERIC, setlocale_init)))
98994639
HS
435 setlocale_failure = TRUE;
436 else
437 curnum = savepv(curnum);
438#endif /* USE_LOCALE_NUMERIC */
439 }
440
441 if (setlocale_failure) {
442 char *p;
0bd48802 443 const bool locwarn = (printwarn > 1 ||
98994639
HS
444 (printwarn &&
445 (!(p = PerlEnv_getenv("PERL_BADLANG")) || atoi(p))));
446
447 if (locwarn) {
448#ifdef LC_ALL
449
450 PerlIO_printf(Perl_error_log,
451 "perl: warning: Setting locale failed.\n");
452
453#else /* !LC_ALL */
454
455 PerlIO_printf(Perl_error_log,
456 "perl: warning: Setting locale failed for the categories:\n\t");
457#ifdef USE_LOCALE_CTYPE
458 if (! curctype)
459 PerlIO_printf(Perl_error_log, "LC_CTYPE ");
460#endif /* USE_LOCALE_CTYPE */
461#ifdef USE_LOCALE_COLLATE
462 if (! curcoll)
463 PerlIO_printf(Perl_error_log, "LC_COLLATE ");
464#endif /* USE_LOCALE_COLLATE */
465#ifdef USE_LOCALE_NUMERIC
466 if (! curnum)
467 PerlIO_printf(Perl_error_log, "LC_NUMERIC ");
468#endif /* USE_LOCALE_NUMERIC */
469 PerlIO_printf(Perl_error_log, "\n");
470
471#endif /* LC_ALL */
472
473 PerlIO_printf(Perl_error_log,
474 "perl: warning: Please check that your locale settings:\n");
475
476#ifdef __GLIBC__
477 PerlIO_printf(Perl_error_log,
478 "\tLANGUAGE = %c%s%c,\n",
479 language ? '"' : '(',
480 language ? language : "unset",
481 language ? '"' : ')');
482#endif
483
484 PerlIO_printf(Perl_error_log,
485 "\tLC_ALL = %c%s%c,\n",
486 lc_all ? '"' : '(',
487 lc_all ? lc_all : "unset",
488 lc_all ? '"' : ')');
489
490#if defined(USE_ENVIRON_ARRAY)
491 {
492 char **e;
493 for (e = environ; *e; e++) {
494 if (strnEQ(*e, "LC_", 3)
495 && strnNE(*e, "LC_ALL=", 7)
496 && (p = strchr(*e, '=')))
497 PerlIO_printf(Perl_error_log, "\t%.*s = \"%s\",\n",
498 (int)(p - *e), *e, p + 1);
499 }
500 }
501#else
502 PerlIO_printf(Perl_error_log,
503 "\t(possibly more locale environment variables)\n");
504#endif
505
506 PerlIO_printf(Perl_error_log,
507 "\tLANG = %c%s%c\n",
508 lang ? '"' : '(',
509 lang ? lang : "unset",
510 lang ? '"' : ')');
511
512 PerlIO_printf(Perl_error_log,
513 " are supported and installed on your system.\n");
514 }
515
516#ifdef LC_ALL
517
518 if (setlocale(LC_ALL, "C")) {
519 if (locwarn)
520 PerlIO_printf(Perl_error_log,
521 "perl: warning: Falling back to the standard locale (\"C\").\n");
522 ok = 0;
523 }
524 else {
525 if (locwarn)
526 PerlIO_printf(Perl_error_log,
527 "perl: warning: Failed to fall back to the standard locale (\"C\").\n");
528 ok = -1;
529 }
530
531#else /* ! LC_ALL */
532
533 if (0
534#ifdef USE_LOCALE_CTYPE
535 || !(curctype || setlocale(LC_CTYPE, "C"))
536#endif /* USE_LOCALE_CTYPE */
537#ifdef USE_LOCALE_COLLATE
538 || !(curcoll || setlocale(LC_COLLATE, "C"))
539#endif /* USE_LOCALE_COLLATE */
540#ifdef USE_LOCALE_NUMERIC
541 || !(curnum || setlocale(LC_NUMERIC, "C"))
542#endif /* USE_LOCALE_NUMERIC */
543 )
544 {
545 if (locwarn)
546 PerlIO_printf(Perl_error_log,
547 "perl: warning: Cannot fall back to the standard locale (\"C\").\n");
548 ok = -1;
549 }
550
551#endif /* ! LC_ALL */
552
553#ifdef USE_LOCALE_CTYPE
56279a21 554 Safefree(curctype);
bd61b366 555 curctype = savepv(setlocale(LC_CTYPE, NULL));
98994639
HS
556#endif /* USE_LOCALE_CTYPE */
557#ifdef USE_LOCALE_COLLATE
56279a21 558 Safefree(curcoll);
bd61b366 559 curcoll = savepv(setlocale(LC_COLLATE, NULL));
98994639
HS
560#endif /* USE_LOCALE_COLLATE */
561#ifdef USE_LOCALE_NUMERIC
56279a21 562 Safefree(curnum);
bd61b366 563 curnum = savepv(setlocale(LC_NUMERIC, NULL));
98994639
HS
564#endif /* USE_LOCALE_NUMERIC */
565 }
566 else {
567
568#ifdef USE_LOCALE_CTYPE
569 new_ctype(curctype);
570#endif /* USE_LOCALE_CTYPE */
571
572#ifdef USE_LOCALE_COLLATE
573 new_collate(curcoll);
574#endif /* USE_LOCALE_COLLATE */
575
576#ifdef USE_LOCALE_NUMERIC
577 new_numeric(curnum);
578#endif /* USE_LOCALE_NUMERIC */
b310b053 579
98994639
HS
580 }
581
8ef6e574 582#if defined(USE_PERLIO) && defined(USE_LOCALE_CTYPE)
b310b053 583 {
fde18df1 584 /* Set PL_utf8locale to TRUE if using PerlIO _and_
7d74bb61 585 the current LC_CTYPE locale is UTF-8.
8aa8f774 586 If PL_utf8locale and PL_unicode (set by -C or by $ENV{PERL_UNICODE})
a05d7ebb 587 are true, perl.c:S_parse_body() will turn on the PerlIO :utf8 layer
fde18df1 588 on STDIN, STDOUT, STDERR, _and_ the default open discipline.
085a54d9 589 */
7d74bb61 590 PL_utf8locale = is_cur_LC_category_utf8(LC_CTYPE);
fde18df1 591 }
a05d7ebb 592 /* Set PL_unicode to $ENV{PERL_UNICODE} if using PerlIO.
fde18df1
JH
593 This is an alternative to using the -C command line switch
594 (the -C if present will override this). */
595 {
dd374669 596 const char *p = PerlEnv_getenv("PERL_UNICODE");
a05d7ebb 597 PL_unicode = p ? parse_unicode_opts(&p) : 0;
5a22a2bb
NC
598 if (PL_unicode & PERL_UNICODE_UTF8CACHEASSERT_FLAG)
599 PL_utf8cache = -1;
b310b053 600 }
ec71e770 601#endif
b310b053 602
98994639 603#ifdef USE_LOCALE_CTYPE
43c5f42d 604 Safefree(curctype);
98994639
HS
605#endif /* USE_LOCALE_CTYPE */
606#ifdef USE_LOCALE_COLLATE
43c5f42d 607 Safefree(curcoll);
98994639
HS
608#endif /* USE_LOCALE_COLLATE */
609#ifdef USE_LOCALE_NUMERIC
43c5f42d 610 Safefree(curnum);
98994639 611#endif /* USE_LOCALE_NUMERIC */
8ef6e574
KW
612
613#endif /* USE_LOCALE */
614
98994639
HS
615 return ok;
616}
617
98994639
HS
618#ifdef USE_LOCALE_COLLATE
619
620/*
621 * mem_collxfrm() is a bit like strxfrm() but with two important
622 * differences. First, it handles embedded NULs. Second, it allocates
623 * a bit more memory than needed for the transformed data itself.
624 * The real transformed data begins at offset sizeof(collationix).
625 * Please see sv_collxfrm() to see how this is used.
626 */
166f8a29 627
98994639
HS
628char *
629Perl_mem_collxfrm(pTHX_ const char *s, STRLEN len, STRLEN *xlen)
630{
97aff369 631 dVAR;
98994639
HS
632 char *xbuf;
633 STRLEN xAlloc, xin, xout; /* xalloc is a reserved word in VC */
634
7918f24d
NC
635 PERL_ARGS_ASSERT_MEM_COLLXFRM;
636
98994639
HS
637 /* the first sizeof(collationix) bytes are used by sv_collxfrm(). */
638 /* the +1 is for the terminating NUL. */
639
640 xAlloc = sizeof(PL_collation_ix) + PL_collxfrm_base + (PL_collxfrm_mult * len) + 1;
a02a5408 641 Newx(xbuf, xAlloc, char);
98994639
HS
642 if (! xbuf)
643 goto bad;
644
645 *(U32*)xbuf = PL_collation_ix;
646 xout = sizeof(PL_collation_ix);
647 for (xin = 0; xin < len; ) {
224e8ef5 648 Size_t xused;
98994639
HS
649
650 for (;;) {
651 xused = strxfrm(xbuf + xout, s + xin, xAlloc - xout);
224e8ef5 652 if (xused >= PERL_INT_MAX)
98994639 653 goto bad;
eb160463 654 if ((STRLEN)xused < xAlloc - xout)
98994639
HS
655 break;
656 xAlloc = (2 * xAlloc) + 1;
657 Renew(xbuf, xAlloc, char);
658 if (! xbuf)
659 goto bad;
660 }
661
662 xin += strlen(s + xin) + 1;
663 xout += xused;
664
665 /* Embedded NULs are understood but silently skipped
666 * because they make no sense in locale collation. */
667 }
668
669 xbuf[xout] = '\0';
670 *xlen = xout - sizeof(PL_collation_ix);
671 return xbuf;
672
673 bad:
674 Safefree(xbuf);
675 *xlen = 0;
676 return NULL;
677}
678
679#endif /* USE_LOCALE_COLLATE */
680
8ef6e574
KW
681#ifdef USE_LOCALE
682
637917bb 683STATIC bool
7d74bb61
KW
684S_is_cur_LC_category_utf8(pTHX_ int category)
685{
686 /* Returns TRUE if the current locale for 'category' is UTF-8; FALSE
687 * otherwise. 'category' may not be LC_ALL. If the platform doesn't have
119ee68b
KW
688 * nl_langinfo(), nor MB_CUR_MAX, this employs a heuristic, which hence
689 * could give the wrong result. It errs on the side of not being a UTF-8
690 * locale. */
7d74bb61
KW
691
692 char *save_input_locale = NULL;
7d74bb61
KW
693 STRLEN final_pos;
694
8ef6e574 695#ifdef LC_ALL
7d74bb61 696 assert(category != LC_ALL);
8ef6e574 697#endif
7d74bb61
KW
698
699 /* First dispose of the trivial cases */
b07fffd1 700 save_input_locale = setlocale(category, NULL);
7d74bb61
KW
701 if (! save_input_locale) {
702 return FALSE; /* XXX maybe should croak */
703 }
b07fffd1 704 save_input_locale = stdize_locale(savepv(save_input_locale));
7d74bb61
KW
705 if ((*save_input_locale == 'C' && save_input_locale[1] == '\0')
706 || strEQ(save_input_locale, "POSIX"))
707 {
b07fffd1 708 Safefree(save_input_locale);
7d74bb61
KW
709 return FALSE;
710 }
711
1d958db2
KW
712#if defined(USE_LOCALE_CTYPE) \
713 && (defined(MB_CUR_MAX) || (defined(HAS_NL_LANGINFO) && defined(CODESET)))
7d74bb61 714
1d958db2 715 { /* Next try nl_langinfo or MB_CUR_MAX if available */
7d74bb61
KW
716
717 char *save_ctype_locale = NULL;
119ee68b 718 bool is_utf8;
7d74bb61 719
119ee68b 720 if (category != LC_CTYPE) { /* These work only on LC_CTYPE */
7d74bb61
KW
721
722 /* Get the current LC_CTYPE locale */
723 save_ctype_locale = stdize_locale(savepv(setlocale(LC_CTYPE, NULL)));
724 if (! save_ctype_locale) {
725 goto cant_use_nllanginfo;
726 }
727
728 /* If LC_CTYPE and the desired category use the same locale, this
729 * means that finding the value for LC_CTYPE is the same as finding
730 * the value for the desired category. Otherwise, switch LC_CTYPE
731 * to the desired category's locale */
732 if (strEQ(save_ctype_locale, save_input_locale)) {
733 Safefree(save_ctype_locale);
734 save_ctype_locale = NULL;
735 }
736 else if (! setlocale(LC_CTYPE, save_input_locale)) {
737 Safefree(save_ctype_locale);
738 goto cant_use_nllanginfo;
739 }
740 }
741
742 /* Here the current LC_CTYPE is set to the locale of the category whose
1d958db2
KW
743 * information is desired. This means that nl_langinfo() and MB_CUR_MAX
744 * should give the correct results */
119ee68b 745
1d958db2
KW
746# if defined(HAS_NL_LANGINFO) && defined(CODESET)
747 {
748 char *codeset = savepv(nl_langinfo(CODESET));
749 if (codeset && strNE(codeset, "")) {
119ee68b 750
1d958db2
KW
751 /* If we switched LC_CTYPE, switch back */
752 if (save_ctype_locale) {
753 setlocale(LC_CTYPE, save_ctype_locale);
754 Safefree(save_ctype_locale);
755 }
756
757 is_utf8 = foldEQ(codeset, STR_WITH_LEN("UTF-8"))
758 || foldEQ(codeset, STR_WITH_LEN("UTF8"));
759
760 Safefree(codeset);
761 Safefree(save_input_locale);
762 return is_utf8;
763 }
119ee68b
KW
764 }
765
1d958db2
KW
766# endif
767# ifdef MB_CUR_MAX
768
769 /* Here, either we don't have nl_langinfo, or it didn't return a
770 * codeset. Try MB_CUR_MAX */
771
119ee68b
KW
772 /* Standard UTF-8 needs at least 4 bytes to represent the maximum
773 * Unicode code point. Since UTF-8 is the only non-single byte
774 * encoding we handle, we just say any such encoding is UTF-8, and if
775 * turns out to be wrong, other things will fail */
776 is_utf8 = MB_CUR_MAX >= 4;
777
778 Safefree(save_input_locale);
779
780# ifdef HAS_MBTOWC
781
782 /* ... But, most system that have MB_CUR_MAX will also have mbtowc(),
783 * since they are both in the C99 standard. We can feed a known byte
784 * string to the latter function, and check that it gives the expected
785 * result */
786 if (is_utf8) {
787 wchar_t wc;
1d958db2 788 (void) mbtowc(&wc, NULL, 0); /* Reset any shift state */
119ee68b
KW
789 if (mbtowc(&wc, HYPHEN_UTF8, strlen(HYPHEN_UTF8))
790 != strlen(HYPHEN_UTF8)
791 || wc != (wchar_t) 0x2010)
792 {
793 is_utf8 = FALSE;
794 }
795 }
119ee68b
KW
796# endif
797
1d958db2
KW
798 /* If we switched LC_CTYPE, switch back */
799 if (save_ctype_locale) {
800 setlocale(LC_CTYPE, save_ctype_locale);
801 Safefree(save_ctype_locale);
119ee68b 802 }
7d74bb61 803
1d958db2 804 return is_utf8;
119ee68b 805# endif
7d74bb61 806 }
119ee68b 807
7d74bb61
KW
808 cant_use_nllanginfo:
809
810#endif /* HAS_NL_LANGINFO etc */
811
812 /* nl_langinfo not available or failed somehow. Look at the locale name to
751e9426 813 * see if it matches qr/UTF -? 8 /ix */
7d74bb61
KW
814
815 final_pos = strlen(save_input_locale) - 1;
751e9426
KW
816 if (final_pos >= 3) {
817 char *name = save_input_locale;
818
819 /* Find next 'U' or 'u' and look from there */
820 while ((name += strcspn(name, "Uu") + 1)
821 <= save_input_locale + final_pos - 2)
7d74bb61 822 {
751e9426
KW
823 if (toFOLD(*(name)) != 't'
824 || toFOLD(*(name + 1)) != 'f')
825 {
826 continue;
827 }
828 name += 2;
829 if (*(name) == '-') {
830 if ((name > save_input_locale + final_pos - 1)) {
831 break;
832 }
833 name++;
834 }
835 if (*(name) == '8') {
836 Safefree(save_input_locale);
837 return TRUE;
838 }
7d74bb61
KW
839 }
840 }
841
842#ifdef WIN32
843 /* http://msdn.microsoft.com/en-us/library/windows/desktop/dd317756.aspx */
844 if (final_pos >= 4
845 && *(save_input_locale + final_pos - 0) == '1'
846 && *(save_input_locale + final_pos - 1) == '0'
847 && *(save_input_locale + final_pos - 2) == '0'
848 && *(save_input_locale + final_pos - 3) == '5'
849 && *(save_input_locale + final_pos - 4) == '6')
850 {
851 Safefree(save_input_locale);
852 return TRUE;
853 }
854#endif
855
fa9b773e
KW
856 /* Other common encodings are the ISO 8859 series, which aren't UTF-8 */
857 if (instr(save_input_locale, "8859")) {
858 Safefree(save_input_locale);
859 return FALSE;
860 }
861
862#ifdef HAS_LOCALECONV
863
864# ifdef USE_LOCALE_MONETARY
865
866 /* Here, there is nothing in the locale name to indicate whether the locale
867 * is UTF-8 or not. This "name", the return of setlocale(), is actually
868 * defined to be opaque, so we can't really rely on the absence of various
869 * substrings in the name to indicate its UTF-8ness. Look at the locale's
870 * currency symbol. Often that will be in the native script, and if the
871 * symbol isn't in UTF-8, we know that the locale isn't. If it is
872 * non-ASCII UTF-8, we infer that the locale is too.
873 * To do this, like above for LC_CTYPE, we first set LC_MONETARY to the
874 * locale of the desired category, if it isn't that locale already */
875
876 {
877 char *save_monetary_locale = NULL;
878 bool illegal_utf8 = FALSE;
879 bool only_ascii = FALSE;
880 const struct lconv* const lc = localeconv();
881
882 if (category != LC_MONETARY) {
883
884 save_monetary_locale = stdize_locale(savepv(setlocale(LC_MONETARY,
885 NULL)));
886 if (! save_monetary_locale) {
887 goto cant_use_monetary;
888 }
889
890 if (strNE(save_monetary_locale, save_input_locale)) {
891 if (! setlocale(LC_MONETARY, save_input_locale)) {
892 Safefree(save_monetary_locale);
893 goto cant_use_monetary;
894 }
895 }
896 }
897
898 /* Here the current LC_MONETARY is set to the locale of the category
899 * whose information is desired. */
900
901 if (lc && lc->currency_symbol) {
902 if (! is_utf8_string((U8 *) lc->currency_symbol, 0)) {
903 illegal_utf8 = TRUE;
904 }
905 else if (is_ascii_string((U8 *) lc->currency_symbol, 0)) {
906 only_ascii = TRUE;
907 }
908 }
909
910 /* If we changed it, restore LC_MONETARY to its original locale */
911 if (save_monetary_locale) {
912 setlocale(LC_MONETARY, save_monetary_locale);
913 Safefree(save_monetary_locale);
914 }
915
916 Safefree(save_input_locale);
917
918 /* It isn't a UTF-8 locale if the symbol is not legal UTF-8; otherwise
919 * assume the locale is UTF-8 if and only if the symbol is non-ascii
920 * UTF-8. (We can't really tell if the locale is UTF-8 or not if the
921 * symbol is just a '$', so we err on the side of it not being UTF-8)
922 * */
923 return (illegal_utf8)
924 ? FALSE
925 : ! only_ascii;
926
927 }
928 cant_use_monetary:
929
930# endif /* USE_LOCALE_MONETARY */
931#endif /* HAS_LOCALECONV */
932
933#if 0 && defined(HAS_STRERROR) && defined(USE_LOCALE_MESSAGES)
934
935/* This code is ifdefd out because it was found to not be necessary in testing
936 * on our dromedary test machine, which has over 700 locales. There, looking
937 * at just the currency symbol gave essentially the same results as doing this
938 * extra work. Executing this also caused segfaults in miniperl. I left it in
939 * so as to avoid rewriting it if real-world experience indicates that
940 * dromedary is an outlier. Essentially, instead of returning abpve if we
941 * haven't found illegal utf8, we continue on and examine all the strerror()
942 * messages on the platform for utf8ness. If all are ASCII, we still don't
943 * know the answer; but otherwise we have a pretty good indication of the
944 * utf8ness. The reason this doesn't necessarily help much is that the
945 * messages may not have been translated into the locale. The currency symbol
946 * is much more likely to have been translated. The code below would need to
947 * be altered somewhat to just be a continuation of testing the currency
948 * symbol. */
949 int e;
950 unsigned int failures = 0, non_ascii = 0;
951 char *save_messages_locale = NULL;
952
953 /* Like above for LC_CTYPE, we set LC_MESSAGES to the locale of the
954 * desired category, if it isn't that locale already */
955
956 if (category != LC_MESSAGES) {
957
958 save_messages_locale = stdize_locale(savepv(setlocale(LC_MESSAGES,
959 NULL)));
960 if (! save_messages_locale) {
961 goto cant_use_messages;
962 }
963
964 if (strEQ(save_messages_locale, save_input_locale)) {
965 Safefree(save_input_locale);
966 }
967 else if (! setlocale(LC_MESSAGES, save_input_locale)) {
968 Safefree(save_messages_locale);
969 goto cant_use_messages;
970 }
971 }
972
973 /* Here the current LC_MESSAGES is set to the locale of the category
974 * whose information is desired. Look through all the messages */
975
976 for (e = 0;
977#ifdef HAS_SYS_ERRLIST
978 e <= sys_nerr
979#endif
980 ; e++)
981 {
982 const U8* const errmsg = (U8 *) Strerror(e) ;
983 if (!errmsg)
984 break;
985 if (! is_utf8_string(errmsg, 0)) {
986 failures++;
987 break;
988 }
989 else if (! is_ascii_string(errmsg, 0)) {
990 non_ascii++;
991 }
992 }
993
994 /* And, if we changed it, restore LC_MESSAGES to its original locale */
995 if (save_messages_locale) {
996 setlocale(LC_MESSAGES, save_messages_locale);
997 Safefree(save_messages_locale);
998 }
999
1000 /* Any non-UTF-8 message means not a UTF-8 locale; if all are valid,
1001 * any non-ascii means it is one; otherwise we assume it isn't */
1002 return (failures) ? FALSE : non_ascii;
1003
1004 }
1005 cant_use_messages:
1006
1007#endif
1008
1009 Safefree(save_input_locale);
7d74bb61
KW
1010 return FALSE;
1011}
1012
8ef6e574 1013#endif
7d74bb61 1014
66610fdd
RGS
1015/*
1016 * Local variables:
1017 * c-indentation-style: bsd
1018 * c-basic-offset: 4
14d04a33 1019 * indent-tabs-mode: nil
66610fdd
RGS
1020 * End:
1021 *
14d04a33 1022 * ex: set ts=8 sts=4 sw=4 et:
37442d52 1023 */