https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120358
--- Comment #27 from Holger Hoffstätte <hol...@applied-asynchrony.com> ---
If I add lifetime-extending printfs to QStringTokenizerBase::next() like this:
--- qstringtokenizer.h 2025-07-04 17:56:28.523676630 +0200
+++ qstringtokenizer-printf.h 2025-07-04 17:56:03.998840901 +0200
@@ -389,11 +389,14 @@ auto QStringTokenizerBase<Haystack, Need
const auto ns = QtPrivate::Tok::tokenSize(m_needle);
state.start = state.end + ns;
state.extra = (ns == 0 ? 1 : 0);
+ __builtin_printf("\nsliced: '%s'\n",
result.toString().toStdString().c_str());
} else {
// token separator not found => return final element:
result = m_haystack.sliced(state.start);
+ __builtin_printf("\nfinal: '%s'\n",
result.toString().toStdString().c_str());
}
if ((m_sb & Qt::SkipEmptyParts) && result.isEmpty()) {
+ __builtin_printf("\nskipping empty result: '%s'\n",
result.toString().toStdString().c_str());
continue;
}
return {result, true, state};
..it starts to work. Removing any one of these printfs breaks the lifetime of
the
intermediate "Haystack result" and leads to garbage when trying to find the
last element.
