sv_setpv(PL_numeric_radix_sv, lc->decimal_point);
else
PL_numeric_radix_sv = newSVpv(lc->decimal_point, 0);
- if (! is_invariant_string((U8 *) lc->decimal_point, 0)
+ if (! is_utf8_invariant_string((U8 *) lc->decimal_point, 0)
&& is_utf8_string((U8 *) lc->decimal_point, 0)
&& _is_cur_LC_category_utf8(LC_NUMERIC))
{
* NUL */
char bad_chars_list[ (94 * 4) + (3 * 5) + 1 ];
- bool check_for_problems = ckWARN_d(WARN_LOCALE); /* No warnings means
- no check */
+ /* Don't check for problems if we are suppressing the warnings */
+ bool check_for_problems = ckWARN_d(WARN_LOCALE)
+ || UNLIKELY(DEBUG_L_TEST);
bool multi_byte_locale = FALSE; /* Assume is a single-byte locale
to start */
unsigned int bad_count = 0; /* Count of bad characters */
? bad_chars_list
: ""
);
- /* If we are actually in the scope of the locale, output the
- * message now. Otherwise we save it to be output at the first
- * operation using this locale, if that actually happens. Most
- * programs don't use locales, so they are immune to bad ones */
- if (IN_LC(LC_CTYPE)) {
+ /* If we are actually in the scope of the locale or are debugging,
+ * output the message now. If not in that scope, we save the
+ * message to be output at the first operation using this locale,
+ * if that actually happens. Most programs don't use locales, so
+ * they are immune to bad ones. */
+ if (IN_LC(LC_CTYPE) || UNLIKELY(DEBUG_L_TEST)) {
/* We have to save 'newctype' because the setlocale() just
* below may destroy it. The next setlocale() further down
/* The '0' below suppresses a bogus gcc compiler warning */
Perl_warner(aTHX_ packWARN(WARN_LOCALE), SvPVX(PL_warn_locale), 0);
+
setlocale(LC_CTYPE, badlocale);
Safefree(badlocale);
- SvREFCNT_dec_NN(PL_warn_locale);
- PL_warn_locale = NULL;
+
+ if (IN_LC(LC_CTYPE)) {
+ SvREFCNT_dec_NN(PL_warn_locale);
+ PL_warn_locale = NULL;
+ }
}
}
}
PL_collxfrm_base = 0;
PL_collxfrm_mult = 2;
PL_in_utf8_COLLATE_locale = FALSE;
- *PL_strxfrm_min_char = '\0';
+ PL_strxfrm_NUL_replacement = '\0';
PL_strxfrm_max_cp = 0;
return;
}
}
PL_in_utf8_COLLATE_locale = _is_cur_LC_category_utf8(LC_COLLATE);
- *PL_strxfrm_min_char = '\0';
+ PL_strxfrm_NUL_replacement = '\0';
PL_strxfrm_max_cp = 0;
/* A locale collation definition includes primary, secondary, tertiary,
* This has the desired effect that strcmp() will look at the secondary
* or tertiary weights only if the strings compare equal at all higher
* priority weights. The spaces shown here, like in
- * "A¹B¹C¹ * A²B²C² "
+ * "A¹B¹C¹ A²B²C² "
* are not just for readability. In the general case, these must
* actually be bytes, which we will call here 'separator weights'; and
* they must be smaller than any other weight value, but since these
: NULL;
sl_result = my_setlocale(LC_MESSAGES, locale_param);
DEBUG_LOCALE_INIT(LC_MESSAGES, locale_param, sl_result);
- if (! sl_result)
+ if (! sl_result) {
setlocale_failure = TRUE;
}
# endif /* USE_LOCALE_MESSAGES */
{
char **e;
for (e = environ; *e; e++) {
- if (strnEQ(*e, "LC_", 3)
- && strnNE(*e, "LC_ALL=", 7)
+ if (strEQs(*e, "LC_")
+ && strNEs(*e, "LC_ALL=")
&& (p = strchr(*e, '=')))
PerlIO_printf(Perl_error_log, "\t%.*s = \"%s\",\n",
(int)(p - *e), *e, p + 1);
* This will give as good as possible results on strings that don't
* otherwise contain that character, but otherwise there may be
* less-than-perfect results with that character and NUL. This is
- * unavoidable unless we replace strxfrm with our own implementation.
- *
- * This is one of the few places in the perl core, where we can use
- * standard functions like strlen() and strcat(). It's because we're
- * looking for NULs. */
- if (s_strlen < len) {
+ * unavoidable unless we replace strxfrm with our own implementation. */
+ if (s_strlen < len) { /* Only execute if there is an embedded NUL */
char * e = s + len;
char * sans_nuls;
- STRLEN cur_min_char_len;
+ STRLEN sans_nuls_len;
+ STRLEN sans_nuls_pos;
int try_non_controls;
-
- /* If we don't know what control character sorts lowest for this
- * locale, find it */
- if (*PL_strxfrm_min_char == '\0') {
+ char this_replacement_char[] = "?\0"; /* Room for a two-byte string,
+ making sure 2nd byte is NUL.
+ */
+ STRLEN this_replacement_len;
+
+ /* If we don't know what non-NUL control character sorts lowest for
+ * this locale, find it */
+ if (PL_strxfrm_NUL_replacement == '\0') {
int j;
-#ifdef DEBUGGING
- U8 cur_min_cp = 1; /* The code point that sorts lowest, so far */
-#endif
- char * cur_min_x = NULL; /* And its xfrm, (except it also
+ char * cur_min_x = NULL; /* The min_char's xfrm, (except it also
includes the collation index
prefixed. */
DEBUG_Lv(PerlIO_printf(Perl_debug_log, "Looking to replace NUL\n"));
/* Unlikely, but it may be that no control will work to replace
- * NUL, in which case we instead look for any character */
+ * NUL, in which case we instead look for any character. Controls
+ * are preferred because collation order is, in general, context
+ * sensitive, with adjoining characters affecting the order, and
+ * controls are less likely to have such interactions, allowing the
+ * NUL-replacement to stand on its own. (Another way to look at it
+ * is to imagine what would happen if the NUL were replaced by a
+ * combining character; it wouldn't work out all that well.) */
for (try_non_controls = 0;
try_non_controls < 2;
try_non_controls++)
STRLEN x_len; /* length of 'x' */
STRLEN trial_len = 1;
- /* Create a 1 byte string of the current code point, but
- * with room to be 2 bytes */
- char cur_source[] = { (char) j, '\0' , '\0' };
-
- if (PL_in_utf8_COLLATE_locale) {
- if (! try_non_controls && ! isCNTRL_L1(j)) {
- continue;
- }
+ /* Create a 1 byte string of the current code point */
+ char cur_source[] = { (char) j, '\0' };
- /* If needs to be 2 bytes, find them */
- if (! UVCHR_IS_INVARIANT(j)) {
- char * d = cur_source;
- append_utf8_from_native_byte((U8) j, (U8 **) &d);
- trial_len = 2;
- }
- }
- else if (! try_non_controls && ! isCNTRL_LC(j)) {
+ if (! try_non_controls && (PL_in_utf8_COLLATE_locale)
+ ? ! isCNTRL_L1(j)
+ : ! isCNTRL_LC(j))
+ {
continue;
}
/* Then transform it */
x = _mem_collxfrm(cur_source, trial_len, &x_len,
- PL_in_utf8_COLLATE_locale);
+ 0 /* The string is not in UTF-8 */);
- /* Ignore any character that didn't successfully transform
+ /* Ignore any character that didn't successfully transform.
* */
if (! x) {
continue;
|| strLT(x + COLLXFRM_HDR_LEN,
cur_min_x + COLLXFRM_HDR_LEN))
{
- PL_strxfrm_min_char[0] = cur_source[0];
- PL_strxfrm_min_char[1] = cur_source[1];
- PL_strxfrm_min_char[2] = cur_source[2];
+ PL_strxfrm_NUL_replacement = j;
cur_min_x = x;
-#ifdef DEBUGGING
- cur_min_cp = j;
-#endif
}
else {
Safefree(x);
}
- } /* end of loop through all bytes */
+ } /* end of loop through all 255 characters */
+ /* Stop looking if found */
if (cur_min_x) {
break;
}
* character that works */
DEBUG_L(PerlIO_printf(Perl_debug_log,
"_mem_collxfrm: No control worked. Trying non-controls\n"));
- }
+ } /* End of loop to try first the controls, then any char */
if (! cur_min_x) {
DEBUG_L(PerlIO_printf(Perl_debug_log,
DEBUG_L(PerlIO_printf(Perl_debug_log,
"_mem_collxfrm: Replacing embedded NULs in locale %s with "
- "0x%02X\n", PL_collation_name, cur_min_cp));
+ "0x%02X\n", PL_collation_name, PL_strxfrm_NUL_replacement));
Safefree(cur_min_x);
+ } /* End of determining the character that is to replace NULs */
+
+ /* If the replacement is variant under UTF-8, it must match the
+ * UTF8-ness as the original */
+ if ( ! UVCHR_IS_INVARIANT(PL_strxfrm_NUL_replacement) && utf8) {
+ this_replacement_char[0] =
+ UTF8_EIGHT_BIT_HI(PL_strxfrm_NUL_replacement);
+ this_replacement_char[1] =
+ UTF8_EIGHT_BIT_LO(PL_strxfrm_NUL_replacement);
+ this_replacement_len = 2;
+ }
+ else {
+ this_replacement_char[0] = PL_strxfrm_NUL_replacement;
+ /* this_replacement_char[1] = '\0' was done at initialization */
+ this_replacement_len = 1;
}
/* The worst case length for the replaced string would be if every
* character in it is NUL. Multiply that by the length of each
* replacement, and allow for a trailing NUL */
- cur_min_char_len = strlen(PL_strxfrm_min_char);
- Newx(sans_nuls, (len * cur_min_char_len) + 1, char);
+ sans_nuls_len = (len * this_replacement_len) + 1;
+ Newx(sans_nuls, sans_nuls_len, char);
*sans_nuls = '\0';
+ sans_nuls_pos = 0;
/* Replace each NUL with the lowest collating control. Loop until have
* exhausted all the NULs */
while (s + s_strlen < e) {
- strcat(sans_nuls, s);
+ sans_nuls_pos = my_strlcat(sans_nuls + sans_nuls_pos,
+ s,
+ sans_nuls_len);
/* Do the actual replacement */
- strcat(sans_nuls, PL_strxfrm_min_char);
+ sans_nuls_pos = my_strlcat(sans_nuls + sans_nuls_pos,
+ this_replacement_char,
+ sans_nuls_len);
/* Move past the input NUL */
s += s_strlen + 1;
}
/* And add anything that trails the final NUL */
- strcat(sans_nuls, s);
+ my_strlcat(sans_nuls + sans_nuls_pos, s, sans_nuls_len);
/* Switch so below we transform this modified string */
s = sans_nuls;
len = strlen(s);
- }
+ } /* End of replacing NULs */
/* Make sure the UTF8ness of the string and locale match */
if (utf8 != PL_in_utf8_COLLATE_locale) {
{
STRLEN i;
STRLEN d= 0;
+ char * e = (char *) t + len;
for (i = 0; i < len; i+= UTF8SKIP(t + i)) {
U8 cur_char = t[i];
if (UTF8_IS_INVARIANT(cur_char)) {
s[d++] = cur_char;
}
- else if (UTF8_IS_DOWNGRADEABLE_START(cur_char)) {
+ else if (UTF8_IS_NEXT_CHAR_DOWNGRADEABLE(t + i, e)) {
s[d++] = EIGHT_BIT_UTF8_TO_NATIVE(cur_char, t[i+1]);
}
else { /* Replace illegal cp with highest collating
#ifdef DEBUGGING
-void
+STATIC void
S_print_collxfrm_input_and_return(pTHX_
const char * const s,
const char * const e,
PERL_ARGS_ASSERT_PRINT_COLLXFRM_INPUT_AND_RETURN;
- PerlIO_printf(Perl_debug_log, "_mem_collxfrm[%d]: returning ",
+ PerlIO_printf(Perl_debug_log, "_mem_collxfrm[%u]: returning ",
PL_collation_ix);
if (xlen) {
- PerlIO_printf(Perl_debug_log, "%"UVuf"", (UV) *xlen);
+ PerlIO_printf(Perl_debug_log, "%" UVuf, (UV) *xlen);
}
else {
PerlIO_printf(Perl_debug_log, "NULL");
if (! first_time) {
PerlIO_printf(Perl_debug_log, " ");
}
- PerlIO_printf(Perl_debug_log, "%02"UVXf"", cp);
+ PerlIO_printf(Perl_debug_log, "%02" UVXf, cp);
prev_was_printable = FALSE;
}
t += (is_utf8) ? UTF8SKIP(t) : 1;
lc = localeconv();
if (! lc
|| ! lc->currency_symbol
- || is_invariant_string((U8 *) lc->currency_symbol, 0))
+ || is_utf8_invariant_string((U8 *) lc->currency_symbol, 0))
{
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));
only_ascii = TRUE;
for (i = 0; i < 7 + 12; i++) { /* 7 days; 12 months */
formatted_time = my_strftime("%A %B %Z %p",
0, 0, hour, dom, month, 112, 0, 0, is_dst);
- if (! formatted_time || is_invariant_string((U8 *) formatted_time, 0)) {
+ if ( ! formatted_time
+ || is_utf8_invariant_string((U8 *) formatted_time, 0))
+ {
/* Here, we didn't find a non-ASCII. Try the next time through
* with the complemented dst and am/pm, and try with the next
break;
}
errmsg = savepv(errmsg);
- if (! is_invariant_string((U8 *) errmsg, 0)) {
+ if (! is_utf8_invariant_string((U8 *) errmsg, 0)) {
non_ascii = TRUE;
is_utf8 = is_utf8_string((U8 *) errmsg, 0);
break;
dVAR;
# ifdef USE_THREAD_SAFE_LOCALE
- locale_t save_locale;
+ locale_t save_locale = NULL;
# else
- char * save_locale;
+ char * save_locale = NULL;
bool locale_is_C = FALSE;
/* We have a critical section to prevent another thread from changing the
/* initialise to a non-null value to keep it out of BSS and so keep
* -DPERL_GLOBAL_STRUCT_PRIVATE happy */
- static char ret[128] = "x";
-
+ static char ret[128] = "If you can read this, thank your buggy C"
+ " library strlcpy(), and change your hints file"
+ " to undef it";
my_strlcpy(ret, "setlocale(", sizeof(ret));
switch (category) {