tech-pkg archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: lang/guile30 violates basic pkgsrc mechanisms to work around an ancient bug



https://gitlab.com/gnutls/gnutls/-/issues/996

I ran into this with Guile via GnuTLS back in 2020, and while my
communication could've been better (and I haven't fact checked
my replies today) the insistence on how a
valid target should look like was there as well (on the guile
compiler side). If you know Guile, it's doable to patch this
behavior, but who knows if they are open to accepting
patches on this behavior.

On 4/12/22 22:10, Joerg Sonnenberger wrote:
Am Tue, Apr 12, 2022 at 12:20:34PM -0700 schrieb Greg A. Woods:
So, clearly the target string must have three hyphen-separated parts,
and none of them can be null strings!  Though it's not as clearly
documented as one might like, this is how I would read the specification
for the values of '--host' et al:
You can have three or four components. There are two reasons for using
four components: you are pissed off by an OS kernel being more popular
than the userland or because you want to distinguish between important
ABI variants. Guile is plainly wrong to insist on the four component
variant. It is also wrong to insist on a non-empty vendor field -
outside a few rare cases, it is useless and forcing a random junk value
creates a lot more harm. Please fix Guile and don't change the rest of
the universe to match this misguided mess.

Joerg


Home | Main Index | Thread Index | Old Index