Change 14194 by jhi@alpha on 2002/01/11 20:04:26
Fix for
Subject: UTF-8 sprintf bug in bleadperl
From: [EMAIL PROTECTED] (Andreas J. Koenig)
Date: 11 Jan 2002 15:47:15 +0100
Message-ID: <[EMAIL PROTECTED]>
I have tests, have just to figure out where to put them
since op/sprintf looks a bit unfriendly for tests of
somewhat freer form.
Affected files ...
.... //depot/perl/doop.c#119 edit
.... //depot/perl/sv.c#506 edit
Differences ...
==== //depot/perl/doop.c#119 (text) ====
Index: perl/doop.c
--- perl/doop.c.~1~ Fri Jan 11 13:15:05 2002
+++ perl/doop.c Fri Jan 11 13:15:05 2002
@@ -697,6 +697,8 @@
char *pat = SvPV(*sarg, patlen);
bool do_taint = FALSE;
+ if (DO_UTF8(*sarg))
+ SvUTF8_on(sv);
sv_vsetpvfn(sv, pat, patlen, Null(va_list*), sarg + 1, len - 1, &do_taint);
SvSETMAGIC(sv);
if (do_taint)
==== //depot/perl/sv.c#506 (text) ====
Index: perl/sv.c
--- perl/sv.c.~1~ Fri Jan 11 13:15:05 2002
+++ perl/sv.c Fri Jan 11 13:15:05 2002
@@ -7655,6 +7655,7 @@
I32 svix = 0;
static char nullstr[] = "(null)";
SV *argsv = Nullsv;
+ bool has_utf8 = FALSE; /* has the result utf8? */
/* no matter what, this is a string now */
(void)SvPV_force(sv, origlen);
@@ -7688,13 +7689,16 @@
}
}
+ if (!args && svix < svmax && DO_UTF8(*svargs))
+ has_utf8 = TRUE;
+
patend = (char*)pat + patlen;
for (p = (char*)pat; p < patend; p = q) {
bool alt = FALSE;
bool left = FALSE;
bool vectorize = FALSE;
bool vectorarg = FALSE;
- bool vec_utf = FALSE;
+ bool vec_utf8 = FALSE;
char fill = ' ';
char plus = 0;
char intsize = 0;
@@ -7702,7 +7706,7 @@
STRLEN zeros = 0;
bool has_precis = FALSE;
STRLEN precis = 0;
- bool is_utf = FALSE;
+ bool is_utf8 = FALSE; /* is this item utf8? */
char esignbuf[4];
U8 utf8buf[UTF8_MAXLEN+1];
@@ -7827,17 +7831,17 @@
svargs[ewix ? ewix-1 : svix++] : &PL_sv_undef;
dotstr = SvPVx(vecsv, dotstrlen);
if (DO_UTF8(vecsv))
- is_utf = TRUE;
+ is_utf8 = TRUE;
}
if (args) {
vecsv = va_arg(*args, SV*);
vecstr = (U8*)SvPVx(vecsv,veclen);
- vec_utf = DO_UTF8(vecsv);
+ vec_utf8 = DO_UTF8(vecsv);
}
else if (efix ? efix <= svmax : svix < svmax) {
vecsv = svargs[efix ? efix-1 : svix++];
vecstr = (U8*)SvPVx(vecsv,veclen);
- vec_utf = DO_UTF8(vecsv);
+ vec_utf8 = DO_UTF8(vecsv);
}
else {
vecstr = (U8*)"";
@@ -7931,7 +7935,7 @@
&& !IN_BYTES) {
eptr = (char*)utf8buf;
elen = uvchr_to_utf8((U8*)eptr, uv) - utf8buf;
- is_utf = TRUE;
+ is_utf8 = TRUE;
}
else {
c = (char)uv;
@@ -7967,7 +7971,7 @@
if (width) { /* fudge width (can't fudge elen) */
width += elen - sv_len_utf8(argsv);
}
- is_utf = TRUE;
+ is_utf8 = TRUE;
}
}
goto string;
@@ -7983,7 +7987,7 @@
argsv = va_arg(*args, SV*);
eptr = SvPVx(argsv, elen);
if (DO_UTF8(argsv))
- is_utf = TRUE;
+ is_utf8 = TRUE;
string:
vectorize = FALSE;
@@ -8013,8 +8017,9 @@
STRLEN ulen;
if (!veclen)
continue;
- if (vec_utf)
- uv = utf8n_to_uvchr(vecstr, veclen, &ulen, UTF8_ALLOW_ANYUV);
+ if (vec_utf8)
+ uv = utf8n_to_uvchr(vecstr, veclen, &ulen,
+ UTF8_ALLOW_ANYUV);
else {
uv = *vecstr;
ulen = 1;
@@ -8098,8 +8103,9 @@
vector:
if (!veclen)
continue;
- if (vec_utf)
- uv = utf8n_to_uvchr(vecstr, veclen, &ulen, UTF8_ALLOW_ANYUV);
+ if (vec_utf8)
+ uv = utf8n_to_uvchr(vecstr, veclen, &ulen,
+ UTF8_ALLOW_ANYUV);
else {
uv = *vecstr;
ulen = 1;
@@ -8354,6 +8360,20 @@
*p++ = '0';
}
if (elen) {
+ if (is_utf8 != has_utf8) {
+ if (is_utf8) {
+ if (SvCUR(sv)) {
+ sv_utf8_upgrade(sv);
+ p = SvEND(sv);
+ }
+ }
+ else {
+ SV *nsv = sv_2mortal(newSVpvn(eptr, elen));
+ sv_utf8_upgrade(nsv);
+ eptr = SvPVX(nsv);
+ elen = SvCUR(nsv);
+ }
+ }
Copy(eptr, p, elen, char);
p += elen;
}
@@ -8369,7 +8389,9 @@
else
vectorize = FALSE; /* done iterating over vecstr */
}
- if (is_utf)
+ if (is_utf8)
+ has_utf8 = TRUE;
+ if (has_utf8)
SvUTF8_on(sv);
*p = '\0';
SvCUR(sv) = p - SvPVX(sv);
End of Patch.
