Hi All, Fixed on mainline asr16-5542 and closed.
Cheers Paul On Fri, 21 Nov 2025 at 21:21, Paul Richard Thomas < [email protected]> wrote: > Hi Jerry, > > You didn't see either ChangeLog or patch because they weren't attached. > Remedied for everyone else. > > Cheers > > Paul > > > On Fri, 21 Nov 2025 at 20:07, Jerry D <[email protected]> wrote: > >> On 11/21/25 8:45 AM, Paul Richard Thomas wrote: >> > Hi All, >> > >> > I have to confess that I had never spotted the two PDT examples in the >> > F2018 standard. To my horror, they both fail. >> > >> > The second example involves finalizers of PDTs and is already caught by >> > 104650. I am onto that one as well :-) >> > >> > A PDT typespec in a function declaration such as; >> > >> > CONTAINS >> > TYPE(t(single)) FUNCTION real_to_t1(x) >> > REAL(single) x >> > real_to_t1%value = x >> > END FUNCTION >> > >> > causes gfc_use_derived to emit, "Derived type ‘pdtt_4’ at (1) is being >> > used before it is defined", whereas this; >> > >> > CONTAINS >> > FUNCTION real_to_t1(x) >> > TYPE(t(single)) real_to_t1 >> > REAL(single) x >> > real_to_t1%value = x >> > END FUNCTION >> > >> > compiled and ran successfully. The attached patch fixes this problem >> and >> > is explained by the ChangeLog and the comment in the patch. >> > >> > Regtest OK on FC43/x86_64. OK for mainline? >> > >> > Paul >> >> I did not see the patch attached, however I tested it off of bugzilla. >> >> OK from my point of view. >> >> Jerry >> >
