Change 18247 by jhi@lyta on 2002/12/06 13:40:10
More comments and robustness in pp_ucfirst() and pp_lcfirst().
Affected files ...
.... //depot/maint-5.8/perl/pp.c#4 edit
Differences ...
==== //depot/maint-5.8/perl/pp.c#4 (text) ====
Index: perl/pp.c
--- perl/pp.c#3~18126~ Sat Nov 9 11:00:47 2002
+++ perl/pp.c Fri Dec 6 05:40:10 2002
@@ -3325,7 +3325,9 @@
STRLEN slen;
SvGETMAGIC(sv);
- if (DO_UTF8(sv) && (s = (U8*)SvPV_nomg(sv, slen)) && slen && UTF8_IS_START(*s)) {
+ if (DO_UTF8(sv) &&
+ (s = (U8*)SvPV_nomg(sv, slen)) && slen &&
+ UTF8_IS_START(*s)) {
U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
STRLEN ulen;
STRLEN tculen;
@@ -3336,8 +3338,16 @@
if (!SvPADTMP(sv) || SvREADONLY(sv)) {
dTARGET;
+ /* slen is the byte length of the whole SV.
+ * ulen is the byte length of the original Unicode character
+ * stored as UTF-8 at s.
+ * tculen is the byte length of the freshly titlecased
+ * Unicode character stored as UTF-8 at tmpbuf.
+ * We first set the result to be the titlecased character,
+ * and then append the rest of the SV data. */
sv_setpvn(TARG, (char*)tmpbuf, tculen);
- sv_catpvn(TARG, (char*)(s + ulen), slen - ulen);
+ if (slen > ulen)
+ sv_catpvn(TARG, (char*)(s + ulen), slen - ulen);
SvUTF8_on(TARG);
SETs(TARG);
}
@@ -3377,7 +3387,9 @@
STRLEN slen;
SvGETMAGIC(sv);
- if (DO_UTF8(sv) && (s = (U8*)SvPV_nomg(sv, slen)) && slen && UTF8_IS_START(*s)) {
+ if (DO_UTF8(sv) &&
+ (s = (U8*)SvPV_nomg(sv, slen)) && slen &&
+ UTF8_IS_START(*s)) {
STRLEN ulen;
U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
U8 *tend;
@@ -3390,7 +3402,8 @@
if (!SvPADTMP(sv) || (STRLEN)(tend - tmpbuf) != ulen || SvREADONLY(sv)) {
dTARGET;
sv_setpvn(TARG, (char*)tmpbuf, tend - tmpbuf);
- sv_catpvn(TARG, (char*)(s + ulen), slen - ulen);
+ if (slen > ulen)
+ sv_catpvn(TARG, (char*)(s + ulen), slen - ulen);
SvUTF8_on(TARG);
SETs(TARG);
}
End of Patch.
