3 * Copyright (C) 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001,
4 * 2002, 2003, 2005, 2006, 2007, 2008 by Larry Wall and others
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.
12 * A Elbereth Gilthoniel,
13 * silivren penna míriel
14 * o menel aglar elenath!
15 * Na-chaered palan-díriel
16 * o galadhremmin ennorath,
17 * Fanuilos, le linnathon
18 * nef aear, si nef aearon!
20 * [p.238 of _The Lord of the Rings_, II/i: "Many Meetings"]
23 /* utility functions for handling locale-specific stuff like what
24 * character represents the decimal point.
26 * All C programs have an underlying locale. Perl code generally doesn't pay
27 * any 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. And, LC_MESSAGES is
33 * switched to the C locale for outputting the message unless within the scope
36 * This code now has multi-thread-safe locale handling on systems that support
37 * that. This is completely transparent to most XS code. On earlier systems,
38 * it would be possible to emulate thread-safe locales, but this likely would
39 * involve a lot of locale switching, and would require XS code changes.
40 * Macros could be written so that the code wouldn't have to know which type of
41 * system is being used. It's unlikely that we would ever do that, since most
42 * modern systems support thread-safe locales, but there was code written to
43 * this end, and is retained, #ifdef'd out.
47 #define PERL_IN_LOCALE_C
48 #include "perl_langinfo.h"
60 /* If the environment says to, we can output debugging information during
61 * initialization. This is done before option parsing, and before any thread
62 * creation, so can be a file-level static */
63 #if ! defined(DEBUGGING)
64 # define debug_initialization 0
65 # define DEBUG_INITIALIZATION_set(v)
67 static bool debug_initialization = FALSE;
68 # define DEBUG_INITIALIZATION_set(v) (debug_initialization = v)
72 /* Returns the Unix errno portion; ignoring any others. This is a macro here
73 * instead of putting it into perl.h, because unclear to khw what should be
75 #define GET_ERRNO saved_errno
77 /* strlen() of a literal string constant. We might want this more general,
78 * but using it in just this file for now. A problem with more generality is
79 * the compiler warnings about comparing unlike signs */
80 #define STRLENs(s) (sizeof("" s "") - 1)
82 /* Is the C string input 'name' "C" or "POSIX"? If so, and 'name' is the
83 * return of setlocale(), then this is extremely likely to be the C or POSIX
84 * locale. However, the output of setlocale() is documented to be opaque, but
85 * the odds are extremely small that it would return these two strings for some
86 * other locale. Note that VMS in these two locales includes many non-ASCII
87 * characters as controls and punctuation (below are hex bytes):
89 * punct: A1-A3 A5 A7-AB B0-B3 B5-B7 B9-BD BF-CF D1-DD DF-EF F1-FD
90 * Oddly, none there are listed as alphas, though some represent alphabetics
91 * http://www.nntp.perl.org/group/perl.perl5.porters/2013/02/msg198753.html */
92 #define isNAME_C_OR_POSIX(name) \
94 && (( *(name) == 'C' && (*(name + 1)) == '\0') \
95 || strEQ((name), "POSIX")))
99 /* This code keeps a LRU cache of the UTF-8ness of the locales it has so-far
100 * looked up. This is in the form of a C string: */
102 #define UTF8NESS_SEP "\v"
103 #define UTF8NESS_PREFIX "\f"
105 /* So, the string looks like:
107 * \vC\a0\vPOSIX\a0\vam_ET\a0\vaf_ZA.utf8\a1\ven_US.UTF-8\a1\0
109 * where the digit 0 after the \a indicates that the locale starting just
110 * after the preceding \v is not UTF-8, and the digit 1 mean it is. */
112 STATIC_ASSERT_DECL(STRLENs(UTF8NESS_SEP) == 1);
113 STATIC_ASSERT_DECL(STRLENs(UTF8NESS_PREFIX) == 1);
115 #define C_and_POSIX_utf8ness UTF8NESS_SEP "C" UTF8NESS_PREFIX "0" \
116 UTF8NESS_SEP "POSIX" UTF8NESS_PREFIX "0"
118 /* The cache is initialized to C_and_POSIX_utf8ness at start up. These are
119 * kept there always. The remining portion of the cache is LRU, with the
120 * oldest looked-up locale at the tail end */
123 S_stdize_locale(pTHX_ char *locs)
125 /* Standardize the locale name from a string returned by 'setlocale',
126 * possibly modifying that string.
128 * The typical return value of setlocale() is either
129 * (1) "xx_YY" if the first argument of setlocale() is not LC_ALL
130 * (2) "xa_YY xb_YY ..." if the first argument of setlocale() is LC_ALL
131 * (the space-separated values represent the various sublocales,
132 * in some unspecified order). This is not handled by this function.
134 * In some platforms it has a form like "LC_SOMETHING=Lang_Country.866\n",
135 * which is harmful for further use of the string in setlocale(). This
136 * function removes the trailing new line and everything up through the '='
139 const char * const s = strchr(locs, '=');
142 PERL_ARGS_ASSERT_STDIZE_LOCALE;
145 const char * const t = strchr(s, '.');
148 const char * const u = strchr(t, '\n');
149 if (u && (u[1] == 0)) {
150 const STRLEN len = u - s;
151 Move(s + 1, locs, len, char);
159 Perl_croak(aTHX_ "Can't fix broken locale name \"%s\"", locs);
164 /* Two parallel arrays; first the locale categories Perl uses on this system;
165 * the second array is their names. These arrays are in mostly arbitrary
168 const int categories[] = {
170 # ifdef USE_LOCALE_NUMERIC
173 # ifdef USE_LOCALE_CTYPE
176 # ifdef USE_LOCALE_COLLATE
179 # ifdef USE_LOCALE_TIME
182 # ifdef USE_LOCALE_MESSAGES
185 # ifdef USE_LOCALE_MONETARY
188 # ifdef USE_LOCALE_ADDRESS
191 # ifdef USE_LOCALE_IDENTIFICATION
194 # ifdef USE_LOCALE_MEASUREMENT
197 # ifdef USE_LOCALE_PAPER
200 # ifdef USE_LOCALE_TELEPHONE
203 # ifdef USE_LOCALE_SYNTAX
206 # ifdef USE_LOCALE_TOD
212 -1 /* Placeholder because C doesn't allow a
213 trailing comma, and it would get complicated
214 with all the #ifdef's */
217 /* The top-most real element is LC_ALL */
219 const char * const category_names[] = {
221 # ifdef USE_LOCALE_NUMERIC
224 # ifdef USE_LOCALE_CTYPE
227 # ifdef USE_LOCALE_COLLATE
230 # ifdef USE_LOCALE_TIME
233 # ifdef USE_LOCALE_MESSAGES
236 # ifdef USE_LOCALE_MONETARY
239 # ifdef USE_LOCALE_ADDRESS
242 # ifdef USE_LOCALE_IDENTIFICATION
245 # ifdef USE_LOCALE_MEASUREMENT
248 # ifdef USE_LOCALE_PAPER
251 # ifdef USE_LOCALE_TELEPHONE
254 # ifdef USE_LOCALE_SYNTAX
257 # ifdef USE_LOCALE_TOD
263 NULL /* Placeholder */
268 /* On systems with LC_ALL, it is kept in the highest index position. (-2
269 * to account for the final unused placeholder element.) */
270 # define NOMINAL_LC_ALL_INDEX (C_ARRAY_LENGTH(categories) - 2)
274 /* On systems without LC_ALL, we pretend it is there, one beyond the real
275 * top element, hence in the unused placeholder element. */
276 # define NOMINAL_LC_ALL_INDEX (C_ARRAY_LENGTH(categories) - 1)
280 /* Pretending there is an LC_ALL element just above allows us to avoid most
281 * special cases. Most loops through these arrays in the code below are
282 * written like 'for (i = 0; i < NOMINAL_LC_ALL_INDEX; i++)'. They will work
283 * on either type of system. But the code must be written to not access the
284 * element at 'LC_ALL_INDEX' except on platforms that have it. This can be
285 * checked for at compile time by using the #define LC_ALL_INDEX which is only
286 * defined if we do have LC_ALL. */
289 S_category_name(const int category)
295 if (category == LC_ALL) {
301 for (i = 0; i < NOMINAL_LC_ALL_INDEX; i++) {
302 if (category == categories[i]) {
303 return category_names[i];
308 const char suffix[] = " (unknown)";
310 Size_t length = sizeof(suffix) + 1;
319 /* Calculate the number of digits */
325 Newx(unknown, length, char);
326 my_snprintf(unknown, length, "%d%s", category, suffix);
332 /* Now create LC_foo_INDEX #defines for just those categories on this system */
333 # ifdef USE_LOCALE_NUMERIC
334 # define LC_NUMERIC_INDEX 0
335 # define _DUMMY_NUMERIC LC_NUMERIC_INDEX
337 # define _DUMMY_NUMERIC -1
339 # ifdef USE_LOCALE_CTYPE
340 # define LC_CTYPE_INDEX _DUMMY_NUMERIC + 1
341 # define _DUMMY_CTYPE LC_CTYPE_INDEX
343 # define _DUMMY_CTYPE _DUMMY_NUMERIC
345 # ifdef USE_LOCALE_COLLATE
346 # define LC_COLLATE_INDEX _DUMMY_CTYPE + 1
347 # define _DUMMY_COLLATE LC_COLLATE_INDEX
349 # define _DUMMY_COLLATE _DUMMY_CTYPE
351 # ifdef USE_LOCALE_TIME
352 # define LC_TIME_INDEX _DUMMY_COLLATE + 1
353 # define _DUMMY_TIME LC_TIME_INDEX
355 # define _DUMMY_TIME _DUMMY_COLLATE
357 # ifdef USE_LOCALE_MESSAGES
358 # define LC_MESSAGES_INDEX _DUMMY_TIME + 1
359 # define _DUMMY_MESSAGES LC_MESSAGES_INDEX
361 # define _DUMMY_MESSAGES _DUMMY_TIME
363 # ifdef USE_LOCALE_MONETARY
364 # define LC_MONETARY_INDEX _DUMMY_MESSAGES + 1
365 # define _DUMMY_MONETARY LC_MONETARY_INDEX
367 # define _DUMMY_MONETARY _DUMMY_MESSAGES
369 # ifdef USE_LOCALE_ADDRESS
370 # define LC_ADDRESS_INDEX _DUMMY_MONETARY + 1
371 # define _DUMMY_ADDRESS LC_ADDRESS_INDEX
373 # define _DUMMY_ADDRESS _DUMMY_MONETARY
375 # ifdef USE_LOCALE_IDENTIFICATION
376 # define LC_IDENTIFICATION_INDEX _DUMMY_ADDRESS + 1
377 # define _DUMMY_IDENTIFICATION LC_IDENTIFICATION_INDEX
379 # define _DUMMY_IDENTIFICATION _DUMMY_ADDRESS
381 # ifdef USE_LOCALE_MEASUREMENT
382 # define LC_MEASUREMENT_INDEX _DUMMY_IDENTIFICATION + 1
383 # define _DUMMY_MEASUREMENT LC_MEASUREMENT_INDEX
385 # define _DUMMY_MEASUREMENT _DUMMY_IDENTIFICATION
387 # ifdef USE_LOCALE_PAPER
388 # define LC_PAPER_INDEX _DUMMY_MEASUREMENT + 1
389 # define _DUMMY_PAPER LC_PAPER_INDEX
391 # define _DUMMY_PAPER _DUMMY_MEASUREMENT
393 # ifdef USE_LOCALE_TELEPHONE
394 # define LC_TELEPHONE_INDEX _DUMMY_PAPER + 1
395 # define _DUMMY_TELEPHONE LC_TELEPHONE_INDEX
397 # define _DUMMY_TELEPHONE _DUMMY_PAPER
399 # ifdef USE_LOCALE_SYNTAX
400 # define LC_SYNTAX_INDEX _DUMMY_TELEPHONE + 1
401 # define _DUMMY_SYNTAX LC_SYNTAX_INDEX
403 # define _DUMMY_SYNTAX _DUMMY_TELEPHONE
405 # ifdef USE_LOCALE_TOD
406 # define LC_TOD_INDEX _DUMMY_SYNTAX + 1
407 # define _DUMMY_TOD LC_TOD_INDEX
409 # define _DUMMY_TOD _DUMMY_SYNTAX
412 # define LC_ALL_INDEX _DUMMY_TOD + 1
414 #endif /* ifdef USE_LOCALE */
416 /* Windows requres a customized base-level setlocale() */
418 # define my_setlocale(cat, locale) win32_setlocale(cat, locale)
420 # define my_setlocale(cat, locale) setlocale(cat, locale)
423 #ifndef USE_POSIX_2008_LOCALE
425 /* "do_setlocale_c" is intended to be called when the category is a constant
426 * known at compile time; "do_setlocale_r", not known until run time */
427 # define do_setlocale_c(cat, locale) my_setlocale(cat, locale)
428 # define do_setlocale_r(cat, locale) my_setlocale(cat, locale)
429 # define FIX_GLIBC_LC_MESSAGES_BUG(i)
431 #else /* Below uses POSIX 2008 */
433 /* We emulate setlocale with our own function. LC_foo is not valid for the
434 * POSIX 2008 functions. Instead LC_foo_MASK is used, which we use an array
435 * lookup to convert to. At compile time we have defined LC_foo_INDEX as the
436 * proper offset into the array 'category_masks[]'. At runtime, we have to
437 * search through the array (as the actual numbers may not be small contiguous
438 * positive integers which would lend themselves to array lookup). */
439 # define do_setlocale_c(cat, locale) \
440 emulate_setlocale(cat, locale, cat ## _INDEX, TRUE)
441 # define do_setlocale_r(cat, locale) emulate_setlocale(cat, locale, 0, FALSE)
443 # if ! defined(__GLIBC__) || ! defined(USE_LOCALE_MESSAGES)
445 # define FIX_GLIBC_LC_MESSAGES_BUG(i)
447 # else /* Invalidate glibc cache of loaded translations, see [perl #134264] */
449 # include <libintl.h>
450 # define FIX_GLIBC_LC_MESSAGES_BUG(i) \
452 if ((i) == LC_MESSAGES_INDEX) { \
453 textdomain(textdomain(NULL)); \
459 /* A third array, parallel to the ones above to map from category to its
461 const int category_masks[] = {
462 # ifdef USE_LOCALE_NUMERIC
465 # ifdef USE_LOCALE_CTYPE
468 # ifdef USE_LOCALE_COLLATE
471 # ifdef USE_LOCALE_TIME
474 # ifdef USE_LOCALE_MESSAGES
477 # ifdef USE_LOCALE_MONETARY
480 # ifdef USE_LOCALE_ADDRESS
483 # ifdef USE_LOCALE_IDENTIFICATION
484 LC_IDENTIFICATION_MASK,
486 # ifdef USE_LOCALE_MEASUREMENT
489 # ifdef USE_LOCALE_PAPER
492 # ifdef USE_LOCALE_TELEPHONE
495 # ifdef USE_LOCALE_SYNTAX
498 # ifdef USE_LOCALE_TOD
501 /* LC_ALL can't be turned off by a Configure
502 * option, and in Posix 2008, should always be
503 * here, so compile it in unconditionally.
504 * This could catch some glitches at compile
510 S_emulate_setlocale(const int category,
513 const bool is_index_valid
516 /* This function effectively performs a setlocale() on just the current
517 * thread; thus it is thread-safe. It does this by using the POSIX 2008
518 * locale functions to emulate the behavior of setlocale(). Similar to
519 * regular setlocale(), the return from this function points to memory that
520 * can be overwritten by other system calls, so needs to be copied
521 * immediately if you need to retain it. The difference here is that
522 * system calls besides another setlocale() can overwrite it.
524 * By doing this, most locale-sensitive functions become thread-safe. The
525 * exceptions are mostly those that return a pointer to static memory.
527 * This function takes the same parameters, 'category' and 'locale', that
528 * the regular setlocale() function does, but it also takes two additional
529 * ones. This is because the 2008 functions don't use a category; instead
530 * they use a corresponding mask. Because this function operates in both
531 * worlds, it may need one or the other or both. This function can
532 * calculate the mask from the input category, but to avoid this
533 * calculation, if the caller knows at compile time what the mask is, it
534 * can pass it, setting 'is_index_valid' to TRUE; otherwise the mask
535 * parameter is ignored.
537 * POSIX 2008, for some sick reason, chose not to provide a method to find
538 * the category name of a locale. Some vendors have created a
539 * querylocale() function to do just that. This function is a lot simpler
540 * to implement on systems that have this. Otherwise, we have to keep
541 * track of what the locale has been set to, so that we can return its
542 * name to emulate setlocale(). It's also possible for C code in some
543 * library to change the locale without us knowing it, though as of
544 * September 2017, there are no occurrences in CPAN of uselocale(). Some
545 * libraries do use setlocale(), but that changes the global locale, and
546 * threads using per-thread locales will just ignore those changes.
547 * Another problem is that without querylocale(), we have to guess at what
548 * was meant by setting a locale of "". We handle this by not actually
549 * ever setting to "" (unless querylocale exists), but to emulate what we
550 * think should happen for "".
556 const char * safelocale = locale ? locale : "(null)";
561 if (DEBUG_Lv_TEST || debug_initialization) {
562 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale input=%d (%s), \"%s\", %d, %d\n", __FILE__, __LINE__, category, category_name(category), safelocale, index, is_index_valid);
567 /* If the input mask might be incorrect, calculate the correct one */
568 if (! is_index_valid) {
573 if (DEBUG_Lv_TEST || debug_initialization) {
574 PerlIO_printf(Perl_debug_log, "%s:%d: finding index of category %d (%s)\n", __FILE__, __LINE__, category, category_name(category));
579 for (i = 0; i <= LC_ALL_INDEX; i++) {
580 if (category == categories[i]) {
586 /* Here, we don't know about this category, so can't handle it.
587 * Fallback to the early POSIX usages */
588 Perl_warner(aTHX_ packWARN(WARN_LOCALE),
589 "Unknown locale category %d; can't set it to %s\n",
590 category, safelocale);
597 if (DEBUG_Lv_TEST || debug_initialization) {
598 PerlIO_printf(Perl_debug_log, "%s:%d: index is %d for %s\n", __FILE__, __LINE__, index, category_name(category));
605 mask = category_masks[index];
609 if (DEBUG_Lv_TEST || debug_initialization) {
610 PerlIO_printf(Perl_debug_log, "%s:%d: category name is %s; mask is 0x%x\n", __FILE__, __LINE__, category_names[index], mask);
615 /* If just querying what the existing locale is ... */
616 if (locale == NULL) {
617 locale_t cur_obj = uselocale((locale_t) 0);
621 if (DEBUG_Lv_TEST || debug_initialization) {
622 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale querying %p\n", __FILE__, __LINE__, cur_obj);
627 if (cur_obj == LC_GLOBAL_LOCALE) {
628 return my_setlocale(category, NULL);
631 # ifdef HAS_QUERYLOCALE
633 return (char *) querylocale(mask, cur_obj);
637 /* If this assert fails, adjust the size of curlocales in intrpvar.h */
638 STATIC_ASSERT_STMT(C_ARRAY_LENGTH(PL_curlocales) > LC_ALL_INDEX);
640 # if defined(_NL_LOCALE_NAME) \
641 && defined(DEBUGGING) \
642 /* On systems that accept any locale name, the real underlying \
643 * locale is often returned by this internal function, so we \
645 && ! defined(SETLOCALE_ACCEPTS_ANY_LOCALE_NAME)
647 /* Internal glibc for querylocale(), but doesn't handle
648 * empty-string ("") locale properly; who knows what other
649 * glitches. Check for it now, under debug. */
651 char * temp_name = nl_langinfo_l(_NL_LOCALE_NAME(category),
652 uselocale((locale_t) 0));
654 PerlIO_printf(Perl_debug_log, "%s:%d: temp_name=%s\n", __FILE__, __LINE__, temp_name ? temp_name : "NULL");
655 PerlIO_printf(Perl_debug_log, "%s:%d: index=%d\n", __FILE__, __LINE__, index);
656 PerlIO_printf(Perl_debug_log, "%s:%d: PL_curlocales[index]=%s\n", __FILE__, __LINE__, PL_curlocales[index]);
658 if (temp_name && PL_curlocales[index] && strNE(temp_name, "")) {
659 if ( strNE(PL_curlocales[index], temp_name)
660 && ! ( isNAME_C_OR_POSIX(temp_name)
661 && isNAME_C_OR_POSIX(PL_curlocales[index]))) {
663 # ifdef USE_C_BACKTRACE
665 dump_c_backtrace(Perl_debug_log, 20, 1);
669 Perl_croak(aTHX_ "panic: Mismatch between what Perl thinks %s is"
670 " (%s) and what internal glibc thinks"
671 " (%s)\n", category_names[index],
672 PL_curlocales[index], temp_name);
681 /* Without querylocale(), we have to use our record-keeping we've
684 if (category != LC_ALL) {
688 if (DEBUG_Lv_TEST || debug_initialization) {
689 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale returning %s\n", __FILE__, __LINE__, PL_curlocales[index]);
694 return PL_curlocales[index];
696 else { /* For LC_ALL */
698 Size_t names_len = 0;
700 bool are_all_categories_the_same_locale = TRUE;
702 /* If we have a valid LC_ALL value, just return it */
703 if (PL_curlocales[LC_ALL_INDEX]) {
707 if (DEBUG_Lv_TEST || debug_initialization) {
708 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale returning %s\n", __FILE__, __LINE__, PL_curlocales[LC_ALL_INDEX]);
713 return PL_curlocales[LC_ALL_INDEX];
716 /* Otherwise, we need to construct a string of name=value pairs.
717 * We use the glibc syntax, like
718 * LC_NUMERIC=C;LC_TIME=en_US.UTF-8;...
719 * First calculate the needed size. Along the way, check if all
720 * the locale names are the same */
721 for (i = 0; i < LC_ALL_INDEX; i++) {
725 if (DEBUG_Lv_TEST || debug_initialization) {
726 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale i=%d, name=%s, locale=%s\n", __FILE__, __LINE__, i, category_names[i], PL_curlocales[i]);
731 names_len += strlen(category_names[i])
733 + strlen(PL_curlocales[i])
736 if (i > 0 && strNE(PL_curlocales[i], PL_curlocales[i-1])) {
737 are_all_categories_the_same_locale = FALSE;
741 /* If they are the same, we don't actually have to construct the
742 * string; we just make the entry in LC_ALL_INDEX valid, and be
743 * that single name */
744 if (are_all_categories_the_same_locale) {
745 PL_curlocales[LC_ALL_INDEX] = savepv(PL_curlocales[0]);
746 return PL_curlocales[LC_ALL_INDEX];
749 names_len++; /* Trailing '\0' */
750 SAVEFREEPV(Newx(all_string, names_len, char));
753 /* Then fill in the string */
754 for (i = 0; i < LC_ALL_INDEX; i++) {
758 if (DEBUG_Lv_TEST || debug_initialization) {
759 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale i=%d, name=%s, locale=%s\n", __FILE__, __LINE__, i, category_names[i], PL_curlocales[i]);
764 my_strlcat(all_string, category_names[i], names_len);
765 my_strlcat(all_string, "=", names_len);
766 my_strlcat(all_string, PL_curlocales[i], names_len);
767 my_strlcat(all_string, ";", names_len);
772 if (DEBUG_L_TEST || debug_initialization) {
773 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale returning %s\n", __FILE__, __LINE__, all_string);
783 SETERRNO(EINVAL, LIB_INVARG);
791 } /* End of this being setlocale(LC_foo, NULL) */
793 /* Here, we are switching locales. */
795 # ifndef HAS_QUERYLOCALE
797 if (strEQ(locale, "")) {
799 /* For non-querylocale() systems, we do the setting of "" ourselves to
800 * be sure that we really know what's going on. We follow the Linux
801 * documented behavior (but if that differs from the actual behavior,
802 * this won't work exactly as the OS implements). We go out and
803 * examine the environment based on our understanding of how the system
804 * works, and use that to figure things out */
806 const char * const lc_all = PerlEnv_getenv("LC_ALL");
808 /* Use any "LC_ALL" environment variable, as it overrides everything
810 if (lc_all && strNE(lc_all, "")) {
815 /* Otherwise, we need to dig deeper. Unless overridden, the
816 * default is the LANG environment variable; if it doesn't exist,
819 const char * default_name;
821 default_name = PerlEnv_getenv("LANG");
823 if (! default_name || strEQ(default_name, "")) {
827 if (category != LC_ALL) {
828 const char * const name = PerlEnv_getenv(category_names[index]);
830 /* Here we are setting a single category. Assume will have the
832 locale = default_name;
834 /* But then look for an overriding environment variable */
835 if (name && strNE(name, "")) {
840 bool did_override = FALSE;
843 /* Here, we are getting LC_ALL. Any categories that don't have
844 * a corresponding environment variable set should be set to
845 * LANG, or to "C" if there is no LANG. If no individual
846 * categories differ from this, we can just set LC_ALL. This
847 * is buggy on systems that have extra categories that we don't
848 * know about. If there is an environment variable that sets
849 * that category, we won't know to look for it, and so our use
850 * of LANG or "C" improperly overrides it. On the other hand,
851 * if we don't do what is done here, and there is no
852 * environment variable, the category's locale should be set to
853 * LANG or "C". So there is no good solution. khw thinks the
854 * best is to look at systems to see what categories they have,
855 * and include them, and then to assume that we know the
858 for (i = 0; i < LC_ALL_INDEX; i++) {
859 const char * const env_override
860 = PerlEnv_getenv(category_names[i]);
861 const char * this_locale = ( env_override
862 && strNE(env_override, ""))
865 if (! emulate_setlocale(categories[i], this_locale, i, TRUE))
870 if (strNE(this_locale, default_name)) {
875 /* If all the categories are the same, we can set LC_ALL to
877 if (! did_override) {
878 locale = default_name;
882 /* Here, LC_ALL is no longer valid, as some individual
883 * categories don't match it. We call ourselves
884 * recursively, as that will execute the code that
885 * generates the proper locale string for this situation.
886 * We don't do the remainder of this function, as that is
887 * to update our records, and we've just done that for the
888 * individual categories in the loop above, and doing so
889 * would cause LC_ALL to be done as well */
890 return emulate_setlocale(LC_ALL, NULL, LC_ALL_INDEX, TRUE);
894 } /* End of this being setlocale(LC_foo, "") */
895 else if (strchr(locale, ';')) {
897 /* LC_ALL may actually incude a conglomeration of various categories.
898 * Without querylocale, this code uses the glibc (as of this writing)
899 * syntax for representing that, but that is not a stable API, and
900 * other platforms do it differently, so we have to handle all cases
904 const char * s = locale;
905 const char * e = locale + strlen(locale);
907 const char * category_end;
908 const char * name_start;
909 const char * name_end;
911 /* If the string that gives what to set doesn't include all categories,
912 * the omitted ones get set to "C". To get this behavior, first set
913 * all the individual categories to "C", and override the furnished
915 for (i = 0; i < LC_ALL_INDEX; i++) {
916 if (! emulate_setlocale(categories[i], "C", i, TRUE)) {
923 /* Parse through the category */
924 while (isWORDCHAR(*p)) {
931 "panic: %s: %d: Unexpected character in locale name '%02X",
932 __FILE__, __LINE__, *(p-1));
935 /* Parse through the locale name */
937 while (p < e && *p != ';') {
940 "panic: %s: %d: Unexpected character in locale name '%02X",
941 __FILE__, __LINE__, *(p-1));
947 /* Space past the semi-colon */
952 /* Find the index of the category name in our lists */
953 for (i = 0; i < LC_ALL_INDEX; i++) {
954 char * individ_locale;
956 /* Keep going if this isn't the index. The strnNE() avoids a
957 * Perl_form(), but would fail if ever a category name could be
958 * a substring of another one, like if there were a
960 if strnNE(s, category_names[i], category_end - s) {
964 /* If this index is for the single category we're changing, we
965 * have found the locale to set it to. */
966 if (category == categories[i]) {
967 locale = Perl_form(aTHX_ "%.*s",
968 (int) (name_end - name_start),
973 assert(category == LC_ALL);
974 individ_locale = Perl_form(aTHX_ "%.*s",
975 (int) (name_end - name_start), name_start);
976 if (! emulate_setlocale(categories[i], individ_locale, i, TRUE))
985 /* Here we have set all the individual categories by recursive calls.
986 * These collectively should have fixed up LC_ALL, so can just query
987 * what that now is */
988 assert(category == LC_ALL);
990 return do_setlocale_c(LC_ALL, NULL);
991 } /* End of this being setlocale(LC_ALL,
992 "LC_CTYPE=foo;LC_NUMERIC=bar;...") */
996 /* Here at the end of having to deal with the absence of querylocale().
997 * Some cases have already been fully handled by recursive calls to this
998 * function. But at this point, we haven't dealt with those, but are now
999 * prepared to, knowing what the locale name to set this category to is.
1000 * This would have come for free if this system had had querylocale() */
1002 # endif /* end of ! querylocale */
1004 assert(PL_C_locale_obj);
1006 /* Switching locales generally entails freeing the current one's space (at
1007 * the C library's discretion). We need to stop using that locale before
1008 * the switch. So switch to a known locale object that we don't otherwise
1009 * mess with. This returns the locale object in effect at the time of the
1011 old_obj = uselocale(PL_C_locale_obj);
1015 if (DEBUG_Lv_TEST || debug_initialization) {
1016 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale was using %p\n", __FILE__, __LINE__, old_obj);
1025 if (DEBUG_L_TEST || debug_initialization) {
1027 PerlIO_printf(Perl_debug_log, "%s:%d: emulate_setlocale switching to C failed: %d\n", __FILE__, __LINE__, GET_ERRNO);
1038 if (DEBUG_Lv_TEST || debug_initialization) {
1039 PerlIO_printf(Perl_debug_log,
1040 "%s:%d: emulate_setlocale now using %p\n",
1041 __FILE__, __LINE__, PL_C_locale_obj);
1046 /* If this call is to switch to the LC_ALL C locale, it already exists, and
1047 * in fact, we already have switched to it (in preparation for what
1048 * normally is to come). But since we're already there, continue to use
1049 * it instead of trying to create a new locale */
1050 if (mask == LC_ALL_MASK && isNAME_C_OR_POSIX(locale)) {
1054 if (DEBUG_Lv_TEST || debug_initialization) {
1055 PerlIO_printf(Perl_debug_log,
1056 "%s:%d: will stay in C object\n", __FILE__, __LINE__);
1061 new_obj = PL_C_locale_obj;
1063 /* We already had switched to the C locale in preparation for freeing
1065 if (old_obj != LC_GLOBAL_LOCALE && old_obj != PL_C_locale_obj) {
1066 freelocale(old_obj);
1070 /* If we weren't in a thread safe locale, set so that newlocale() below
1071 * which uses 'old_obj', uses an empty one. Same for our reserved C
1072 * object. The latter is defensive coding, so that, even if there is
1073 * some bug, we will never end up trying to modify either of these, as
1074 * if passed to newlocale(), they can be. */
1075 if (old_obj == LC_GLOBAL_LOCALE || old_obj == PL_C_locale_obj) {
1076 old_obj = (locale_t) 0;
1079 /* Ready to create a new locale by modification of the exising one */
1080 new_obj = newlocale(mask, locale, old_obj);
1087 if (DEBUG_L_TEST || debug_initialization) {
1088 PerlIO_printf(Perl_debug_log,
1089 "%s:%d: emulate_setlocale creating new object"
1090 " failed: %d\n", __FILE__, __LINE__, GET_ERRNO);
1095 if (! uselocale(old_obj)) {
1099 if (DEBUG_L_TEST || debug_initialization) {
1100 PerlIO_printf(Perl_debug_log,
1101 "%s:%d: switching back failed: %d\n",
1102 __FILE__, __LINE__, GET_ERRNO);
1114 if (DEBUG_Lv_TEST || debug_initialization) {
1115 PerlIO_printf(Perl_debug_log,
1116 "%s:%d: emulate_setlocale created %p",
1117 __FILE__, __LINE__, new_obj);
1119 PerlIO_printf(Perl_debug_log,
1120 "; should have freed %p", old_obj);
1122 PerlIO_printf(Perl_debug_log, "\n");
1127 /* And switch into it */
1128 if (! uselocale(new_obj)) {
1133 if (DEBUG_L_TEST || debug_initialization) {
1134 PerlIO_printf(Perl_debug_log,
1135 "%s:%d: emulate_setlocale switching to new object"
1136 " failed\n", __FILE__, __LINE__);
1141 if (! uselocale(old_obj)) {
1145 if (DEBUG_L_TEST || debug_initialization) {
1146 PerlIO_printf(Perl_debug_log,
1147 "%s:%d: switching back failed: %d\n",
1148 __FILE__, __LINE__, GET_ERRNO);
1154 freelocale(new_obj);
1162 if (DEBUG_Lv_TEST || debug_initialization) {
1163 PerlIO_printf(Perl_debug_log,
1164 "%s:%d: emulate_setlocale now using %p\n",
1165 __FILE__, __LINE__, new_obj);
1170 /* We are done, except for updating our records (if the system doesn't keep
1171 * them) and in the case of locale "", we don't actually know what the
1172 * locale that got switched to is, as it came from the environment. So
1173 * have to find it */
1175 # ifdef HAS_QUERYLOCALE
1177 if (strEQ(locale, "")) {
1178 locale = querylocale(mask, new_obj);
1183 /* Here, 'locale' is the return value */
1185 /* Without querylocale(), we have to update our records */
1187 if (category == LC_ALL) {
1190 /* For LC_ALL, we change all individual categories to correspond */
1191 /* PL_curlocales is a parallel array, so has same
1192 * length as 'categories' */
1193 for (i = 0; i <= LC_ALL_INDEX; i++) {
1194 Safefree(PL_curlocales[i]);
1195 PL_curlocales[i] = savepv(locale);
1198 FIX_GLIBC_LC_MESSAGES_BUG(LC_MESSAGES_INDEX);
1202 /* For a single category, if it's not the same as the one in LC_ALL, we
1205 if (PL_curlocales[LC_ALL_INDEX] && strNE(PL_curlocales[LC_ALL_INDEX], locale)) {
1206 Safefree(PL_curlocales[LC_ALL_INDEX]);
1207 PL_curlocales[LC_ALL_INDEX] = NULL;
1210 /* Then update the category's record */
1211 Safefree(PL_curlocales[index]);
1212 PL_curlocales[index] = savepv(locale);
1214 FIX_GLIBC_LC_MESSAGES_BUG(index);
1222 #endif /* USE_POSIX_2008_LOCALE */
1227 S_set_numeric_radix(pTHX_ const bool use_locale)
1229 /* If 'use_locale' is FALSE, set to use a dot for the radix character. If
1230 * TRUE, use the radix character derived from the current locale */
1232 #if defined(USE_LOCALE_NUMERIC) && ( defined(HAS_LOCALECONV) \
1233 || defined(HAS_NL_LANGINFO))
1235 const char * radix = (use_locale)
1236 ? my_nl_langinfo(RADIXCHAR, FALSE)
1237 /* FALSE => already in dest locale */
1240 sv_setpv(PL_numeric_radix_sv, radix);
1242 /* If this is valid UTF-8 that isn't totally ASCII, and we are in
1243 * a UTF-8 locale, then mark the radix as being in UTF-8 */
1244 if (is_utf8_non_invariant_string((U8 *) SvPVX(PL_numeric_radix_sv),
1245 SvCUR(PL_numeric_radix_sv))
1246 && _is_cur_LC_category_utf8(LC_NUMERIC))
1248 SvUTF8_on(PL_numeric_radix_sv);
1253 if (DEBUG_L_TEST || debug_initialization) {
1254 PerlIO_printf(Perl_debug_log, "Locale radix is '%s', ?UTF-8=%d\n",
1255 SvPVX(PL_numeric_radix_sv),
1256 cBOOL(SvUTF8(PL_numeric_radix_sv)));
1262 PERL_UNUSED_ARG(use_locale);
1264 #endif /* USE_LOCALE_NUMERIC and can find the radix char */
1269 S_new_numeric(pTHX_ const char *newnum)
1272 #ifndef USE_LOCALE_NUMERIC
1274 PERL_UNUSED_ARG(newnum);
1278 /* Called after each libc setlocale() call affecting LC_NUMERIC, to tell
1279 * core Perl this and that 'newnum' is the name of the new locale.
1280 * It installs this locale as the current underlying default.
1282 * The default locale and the C locale can be toggled between by use of the
1283 * set_numeric_underlying() and set_numeric_standard() functions, which
1284 * should probably not be called directly, but only via macros like
1285 * SET_NUMERIC_STANDARD() in perl.h.
1287 * The toggling is necessary mainly so that a non-dot radix decimal point
1288 * character can be output, while allowing internal calculations to use a
1291 * This sets several interpreter-level variables:
1292 * PL_numeric_name The underlying locale's name: a copy of 'newnum'
1293 * PL_numeric_underlying A boolean indicating if the toggled state is such
1294 * that the current locale is the program's underlying
1296 * PL_numeric_standard An int indicating if the toggled state is such
1297 * that the current locale is the C locale or
1298 * indistinguishable from the C locale. If non-zero, it
1299 * is in C; if > 1, it means it may not be toggled away
1301 * PL_numeric_underlying_is_standard A bool kept by this function
1302 * indicating that the underlying locale and the standard
1303 * C locale are indistinguishable for the purposes of
1304 * LC_NUMERIC. This happens when both of the above two
1305 * variables are true at the same time. (Toggling is a
1306 * no-op under these circumstances.) This variable is
1307 * used to avoid having to recalculate.
1313 Safefree(PL_numeric_name);
1314 PL_numeric_name = NULL;
1315 PL_numeric_standard = TRUE;
1316 PL_numeric_underlying = TRUE;
1317 PL_numeric_underlying_is_standard = TRUE;
1321 save_newnum = stdize_locale(savepv(newnum));
1322 PL_numeric_underlying = TRUE;
1323 PL_numeric_standard = isNAME_C_OR_POSIX(save_newnum);
1325 #ifndef TS_W32_BROKEN_LOCALECONV
1327 /* If its name isn't C nor POSIX, it could still be indistinguishable from
1328 * them. But on broken Windows systems calling my_nl_langinfo() for
1329 * THOUSEP can currently (but rarely) cause a race, so avoid doing that,
1330 * and just always change the locale if not C nor POSIX on those systems */
1331 if (! PL_numeric_standard) {
1332 PL_numeric_standard = cBOOL(strEQ(".", my_nl_langinfo(RADIXCHAR,
1333 FALSE /* Don't toggle locale */ ))
1334 && strEQ("", my_nl_langinfo(THOUSEP, FALSE)));
1339 /* Save the new name if it isn't the same as the previous one, if any */
1340 if (! PL_numeric_name || strNE(PL_numeric_name, save_newnum)) {
1341 Safefree(PL_numeric_name);
1342 PL_numeric_name = save_newnum;
1345 Safefree(save_newnum);
1348 PL_numeric_underlying_is_standard = PL_numeric_standard;
1350 # ifdef HAS_POSIX_2008_LOCALE
1352 PL_underlying_numeric_obj = newlocale(LC_NUMERIC_MASK,
1354 PL_underlying_numeric_obj);
1358 if (DEBUG_L_TEST || debug_initialization) {
1359 PerlIO_printf(Perl_debug_log, "Called new_numeric with %s, PL_numeric_name=%s\n", newnum, PL_numeric_name);
1362 /* Keep LC_NUMERIC in the C locale. This is for XS modules, so they don't
1363 * have to worry about the radix being a non-dot. (Core operations that
1364 * need the underlying locale change to it temporarily). */
1365 if (PL_numeric_standard) {
1366 set_numeric_radix(0);
1369 set_numeric_standard();
1372 #endif /* USE_LOCALE_NUMERIC */
1377 Perl_set_numeric_standard(pTHX)
1380 #ifdef USE_LOCALE_NUMERIC
1382 /* Toggle the LC_NUMERIC locale to C. Most code should use the macros like
1383 * SET_NUMERIC_STANDARD() in perl.h instead of calling this directly. The
1384 * macro avoids calling this routine if toggling isn't necessary according
1385 * to our records (which could be wrong if some XS code has changed the
1386 * locale behind our back) */
1390 if (DEBUG_L_TEST || debug_initialization) {
1391 PerlIO_printf(Perl_debug_log,
1392 "Setting LC_NUMERIC locale to standard C\n");
1397 do_setlocale_c(LC_NUMERIC, "C");
1398 PL_numeric_standard = TRUE;
1399 PL_numeric_underlying = PL_numeric_underlying_is_standard;
1400 set_numeric_radix(0);
1402 #endif /* USE_LOCALE_NUMERIC */
1407 Perl_set_numeric_underlying(pTHX)
1410 #ifdef USE_LOCALE_NUMERIC
1412 /* Toggle the LC_NUMERIC locale to the current underlying default. Most
1413 * code should use the macros like SET_NUMERIC_UNDERLYING() in perl.h
1414 * instead of calling this directly. The macro avoids calling this routine
1415 * if toggling isn't necessary according to our records (which could be
1416 * wrong if some XS code has changed the locale behind our back) */
1420 if (DEBUG_L_TEST || debug_initialization) {
1421 PerlIO_printf(Perl_debug_log,
1422 "Setting LC_NUMERIC locale to %s\n",
1428 do_setlocale_c(LC_NUMERIC, PL_numeric_name);
1429 PL_numeric_standard = PL_numeric_underlying_is_standard;
1430 PL_numeric_underlying = TRUE;
1431 set_numeric_radix(! PL_numeric_standard);
1433 #endif /* USE_LOCALE_NUMERIC */
1438 * Set up for a new ctype locale.
1441 S_new_ctype(pTHX_ const char *newctype)
1444 #ifndef USE_LOCALE_CTYPE
1446 PERL_UNUSED_ARG(newctype);
1447 PERL_UNUSED_CONTEXT;
1451 /* Called after each libc setlocale() call affecting LC_CTYPE, to tell
1452 * core Perl this and that 'newctype' is the name of the new locale.
1454 * This function sets up the folding arrays for all 256 bytes, assuming
1455 * that tofold() is tolc() since fold case is not a concept in POSIX,
1457 * Any code changing the locale (outside this file) should use
1458 * Perl_setlocale or POSIX::setlocale, which call this function. Therefore
1459 * this function should be called directly only from this file and from
1460 * POSIX::setlocale() */
1464 /* Don't check for problems if we are suppressing the warnings */
1465 bool check_for_problems = ckWARN_d(WARN_LOCALE) || UNLIKELY(DEBUG_L_TEST);
1466 bool maybe_utf8_turkic = FALSE;
1468 PERL_ARGS_ASSERT_NEW_CTYPE;
1470 /* We will replace any bad locale warning with 1) nothing if the new one is
1471 * ok; or 2) a new warning for the bad new locale */
1472 if (PL_warn_locale) {
1473 SvREFCNT_dec_NN(PL_warn_locale);
1474 PL_warn_locale = NULL;
1477 PL_in_utf8_CTYPE_locale = _is_cur_LC_category_utf8(LC_CTYPE);
1479 /* A UTF-8 locale gets standard rules. But note that code still has to
1480 * handle this specially because of the three problematic code points */
1481 if (PL_in_utf8_CTYPE_locale) {
1482 Copy(PL_fold_latin1, PL_fold_locale, 256, U8);
1484 /* UTF-8 locales can have special handling for 'I' and 'i' if they are
1485 * Turkic. Make sure these two are the only anomalies. (We don't use
1486 * towupper and towlower because they aren't in C89.) */
1488 #if defined(HAS_TOWUPPER) && defined (HAS_TOWLOWER)
1490 if (towupper('i') == 0x130 && towlower('I') == 0x131) {
1494 if (toupper('i') == 'i' && tolower('I') == 'I') {
1497 check_for_problems = TRUE;
1498 maybe_utf8_turkic = TRUE;
1502 /* We don't populate the other lists if a UTF-8 locale, but do check that
1503 * everything works as expected, unless checking turned off */
1504 if (check_for_problems || ! PL_in_utf8_CTYPE_locale) {
1505 /* Assume enough space for every character being bad. 4 spaces each
1506 * for the 94 printable characters that are output like "'x' "; and 5
1507 * spaces each for "'\\' ", "'\t' ", and "'\n' "; plus a terminating
1509 char bad_chars_list[ (94 * 4) + (3 * 5) + 1 ] = { '\0' };
1510 bool multi_byte_locale = FALSE; /* Assume is a single-byte locale
1512 unsigned int bad_count = 0; /* Count of bad characters */
1514 for (i = 0; i < 256; i++) {
1515 if (! PL_in_utf8_CTYPE_locale) {
1517 PL_fold_locale[i] = (U8) tolower(i);
1518 else if (islower(i))
1519 PL_fold_locale[i] = (U8) toupper(i);
1521 PL_fold_locale[i] = (U8) i;
1524 /* If checking for locale problems, see if the native ASCII-range
1525 * printables plus \n and \t are in their expected categories in
1526 * the new locale. If not, this could mean big trouble, upending
1527 * Perl's and most programs' assumptions, like having a
1528 * metacharacter with special meaning become a \w. Fortunately,
1529 * it's very rare to find locales that aren't supersets of ASCII
1530 * nowadays. It isn't a problem for most controls to be changed
1531 * into something else; we check only \n and \t, though perhaps \r
1532 * could be an issue as well. */
1533 if ( check_for_problems
1534 && (isGRAPH_A(i) || isBLANK_A(i) || i == '\n'))
1536 bool is_bad = FALSE;
1537 char name[4] = { '\0' };
1539 /* Convert the name into a string */
1544 else if (i == '\n') {
1545 my_strlcpy(name, "\\n", sizeof(name));
1547 else if (i == '\t') {
1548 my_strlcpy(name, "\\t", sizeof(name));
1552 my_strlcpy(name, "' '", sizeof(name));
1555 /* Check each possibe class */
1556 if (UNLIKELY(cBOOL(isalnum(i)) != cBOOL(isALPHANUMERIC_A(i)))) {
1558 DEBUG_L(PerlIO_printf(Perl_debug_log,
1559 "isalnum('%s') unexpectedly is %d\n",
1560 name, cBOOL(isalnum(i))));
1562 if (UNLIKELY(cBOOL(isalpha(i)) != cBOOL(isALPHA_A(i)))) {
1564 DEBUG_L(PerlIO_printf(Perl_debug_log,
1565 "isalpha('%s') unexpectedly is %d\n",
1566 name, cBOOL(isalpha(i))));
1568 if (UNLIKELY(cBOOL(isdigit(i)) != cBOOL(isDIGIT_A(i)))) {
1570 DEBUG_L(PerlIO_printf(Perl_debug_log,
1571 "isdigit('%s') unexpectedly is %d\n",
1572 name, cBOOL(isdigit(i))));
1574 if (UNLIKELY(cBOOL(isgraph(i)) != cBOOL(isGRAPH_A(i)))) {
1576 DEBUG_L(PerlIO_printf(Perl_debug_log,
1577 "isgraph('%s') unexpectedly is %d\n",
1578 name, cBOOL(isgraph(i))));
1580 if (UNLIKELY(cBOOL(islower(i)) != cBOOL(isLOWER_A(i)))) {
1582 DEBUG_L(PerlIO_printf(Perl_debug_log,
1583 "islower('%s') unexpectedly is %d\n",
1584 name, cBOOL(islower(i))));
1586 if (UNLIKELY(cBOOL(isprint(i)) != cBOOL(isPRINT_A(i)))) {
1588 DEBUG_L(PerlIO_printf(Perl_debug_log,
1589 "isprint('%s') unexpectedly is %d\n",
1590 name, cBOOL(isprint(i))));
1592 if (UNLIKELY(cBOOL(ispunct(i)) != cBOOL(isPUNCT_A(i)))) {
1594 DEBUG_L(PerlIO_printf(Perl_debug_log,
1595 "ispunct('%s') unexpectedly is %d\n",
1596 name, cBOOL(ispunct(i))));
1598 if (UNLIKELY(cBOOL(isspace(i)) != cBOOL(isSPACE_A(i)))) {
1600 DEBUG_L(PerlIO_printf(Perl_debug_log,
1601 "isspace('%s') unexpectedly is %d\n",
1602 name, cBOOL(isspace(i))));
1604 if (UNLIKELY(cBOOL(isupper(i)) != cBOOL(isUPPER_A(i)))) {
1606 DEBUG_L(PerlIO_printf(Perl_debug_log,
1607 "isupper('%s') unexpectedly is %d\n",
1608 name, cBOOL(isupper(i))));
1610 if (UNLIKELY(cBOOL(isxdigit(i))!= cBOOL(isXDIGIT_A(i)))) {
1612 DEBUG_L(PerlIO_printf(Perl_debug_log,
1613 "isxdigit('%s') unexpectedly is %d\n",
1614 name, cBOOL(isxdigit(i))));
1616 if (UNLIKELY(tolower(i) != (int) toLOWER_A(i))) {
1618 DEBUG_L(PerlIO_printf(Perl_debug_log,
1619 "tolower('%s')=0x%x instead of the expected 0x%x\n",
1620 name, tolower(i), (int) toLOWER_A(i)));
1622 if (UNLIKELY(toupper(i) != (int) toUPPER_A(i))) {
1624 DEBUG_L(PerlIO_printf(Perl_debug_log,
1625 "toupper('%s')=0x%x instead of the expected 0x%x\n",
1626 name, toupper(i), (int) toUPPER_A(i)));
1628 if (UNLIKELY((i == '\n' && ! isCNTRL_LC(i)))) {
1630 DEBUG_L(PerlIO_printf(Perl_debug_log,
1631 "'\\n' (=%02X) is not a control\n", (int) i));
1634 /* Add to the list; Separate multiple entries with a blank */
1637 my_strlcat(bad_chars_list, " ", sizeof(bad_chars_list));
1639 my_strlcat(bad_chars_list, name, sizeof(bad_chars_list));
1645 if (bad_count == 2 && maybe_utf8_turkic) {
1647 *bad_chars_list = '\0';
1648 PL_fold_locale['I'] = 'I';
1649 PL_fold_locale['i'] = 'i';
1650 PL_in_utf8_turkic_locale = TRUE;
1651 DEBUG_L(PerlIO_printf(Perl_debug_log, "%s:%d: %s is turkic\n",
1652 __FILE__, __LINE__, newctype));
1655 PL_in_utf8_turkic_locale = FALSE;
1660 /* We only handle single-byte locales (outside of UTF-8 ones; so if
1661 * this locale requires more than one byte, there are going to be
1663 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
1664 "%s:%d: check_for_problems=%d, MB_CUR_MAX=%d\n",
1665 __FILE__, __LINE__, check_for_problems, (int) MB_CUR_MAX));
1667 if ( check_for_problems && MB_CUR_MAX > 1
1668 && ! PL_in_utf8_CTYPE_locale
1670 /* Some platforms return MB_CUR_MAX > 1 for even the "C"
1671 * locale. Just assume that the implementation for them (plus
1672 * for POSIX) is correct and the > 1 value is spurious. (Since
1673 * these are specially handled to never be considered UTF-8
1674 * locales, as long as this is the only problem, everything
1675 * should work fine */
1676 && strNE(newctype, "C") && strNE(newctype, "POSIX"))
1678 multi_byte_locale = TRUE;
1683 /* If we found problems and we want them output, do so */
1684 if ( (UNLIKELY(bad_count) || UNLIKELY(multi_byte_locale))
1685 && (LIKELY(ckWARN_d(WARN_LOCALE)) || UNLIKELY(DEBUG_L_TEST)))
1687 if (UNLIKELY(bad_count) && PL_in_utf8_CTYPE_locale) {
1688 PL_warn_locale = Perl_newSVpvf(aTHX_
1689 "Locale '%s' contains (at least) the following characters"
1690 " which have\nunexpected meanings: %s\nThe Perl program"
1691 " will use the expected meanings",
1692 newctype, bad_chars_list);
1695 PL_warn_locale = Perl_newSVpvf(aTHX_
1696 "Locale '%s' may not work well.%s%s%s\n",
1699 ? " Some characters in it are not recognized by"
1703 ? "\nThe following characters (and maybe others)"
1704 " may not have the same meaning as the Perl"
1705 " program expects:\n"
1713 # ifdef HAS_NL_LANGINFO
1715 Perl_sv_catpvf(aTHX_ PL_warn_locale, "; codeset=%s",
1716 /* parameter FALSE is a don't care here */
1717 my_nl_langinfo(CODESET, FALSE));
1721 Perl_sv_catpvf(aTHX_ PL_warn_locale, "\n");
1723 /* If we are actually in the scope of the locale or are debugging,
1724 * output the message now. If not in that scope, we save the
1725 * message to be output at the first operation using this locale,
1726 * if that actually happens. Most programs don't use locales, so
1727 * they are immune to bad ones. */
1728 if (IN_LC(LC_CTYPE) || UNLIKELY(DEBUG_L_TEST)) {
1730 /* The '0' below suppresses a bogus gcc compiler warning */
1731 Perl_warner(aTHX_ packWARN(WARN_LOCALE), SvPVX(PL_warn_locale), 0);
1733 if (IN_LC(LC_CTYPE)) {
1734 SvREFCNT_dec_NN(PL_warn_locale);
1735 PL_warn_locale = NULL;
1741 #endif /* USE_LOCALE_CTYPE */
1746 Perl__warn_problematic_locale()
1749 #ifdef USE_LOCALE_CTYPE
1753 /* Internal-to-core function that outputs the message in PL_warn_locale,
1754 * and then NULLS it. Should be called only through the macro
1755 * _CHECK_AND_WARN_PROBLEMATIC_LOCALE */
1757 if (PL_warn_locale) {
1758 Perl_ck_warner(aTHX_ packWARN(WARN_LOCALE),
1759 SvPVX(PL_warn_locale),
1760 0 /* dummy to avoid compiler warning */ );
1761 SvREFCNT_dec_NN(PL_warn_locale);
1762 PL_warn_locale = NULL;
1770 S_new_collate(pTHX_ const char *newcoll)
1773 #ifndef USE_LOCALE_COLLATE
1775 PERL_UNUSED_ARG(newcoll);
1776 PERL_UNUSED_CONTEXT;
1780 /* Called after each libc setlocale() call affecting LC_COLLATE, to tell
1781 * core Perl this and that 'newcoll' is the name of the new locale.
1783 * The design of locale collation is that every locale change is given an
1784 * index 'PL_collation_ix'. The first time a string particpates in an
1785 * operation that requires collation while locale collation is active, it
1786 * is given PERL_MAGIC_collxfrm magic (via sv_collxfrm_flags()). That
1787 * magic includes the collation index, and the transformation of the string
1788 * by strxfrm(), q.v. That transformation is used when doing comparisons,
1789 * instead of the string itself. If a string changes, the magic is
1790 * cleared. The next time the locale changes, the index is incremented,
1791 * and so we know during a comparison that the transformation is not
1792 * necessarily still valid, and so is recomputed. Note that if the locale
1793 * changes enough times, the index could wrap (a U32), and it is possible
1794 * that a transformation would improperly be considered valid, leading to
1795 * an unlikely bug */
1798 if (PL_collation_name) {
1800 Safefree(PL_collation_name);
1801 PL_collation_name = NULL;
1803 PL_collation_standard = TRUE;
1804 is_standard_collation:
1805 PL_collxfrm_base = 0;
1806 PL_collxfrm_mult = 2;
1807 PL_in_utf8_COLLATE_locale = FALSE;
1808 PL_strxfrm_NUL_replacement = '\0';
1809 PL_strxfrm_max_cp = 0;
1813 /* If this is not the same locale as currently, set the new one up */
1814 if (! PL_collation_name || strNE(PL_collation_name, newcoll)) {
1816 Safefree(PL_collation_name);
1817 PL_collation_name = stdize_locale(savepv(newcoll));
1818 PL_collation_standard = isNAME_C_OR_POSIX(newcoll);
1819 if (PL_collation_standard) {
1820 goto is_standard_collation;
1823 PL_in_utf8_COLLATE_locale = _is_cur_LC_category_utf8(LC_COLLATE);
1824 PL_strxfrm_NUL_replacement = '\0';
1825 PL_strxfrm_max_cp = 0;
1827 /* A locale collation definition includes primary, secondary, tertiary,
1828 * etc. weights for each character. To sort, the primary weights are
1829 * used, and only if they compare equal, then the secondary weights are
1830 * used, and only if they compare equal, then the tertiary, etc.
1832 * strxfrm() works by taking the input string, say ABC, and creating an
1833 * output transformed string consisting of first the primary weights,
1834 * A¹B¹C¹ followed by the secondary ones, A²B²C²; and then the
1835 * tertiary, etc, yielding A¹B¹C¹ A²B²C² A³B³C³ .... Some characters
1836 * may not have weights at every level. In our example, let's say B
1837 * doesn't have a tertiary weight, and A doesn't have a secondary
1838 * weight. The constructed string is then going to be
1839 * A¹B¹C¹ B²C² A³C³ ....
1840 * This has the desired effect that strcmp() will look at the secondary
1841 * or tertiary weights only if the strings compare equal at all higher
1842 * priority weights. The spaces shown here, like in
1844 * are not just for readability. In the general case, these must
1845 * actually be bytes, which we will call here 'separator weights'; and
1846 * they must be smaller than any other weight value, but since these
1847 * are C strings, only the terminating one can be a NUL (some
1848 * implementations may include a non-NUL separator weight just before
1849 * the NUL). Implementations tend to reserve 01 for the separator
1850 * weights. They are needed so that a shorter string's secondary
1851 * weights won't be misconstrued as primary weights of a longer string,
1852 * etc. By making them smaller than any other weight, the shorter
1853 * string will sort first. (Actually, if all secondary weights are
1854 * smaller than all primary ones, there is no need for a separator
1855 * weight between those two levels, etc.)
1857 * The length of the transformed string is roughly a linear function of
1858 * the input string. It's not exactly linear because some characters
1859 * don't have weights at all levels. When we call strxfrm() we have to
1860 * allocate some memory to hold the transformed string. The
1861 * calculations below try to find coefficients 'm' and 'b' for this
1862 * locale so that m*x + b equals how much space we need, given the size
1863 * of the input string in 'x'. If we calculate too small, we increase
1864 * the size as needed, and call strxfrm() again, but it is better to
1865 * get it right the first time to avoid wasted expensive string
1866 * transformations. */
1869 /* We use the string below to find how long the tranformation of it
1870 * is. Almost all locales are supersets of ASCII, or at least the
1871 * ASCII letters. We use all of them, half upper half lower,
1872 * because if we used fewer, we might hit just the ones that are
1873 * outliers in a particular locale. Most of the strings being
1874 * collated will contain a preponderance of letters, and even if
1875 * they are above-ASCII, they are likely to have the same number of
1876 * weight levels as the ASCII ones. It turns out that digits tend
1877 * to have fewer levels, and some punctuation has more, but those
1878 * are relatively sparse in text, and khw believes this gives a
1879 * reasonable result, but it could be changed if experience so
1881 const char longer[] = "ABCDEFGHIJKLMnopqrstuvwxyz";
1882 char * x_longer; /* Transformed 'longer' */
1883 Size_t x_len_longer; /* Length of 'x_longer' */
1885 char * x_shorter; /* We also transform a substring of 'longer' */
1886 Size_t x_len_shorter;
1888 /* _mem_collxfrm() is used get the transformation (though here we
1889 * are interested only in its length). It is used because it has
1890 * the intelligence to handle all cases, but to work, it needs some
1891 * values of 'm' and 'b' to get it started. For the purposes of
1892 * this calculation we use a very conservative estimate of 'm' and
1893 * 'b'. This assumes a weight can be multiple bytes, enough to
1894 * hold any UV on the platform, and there are 5 levels, 4 weight
1895 * bytes, and a trailing NUL. */
1896 PL_collxfrm_base = 5;
1897 PL_collxfrm_mult = 5 * sizeof(UV);
1899 /* Find out how long the transformation really is */
1900 x_longer = _mem_collxfrm(longer,
1904 /* We avoid converting to UTF-8 in the
1905 * called function by telling it the
1906 * string is in UTF-8 if the locale is a
1907 * UTF-8 one. Since the string passed
1908 * here is invariant under UTF-8, we can
1909 * claim it's UTF-8 even though it isn't.
1911 PL_in_utf8_COLLATE_locale);
1914 /* Find out how long the transformation of a substring of 'longer'
1915 * is. Together the lengths of these transformations are
1916 * sufficient to calculate 'm' and 'b'. The substring is all of
1917 * 'longer' except the first character. This minimizes the chances
1918 * of being swayed by outliers */
1919 x_shorter = _mem_collxfrm(longer + 1,
1922 PL_in_utf8_COLLATE_locale);
1923 Safefree(x_shorter);
1925 /* If the results are nonsensical for this simple test, the whole
1926 * locale definition is suspect. Mark it so that locale collation
1927 * is not active at all for it. XXX Should we warn? */
1928 if ( x_len_shorter == 0
1929 || x_len_longer == 0
1930 || x_len_shorter >= x_len_longer)
1932 PL_collxfrm_mult = 0;
1933 PL_collxfrm_base = 0;
1936 SSize_t base; /* Temporary */
1938 /* We have both: m * strlen(longer) + b = x_len_longer
1939 * m * strlen(shorter) + b = x_len_shorter;
1940 * subtracting yields:
1941 * m * (strlen(longer) - strlen(shorter))
1942 * = x_len_longer - x_len_shorter
1943 * But we have set things up so that 'shorter' is 1 byte smaller
1944 * than 'longer'. Hence:
1945 * m = x_len_longer - x_len_shorter
1947 * But if something went wrong, make sure the multiplier is at
1950 if (x_len_longer > x_len_shorter) {
1951 PL_collxfrm_mult = (STRLEN) x_len_longer - x_len_shorter;
1954 PL_collxfrm_mult = 1;
1959 * but in case something has gone wrong, make sure it is
1961 base = x_len_longer - PL_collxfrm_mult * (sizeof(longer) - 1);
1966 /* Add 1 for the trailing NUL */
1967 PL_collxfrm_base = base + 1;
1972 if (DEBUG_L_TEST || debug_initialization) {
1973 PerlIO_printf(Perl_debug_log,
1974 "%s:%d: ?UTF-8 locale=%d; x_len_shorter=%zu, "
1976 " collate multipler=%zu, collate base=%zu\n",
1978 PL_in_utf8_COLLATE_locale,
1979 x_len_shorter, x_len_longer,
1980 PL_collxfrm_mult, PL_collxfrm_base);
1987 #endif /* USE_LOCALE_COLLATE */
1995 #define USE_WSETLOCALE
1997 #ifdef USE_WSETLOCALE
2000 S_wrap_wsetlocale(pTHX_ int category, const char *locale) {
2007 MultiByteToWideChar(CP_UTF8, 0, locale, -1, NULL, 0);
2014 Newx(wlocale, req_size, wchar_t);
2015 if (!MultiByteToWideChar(CP_UTF8, 0, locale, -1, wlocale, req_size)) {
2024 wresult = _wsetlocale(category, wlocale);
2028 WideCharToMultiByte(CP_UTF8, 0, wresult, -1, NULL, 0, NULL, NULL);
2029 Newx(result, req_size, char);
2030 SAVEFREEPV(result); /* is there something better we can do here? */
2031 if (!WideCharToMultiByte(CP_UTF8, 0, wresult, -1,
2032 result, req_size, NULL, NULL)) {
2047 S_win32_setlocale(pTHX_ int category, const char* locale)
2049 /* This, for Windows, emulates POSIX setlocale() behavior. There is no
2050 * difference between the two unless the input locale is "", which normally
2051 * means on Windows to get the machine default, which is set via the
2052 * computer's "Regional and Language Options" (or its current equivalent).
2053 * In POSIX, it instead means to find the locale from the user's
2054 * environment. This routine changes the Windows behavior to first look in
2055 * the environment, and, if anything is found, use that instead of going to
2056 * the machine default. If there is no environment override, the machine
2057 * default is used, by calling the real setlocale() with "".
2059 * The POSIX behavior is to use the LC_ALL variable if set; otherwise to
2060 * use the particular category's variable if set; otherwise to use the LANG
2063 bool override_LC_ALL = FALSE;
2067 if (locale && strEQ(locale, "")) {
2071 locale = PerlEnv_getenv("LC_ALL");
2073 if (category == LC_ALL) {
2074 override_LC_ALL = TRUE;
2080 for (i = 0; i < NOMINAL_LC_ALL_INDEX; i++) {
2081 if (category == categories[i]) {
2082 locale = PerlEnv_getenv(category_names[i]);
2087 locale = PerlEnv_getenv("LANG");
2103 #ifdef USE_WSETLOCALE
2104 result = S_wrap_wsetlocale(aTHX_ category, locale);
2106 result = setlocale(category, locale);
2108 DEBUG_L(STMT_START {
2110 PerlIO_printf(Perl_debug_log, "%s:%d: %s\n", __FILE__, __LINE__,
2111 setlocale_debug_string(category, locale, result));
2115 if (! override_LC_ALL) {
2119 /* Here the input category was LC_ALL, and we have set it to what is in the
2120 * LANG variable or the system default if there is no LANG. But these have
2121 * lower priority than the other LC_foo variables, so override it for each
2122 * one that is set. (If they are set to "", it means to use the same thing
2123 * we just set LC_ALL to, so can skip) */
2125 for (i = 0; i < LC_ALL_INDEX; i++) {
2126 result = PerlEnv_getenv(category_names[i]);
2127 if (result && strNE(result, "")) {
2128 #ifdef USE_WSETLOCALE
2129 S_wrap_wsetlocale(aTHX_ categories[i], result);
2131 setlocale(categories[i], result);
2133 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "%s:%d: %s\n",
2135 setlocale_debug_string(categories[i], result, "not captured")));
2139 result = setlocale(LC_ALL, NULL);
2140 DEBUG_L(STMT_START {
2142 PerlIO_printf(Perl_debug_log, "%s:%d: %s\n",
2144 setlocale_debug_string(LC_ALL, NULL, result));
2154 =for apidoc Perl_setlocale
2156 This is an (almost) drop-in replacement for the system L<C<setlocale(3)>>,
2157 taking the same parameters, and returning the same information, except that it
2158 returns the correct underlying C<LC_NUMERIC> locale. Regular C<setlocale> will
2159 instead return C<C> if the underlying locale has a non-dot decimal point
2160 character, or a non-empty thousands separator for displaying floating point
2161 numbers. This is because perl keeps that locale category such that it has a
2162 dot and empty separator, changing the locale briefly during the operations
2163 where the underlying one is required. C<Perl_setlocale> knows about this, and
2164 compensates; regular C<setlocale> doesn't.
2166 Another reason it isn't completely a drop-in replacement is that it is
2167 declared to return S<C<const char *>>, whereas the system setlocale omits the
2168 C<const> (presumably because its API was specified long ago, and can't be
2169 updated; it is illegal to change the information C<setlocale> returns; doing
2170 so leads to segfaults.)
2172 Finally, C<Perl_setlocale> works under all circumstances, whereas plain
2173 C<setlocale> can be completely ineffective on some platforms under some
2176 C<Perl_setlocale> should not be used to change the locale except on systems
2177 where the predefined variable C<${^SAFE_LOCALES}> is 1. On some such systems,
2178 the system C<setlocale()> is ineffective, returning the wrong information, and
2179 failing to actually change the locale. C<Perl_setlocale>, however works
2180 properly in all circumstances.
2182 The return points to a per-thread static buffer, which is overwritten the next
2183 time C<Perl_setlocale> is called from the same thread.
2190 Perl_setlocale(const int category, const char * locale)
2192 /* This wraps POSIX::setlocale() */
2196 PERL_UNUSED_ARG(category);
2197 PERL_UNUSED_ARG(locale);
2203 const char * retval;
2204 const char * newlocale;
2207 DECLARATION_FOR_LC_NUMERIC_MANIPULATION;
2209 #ifdef USE_LOCALE_NUMERIC
2211 /* A NULL locale means only query what the current one is. We have the
2212 * LC_NUMERIC name saved, because we are normally switched into the C
2213 * (or equivalent) locale for it. For an LC_ALL query, switch back to get
2214 * the correct results. All other categories don't require special
2216 if (locale == NULL) {
2217 if (category == LC_NUMERIC) {
2219 /* We don't have to copy this return value, as it is a per-thread
2220 * variable, and won't change until a future setlocale */
2221 return PL_numeric_name;
2226 else if (category == LC_ALL) {
2227 STORE_LC_NUMERIC_FORCE_TO_UNDERLYING();
2236 retval = save_to_buffer(do_setlocale_r(category, locale),
2237 &PL_setlocale_buf, &PL_setlocale_bufsize, 0);
2240 #if defined(USE_LOCALE_NUMERIC) && defined(LC_ALL)
2242 if (locale == NULL && category == LC_ALL) {
2243 RESTORE_LC_NUMERIC();
2248 DEBUG_L(PerlIO_printf(Perl_debug_log,
2249 "%s:%d: %s\n", __FILE__, __LINE__,
2250 setlocale_debug_string(category, locale, retval)));
2258 /* If locale == NULL, we are just querying the state */
2259 if (locale == NULL) {
2263 /* Now that have switched locales, we have to update our records to
2268 #ifdef USE_LOCALE_CTYPE
2275 #ifdef USE_LOCALE_COLLATE
2278 new_collate(retval);
2282 #ifdef USE_LOCALE_NUMERIC
2285 new_numeric(retval);
2293 /* LC_ALL updates all the things we care about. The values may not
2294 * be the same as 'retval', as the locale "" may have set things
2297 # ifdef USE_LOCALE_CTYPE
2299 newlocale = savepv(do_setlocale_c(LC_CTYPE, NULL));
2300 new_ctype(newlocale);
2301 Safefree(newlocale);
2303 # endif /* USE_LOCALE_CTYPE */
2304 # ifdef USE_LOCALE_COLLATE
2306 newlocale = savepv(do_setlocale_c(LC_COLLATE, NULL));
2307 new_collate(newlocale);
2308 Safefree(newlocale);
2311 # ifdef USE_LOCALE_NUMERIC
2313 newlocale = savepv(do_setlocale_c(LC_NUMERIC, NULL));
2314 new_numeric(newlocale);
2315 Safefree(newlocale);
2317 # endif /* USE_LOCALE_NUMERIC */
2330 PERL_STATIC_INLINE const char *
2331 S_save_to_buffer(const char * string, char **buf, Size_t *buf_size, const Size_t offset)
2333 /* Copy the NUL-terminated 'string' to 'buf' + 'offset'. 'buf' has size 'buf_size',
2334 * growing it if necessary */
2338 PERL_ARGS_ASSERT_SAVE_TO_BUFFER;
2344 string_size = strlen(string) + offset + 1;
2346 if (*buf_size == 0) {
2347 Newx(*buf, string_size, char);
2348 *buf_size = string_size;
2350 else if (string_size > *buf_size) {
2351 Renew(*buf, string_size, char);
2352 *buf_size = string_size;
2355 Copy(string, *buf + offset, string_size - offset, char);
2361 =for apidoc Perl_langinfo
2363 This is an (almost) drop-in replacement for the system C<L<nl_langinfo(3)>>,
2364 taking the same C<item> parameter values, and returning the same information.
2365 But it is more thread-safe than regular C<nl_langinfo()>, and hides the quirks
2366 of Perl's locale handling from your code, and can be used on systems that lack
2367 a native C<nl_langinfo>.
2375 The reason it isn't quite a drop-in replacement is actually an advantage. The
2376 only difference is that it returns S<C<const char *>>, whereas plain
2377 C<nl_langinfo()> returns S<C<char *>>, but you are (only by documentation)
2378 forbidden to write into the buffer. By declaring this C<const>, the compiler
2379 enforces this restriction, so if it is violated, you know at compilation time,
2380 rather than getting segfaults at runtime.
2384 It delivers the correct results for the C<RADIXCHAR> and C<THOUSEP> items,
2385 without you having to write extra code. The reason for the extra code would be
2386 because these are from the C<LC_NUMERIC> locale category, which is normally
2387 kept set by Perl so that the radix is a dot, and the separator is the empty
2388 string, no matter what the underlying locale is supposed to be, and so to get
2389 the expected results, you have to temporarily toggle into the underlying
2390 locale, and later toggle back. (You could use plain C<nl_langinfo> and
2391 C<L</STORE_LC_NUMERIC_FORCE_TO_UNDERLYING>> for this but then you wouldn't get
2392 the other advantages of C<Perl_langinfo()>; not keeping C<LC_NUMERIC> in the C
2393 (or equivalent) locale would break a lot of CPAN, which is expecting the radix
2394 (decimal point) character to be a dot.)
2398 The system function it replaces can have its static return buffer trashed,
2399 not only by a subsequent call to that function, but by a C<freelocale>,
2400 C<setlocale>, or other locale change. The returned buffer of this function is
2401 not changed until the next call to it, so the buffer is never in a trashed
2406 Its return buffer is per-thread, so it also is never overwritten by a call to
2407 this function from another thread; unlike the function it replaces.
2411 But most importantly, it works on systems that don't have C<nl_langinfo>, such
2412 as Windows, hence makes your code more portable. Of the fifty-some possible
2413 items specified by the POSIX 2008 standard,
2414 L<http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/langinfo.h.html>,
2415 only one is completely unimplemented, though on non-Windows platforms, another
2416 significant one is also not implemented). It uses various techniques to
2417 recover the other items, including calling C<L<localeconv(3)>>, and
2418 C<L<strftime(3)>>, both of which are specified in C89, so should be always be
2419 available. Later C<strftime()> versions have additional capabilities; C<""> is
2420 returned for those not available on your system.
2422 It is important to note that when called with an item that is recovered by
2423 using C<localeconv>, the buffer from any previous explicit call to
2424 C<localeconv> will be overwritten. This means you must save that buffer's
2425 contents if you need to access them after a call to this function. (But note
2426 that you might not want to be using C<localeconv()> directly anyway, because of
2427 issues like the ones listed in the second item of this list (above) for
2428 C<RADIXCHAR> and C<THOUSEP>. You can use the methods given in L<perlcall> to
2429 call L<POSIX/localeconv> and avoid all the issues, but then you have a hash to
2432 The details for those items which may deviate from what this emulation returns
2433 and what a native C<nl_langinfo()> would return are specified in
2438 When using C<Perl_langinfo> on systems that don't have a native
2439 C<nl_langinfo()>, you must
2441 #include "perl_langinfo.h"
2443 before the C<perl.h> C<#include>. You can replace your C<langinfo.h>
2444 C<#include> with this one. (Doing it this way keeps out the symbols that plain
2445 C<langinfo.h> would try to import into the namespace for code that doesn't need
2448 The original impetus for C<Perl_langinfo()> was so that code that needs to
2449 find out the current currency symbol, floating point radix character, or digit
2450 grouping separator can use, on all systems, the simpler and more
2451 thread-friendly C<nl_langinfo> API instead of C<L<localeconv(3)>> which is a
2452 pain to make thread-friendly. For other fields returned by C<localeconv>, it
2453 is better to use the methods given in L<perlcall> to call
2454 L<C<POSIX::localeconv()>|POSIX/localeconv>, which is thread-friendly.
2461 #ifdef HAS_NL_LANGINFO
2462 Perl_langinfo(const nl_item item)
2464 Perl_langinfo(const int item)
2467 return my_nl_langinfo(item, TRUE);
2471 #ifdef HAS_NL_LANGINFO
2472 S_my_nl_langinfo(const nl_item item, bool toggle)
2474 S_my_nl_langinfo(const int item, bool toggle)
2478 const char * retval;
2480 #ifdef USE_LOCALE_NUMERIC
2482 /* We only need to toggle into the underlying LC_NUMERIC locale for these
2483 * two items, and only if not already there */
2484 if (toggle && (( item != RADIXCHAR && item != THOUSEP)
2485 || PL_numeric_underlying))
2487 #endif /* No toggling needed if not using LC_NUMERIC */
2491 #if defined(HAS_NL_LANGINFO) /* nl_langinfo() is available. */
2492 # if ! defined(HAS_THREAD_SAFE_NL_LANGINFO_L) \
2493 || ! defined(HAS_POSIX_2008_LOCALE)
2495 /* Here, use plain nl_langinfo(), switching to the underlying LC_NUMERIC
2496 * for those items dependent on it. This must be copied to a buffer before
2497 * switching back, as some systems destroy the buffer when setlocale() is
2501 DECLARATION_FOR_LC_NUMERIC_MANIPULATION;
2504 STORE_LC_NUMERIC_FORCE_TO_UNDERLYING();
2507 /* Prevent interference from another thread executing this code
2511 /* Copy to a per-thread buffer, which is also one that won't be
2512 * destroyed by a subsequent setlocale(), such as the
2513 * RESTORE_LC_NUMERIC may do just below. */
2514 retval = save_to_buffer(nl_langinfo(item),
2515 &PL_langinfo_buf, &PL_langinfo_bufsize, 0);
2519 RESTORE_LC_NUMERIC();
2523 # else /* Use nl_langinfo_l(), avoiding both a mutex and changing the locale */
2526 bool do_free = FALSE;
2527 locale_t cur = uselocale((locale_t) 0);
2529 if (cur == LC_GLOBAL_LOCALE) {
2530 cur = duplocale(LC_GLOBAL_LOCALE);
2534 # ifdef USE_LOCALE_NUMERIC
2537 if (PL_underlying_numeric_obj) {
2538 cur = PL_underlying_numeric_obj;
2541 cur = newlocale(LC_NUMERIC_MASK, PL_numeric_name, cur);
2548 /* We have to save it to a buffer, because the freelocale() just below
2549 * can invalidate the internal one */
2550 retval = save_to_buffer(nl_langinfo_l(item, cur),
2551 &PL_langinfo_buf, &PL_langinfo_bufsize, 0);
2560 if (strEQ(retval, "")) {
2561 if (item == YESSTR) {
2564 if (item == NOSTR) {
2571 #else /* Below, emulate nl_langinfo as best we can */
2575 # ifdef HAS_LOCALECONV
2577 const struct lconv* lc;
2579 DECLARATION_FOR_LC_NUMERIC_MANIPULATION;
2581 # ifdef TS_W32_BROKEN_LOCALECONV
2583 const char * save_global;
2584 const char * save_thread;
2592 # ifdef HAS_STRFTIME
2595 bool return_format = FALSE; /* Return the %format, not the value */
2596 const char * format;
2600 /* We copy the results to a per-thread buffer, even if not
2601 * multi-threaded. This is in part to simplify this code, and partly
2602 * because we need a buffer anyway for strftime(), and partly because a
2603 * call of localeconv() could otherwise wipe out the buffer, and the
2604 * programmer would not be expecting this, as this is a nl_langinfo()
2605 * substitute after all, so s/he might be thinking their localeconv()
2606 * is safe until another localeconv() call. */
2611 /* This is unimplemented */
2612 case ERA: /* For use with strftime() %E modifier */
2617 /* We use only an English set, since we don't know any more */
2618 case YESEXPR: return "^[+1yY]";
2619 case YESSTR: return "yes";
2620 case NOEXPR: return "^[-0nN]";
2621 case NOSTR: return "no";
2627 /* On non-windows, this is unimplemented, in part because of
2628 * inconsistencies between vendors. The Darwin native
2629 * nl_langinfo() implementation simply looks at everything past
2630 * any dot in the name, but that doesn't work for other
2631 * vendors. Many Linux locales that don't have UTF-8 in their
2632 * names really are UTF-8, for example; z/OS locales that do
2633 * have UTF-8 in their names, aren't really UTF-8 */
2638 { /* But on Windows, the name does seem to be consistent, so
2643 const char * name = my_setlocale(LC_CTYPE, NULL);
2645 if (isNAME_C_OR_POSIX(name)) {
2646 return "ANSI_X3.4-1968";
2649 /* Find the dot in the locale name */
2650 first = (const char *) strchr(name, '.');
2656 /* Look at everything past the dot */
2661 if (! isDIGIT(*p)) {
2668 /* Here everything past the dot is a digit. Treat it as a
2670 retval = save_to_buffer("CP", &PL_langinfo_buf,
2671 &PL_langinfo_bufsize, 0);
2672 offset = STRLENs("CP");
2676 retval = save_to_buffer(first, &PL_langinfo_buf,
2677 &PL_langinfo_bufsize, offset);
2683 # ifdef HAS_LOCALECONV
2687 /* We don't bother with localeconv_l() because any system that
2688 * has it is likely to also have nl_langinfo() */
2690 LOCALECONV_LOCK; /* Prevent interference with other threads
2691 using localeconv() */
2693 # ifdef TS_W32_BROKEN_LOCALECONV
2695 /* This is a workaround for a Windows bug prior to VS 15.
2696 * What we do here is, while locked, switch to the global
2697 * locale so localeconv() works; then switch back just before
2698 * the unlock. This can screw things up if some thread is
2699 * already using the global locale while assuming no other is.
2700 * A different workaround would be to call GetCurrencyFormat on
2701 * a known value, and parse it; patches welcome
2703 * We have to use LC_ALL instead of LC_MONETARY because of
2704 * another bug in Windows */
2706 save_thread = savepv(my_setlocale(LC_ALL, NULL));
2707 _configthreadlocale(_DISABLE_PER_THREAD_LOCALE);
2708 save_global= savepv(my_setlocale(LC_ALL, NULL));
2709 my_setlocale(LC_ALL, save_thread);
2715 || ! lc->currency_symbol
2716 || strEQ("", lc->currency_symbol))
2722 /* Leave the first spot empty to be filled in below */
2723 retval = save_to_buffer(lc->currency_symbol, &PL_langinfo_buf,
2724 &PL_langinfo_bufsize, 1);
2725 if (lc->mon_decimal_point && strEQ(lc->mon_decimal_point, ""))
2726 { /* khw couldn't figure out how the localedef specifications
2727 would show that the $ should replace the radix; this is
2728 just a guess as to how it might work.*/
2729 PL_langinfo_buf[0] = '.';
2731 else if (lc->p_cs_precedes) {
2732 PL_langinfo_buf[0] = '-';
2735 PL_langinfo_buf[0] = '+';
2738 # ifdef TS_W32_BROKEN_LOCALECONV
2740 my_setlocale(LC_ALL, save_global);
2741 _configthreadlocale(_ENABLE_PER_THREAD_LOCALE);
2742 my_setlocale(LC_ALL, save_thread);
2743 Safefree(save_global);
2744 Safefree(save_thread);
2751 # ifdef TS_W32_BROKEN_LOCALECONV
2755 /* For this, we output a known simple floating point number to
2756 * a buffer, and parse it, looking for the radix */
2759 STORE_LC_NUMERIC_FORCE_TO_UNDERLYING();
2762 if (PL_langinfo_bufsize < 10) {
2763 PL_langinfo_bufsize = 10;
2764 Renew(PL_langinfo_buf, PL_langinfo_bufsize, char);
2767 needed_size = my_snprintf(PL_langinfo_buf, PL_langinfo_bufsize,
2769 if (needed_size >= (int) PL_langinfo_bufsize) {
2770 PL_langinfo_bufsize = needed_size + 1;
2771 Renew(PL_langinfo_buf, PL_langinfo_bufsize, char);
2772 needed_size = my_snprintf(PL_langinfo_buf, PL_langinfo_bufsize,
2774 assert(needed_size < (int) PL_langinfo_bufsize);
2777 ptr = PL_langinfo_buf;
2778 e = PL_langinfo_buf + PL_langinfo_bufsize;
2781 while (ptr < e && *ptr != '1') {
2788 while (ptr < e && *ptr != '5') {
2792 /* Everything in between is the radix string */
2794 PL_langinfo_buf[0] = '?';
2795 PL_langinfo_buf[1] = '\0';
2799 Move(item_start, PL_langinfo_buf, ptr - PL_langinfo_buf, char);
2803 RESTORE_LC_NUMERIC();
2806 retval = PL_langinfo_buf;
2811 case RADIXCHAR: /* No special handling needed */
2818 STORE_LC_NUMERIC_FORCE_TO_UNDERLYING();
2821 LOCALECONV_LOCK; /* Prevent interference with other threads
2822 using localeconv() */
2824 # ifdef TS_W32_BROKEN_LOCALECONV
2826 /* This should only be for the thousands separator. A
2827 * different work around would be to use GetNumberFormat on a
2828 * known value and parse the result to find the separator */
2829 save_thread = savepv(my_setlocale(LC_ALL, NULL));
2830 _configthreadlocale(_DISABLE_PER_THREAD_LOCALE);
2831 save_global = savepv(my_setlocale(LC_ALL, NULL));
2832 my_setlocale(LC_ALL, save_thread);
2834 /* This is the start of code that for broken Windows replaces
2835 * the above and below code, and instead calls
2836 * GetNumberFormat() and then would parse that to find the
2837 * thousands separator. It needs to handle UTF-16 vs -8
2840 needed_size = GetNumberFormatEx(PL_numeric_name, 0, "1234.5", NULL, PL_langinfo_buf, PL_langinfo_bufsize);
2841 DEBUG_L(PerlIO_printf(Perl_debug_log,
2842 "%s: %d: return from GetNumber, count=%d, val=%s\n",
2843 __FILE__, __LINE__, needed_size, PL_langinfo_buf));
2853 temp = (item == RADIXCHAR)
2855 : lc->thousands_sep;
2861 retval = save_to_buffer(temp, &PL_langinfo_buf,
2862 &PL_langinfo_bufsize, 0);
2864 # ifdef TS_W32_BROKEN_LOCALECONV
2866 my_setlocale(LC_ALL, save_global);
2867 _configthreadlocale(_ENABLE_PER_THREAD_LOCALE);
2868 my_setlocale(LC_ALL, save_thread);
2869 Safefree(save_global);
2870 Safefree(save_thread);
2877 RESTORE_LC_NUMERIC();
2883 # ifdef HAS_STRFTIME
2885 /* These are defined by C89, so we assume that strftime supports
2886 * them, and so are returned unconditionally; they may not be what
2887 * the locale actually says, but should give good enough results
2888 * for someone using them as formats (as opposed to trying to parse
2889 * them to figure out what the locale says). The other format
2890 * items are actually tested to verify they work on the platform */
2891 case D_FMT: return "%x";
2892 case T_FMT: return "%X";
2893 case D_T_FMT: return "%c";
2895 /* These formats are only available in later strfmtime's */
2896 case ERA_D_FMT: case ERA_T_FMT: case ERA_D_T_FMT: case T_FMT_AMPM:
2898 /* The rest can be gotten from most versions of strftime(). */
2899 case ABDAY_1: case ABDAY_2: case ABDAY_3:
2900 case ABDAY_4: case ABDAY_5: case ABDAY_6: case ABDAY_7:
2902 case AM_STR: case PM_STR:
2903 case ABMON_1: case ABMON_2: case ABMON_3: case ABMON_4:
2904 case ABMON_5: case ABMON_6: case ABMON_7: case ABMON_8:
2905 case ABMON_9: case ABMON_10: case ABMON_11: case ABMON_12:
2906 case DAY_1: case DAY_2: case DAY_3: case DAY_4:
2907 case DAY_5: case DAY_6: case DAY_7:
2908 case MON_1: case MON_2: case MON_3: case MON_4:
2909 case MON_5: case MON_6: case MON_7: case MON_8:
2910 case MON_9: case MON_10: case MON_11: case MON_12:
2912 init_tm(&tm); /* Precaution against core dumps */
2916 tm.tm_year = 2017 - 1900;
2920 GCC_DIAG_IGNORE_STMT(-Wimplicit-fallthrough);
2925 "panic: %s: %d: switch case: %d problem",
2926 __FILE__, __LINE__, item);
2927 NOT_REACHED; /* NOTREACHED */
2929 case PM_STR: tm.tm_hour = 18;
2934 case ABDAY_7: tm.tm_wday++;
2935 case ABDAY_6: tm.tm_wday++;
2936 case ABDAY_5: tm.tm_wday++;
2937 case ABDAY_4: tm.tm_wday++;
2938 case ABDAY_3: tm.tm_wday++;
2939 case ABDAY_2: tm.tm_wday++;
2944 case DAY_7: tm.tm_wday++;
2945 case DAY_6: tm.tm_wday++;
2946 case DAY_5: tm.tm_wday++;
2947 case DAY_4: tm.tm_wday++;
2948 case DAY_3: tm.tm_wday++;
2949 case DAY_2: tm.tm_wday++;
2954 case ABMON_12: tm.tm_mon++;
2955 case ABMON_11: tm.tm_mon++;
2956 case ABMON_10: tm.tm_mon++;
2957 case ABMON_9: tm.tm_mon++;
2958 case ABMON_8: tm.tm_mon++;
2959 case ABMON_7: tm.tm_mon++;
2960 case ABMON_6: tm.tm_mon++;
2961 case ABMON_5: tm.tm_mon++;
2962 case ABMON_4: tm.tm_mon++;
2963 case ABMON_3: tm.tm_mon++;
2964 case ABMON_2: tm.tm_mon++;
2969 case MON_12: tm.tm_mon++;
2970 case MON_11: tm.tm_mon++;
2971 case MON_10: tm.tm_mon++;
2972 case MON_9: tm.tm_mon++;
2973 case MON_8: tm.tm_mon++;
2974 case MON_7: tm.tm_mon++;
2975 case MON_6: tm.tm_mon++;
2976 case MON_5: tm.tm_mon++;
2977 case MON_4: tm.tm_mon++;
2978 case MON_3: tm.tm_mon++;
2979 case MON_2: tm.tm_mon++;
2986 return_format = TRUE;
2991 return_format = TRUE;
2996 return_format = TRUE;
3001 return_format = TRUE;
3006 format = "%Ow"; /* Find the alternate digit for 0 */
3010 GCC_DIAG_RESTORE_STMT;
3012 /* We can't use my_strftime() because it doesn't look at
3014 while (0 == strftime(PL_langinfo_buf, PL_langinfo_bufsize,
3017 /* A zero return means one of:
3018 * a) there wasn't enough space in PL_langinfo_buf
3019 * b) the format, like a plain %p, returns empty
3020 * c) it was an illegal format, though some
3021 * implementations of strftime will just return the
3022 * illegal format as a plain character sequence.
3024 * To quickly test for case 'b)', try again but precede
3025 * the format with a plain character. If that result is
3026 * still empty, the problem is either 'a)' or 'c)' */
3028 Size_t format_size = strlen(format) + 1;
3029 Size_t mod_size = format_size + 1;
3033 Newx(mod_format, mod_size, char);
3034 Newx(temp_result, PL_langinfo_bufsize, char);
3036 my_strlcpy(mod_format + 1, format, mod_size);
3037 len = strftime(temp_result,
3038 PL_langinfo_bufsize,
3040 Safefree(mod_format);
3041 Safefree(temp_result);
3043 /* If 'len' is non-zero, it means that we had a case like
3044 * %p which means the current locale doesn't use a.m. or
3045 * p.m., and that is valid */
3048 /* Here, still didn't work. If we get well beyond a
3049 * reasonable size, bail out to prevent an infinite
3052 if (PL_langinfo_bufsize > 100 * format_size) {
3053 *PL_langinfo_buf = '\0';
3056 /* Double the buffer size to retry; Add 1 in case
3057 * original was 0, so we aren't stuck at 0. */
3058 PL_langinfo_bufsize *= 2;
3059 PL_langinfo_bufsize++;
3060 Renew(PL_langinfo_buf, PL_langinfo_bufsize, char);
3068 /* Here, we got a result.
3070 * If the item is 'ALT_DIGITS', PL_langinfo_buf contains the
3071 * alternate format for wday 0. If the value is the same as
3072 * the normal 0, there isn't an alternate, so clear the buffer.
3074 if ( item == ALT_DIGITS
3075 && strEQ(PL_langinfo_buf, "0"))
3077 *PL_langinfo_buf = '\0';
3080 /* ALT_DIGITS is problematic. Experiments on it showed that
3081 * strftime() did not always work properly when going from
3082 * alt-9 to alt-10. Only a few locales have this item defined,
3083 * and in all of them on Linux that khw was able to find,
3084 * nl_langinfo() merely returned the alt-0 character, possibly
3085 * doubled. Most Unicode digits are in blocks of 10
3086 * consecutive code points, so that is sufficient information
3087 * for those scripts, as we can infer alt-1, alt-2, .... But
3088 * for a Japanese locale, a CJK ideographic 0 is returned, and
3089 * the CJK digits are not in code point order, so you can't
3090 * really infer anything. The localedef for this locale did
3091 * specify the succeeding digits, so that strftime() works
3092 * properly on them, without needing to infer anything. But
3093 * the nl_langinfo() return did not give sufficient information
3094 * for the caller to understand what's going on. So until
3095 * there is evidence that it should work differently, this
3096 * returns the alt-0 string for ALT_DIGITS.
3098 * wday was chosen because its range is all a single digit.
3099 * Things like tm_sec have two digits as the minimum: '00' */
3101 retval = PL_langinfo_buf;
3103 /* If to return the format, not the value, overwrite the buffer
3104 * with it. But some strftime()s will keep the original format
3105 * if illegal, so change those to "" */
3106 if (return_format) {
3107 if (strEQ(PL_langinfo_buf, format)) {
3108 *PL_langinfo_buf = '\0';
3111 retval = save_to_buffer(format, &PL_langinfo_buf,
3112 &PL_langinfo_bufsize, 0);
3130 * Initialize locale awareness.
3133 Perl_init_i18nl10n(pTHX_ int printwarn)
3137 * 0 if not to output warning when setup locale is bad
3138 * 1 if to output warning based on value of PERL_BADLANG
3139 * >1 if to output regardless of PERL_BADLANG
3142 * 1 = set ok or not applicable,
3143 * 0 = fallback to a locale of lower priority
3144 * -1 = fallback to all locales failed, not even to the C locale
3146 * Under -DDEBUGGING, if the environment variable PERL_DEBUG_LOCALE_INIT is
3147 * set, debugging information is output.
3149 * This looks more complicated than it is, mainly due to the #ifdefs and
3152 * Besides some asserts, data structure initialization, and specific
3153 * platform complications, this routine is effectively just two things.
3155 * a) setlocale(LC_ALL, "");
3157 * which sets LC_ALL to the values in the current environment.
3159 * And for each individual category 'foo' whose value we care about:
3161 * b) save_foo = setlocale(LC_foo, NULL); handle_foo(save_foo);
3163 * (We don't tend to care about categories like LC_PAPER, for example.)
3165 * But there are complications. On systems without LC_ALL, it emulates
3166 * step a) by looping through all the categories, and doing
3168 * setlocale(LC_foo, "");
3172 * And it has to deal with if this is an embedded perl, whose locale
3173 * doesn't come from the environment, but has been set up by the caller.
3174 * This is pretty simply handled: the "" in the setlocale calls is not a
3175 * string constant, but a variable which is set to NULL in the embedded
3178 * But the major complication is handling failure and doing fallback.
3179 * There is an array, trial_locales, the elements of which are looped over
3180 * until the locale is successfully set. The array is initialized with
3181 * just one element, for
3182 * setlocale(LC_ALL, $NULL_or_empty)
3183 * If that works, as it almost always does, there's no more elements and
3184 * the loop iterates just the once. Otherwise elements are added for each
3185 * of the environment variables that POSIX dictates should control the
3186 * program, in priority order, with a final one being "C". The loop is
3187 * repeated until the first one succeeds. If all fail, we limp along with
3188 * whatever state we got to. If there is no LC_ALL, an inner loop is run
3189 * through all categories (making things look complex).
3191 * A further complication is that Windows has an additional fallback, the
3192 * user-default ANSI code page obtained from the operating system. This is
3193 * added as yet another loop iteration, just before the final "C"
3195 * On Ultrix, the locale MUST come from the environment, so there is
3196 * preliminary code to set it. I (khw) am not sure that it is necessary,
3197 * and that this couldn't be folded into the loop, but barring any real
3198 * platforms to test on, it's staying as-is
3205 PERL_UNUSED_ARG(printwarn);
3207 #else /* USE_LOCALE */
3210 const char * const language = PerlEnv_getenv("LANGUAGE");
3214 /* NULL uses the existing already set up locale */
3215 const char * const setlocale_init = (PerlEnv_getenv("PERL_SKIP_LOCALE_INIT"))
3218 const char* trial_locales[5]; /* 5 = 1 each for "", LC_ALL, LANG, "", C */
3219 unsigned int trial_locales_count;
3220 const char * const lc_all = PerlEnv_getenv("LC_ALL");
3221 const char * const lang = PerlEnv_getenv("LANG");
3222 bool setlocale_failure = FALSE;
3225 /* A later getenv() could zap this, so only use here */
3226 const char * const bad_lang_use_once = PerlEnv_getenv("PERL_BADLANG");
3228 const bool locwarn = (printwarn > 1
3230 && ( ! bad_lang_use_once
3232 /* disallow with "" or "0" */
3234 && strNE("0", bad_lang_use_once)))));
3236 /* setlocale() return vals; not copied so must be looked at immediately */
3237 const char * sl_result[NOMINAL_LC_ALL_INDEX + 1];
3239 /* current locale for given category; should have been copied so aren't
3241 const char * curlocales[NOMINAL_LC_ALL_INDEX + 1];
3245 /* In some systems you can find out the system default locale
3246 * and use that as the fallback locale. */
3247 # define SYSTEM_DEFAULT_LOCALE
3249 # ifdef SYSTEM_DEFAULT_LOCALE
3251 const char *system_default_locale = NULL;
3256 # define DEBUG_LOCALE_INIT(a,b,c)
3259 DEBUG_INITIALIZATION_set(cBOOL(PerlEnv_getenv("PERL_DEBUG_LOCALE_INIT")));
3261 # define DEBUG_LOCALE_INIT(category, locale, result) \
3263 if (debug_initialization) { \
3264 PerlIO_printf(Perl_debug_log, \
3266 __FILE__, __LINE__, \
3267 setlocale_debug_string(category, \
3273 /* Make sure the parallel arrays are properly set up */
3274 # ifdef USE_LOCALE_NUMERIC
3275 assert(categories[LC_NUMERIC_INDEX] == LC_NUMERIC);
3276 assert(strEQ(category_names[LC_NUMERIC_INDEX], "LC_NUMERIC"));
3277 # ifdef USE_POSIX_2008_LOCALE
3278 assert(category_masks[LC_NUMERIC_INDEX] == LC_NUMERIC_MASK);
3281 # ifdef USE_LOCALE_CTYPE
3282 assert(categories[LC_CTYPE_INDEX] == LC_CTYPE);
3283 assert(strEQ(category_names[LC_CTYPE_INDEX], "LC_CTYPE"));
3284 # ifdef USE_POSIX_2008_LOCALE
3285 assert(category_masks[LC_CTYPE_INDEX] == LC_CTYPE_MASK);
3288 # ifdef USE_LOCALE_COLLATE
3289 assert(categories[LC_COLLATE_INDEX] == LC_COLLATE);
3290 assert(strEQ(category_names[LC_COLLATE_INDEX], "LC_COLLATE"));
3291 # ifdef USE_POSIX_2008_LOCALE
3292 assert(category_masks[LC_COLLATE_INDEX] == LC_COLLATE_MASK);
3295 # ifdef USE_LOCALE_TIME
3296 assert(categories[LC_TIME_INDEX] == LC_TIME);
3297 assert(strEQ(category_names[LC_TIME_INDEX], "LC_TIME"));
3298 # ifdef USE_POSIX_2008_LOCALE
3299 assert(category_masks[LC_TIME_INDEX] == LC_TIME_MASK);
3302 # ifdef USE_LOCALE_MESSAGES
3303 assert(categories[LC_MESSAGES_INDEX] == LC_MESSAGES);
3304 assert(strEQ(category_names[LC_MESSAGES_INDEX], "LC_MESSAGES"));
3305 # ifdef USE_POSIX_2008_LOCALE
3306 assert(category_masks[LC_MESSAGES_INDEX] == LC_MESSAGES_MASK);
3309 # ifdef USE_LOCALE_MONETARY
3310 assert(categories[LC_MONETARY_INDEX] == LC_MONETARY);
3311 assert(strEQ(category_names[LC_MONETARY_INDEX], "LC_MONETARY"));
3312 # ifdef USE_POSIX_2008_LOCALE
3313 assert(category_masks[LC_MONETARY_INDEX] == LC_MONETARY_MASK);
3316 # ifdef USE_LOCALE_ADDRESS
3317 assert(categories[LC_ADDRESS_INDEX] == LC_ADDRESS);
3318 assert(strEQ(category_names[LC_ADDRESS_INDEX], "LC_ADDRESS"));
3319 # ifdef USE_POSIX_2008_LOCALE
3320 assert(category_masks[LC_ADDRESS_INDEX] == LC_ADDRESS_MASK);
3323 # ifdef USE_LOCALE_IDENTIFICATION
3324 assert(categories[LC_IDENTIFICATION_INDEX] == LC_IDENTIFICATION);
3325 assert(strEQ(category_names[LC_IDENTIFICATION_INDEX], "LC_IDENTIFICATION"));
3326 # ifdef USE_POSIX_2008_LOCALE
3327 assert(category_masks[LC_IDENTIFICATION_INDEX] == LC_IDENTIFICATION_MASK);
3330 # ifdef USE_LOCALE_MEASUREMENT
3331 assert(categories[LC_MEASUREMENT_INDEX] == LC_MEASUREMENT);
3332 assert(strEQ(category_names[LC_MEASUREMENT_INDEX], "LC_MEASUREMENT"));
3333 # ifdef USE_POSIX_2008_LOCALE
3334 assert(category_masks[LC_MEASUREMENT_INDEX] == LC_MEASUREMENT_MASK);
3337 # ifdef USE_LOCALE_PAPER
3338 assert(categories[LC_PAPER_INDEX] == LC_PAPER);
3339 assert(strEQ(category_names[LC_PAPER_INDEX], "LC_PAPER"));
3340 # ifdef USE_POSIX_2008_LOCALE
3341 assert(category_masks[LC_PAPER_INDEX] == LC_PAPER_MASK);
3344 # ifdef USE_LOCALE_TELEPHONE
3345 assert(categories[LC_TELEPHONE_INDEX] == LC_TELEPHONE);
3346 assert(strEQ(category_names[LC_TELEPHONE_INDEX], "LC_TELEPHONE"));
3347 # ifdef USE_POSIX_2008_LOCALE
3348 assert(category_masks[LC_TELEPHONE_INDEX] == LC_TELEPHONE_MASK);
3351 # ifdef USE_LOCALE_SYNTAX
3352 assert(categories[LC_SYNTAX_INDEX] == LC_SYNTAX);
3353 assert(strEQ(category_names[LC_SYNTAX_INDEX], "LC_SYNTAX"));
3354 # ifdef USE_POSIX_2008_LOCALE
3355 assert(category_masks[LC_SYNTAX_INDEX] == LC_SYNTAX_MASK);
3358 # ifdef USE_LOCALE_TOD
3359 assert(categories[LC_TOD_INDEX] == LC_TOD);
3360 assert(strEQ(category_names[LC_TOD_INDEX], "LC_TOD"));
3361 # ifdef USE_POSIX_2008_LOCALE
3362 assert(category_masks[LC_TOD_INDEX] == LC_TOD_MASK);
3366 assert(categories[LC_ALL_INDEX] == LC_ALL);
3367 assert(strEQ(category_names[LC_ALL_INDEX], "LC_ALL"));
3368 assert(NOMINAL_LC_ALL_INDEX == LC_ALL_INDEX);
3369 # ifdef USE_POSIX_2008_LOCALE
3370 assert(category_masks[LC_ALL_INDEX] == LC_ALL_MASK);
3373 # endif /* DEBUGGING */
3375 /* Initialize the per-thread mbrFOO() state variables. See POSIX.xs for
3376 * why these particular incantations are used. */
3378 memzero(&PL_mbrlen_ps, sizeof(PL_mbrlen_ps));
3381 memzero(&PL_mbrtowc_ps, sizeof(PL_mbrtowc_ps));
3384 wcrtomb(NULL, L'\0', &PL_wcrtomb_ps);
3387 /* Initialize the cache of the program's UTF-8ness for the always known
3388 * locales C and POSIX */
3389 my_strlcpy(PL_locale_utf8ness, C_and_POSIX_utf8ness,
3390 sizeof(PL_locale_utf8ness));
3392 /* See https://github.com/Perl/perl5/issues/17824 */
3393 Zero(curlocales, NOMINAL_LC_ALL_INDEX, char *);
3395 # ifdef USE_THREAD_SAFE_LOCALE
3398 _configthreadlocale(_ENABLE_PER_THREAD_LOCALE);
3402 # ifdef USE_POSIX_2008_LOCALE
3404 PL_C_locale_obj = newlocale(LC_ALL_MASK, "C", (locale_t) 0);
3405 if (! PL_C_locale_obj) {
3406 Perl_croak_nocontext(
3407 "panic: Cannot create POSIX 2008 C locale object; errno=%d", errno);
3409 if (DEBUG_Lv_TEST || debug_initialization) {
3410 PerlIO_printf(Perl_debug_log, "%s:%d: created C object %p\n", __FILE__, __LINE__, PL_C_locale_obj);
3415 # ifdef USE_LOCALE_NUMERIC
3417 PL_numeric_radix_sv = newSVpvs(".");
3421 # if defined(USE_POSIX_2008_LOCALE) && ! defined(HAS_QUERYLOCALE)
3423 /* Initialize our records. If we have POSIX 2008, we have LC_ALL */
3424 do_setlocale_c(LC_ALL, my_setlocale(LC_ALL, NULL));
3427 # ifdef LOCALE_ENVIRON_REQUIRED
3430 * Ultrix setlocale(..., "") fails if there are no environment
3431 * variables from which to get a locale name.
3435 # error Ultrix without LC_ALL not implemented
3441 sl_result[LC_ALL_INDEX] = do_setlocale_c(LC_ALL, setlocale_init);
3442 DEBUG_LOCALE_INIT(LC_ALL, setlocale_init, sl_result[LC_ALL_INDEX]);
3443 if (sl_result[LC_ALL_INDEX])
3446 setlocale_failure = TRUE;
3448 if (! setlocale_failure) {
3449 const char * locale_param;
3450 for (i = 0; i < LC_ALL_INDEX; i++) {
3451 locale_param = (! done && (lang || PerlEnv_getenv(category_names[i])))
3454 sl_result[i] = do_setlocale_r(categories[i], locale_param);
3455 if (! sl_result[i]) {
3456 setlocale_failure = TRUE;
3458 DEBUG_LOCALE_INIT(categories[i], locale_param, sl_result[i]);
3463 # endif /* LC_ALL */
3464 # endif /* LOCALE_ENVIRON_REQUIRED */
3466 /* We try each locale in the list until we get one that works, or exhaust
3467 * the list. Normally the loop is executed just once. But if setting the
3468 * locale fails, inside the loop we add fallback trials to the array and so
3469 * will execute the loop multiple times */
3470 trial_locales[0] = setlocale_init;
3471 trial_locales_count = 1;
3473 for (i= 0; i < trial_locales_count; i++) {
3474 const char * trial_locale = trial_locales[i];
3478 /* XXX This is to preserve old behavior for LOCALE_ENVIRON_REQUIRED
3479 * when i==0, but I (khw) don't think that behavior makes much
3481 setlocale_failure = FALSE;
3483 # ifdef SYSTEM_DEFAULT_LOCALE
3484 # ifdef WIN32 /* Note that assumes Win32 has LC_ALL */
3486 /* On Windows machines, an entry of "" after the 0th means to use
3487 * the system default locale, which we now proceed to get. */
3488 if (strEQ(trial_locale, "")) {
3491 /* Note that this may change the locale, but we are going to do
3492 * that anyway just below */
3493 system_default_locale = do_setlocale_c(LC_ALL, "");
3494 DEBUG_LOCALE_INIT(LC_ALL, "", system_default_locale);
3496 /* Skip if invalid or if it's already on the list of locales to
3498 if (! system_default_locale) {
3499 goto next_iteration;
3501 for (j = 0; j < trial_locales_count; j++) {
3502 if (strEQ(system_default_locale, trial_locales[j])) {
3503 goto next_iteration;
3507 trial_locale = system_default_locale;
3510 # error SYSTEM_DEFAULT_LOCALE only implemented for Win32
3512 # endif /* SYSTEM_DEFAULT_LOCALE */
3518 sl_result[LC_ALL_INDEX] = do_setlocale_c(LC_ALL, trial_locale);
3519 DEBUG_LOCALE_INIT(LC_ALL, trial_locale, sl_result[LC_ALL_INDEX]);
3520 if (! sl_result[LC_ALL_INDEX]) {
3521 setlocale_failure = TRUE;
3524 /* Since LC_ALL succeeded, it should have changed all the other
3525 * categories it can to its value; so we massage things so that the
3526 * setlocales below just return their category's current values.
3527 * This adequately handles the case in NetBSD where LC_COLLATE may
3528 * not be defined for a locale, and setting it individually will
3529 * fail, whereas setting LC_ALL succeeds, leaving LC_COLLATE set to
3530 * the POSIX locale. */
3531 trial_locale = NULL;
3534 # endif /* LC_ALL */
3536 if (! setlocale_failure) {
3538 for (j = 0; j < NOMINAL_LC_ALL_INDEX; j++) {
3540 = savepv(do_setlocale_r(categories[j], trial_locale));
3541 if (! curlocales[j]) {
3542 setlocale_failure = TRUE;
3544 DEBUG_LOCALE_INIT(categories[j], trial_locale, curlocales[j]);
3547 if (LIKELY(! setlocale_failure)) { /* All succeeded */
3548 break; /* Exit trial_locales loop */
3552 /* Here, something failed; will need to try a fallback. */
3558 if (locwarn) { /* Output failure info only on the first one */
3562 PerlIO_printf(Perl_error_log,
3563 "perl: warning: Setting locale failed.\n");
3565 # else /* !LC_ALL */
3567 PerlIO_printf(Perl_error_log,
3568 "perl: warning: Setting locale failed for the categories:\n\t");
3570 for (j = 0; j < NOMINAL_LC_ALL_INDEX; j++) {
3571 if (! curlocales[j]) {
3572 PerlIO_printf(Perl_error_log, category_names[j]);
3575 Safefree(curlocales[j]);
3579 # endif /* LC_ALL */
3581 PerlIO_printf(Perl_error_log,
3582 "perl: warning: Please check that your locale settings:\n");
3586 PerlIO_printf(Perl_error_log,
3587 "\tLANGUAGE = %c%s%c,\n",
3588 language ? '"' : '(',
3589 language ? language : "unset",
3590 language ? '"' : ')');
3593 PerlIO_printf(Perl_error_log,
3594 "\tLC_ALL = %c%s%c,\n",
3596 lc_all ? lc_all : "unset",
3597 lc_all ? '"' : ')');
3599 # if defined(USE_ENVIRON_ARRAY)
3604 /* Look through the environment for any variables of the
3605 * form qr/ ^ LC_ [A-Z]+ = /x, except LC_ALL which was
3606 * already handled above. These are assumed to be locale
3607 * settings. Output them and their values. */
3608 for (e = environ; *e; e++) {
3609 const STRLEN prefix_len = sizeof("LC_") - 1;
3612 if ( strBEGINs(*e, "LC_")
3613 && ! strBEGINs(*e, "LC_ALL=")
3614 && (uppers_len = strspn(*e + prefix_len,
3615 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"))
3616 && ((*e)[prefix_len + uppers_len] == '='))
3618 PerlIO_printf(Perl_error_log, "\t%.*s = \"%s\",\n",
3619 (int) (prefix_len + uppers_len), *e,
3620 *e + prefix_len + uppers_len + 1);
3627 PerlIO_printf(Perl_error_log,
3628 "\t(possibly more locale environment variables)\n");
3632 PerlIO_printf(Perl_error_log,
3633 "\tLANG = %c%s%c\n",
3635 lang ? lang : "unset",
3638 PerlIO_printf(Perl_error_log,
3639 " are supported and installed on your system.\n");
3642 /* Calculate what fallback locales to try. We have avoided this
3643 * until we have to, because failure is quite unlikely. This will
3644 * usually change the upper bound of the loop we are in.
3646 * Since the system's default way of setting the locale has not
3647 * found one that works, We use Perl's defined ordering: LC_ALL,
3648 * LANG, and the C locale. We don't try the same locale twice, so
3649 * don't add to the list if already there. (On POSIX systems, the
3650 * LC_ALL element will likely be a repeat of the 0th element "",
3651 * but there's no harm done by doing it explicitly.
3653 * Note that this tries the LC_ALL environment variable even on
3654 * systems which have no LC_ALL locale setting. This may or may
3655 * not have been originally intentional, but there's no real need
3656 * to change the behavior. */
3658 for (j = 0; j < trial_locales_count; j++) {
3659 if (strEQ(lc_all, trial_locales[j])) {
3663 trial_locales[trial_locales_count++] = lc_all;
3668 for (j = 0; j < trial_locales_count; j++) {
3669 if (strEQ(lang, trial_locales[j])) {
3673 trial_locales[trial_locales_count++] = lang;
3677 # if defined(WIN32) && defined(LC_ALL)
3679 /* For Windows, we also try the system default locale before "C".
3680 * (If there exists a Windows without LC_ALL we skip this because
3681 * it gets too complicated. For those, the "C" is the next
3682 * fallback possibility). The "" is the same as the 0th element of
3683 * the array, but the code at the loop above knows to treat it
3684 * differently when not the 0th */
3685 trial_locales[trial_locales_count++] = "";
3689 for (j = 0; j < trial_locales_count; j++) {
3690 if (strEQ("C", trial_locales[j])) {
3694 trial_locales[trial_locales_count++] = "C";
3697 } /* end of first time through the loop */
3705 } /* end of looping through the trial locales */
3707 if (ok < 1) { /* If we tried to fallback */
3709 if (! setlocale_failure) { /* fallback succeeded */
3710 msg = "Falling back to";
3712 else { /* fallback failed */
3715 /* We dropped off the end of the loop, so have to decrement i to
3716 * get back to the value the last time through */
3720 msg = "Failed to fall back to";
3722 /* To continue, we should use whatever values we've got */
3724 for (j = 0; j < NOMINAL_LC_ALL_INDEX; j++) {
3725 Safefree(curlocales[j]);
3726 curlocales[j] = savepv(do_setlocale_r(categories[j], NULL));
3727 DEBUG_LOCALE_INIT(categories[j], NULL, curlocales[j]);
3732 const char * description;
3733 const char * name = "";
3734 if (strEQ(trial_locales[i], "C")) {
3735 description = "the standard locale";
3739 # ifdef SYSTEM_DEFAULT_LOCALE
3741 else if (strEQ(trial_locales[i], "")) {
3742 description = "the system default locale";
3743 if (system_default_locale) {
3744 name = system_default_locale;
3748 # endif /* SYSTEM_DEFAULT_LOCALE */
3751 description = "a fallback locale";
3752 name = trial_locales[i];
3754 if (name && strNE(name, "")) {
3755 PerlIO_printf(Perl_error_log,
3756 "perl: warning: %s %s (\"%s\").\n", msg, description, name);
3759 PerlIO_printf(Perl_error_log,
3760 "perl: warning: %s %s.\n", msg, description);
3763 } /* End of tried to fallback */
3765 /* Done with finding the locales; update our records */
3767 # ifdef USE_LOCALE_CTYPE
3769 new_ctype(curlocales[LC_CTYPE_INDEX]);
3772 # ifdef USE_LOCALE_COLLATE
3774 new_collate(curlocales[LC_COLLATE_INDEX]);
3777 # ifdef USE_LOCALE_NUMERIC
3779 new_numeric(curlocales[LC_NUMERIC_INDEX]);
3783 for (i = 0; i < NOMINAL_LC_ALL_INDEX; i++) {
3785 # if defined(USE_ITHREADS) && ! defined(USE_THREAD_SAFE_LOCALE)
3787 /* This caches whether each category's locale is UTF-8 or not. This
3788 * may involve changing the locale. It is ok to do this at
3789 * initialization time before any threads have started, but not later
3790 * unless thread-safe operations are used.
3791 * Caching means that if the program heeds our dictate not to change
3792 * locales in threaded applications, this data will remain valid, and
3793 * it may get queried without having to change locales. If the
3794 * environment is such that all categories have the same locale, this
3795 * isn't needed, as the code will not change the locale; but this
3796 * handles the uncommon case where the environment has disparate
3797 * locales for the categories */
3798 (void) _is_cur_LC_category_utf8(categories[i]);
3802 Safefree(curlocales[i]);
3805 # if defined(USE_PERLIO) && defined(USE_LOCALE_CTYPE)
3807 /* Set PL_utf8locale to TRUE if using PerlIO _and_ the current LC_CTYPE
3808 * locale is UTF-8. The call to new_ctype() just above has already
3809 * calculated the latter value and saved it in PL_in_utf8_CTYPE_locale. If
3810 * both PL_utf8locale and PL_unicode (set by -C or by $ENV{PERL_UNICODE})
3811 * are true, perl.c:S_parse_body() will turn on the PerlIO :utf8 layer on
3812 * STDIN, STDOUT, STDERR, _and_ the default open discipline. */
3813 PL_utf8locale = PL_in_utf8_CTYPE_locale;
3815 /* Set PL_unicode to $ENV{PERL_UNICODE} if using PerlIO.
3816 This is an alternative to using the -C command line switch
3817 (the -C if present will override this). */
3819 const char *p = PerlEnv_getenv("PERL_UNICODE");
3820 PL_unicode = p ? parse_unicode_opts(&p) : 0;
3821 if (PL_unicode & PERL_UNICODE_UTF8CACHEASSERT_FLAG)
3826 #endif /* USE_LOCALE */
3829 /* So won't continue to output stuff */
3830 DEBUG_INITIALIZATION_set(FALSE);
3837 #ifdef USE_LOCALE_COLLATE
3840 Perl__mem_collxfrm(pTHX_ const char *input_string,
3841 STRLEN len, /* Length of 'input_string' */
3842 STRLEN *xlen, /* Set to length of returned string
3843 (not including the collation index
3845 bool utf8 /* Is the input in UTF-8? */
3849 /* _mem_collxfrm() is a bit like strxfrm() but with two important
3850 * differences. First, it handles embedded NULs. Second, it allocates a bit
3851 * more memory than needed for the transformed data itself. The real
3852 * transformed data begins at offset COLLXFRM_HDR_LEN. *xlen is set to
3853 * the length of that, and doesn't include the collation index size.
3854 * Please see sv_collxfrm() to see how this is used. */
3856 #define COLLXFRM_HDR_LEN sizeof(PL_collation_ix)
3858 char * s = (char *) input_string;
3859 STRLEN s_strlen = strlen(input_string);
3861 STRLEN xAlloc; /* xalloc is a reserved word in VC */
3862 STRLEN length_in_chars;
3863 bool first_time = TRUE; /* Cleared after first loop iteration */
3865 PERL_ARGS_ASSERT__MEM_COLLXFRM;
3867 /* Must be NUL-terminated */
3868 assert(*(input_string + len) == '\0');
3870 /* If this locale has defective collation, skip */
3871 if (PL_collxfrm_base == 0 && PL_collxfrm_mult == 0) {
3872 DEBUG_L(PerlIO_printf(Perl_debug_log,
3873 "_mem_collxfrm: locale's collation is defective\n"));
3877 /* Replace any embedded NULs with the control that sorts before any others.
3878 * This will give as good as possible results on strings that don't
3879 * otherwise contain that character, but otherwise there may be
3880 * less-than-perfect results with that character and NUL. This is
3881 * unavoidable unless we replace strxfrm with our own implementation. */
3882 if (UNLIKELY(s_strlen < len)) { /* Only execute if there is an embedded
3886 STRLEN sans_nuls_len;
3887 int try_non_controls;
3888 char this_replacement_char[] = "?\0"; /* Room for a two-byte string,
3889 making sure 2nd byte is NUL.
3891 STRLEN this_replacement_len;
3893 /* If we don't know what non-NUL control character sorts lowest for
3894 * this locale, find it */
3895 if (PL_strxfrm_NUL_replacement == '\0') {
3897 char * cur_min_x = NULL; /* The min_char's xfrm, (except it also
3898 includes the collation index
3901 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "Looking to replace NUL\n"));
3903 /* Unlikely, but it may be that no control will work to replace
3904 * NUL, in which case we instead look for any character. Controls
3905 * are preferred because collation order is, in general, context
3906 * sensitive, with adjoining characters affecting the order, and
3907 * controls are less likely to have such interactions, allowing the
3908 * NUL-replacement to stand on its own. (Another way to look at it
3909 * is to imagine what would happen if the NUL were replaced by a
3910 * combining character; it wouldn't work out all that well.) */
3911 for (try_non_controls = 0;
3912 try_non_controls < 2;
3915 /* Look through all legal code points (NUL isn't) */
3916 for (j = 1; j < 256; j++) {
3917 char * x; /* j's xfrm plus collation index */
3918 STRLEN x_len; /* length of 'x' */
3919 STRLEN trial_len = 1;
3920 char cur_source[] = { '\0', '\0' };
3922 /* Skip non-controls the first time through the loop. The
3923 * controls in a UTF-8 locale are the L1 ones */
3924 if (! try_non_controls && (PL_in_utf8_COLLATE_locale)
3931 /* Create a 1-char string of the current code point */
3932 cur_source[0] = (char) j;
3934 /* Then transform it */
3935 x = _mem_collxfrm(cur_source, trial_len, &x_len,
3936 0 /* The string is not in UTF-8 */);
3938 /* Ignore any character that didn't successfully transform.
3944 /* If this character's transformation is lower than
3945 * the current lowest, this one becomes the lowest */
3946 if ( cur_min_x == NULL
3947 || strLT(x + COLLXFRM_HDR_LEN,
3948 cur_min_x + COLLXFRM_HDR_LEN))
3950 PL_strxfrm_NUL_replacement = j;
3951 Safefree(cur_min_x);
3957 } /* end of loop through all 255 characters */
3959 /* Stop looking if found */
3964 /* Unlikely, but possible, if there aren't any controls that
3965 * work in the locale, repeat the loop, looking for any
3966 * character that works */
3967 DEBUG_L(PerlIO_printf(Perl_debug_log,
3968 "_mem_collxfrm: No control worked. Trying non-controls\n"));
3969 } /* End of loop to try first the controls, then any char */
3972 DEBUG_L(PerlIO_printf(Perl_debug_log,
3973 "_mem_collxfrm: Couldn't find any character to replace"
3974 " embedded NULs in locale %s with", PL_collation_name));
3978 DEBUG_L(PerlIO_printf(Perl_debug_log,
3979 "_mem_collxfrm: Replacing embedded NULs in locale %s with "
3980 "0x%02X\n", PL_collation_name, PL_strxfrm_NUL_replacement));
3982 Safefree(cur_min_x);
3983 } /* End of determining the character that is to replace NULs */
3985 /* If the replacement is variant under UTF-8, it must match the
3986 * UTF8-ness of the original */
3987 if ( ! UVCHR_IS_INVARIANT(PL_strxfrm_NUL_replacement) && utf8) {
3988 this_replacement_char[0] =
3989 UTF8_EIGHT_BIT_HI(PL_strxfrm_NUL_replacement);
3990 this_replacement_char[1] =
3991 UTF8_EIGHT_BIT_LO(PL_strxfrm_NUL_replacement);
3992 this_replacement_len = 2;
3995 this_replacement_char[0] = PL_strxfrm_NUL_replacement;
3996 /* this_replacement_char[1] = '\0' was done at initialization */
3997 this_replacement_len = 1;
4000 /* The worst case length for the replaced string would be if every
4001 * character in it is NUL. Multiply that by the length of each
4002 * replacement, and allow for a trailing NUL */
4003 sans_nuls_len = (len * this_replacement_len) + 1;
4004 Newx(sans_nuls, sans_nuls_len, char);
4007 /* Replace each NUL with the lowest collating control. Loop until have
4008 * exhausted all the NULs */
4009 while (s + s_strlen < e) {
4010 my_strlcat(sans_nuls, s, sans_nuls_len);
4012 /* Do the actual replacement */
4013 my_strlcat(sans_nuls, this_replacement_char, sans_nuls_len);
4015 /* Move past the input NUL */
4017 s_strlen = strlen(s);
4020 /* And add anything that trails the final NUL */
4021 my_strlcat(sans_nuls, s, sans_nuls_len);
4023 /* Switch so below we transform this modified string */
4026 } /* End of replacing NULs */
4028 /* Make sure the UTF8ness of the string and locale match */
4029 if (utf8 != PL_in_utf8_COLLATE_locale) {
4030 /* XXX convert above Unicode to 10FFFF? */
4031 const char * const t = s; /* Temporary so we can later find where the
4034 /* Here they don't match. Change the string's to be what the locale is
4037 if (! utf8) { /* locale is UTF-8, but input isn't; upgrade the input */
4038 s = (char *) bytes_to_utf8((const U8 *) s, &len);
4041 else { /* locale is not UTF-8; but input is; downgrade the input */
4043 s = (char *) bytes_from_utf8((const U8 *) s, &len, &utf8);
4045 /* If the downgrade was successful we are done, but if the input
4046 * contains things that require UTF-8 to represent, have to do
4047 * damage control ... */
4048 if (UNLIKELY(utf8)) {
4050 /* What we do is construct a non-UTF-8 string with
4051 * 1) the characters representable by a single byte converted
4052 * to be so (if necessary);
4053 * 2) and the rest converted to collate the same as the
4054 * highest collating representable character. That makes
4055 * them collate at the end. This is similar to how we
4056 * handle embedded NULs, but we use the highest collating
4057 * code point instead of the smallest. Like the NUL case,
4058 * this isn't perfect, but is the best we can reasonably
4059 * do. Every above-255 code point will sort the same as
4060 * the highest-sorting 0-255 code point. If that code
4061 * point can combine in a sequence with some other code
4062 * points for weight calculations, us changing something to
4063 * be it can adversely affect the results. But in most
4064 * cases, it should work reasonably. And note that this is
4065 * really an illegal situation: using code points above 255
4066 * on a locale where only 0-255 are valid. If two strings
4067 * sort entirely equal, then the sort order for the
4068 * above-255 code points will be in code point order. */
4072 /* If we haven't calculated the code point with the maximum
4073 * collating order for this locale, do so now */
4074 if (! PL_strxfrm_max_cp) {
4077 /* The current transformed string that collates the
4078 * highest (except it also includes the prefixed collation
4080 char * cur_max_x = NULL;
4082 /* Look through all legal code points (NUL isn't) */
4083 for (j = 1; j < 256; j++) {
4086 char cur_source[] = { '\0', '\0' };
4088 /* Create a 1-char string of the current code point */
4089 cur_source[0] = (char) j;
4091 /* Then transform it */
4092 x = _mem_collxfrm(cur_source, 1, &x_len, FALSE);
4094 /* If something went wrong (which it shouldn't), just
4095 * ignore this code point */
4100 /* If this character's transformation is higher than
4101 * the current highest, this one becomes the highest */
4102 if ( cur_max_x == NULL
4103 || strGT(x + COLLXFRM_HDR_LEN,
4104 cur_max_x + COLLXFRM_HDR_LEN))
4106 PL_strxfrm_max_cp = j;
4107 Safefree(cur_max_x);
4116 DEBUG_L(PerlIO_printf(Perl_debug_log,
4117 "_mem_collxfrm: Couldn't find any character to"
4118 " replace above-Latin1 chars in locale %s with",
4119 PL_collation_name));
4123 DEBUG_L(PerlIO_printf(Perl_debug_log,
4124 "_mem_collxfrm: highest 1-byte collating character"
4125 " in locale %s is 0x%02X\n",
4127 PL_strxfrm_max_cp));
4129 Safefree(cur_max_x);
4132 /* Here we know which legal code point collates the highest.
4133 * We are ready to construct the non-UTF-8 string. The length
4134 * will be at least 1 byte smaller than the input string
4135 * (because we changed at least one 2-byte character into a
4136 * single byte), but that is eaten up by the trailing NUL */
4142 char * e = (char *) t + len;
4144 for (i = 0; i < len; i+= UTF8SKIP(t + i)) {
4146 if (UTF8_IS_INVARIANT(cur_char)) {
4149 else if (UTF8_IS_NEXT_CHAR_DOWNGRADEABLE(t + i, e)) {
4150 s[d++] = EIGHT_BIT_UTF8_TO_NATIVE(cur_char, t[i+1]);
4152 else { /* Replace illegal cp with highest collating
4154 s[d++] = PL_strxfrm_max_cp;
4158 Renew(s, d, char); /* Free up unused space */
4163 /* Here, we have constructed a modified version of the input. It could
4164 * be that we already had a modified copy before we did this version.
4165 * If so, that copy is no longer needed */
4166 if (t != input_string) {
4171 length_in_chars = (utf8)
4172 ? utf8_length((U8 *) s, (U8 *) s + len)
4175 /* The first element in the output is the collation id, used by
4176 * sv_collxfrm(); then comes the space for the transformed string. The
4177 * equation should give us a good estimate as to how much is needed */
4178 xAlloc = COLLXFRM_HDR_LEN
4180 + (PL_collxfrm_mult * length_in_chars);
4181 Newx(xbuf, xAlloc, char);
4182 if (UNLIKELY(! xbuf)) {
4183 DEBUG_L(PerlIO_printf(Perl_debug_log,
4184 "_mem_collxfrm: Couldn't malloc %zu bytes\n", xAlloc));
4188 /* Store the collation id */
4189 *(U32*)xbuf = PL_collation_ix;
4191 /* Then the transformation of the input. We loop until successful, or we
4195 *xlen = strxfrm(xbuf + COLLXFRM_HDR_LEN, s, xAlloc - COLLXFRM_HDR_LEN);
4197 /* If the transformed string occupies less space than we told strxfrm()
4198 * was available, it means it successfully transformed the whole
4200 if (*xlen < xAlloc - COLLXFRM_HDR_LEN) {
4202 /* Some systems include a trailing NUL in the returned length.
4203 * Ignore it, using a loop in case multiple trailing NULs are
4206 && *(xbuf + COLLXFRM_HDR_LEN + (*xlen) - 1) == '\0')
4211 /* If the first try didn't get it, it means our prediction was low.
4212 * Modify the coefficients so that we predict a larger value in any
4213 * future transformations */
4215 STRLEN needed = *xlen + 1; /* +1 For trailing NUL */
4216 STRLEN computed_guess = PL_collxfrm_base
4217 + (PL_collxfrm_mult * length_in_chars);
4219 /* On zero-length input, just keep current slope instead of
4221 const STRLEN new_m = (length_in_chars != 0)
4222 ? needed / length_in_chars
4225 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
4226 "%s: %d: initial size of %zu bytes for a length "
4227 "%zu string was insufficient, %zu needed\n",
4229 computed_guess, length_in_chars, needed));
4231 /* If slope increased, use it, but discard this result for
4232 * length 1 strings, as we can't be sure that it's a real slope
4234 if (length_in_chars > 1 && new_m > PL_collxfrm_mult) {
4238 STRLEN old_m = PL_collxfrm_mult;
4239 STRLEN old_b = PL_collxfrm_base;
4243 PL_collxfrm_mult = new_m;
4244 PL_collxfrm_base = 1; /* +1 For trailing NUL */
4245 computed_guess = PL_collxfrm_base
4246 + (PL_collxfrm_mult * length_in_chars);
4247 if (computed_guess < needed) {
4248 PL_collxfrm_base += needed - computed_guess;
4251 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
4252 "%s: %d: slope is now %zu; was %zu, base "
4253 "is now %zu; was %zu\n",
4255 PL_collxfrm_mult, old_m,
4256 PL_collxfrm_base, old_b));
4258 else { /* Slope didn't change, but 'b' did */
4259 const STRLEN new_b = needed
4262 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
4263 "%s: %d: base is now %zu; was %zu\n",
4265 new_b, PL_collxfrm_base));
4266 PL_collxfrm_base = new_b;
4273 if (UNLIKELY(*xlen >= PERL_INT_MAX)) {
4274 DEBUG_L(PerlIO_printf(Perl_debug_log,
4275 "_mem_collxfrm: Needed %zu bytes, max permissible is %u\n",
4276 *xlen, PERL_INT_MAX));
4280 /* A well-behaved strxfrm() returns exactly how much space it needs
4281 * (usually not including the trailing NUL) when it fails due to not
4282 * enough space being provided. Assume that this is the case unless
4283 * it's been proven otherwise */
4284 if (LIKELY(PL_strxfrm_is_behaved) && first_time) {
4285 xAlloc = *xlen + COLLXFRM_HDR_LEN + 1;
4287 else { /* Here, either:
4288 * 1) The strxfrm() has previously shown bad behavior; or
4289 * 2) It isn't the first time through the loop, which means
4290 * that the strxfrm() is now showing bad behavior, because
4291 * we gave it what it said was needed in the previous
4292 * iteration, and it came back saying it needed still more.
4293 * (Many versions of cygwin fit this. When the buffer size
4294 * isn't sufficient, they return the input size instead of
4295 * how much is needed.)
4296 * Increase the buffer size by a fixed percentage and try again.
4298 xAlloc += (xAlloc / 4) + 1;
4299 PL_strxfrm_is_behaved = FALSE;
4303 if (DEBUG_Lv_TEST || debug_initialization) {
4304 PerlIO_printf(Perl_debug_log,
4305 "_mem_collxfrm required more space than previously calculated"
4306 " for locale %s, trying again with new guess=%zu+%zu\n",
4307 PL_collation_name, COLLXFRM_HDR_LEN,
4308 xAlloc - COLLXFRM_HDR_LEN);
4315 Renew(xbuf, xAlloc, char);
4316 if (UNLIKELY(! xbuf)) {
4317 DEBUG_L(PerlIO_printf(Perl_debug_log,
4318 "_mem_collxfrm: Couldn't realloc %zu bytes\n", xAlloc));
4328 if (DEBUG_Lv_TEST || debug_initialization) {
4330 print_collxfrm_input_and_return(s, s + len, xlen, utf8);
4331 PerlIO_printf(Perl_debug_log, "Its xfrm is:");
4332 PerlIO_printf(Perl_debug_log, "%s\n",
4333 _byte_dump_string((U8 *) xbuf + COLLXFRM_HDR_LEN,
4339 /* Free up unneeded space; retain ehough for trailing NUL */
4340 Renew(xbuf, COLLXFRM_HDR_LEN + *xlen + 1, char);
4342 if (s != input_string) {
4352 if (DEBUG_Lv_TEST || debug_initialization) {
4353 print_collxfrm_input_and_return(s, s + len, NULL, utf8);
4359 if (s != input_string) {
4370 S_print_collxfrm_input_and_return(pTHX_
4371 const char * const s,
4372 const char * const e,
4373 const STRLEN * const xlen,
4377 PERL_ARGS_ASSERT_PRINT_COLLXFRM_INPUT_AND_RETURN;
4379 PerlIO_printf(Perl_debug_log, "_mem_collxfrm[%" UVuf "]: returning ",
4380 (UV)PL_collation_ix);
4382 PerlIO_printf(Perl_debug_log, "%zu", *xlen);
4385 PerlIO_printf(Perl_debug_log, "NULL");
4387 PerlIO_printf(Perl_debug_log, " for locale '%s', string='",
4389 print_bytes_for_locale(s, e, is_utf8);
4391 PerlIO_printf(Perl_debug_log, "'\n");
4394 # endif /* DEBUGGING */
4395 #endif /* USE_LOCALE_COLLATE */
4400 S_print_bytes_for_locale(pTHX_
4401 const char * const s,
4402 const char * const e,
4406 bool prev_was_printable = TRUE;
4407 bool first_time = TRUE;
4409 PERL_ARGS_ASSERT_PRINT_BYTES_FOR_LOCALE;
4413 ? utf8_to_uvchr_buf((U8 *) t, e, NULL)
4416 if (! prev_was_printable) {
4417 PerlIO_printf(Perl_debug_log, " ");
4419 PerlIO_printf(Perl_debug_log, "%c", (U8) cp);
4420 prev_was_printable = TRUE;
4424 PerlIO_printf(Perl_debug_log, " ");
4426 PerlIO_printf(Perl_debug_log, "%02" UVXf, cp);
4427 prev_was_printable = FALSE;
4429 t += (is_utf8) ? UTF8SKIP(t) : 1;
4434 # endif /* #ifdef DEBUGGING */
4437 S_switch_category_locale_to_template(pTHX_ const int switch_category, const int template_category, const char * template_locale)
4439 /* Changes the locale for LC_'switch_category" to that of
4440 * LC_'template_category', if they aren't already the same. If not NULL,
4441 * 'template_locale' is the locale that 'template_category' is in.
4443 * Returns a copy of the name of the original locale for 'switch_category'
4444 * so can be switched back to with the companion function
4445 * restore_switched_locale(), (NULL if no restoral is necessary.) */
4447 char * restore_to_locale = NULL;
4449 if (switch_category == template_category) { /* No changes needed */
4453 /* Find the original locale of the category we may need to change, so that
4454 * it can be restored to later */
4455 restore_to_locale = stdize_locale(savepv(do_setlocale_r(switch_category,
4457 if (! restore_to_locale) {
4459 "panic: %s: %d: Could not find current %s locale, errno=%d\n",
4460 __FILE__, __LINE__, category_name(switch_category), errno);
4463 /* If the locale of the template category wasn't passed in, find it now */
4464 if (template_locale == NULL) {
4465 template_locale = do_setlocale_r(template_category, NULL);
4466 if (! template_locale) {
4468 "panic: %s: %d: Could not find current %s locale, errno=%d\n",
4469 __FILE__, __LINE__, category_name(template_category), errno);
4473 /* It the locales are the same, there's nothing to do */
4474 if (strEQ(restore_to_locale, template_locale)) {
4475 Safefree(restore_to_locale);
4477 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "%s locale unchanged as %s\n",
4478 category_name(switch_category), template_locale));
4483 /* Finally, change the locale to the template one */
4484 if (! do_setlocale_r(switch_category, template_locale)) {
4486 "panic: %s: %d: Could not change %s locale to %s, errno=%d\n",
4487 __FILE__, __LINE__, category_name(switch_category),
4488 template_locale, errno);
4491 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "%s locale switched to %s\n",
4492 category_name(switch_category), template_locale));
4494 return restore_to_locale;
4498 S_restore_switched_locale(pTHX_ const int category, const char * const original_locale)
4500 /* Restores the locale for LC_'category' to 'original_locale' (which is a
4501 * copy that will be freed by this function), or do nothing if the latter
4502 * parameter is NULL */
4504 if (original_locale == NULL) {
4508 if (! do_setlocale_r(category, original_locale)) {
4510 "panic: %s: %d: setlocale %s restore to %s failed, errno=%d\n",
4512 category_name(category), original_locale, errno);
4515 Safefree(original_locale);
4518 /* is_cur_LC_category_utf8 uses a small char buffer to avoid malloc/free */
4519 #define CUR_LC_BUFFER_SIZE 64
4522 Perl__is_cur_LC_category_utf8(pTHX_ int category)
4524 /* Returns TRUE if the current locale for 'category' is UTF-8; FALSE
4525 * otherwise. 'category' may not be LC_ALL. If the platform doesn't have
4526 * nl_langinfo(), nor MB_CUR_MAX, this employs a heuristic, which hence
4527 * could give the wrong result. The result will very likely be correct for
4528 * languages that have commonly used non-ASCII characters, but for notably
4529 * English, it comes down to if the locale's name ends in something like
4530 * "UTF-8". It errs on the side of not being a UTF-8 locale.
4532 * If the platform is early C89, not containing mbtowc(), or we are
4533 * compiled to not pay attention to LC_CTYPE, this employs heuristics.
4534 * These work very well for non-Latin locales or those whose currency
4535 * symbol isn't a '$' nor plain ASCII text. But without LC_CTYPE and at
4536 * least MB_CUR_MAX, English locales with an ASCII currency symbol depend
4537 * on the name containing UTF-8 or not. */
4539 /* Name of current locale corresponding to the input category */
4540 const char *save_input_locale = NULL;
4542 bool is_utf8 = FALSE; /* The return value */
4544 /* The variables below are for the cache of previous lookups using this
4545 * function. The cache is a C string, described at the definition for
4546 * 'C_and_POSIX_utf8ness'.
4548 * The first part of the cache is fixed, for the C and POSIX locales. The
4549 * varying part starts just after them. */
4550 char * utf8ness_cache = PL_locale_utf8ness + STRLENs(C_and_POSIX_utf8ness);
4552 Size_t utf8ness_cache_size; /* Size of the varying portion */
4553 Size_t input_name_len; /* Length in bytes of save_input_locale */
4554 Size_t input_name_len_with_overhead; /* plus extra chars used to store
4555 the name in the cache */
4556 char * delimited; /* The name plus the delimiters used to store
4558 char buffer[CUR_LC_BUFFER_SIZE]; /* small buffer */
4559 char * name_pos; /* position of 'delimited' in the cache, or 0
4565 assert(category != LC_ALL);
4569 /* Get the desired category's locale */
4570 save_input_locale = stdize_locale(savepv(do_setlocale_r(category, NULL)));
4571 if (! save_input_locale) {
4573 "panic: %s: %d: Could not find current %s locale, errno=%d\n",
4574 __FILE__, __LINE__, category_name(category), errno);
4577 DEBUG_L(PerlIO_printf(Perl_debug_log,
4578 "Current locale for %s is %s\n",
4579 category_name(category), save_input_locale));
4581 input_name_len = strlen(save_input_locale);
4583 /* In our cache, each name is accompanied by two delimiters and a single
4585 input_name_len_with_overhead = input_name_len + 3;
4587 if ( input_name_len_with_overhead <= CUR_LC_BUFFER_SIZE ) {
4588 /* we can use the buffer, avoid a malloc */
4590 } else { /* need a malloc */
4591 /* Allocate and populate space for a copy of the name surrounded by the
4593 Newx(delimited, input_name_len_with_overhead, char);
4596 delimited[0] = UTF8NESS_SEP[0];
4597 Copy(save_input_locale, delimited + 1, input_name_len, char);
4598 delimited[input_name_len+1] = UTF8NESS_PREFIX[0];
4599 delimited[input_name_len+2] = '\0';
4601 /* And see if that is in the cache */
4602 name_pos = instr(PL_locale_utf8ness, delimited);
4604 is_utf8 = *(name_pos + input_name_len_with_overhead - 1) - '0';
4608 if (DEBUG_Lv_TEST || debug_initialization) {
4609 PerlIO_printf(Perl_debug_log, "UTF8ness for locale %s=%d, \n",
4610 save_input_locale, is_utf8);
4615 /* And, if not already in that position, move it to the beginning of
4616 * the non-constant portion of the list, since it is the most recently
4617 * used. (We don't have to worry about overflow, since just moving
4618 * existing names around) */
4619 if (name_pos > utf8ness_cache) {
4620 Move(utf8ness_cache,
4621 utf8ness_cache + input_name_len_with_overhead,
4622 name_pos - utf8ness_cache, char);
4625 input_name_len_with_overhead - 1, char);
4626 utf8ness_cache[input_name_len_with_overhead - 1] = is_utf8 + '0';
4629 /* free only when not using the buffer */
4630 if ( delimited != buffer ) Safefree(delimited);
4631 Safefree(save_input_locale);
4635 /* Here we don't have stored the utf8ness for the input locale. We have to
4638 # if defined(USE_LOCALE_CTYPE) \
4639 && ( defined(HAS_NL_LANGINFO) \
4640 || (defined(HAS_MBTOWC) || defined(HAS_MBRTOWC)))
4643 const char *original_ctype_locale
4644 = switch_category_locale_to_template(LC_CTYPE,
4648 /* Here the current LC_CTYPE is set to the locale of the category whose
4649 * information is desired. This means that nl_langinfo() and mbtowc()
4650 * should give the correct results */
4652 # ifdef MB_CUR_MAX /* But we can potentially rule out UTF-8ness, avoiding
4653 calling the functions if we have this */
4655 /* Standard UTF-8 needs at least 4 bytes to represent the maximum
4656 * Unicode code point. */
4658 DEBUG_L(PerlIO_printf(Perl_debug_log, "%s: %d: MB_CUR_MAX=%d\n",
4659 __FILE__, __LINE__, (int) MB_CUR_MAX));
4660 if ((unsigned) MB_CUR_MAX < STRLENs(MAX_UNICODE_UTF8)) {
4662 restore_switched_locale(LC_CTYPE, original_ctype_locale);
4663 goto finish_and_return;
4667 # if defined(HAS_NL_LANGINFO)
4669 { /* The task is easiest if the platform has this POSIX 2001 function.
4670 Except on some platforms it can wrongly return "", so have to have
4671 a fallback. And it can return that it's UTF-8, even if there are
4672 variances from that. For example, Turkish locales may use the
4673 alternate dotted I rules, and sometimes it appears to be a
4674 defective locale definition. XXX We should probably check for
4675 these in the Latin1 range and warn (but on glibc, requires
4676 iswalnum() etc. due to their not handling 80-FF correctly */
4677 const char *codeset = my_nl_langinfo(CODESET, FALSE);
4678 /* FALSE => already in dest locale */
4680 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
4681 "\tnllanginfo returned CODESET '%s'\n", codeset));
4683 if (codeset && strNE(codeset, "")) {
4685 /* If the implementation of foldEQ() somehow were
4686 * to change to not go byte-by-byte, this could
4687 * read past end of string, as only one length is
4688 * checked. But currently, a premature NUL will
4689 * compare false, and it will stop there */
4690 is_utf8 = cBOOL( foldEQ(codeset, STR_WITH_LEN("UTF-8"))
4691 || foldEQ(codeset, STR_WITH_LEN("UTF8")));
4693 DEBUG_L(PerlIO_printf(Perl_debug_log,
4694 "\tnllanginfo returned CODESET '%s'; ?UTF8 locale=%d\n",
4696 restore_switched_locale(LC_CTYPE, original_ctype_locale);
4697 goto finish_and_return;
4702 # if defined(HAS_MBTOWC) || defined(HAS_MBRTOWC)
4703 /* We can see if this is a UTF-8-like locale if have mbtowc(). It was a
4704 * late adder to C89, so very likely to have it. However, testing has
4705 * shown that, like nl_langinfo() above, there are locales that are not
4706 * strictly UTF-8 that this will return that they are */
4713 # if defined(HAS_MBRTOWC) && defined(USE_ITHREADS)
4719 /* mbrtowc() and mbtowc() convert a byte string to a wide
4720 * character. Feed a byte string to one of them and check that the
4721 * result is the expected Unicode code point */
4723 # if defined(HAS_MBRTOWC) && defined(USE_ITHREADS)
4724 /* Prefer this function if available, as it's reentrant */
4726 memzero(&ps, sizeof(ps));;
4727 PERL_UNUSED_RESULT(mbrtowc(&wc, NULL, 0, &ps)); /* Reset any shift
4730 len = mbrtowc(&wc, STR_WITH_LEN(REPLACEMENT_CHARACTER_UTF8), &ps);
4736 PERL_UNUSED_RESULT(mbtowc(&wc, NULL, 0));/* Reset any shift state */
4738 len = mbtowc(&wc, STR_WITH_LEN(REPLACEMENT_CHARACTER_UTF8));
4745 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
4746 "\treturn from mbtowc; len=%d; code_point=%x; errno=%d\n",
4747 len, (unsigned int) wc, GET_ERRNO));
4749 is_utf8 = cBOOL( len == STRLENs(REPLACEMENT_CHARACTER_UTF8)
4750 && wc == (wchar_t) UNICODE_REPLACEMENT);
4755 restore_switched_locale(LC_CTYPE, original_ctype_locale);
4756 goto finish_and_return;
4761 /* Here, we must have a C89 compiler that doesn't have mbtowc(). Next
4762 * try looking at the currency symbol to see if it disambiguates
4763 * things. Often that will be in the native script, and if the symbol
4764 * isn't in UTF-8, we know that the locale isn't. If it is non-ASCII
4765 * UTF-8, we infer that the locale is too, as the odds of a non-UTF8
4766 * string being valid UTF-8 are quite small */
4768 # ifdef USE_LOCALE_MONETARY
4770 /* If have LC_MONETARY, we can look at the currency symbol. Often that
4771 * will be in the native script. We do this one first because there is
4772 * just one string to examine, so potentially avoids work */
4775 const char *original_monetary_locale
4776 = switch_category_locale_to_template(LC_MONETARY,
4779 bool only_ascii = FALSE;
4780 const U8 * currency_string
4781 = (const U8 *) my_nl_langinfo(CRNCYSTR, FALSE);
4782 /* 2nd param not relevant for this item */
4783 const U8 * first_variant;
4785 assert( *currency_string == '-'
4786 || *currency_string == '+'
4787 || *currency_string == '.');
4791 if (is_utf8_invariant_string_loc(currency_string, 0, &first_variant))
4793 DEBUG_L(PerlIO_printf(Perl_debug_log, "Couldn't get currency symbol for %s, or contains only ASCII; can't use for determining if UTF-8 locale\n", save_input_locale));
4797 is_utf8 = is_strict_utf8_string(first_variant, 0);
4800 restore_switched_locale(LC_MONETARY, original_monetary_locale);
4804 /* It isn't a UTF-8 locale if the symbol is not legal UTF-8;
4805 * otherwise assume the locale is UTF-8 if and only if the symbol
4806 * is non-ascii UTF-8. */
4807 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "\t?Currency symbol for %s is UTF-8=%d\n",
4808 save_input_locale, is_utf8));
4809 goto finish_and_return;
4813 # endif /* USE_LOCALE_MONETARY */
4814 # if defined(HAS_STRFTIME) && defined(USE_LOCALE_TIME)
4816 /* Still haven't found a non-ASCII string to disambiguate UTF-8 or not. Try
4817 * the names of the months and weekdays, timezone, and am/pm indicator */
4819 const char *original_time_locale
4820 = switch_category_locale_to_template(LC_TIME,
4824 bool is_dst = FALSE;
4828 char * formatted_time;
4830 /* Here the current LC_TIME is set to the locale of the category
4831 * whose information is desired. Look at all the days of the week and
4832 * month names, and the timezone and am/pm indicator for UTF-8 variant
4833 * characters. The first such a one found will tell us if the locale
4834 * is UTF-8 or not */
4836 for (i = 0; i < 7 + 12; i++) { /* 7 days; 12 months */
4837 formatted_time = my_strftime("%A %B %Z %p",
4838 0, 0, hour, dom, month, 2012 - 1900, 0, 0, is_dst);
4839 if ( ! formatted_time
4840 || is_utf8_invariant_string((U8 *) formatted_time, 0))
4843 /* Here, we didn't find a non-ASCII. Try the next time through
4844 * with the complemented dst and am/pm, and try with the next
4845 * weekday. After we have gotten all weekdays, try the next
4848 hour = (hour + 12) % 24;
4856 /* Here, we have a non-ASCII. Return TRUE is it is valid UTF8;
4857 * false otherwise. But first, restore LC_TIME to its original
4858 * locale if we changed it */
4859 restore_switched_locale(LC_TIME, original_time_locale);
4861 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "\t?time-related strings for %s are UTF-8=%d\n",
4863 is_utf8_string((U8 *) formatted_time, 0)));
4864 is_utf8 = is_utf8_string((U8 *) formatted_time, 0);
4865 goto finish_and_return;
4868 /* Falling off the end of the loop indicates all the names were just
4869 * ASCII. Go on to the next test. If we changed it, restore LC_TIME
4870 * to its original locale */
4871 restore_switched_locale(LC_TIME, original_time_locale);
4872 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "All time-related words for %s contain only ASCII; can't use for determining if UTF-8 locale\n", save_input_locale));
4877 # if 0 && defined(USE_LOCALE_MESSAGES) && defined(HAS_SYS_ERRLIST)
4879 /* This code is ifdefd out because it was found to not be necessary in testing
4880 * on our dromedary test machine, which has over 700 locales. There, this
4881 * added no value to looking at the currency symbol and the time strings. I
4882 * left it in so as to avoid rewriting it if real-world experience indicates
4883 * that dromedary is an outlier. Essentially, instead of returning abpve if we
4884 * haven't found illegal utf8, we continue on and examine all the strerror()
4885 * messages on the platform for utf8ness. If all are ASCII, we still don't
4886 * know the answer; but otherwise we have a pretty good indication of the
4887 * utf8ness. The reason this doesn't help much is that the messages may not
4888 * have been translated into the locale. The currency symbol and time strings
4889 * are much more likely to have been translated. */
4892 bool non_ascii = FALSE;
4893 const char *original_messages_locale
4894 = switch_category_locale_to_template(LC_MESSAGES,
4897 const char * errmsg = NULL;
4899 /* Here the current LC_MESSAGES is set to the locale of the category
4900 * whose information is desired. Look through all the messages. We
4901 * can't use Strerror() here because it may expand to code that
4902 * segfaults in miniperl */
4904 for (e = 0; e <= sys_nerr; e++) {
4906 errmsg = sys_errlist[e];
4907 if (errno || !errmsg) {
4910 errmsg = savepv(errmsg);
4911 if (! is_utf8_invariant_string((U8 *) errmsg, 0)) {
4913 is_utf8 = is_utf8_string((U8 *) errmsg, 0);
4919 restore_switched_locale(LC_MESSAGES, original_messages_locale);
4923 /* Any non-UTF-8 message means not a UTF-8 locale; if all are valid,
4924 * any non-ascii means it is one; otherwise we assume it isn't */
4925 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "\t?error messages for %s are UTF-8=%d\n",
4928 goto finish_and_return;
4931 DEBUG_L(PerlIO_printf(Perl_debug_log, "All error messages for %s contain only ASCII; can't use for determining if UTF-8 locale\n", save_input_locale));
4935 # ifndef EBCDIC /* On os390, even if the name ends with "UTF-8', it isn't a
4938 /* As a last resort, look at the locale name to see if it matches
4939 * qr/UTF -? * 8 /ix, or some other common locale names. This "name", the
4940 * return of setlocale(), is actually defined to be opaque, so we can't
4941 * really rely on the absence of various substrings in the name to indicate
4942 * its UTF-8ness, but if it has UTF8 in the name, it is extremely likely to
4943 * be a UTF-8 locale. Similarly for the other common names */
4946 const Size_t final_pos = strlen(save_input_locale) - 1;
4948 if (final_pos >= 3) {
4949 const char *name = save_input_locale;
4951 /* Find next 'U' or 'u' and look from there */
4952 while ((name += strcspn(name, "Uu") + 1)
4953 <= save_input_locale + final_pos - 2)
4955 if ( isALPHA_FOLD_NE(*name, 't')
4956 || isALPHA_FOLD_NE(*(name + 1), 'f'))
4961 if (*(name) == '-') {
4962 if ((name > save_input_locale + final_pos - 1)) {
4967 if (*(name) == '8') {
4968 DEBUG_L(PerlIO_printf(Perl_debug_log,
4969 "Locale %s ends with UTF-8 in name\n",
4970 save_input_locale));
4972 goto finish_and_return;
4975 DEBUG_L(PerlIO_printf(Perl_debug_log,
4976 "Locale %s doesn't end with UTF-8 in name\n",
4977 save_input_locale));
4982 /* http://msdn.microsoft.com/en-us/library/windows/desktop/dd317756.aspx */
4983 if (memENDs(save_input_locale, final_pos, "65001")) {
4984 DEBUG_L(PerlIO_printf(Perl_debug_log,
4985 "Locale %s ends with 65001 in name, is UTF-8 locale\n",
4986 save_input_locale));
4988 goto finish_and_return;
4995 /* Other common encodings are the ISO 8859 series, which aren't UTF-8. But
4996 * since we are about to return FALSE anyway, there is no point in doing
4997 * this extra work */
5000 if (instr(save_input_locale, "8859")) {
5001 DEBUG_L(PerlIO_printf(Perl_debug_log,
5002 "Locale %s has 8859 in name, not UTF-8 locale\n",
5003 save_input_locale));
5005 goto finish_and_return;
5009 DEBUG_L(PerlIO_printf(Perl_debug_log,
5010 "Assuming locale %s is not a UTF-8 locale\n",
5011 save_input_locale));
5014 # endif /* the code that is compiled when no modern LC_CTYPE */
5018 /* Cache this result so we don't have to go through all this next time. */
5019 utf8ness_cache_size = sizeof(PL_locale_utf8ness)
5020 - (utf8ness_cache - PL_locale_utf8ness);
5022 /* But we can't save it if it is too large for the total space available */
5023 if (LIKELY(input_name_len_with_overhead < utf8ness_cache_size)) {
5024 Size_t utf8ness_cache_len = strlen(utf8ness_cache);
5026 /* Here it can fit, but we may need to clear out the oldest cached
5027 * result(s) to do so. Check */
5028 if (utf8ness_cache_len + input_name_len_with_overhead
5029 >= utf8ness_cache_size)
5031 /* Here we have to clear something out to make room for this.
5032 * Start looking at the rightmost place where it could fit and find
5033 * the beginning of the entry that extends past that. */
5034 char * cutoff = (char *) my_memrchr(utf8ness_cache,
5037 - input_name_len_with_overhead);
5040 assert(cutoff >= utf8ness_cache);
5042 /* This and all subsequent entries must be removed */
5044 utf8ness_cache_len = strlen(utf8ness_cache);
5047 /* Make space for the new entry */
5048 Move(utf8ness_cache,
5049 utf8ness_cache + input_name_len_with_overhead,
5050 utf8ness_cache_len + 1 /* Incl. trailing NUL */, char);
5053 Copy(delimited, utf8ness_cache, input_name_len_with_overhead - 1, char);
5054 utf8ness_cache[input_name_len_with_overhead - 1] = is_utf8 + '0';
5056 if ((PL_locale_utf8ness[strlen(PL_locale_utf8ness)-1] & ~1) != '0') {
5058 "panic: %s: %d: Corrupt utf8ness_cache=%s\nlen=%zu,"
5059 " inserted_name=%s, its_len=%zu\n",
5061 PL_locale_utf8ness, strlen(PL_locale_utf8ness),
5062 delimited, input_name_len_with_overhead);
5068 if (DEBUG_Lv_TEST) {
5069 const char * s = PL_locale_utf8ness;
5071 /* Audit the structure */
5072 while (s < PL_locale_utf8ness + strlen(PL_locale_utf8ness)) {
5075 if (*s != UTF8NESS_SEP[0]) {
5077 "panic: %s: %d: Corrupt utf8ness_cache: missing"
5078 " separator %.*s<-- HERE %s\n",
5080 (int) (s - PL_locale_utf8ness), PL_locale_utf8ness,
5084 e = strchr(s, UTF8NESS_PREFIX[0]);
5086 e = PL_locale_utf8ness + strlen(PL_locale_utf8ness);
5088 "panic: %s: %d: Corrupt utf8ness_cache: missing"
5089 " separator %.*s<-- HERE %s\n",
5091 (int) (e - PL_locale_utf8ness), PL_locale_utf8ness,
5095 if (*e != '0' && *e != '1') {
5097 "panic: %s: %d: Corrupt utf8ness_cache: utf8ness"
5098 " must be [01] %.*s<-- HERE %s\n",
5100 (int) (e + 1 - PL_locale_utf8ness),
5101 PL_locale_utf8ness, e + 1);
5103 if (ninstr(PL_locale_utf8ness, s, s-1, e)) {
5105 "panic: %s: %d: Corrupt utf8ness_cache: entry"
5106 " has duplicate %.*s<-- HERE %s\n",
5108 (int) (e - PL_locale_utf8ness), PL_locale_utf8ness,
5115 if (DEBUG_Lv_TEST || debug_initialization) {
5117 PerlIO_printf(Perl_debug_log,
5118 "PL_locale_utf8ness is now %s; returning %d\n",
5119 PL_locale_utf8ness, is_utf8);
5124 /* free only when not using the buffer */
5125 if ( delimited != buffer ) Safefree(delimited);
5126 Safefree(save_input_locale);
5133 Perl__is_in_locale_category(pTHX_ const bool compiling, const int category)
5135 /* Internal function which returns if we are in the scope of a pragma that
5136 * enables the locale category 'category'. 'compiling' should indicate if
5137 * this is during the compilation phase (TRUE) or not (FALSE). */
5139 const COP * const cop = (compiling) ? &PL_compiling : PL_curcop;
5141 SV *these_categories = cop_hints_fetch_pvs(cop, "locale", 0);
5142 if (! these_categories || these_categories == &PL_sv_placeholder) {
5146 /* The pseudo-category 'not_characters' is -1, so just add 1 to each to get
5147 * a valid unsigned */
5148 assert(category >= -1);
5149 return cBOOL(SvUV(these_categories) & (1U << (category + 1)));
5153 Perl_my_strerror(pTHX_ const int errnum)
5155 /* Returns a mortalized copy of the text of the error message associated
5156 * with 'errnum'. It uses the current locale's text unless the platform
5157 * doesn't have the LC_MESSAGES category or we are not being called from
5158 * within the scope of 'use locale'. In the former case, it uses whatever
5159 * strerror returns; in the latter case it uses the text from the C locale.
5161 * The function just calls strerror(), but temporarily switches, if needed,
5162 * to the C locale */
5166 #ifndef USE_LOCALE_MESSAGES
5168 /* If platform doesn't have messages category, we don't do any switching to
5169 * the C locale; we just use whatever strerror() returns */
5171 errstr = savepv(Strerror(errnum));
5173 #else /* Has locale messages */
5175 const bool within_locale_scope = IN_LC(LC_MESSAGES);
5177 # ifndef USE_ITHREADS
5179 /* This function is trivial without threads. */
5180 if (within_locale_scope) {
5181 errstr = savepv(strerror(errnum));
5184 const char * save_locale = savepv(do_setlocale_c(LC_MESSAGES, NULL));
5186 do_setlocale_c(LC_MESSAGES, "C");
5187 errstr = savepv(strerror(errnum));
5188 do_setlocale_c(LC_MESSAGES, save_locale);
5189 Safefree(save_locale);
5192 # elif defined(USE_POSIX_2008_LOCALE) \
5193 && defined(HAS_STRERROR_L)
5195 /* This function is also trivial if we don't have to worry about thread
5196 * safety and have strerror_l(), as it handles the switch of locales so we
5197 * don't have to deal with that. We don't have to worry about thread
5198 * safety if strerror_r() is also available. Both it and strerror_l() are
5199 * thread-safe. Plain strerror() isn't thread safe. But on threaded
5200 * builds when strerror_r() is available, the apparent call to strerror()
5201 * below is actually a macro that behind-the-scenes calls strerror_r(). */
5203 # ifdef HAS_STRERROR_R
5205 if (within_locale_scope) {
5206 errstr = savepv(strerror(errnum));
5209 errstr = savepv(strerror_l(errnum, PL_C_locale_obj));
5214 /* Here we have strerror_l(), but not strerror_r() and we are on a
5215 * threaded-build. We use strerror_l() for everything, constructing a
5216 * locale to pass to it if necessary */
5218 bool do_free = FALSE;
5219 locale_t locale_to_use;
5221 if (within_locale_scope) {
5222 locale_to_use = uselocale((locale_t) 0);
5223 if (locale_to_use == LC_GLOBAL_LOCALE) {
5224 locale_to_use = duplocale(LC_GLOBAL_LOCALE);
5228 else { /* Use C locale if not within 'use locale' scope */
5229 locale_to_use = PL_C_locale_obj;
5232 errstr = savepv(strerror_l(errnum, locale_to_use));
5235 freelocale(locale_to_use);
5239 # else /* Doesn't have strerror_l() */
5241 const char * save_locale = NULL;
5242 bool locale_is_C = FALSE;
5244 /* We have a critical section to prevent another thread from executing this
5245 * same code at the same time. (On thread-safe perls, the LOCK is a
5246 * no-op.) Since this is the only place in core that changes LC_MESSAGES
5247 * (unless the user has called setlocale(), this works to prevent races. */
5250 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
5251 "my_strerror called with errnum %d\n", errnum));
5252 if (! within_locale_scope) {
5253 save_locale = do_setlocale_c(LC_MESSAGES, NULL);
5254 if (! save_locale) {
5257 "panic: %s: %d: Could not find current LC_MESSAGES locale,"
5258 " errno=%d\n", __FILE__, __LINE__, errno);
5261 locale_is_C = isNAME_C_OR_POSIX(save_locale);
5263 /* Switch to the C locale if not already in it */
5264 if (! locale_is_C) {
5266 /* The setlocale() just below likely will zap 'save_locale', so
5268 save_locale = savepv(save_locale);
5269 if (! do_setlocale_c(LC_MESSAGES, "C")) {
5271 /* If, for some reason, the locale change failed, we
5272 * soldier on as best as possible under the circumstances,
5273 * using the current locale, and clear save_locale, so we
5274 * don't try to change back. On z/0S, all setlocale()
5275 * calls fail after you've created a thread. This is their
5276 * way of making sure the entire process is always a single
5277 * locale. This means that 'use locale' is always in place
5278 * for messages under these circumstances. */
5279 Safefree(save_locale);
5284 } /* end of ! within_locale_scope */
5286 DEBUG_Lv(PerlIO_printf(Perl_debug_log, "%s: %d: WITHIN locale scope\n",
5287 __FILE__, __LINE__));
5290 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
5291 "Any locale change has been done; about to call Strerror\n"));
5292 errstr = savepv(Strerror(errnum));
5294 if (! within_locale_scope) {
5295 if (save_locale && ! locale_is_C) {
5296 if (! do_setlocale_c(LC_MESSAGES, save_locale)) {
5299 "panic: %s: %d: setlocale restore to '%s' failed, errno=%d\n",
5300 __FILE__, __LINE__, save_locale, errno);
5302 Safefree(save_locale);
5308 # endif /* End of doesn't have strerror_l */
5311 if (DEBUG_Lv_TEST) {
5312 PerlIO_printf(Perl_debug_log, "Strerror returned; saving a copy: '");
5313 print_bytes_for_locale(errstr, errstr + strlen(errstr), 0);
5314 PerlIO_printf(Perl_debug_log, "'\n");
5318 #endif /* End of does have locale messages */
5326 =for apidoc switch_to_global_locale
5328 On systems without locale support, or on typical single-threaded builds, or on
5329 platforms that do not support per-thread locale operations, this function does
5330 nothing. On such systems that do have locale support, only a locale global to
5331 the whole program is available.
5333 On multi-threaded builds on systems that do have per-thread locale operations,
5334 this function converts the thread it is running in to use the global locale.
5335 This is for code that has not yet or cannot be updated to handle multi-threaded
5336 locale operation. As long as only a single thread is so-converted, everything
5337 works fine, as all the other threads continue to ignore the global one, so only
5338 this thread looks at it.
5340 However, on Windows systems this isn't quite true prior to Visual Studio 15,
5341 at which point Microsoft fixed a bug. A race can occur if you use the
5342 following operations on earlier Windows platforms:
5346 =item L<POSIX::localeconv|POSIX/localeconv>
5348 =item L<I18N::Langinfo>, items C<CRNCYSTR> and C<THOUSEP>
5350 =item L<perlapi/Perl_langinfo>, items C<CRNCYSTR> and C<THOUSEP>
5354 The first item is not fixable (except by upgrading to a later Visual Studio
5355 release), but it would be possible to work around the latter two items by using
5356 the Windows API functions C<GetNumberFormat> and C<GetCurrencyFormat>; patches
5359 Without this function call, threads that use the L<C<setlocale(3)>> system
5360 function will not work properly, as all the locale-sensitive functions will
5361 look at the per-thread locale, and C<setlocale> will have no effect on this
5364 Perl code should convert to either call
5365 L<C<Perl_setlocale>|perlapi/Perl_setlocale> (which is a drop-in for the system
5366 C<setlocale>) or use the methods given in L<perlcall> to call
5367 L<C<POSIX::setlocale>|POSIX/setlocale>. Either one will transparently properly
5368 handle all cases of single- vs multi-thread, POSIX 2008-supported or not.
5370 Non-Perl libraries, such as C<gtk>, that call the system C<setlocale> can
5371 continue to work if this function is called before transferring control to the
5374 Upon return from the code that needs to use the global locale,
5375 L<C<sync_locale()>|perlapi/sync_locale> should be called to restore the safe
5376 multi-thread operation.
5382 Perl_switch_to_global_locale()
5385 #ifdef USE_THREAD_SAFE_LOCALE
5388 _configthreadlocale(_DISABLE_PER_THREAD_LOCALE);
5391 # ifdef HAS_QUERYLOCALE
5393 setlocale(LC_ALL, querylocale(LC_ALL_MASK, uselocale((locale_t) 0)));
5400 for (i = 0; i < LC_ALL_INDEX; i++) {
5401 setlocale(categories[i], do_setlocale_r(categories[i], NULL));
5407 uselocale(LC_GLOBAL_LOCALE);
5416 =for apidoc sync_locale
5418 L<C<Perl_setlocale>|perlapi/Perl_setlocale> can be used at any time to query or
5419 change the locale (though changing the locale is antisocial and dangerous on
5420 multi-threaded systems that don't have multi-thread safe locale operations.
5421 (See L<perllocale/Multi-threaded operation>). Using the system
5422 L<C<setlocale(3)>> should be avoided. Nevertheless, certain non-Perl libraries
5423 called from XS, such as C<Gtk> do so, and this can't be changed. When the
5424 locale is changed by XS code that didn't use
5425 L<C<Perl_setlocale>|perlapi/Perl_setlocale>, Perl needs to be told that the
5426 locale has changed. Use this function to do so, before returning to Perl.
5428 The return value is a boolean: TRUE if the global locale at the time of call
5429 was in effect; and FALSE if a per-thread locale was in effect. This can be
5430 used by the caller that needs to restore things as-they-were to decide whether
5432 L<C<Perl_switch_to_global_locale>|perlapi/switch_to_global_locale>.
5447 const char * newlocale;
5450 # ifdef USE_POSIX_2008_LOCALE
5452 bool was_in_global_locale = FALSE;
5453 locale_t cur_obj = uselocale((locale_t) 0);
5455 /* On Windows, unless the foreign code has turned off the thread-safe
5456 * locale setting, any plain setlocale() will have affected what we see, so
5457 * no need to worry. Otherwise, If the foreign code has done a plain
5458 * setlocale(), it will only affect the global locale on POSIX systems, but
5459 * will affect the */
5460 if (cur_obj == LC_GLOBAL_LOCALE) {
5462 # ifdef HAS_QUERY_LOCALE
5464 do_setlocale_c(LC_ALL, setlocale(LC_ALL, NULL));
5470 /* We can't trust that we can read the LC_ALL format on the
5471 * platform, so do them individually */
5472 for (i = 0; i < LC_ALL_INDEX; i++) {
5473 do_setlocale_r(categories[i], setlocale(categories[i], NULL));
5478 was_in_global_locale = TRUE;
5483 bool was_in_global_locale = TRUE;
5486 # ifdef USE_LOCALE_CTYPE
5488 newlocale = savepv(do_setlocale_c(LC_CTYPE, NULL));
5489 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
5490 "%s:%d: %s\n", __FILE__, __LINE__,
5491 setlocale_debug_string(LC_CTYPE, NULL, newlocale)));
5492 new_ctype(newlocale);
5493 Safefree(newlocale);
5495 # endif /* USE_LOCALE_CTYPE */
5496 # ifdef USE_LOCALE_COLLATE
5498 newlocale = savepv(do_setlocale_c(LC_COLLATE, NULL));
5499 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
5500 "%s:%d: %s\n", __FILE__, __LINE__,
5501 setlocale_debug_string(LC_COLLATE, NULL, newlocale)));
5502 new_collate(newlocale);
5503 Safefree(newlocale);
5506 # ifdef USE_LOCALE_NUMERIC
5508 newlocale = savepv(do_setlocale_c(LC_NUMERIC, NULL));
5509 DEBUG_Lv(PerlIO_printf(Perl_debug_log,
5510 "%s:%d: %s\n", __FILE__, __LINE__,
5511 setlocale_debug_string(LC_NUMERIC, NULL, newlocale)));
5512 new_numeric(newlocale);
5513 Safefree(newlocale);
5515 # endif /* USE_LOCALE_NUMERIC */
5517 return was_in_global_locale;
5523 #if defined(DEBUGGING) && defined(USE_LOCALE)
5526 S_setlocale_debug_string(const int category, /* category number,
5528 const char* const locale, /* locale name */
5530 /* return value from setlocale() when attempting to
5531 * set 'category' to 'locale' */
5532 const char* const retval)
5534 /* Returns a pointer to a NUL-terminated string in static storage with
5535 * added text about the info passed in. This is not thread safe and will
5536 * be overwritten by the next call, so this should be used just to
5537 * formulate a string to immediately print or savepv() on. */
5539 static char ret[256];
5541 my_strlcpy(ret, "setlocale(", sizeof(ret));
5542 my_strlcat(ret, category_name(category), sizeof(ret));
5543 my_strlcat(ret, ", ", sizeof(ret));
5546 my_strlcat(ret, "\"", sizeof(ret));
5547 my_strlcat(ret, locale, sizeof(ret));
5548 my_strlcat(ret, "\"", sizeof(ret));
5551 my_strlcat(ret, "NULL", sizeof(ret));
5554 my_strlcat(ret, ") returned ", sizeof(ret));
5557 my_strlcat(ret, "\"", sizeof(ret));
5558 my_strlcat(ret, retval, sizeof(ret));
5559 my_strlcat(ret, "\"", sizeof(ret));
5562 my_strlcat(ret, "NULL", sizeof(ret));
5565 assert(strlen(ret) < sizeof(ret));
5573 Perl_thread_locale_init()
5575 /* Called from a thread on startup*/
5577 #ifdef USE_THREAD_SAFE_LOCALE
5581 /* C starts the new thread in the global C locale. If we are thread-safe,
5582 * we want to not be in the global locale */
5584 DEBUG_L(PerlIO_printf(Perl_debug_log,
5585 "%s:%d: new thread, initial locale is %s; calling setlocale\n",
5586 __FILE__, __LINE__, setlocale(LC_ALL, NULL)));
5590 _configthreadlocale(_ENABLE_PER_THREAD_LOCALE);
5594 Perl_setlocale(LC_ALL, "C");
5602 Perl_thread_locale_term()
5604 /* Called from a thread as it gets ready to terminate */
5606 #ifdef USE_THREAD_SAFE_LOCALE
5608 /* C starts the new thread in the global C locale. If we are thread-safe,
5609 * we want to not be in the global locale */
5614 locale_t cur_obj = uselocale(LC_GLOBAL_LOCALE);
5615 if (cur_obj != LC_GLOBAL_LOCALE && cur_obj != PL_C_locale_obj) {
5616 freelocale(cur_obj);
5626 * ex: set ts=8 sts=4 sw=4 et: