tech-pkg archive

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

Re: [PATCH] fvwm reproducibility patch

N <> writes:

> Greg Troxel transcribed 682 bytes:
>> My concern is with the pkgsrc maintenance burden of carrying patches, if
>> they belong upstream, and if they don't allow us to actually get
>> reproducible builds.  That was making me think about the general issue
>> and having a way to get reproducible builds for a few packages, before
>> we start applying changes to many packages.  As time goes on, the rest
>> of the world is moving to reproducible builds as normal, and being on
>> the leading edge without actually getting reproducibility seems to be
>> accepting pain for not actually getting the benefit.
>> As I tried to say earlier, for a small number of packages, I don't think
>> it hurts in any measurable way to do this, and I don't object.
> Sorry for waiting so long with an active reply: Would it be okay if
> I went ahead and applied the patch I mentioned in this thread
> (reproducible fvwm)? Upstream is not cooperative (on fvwm2) but acknowledges
> to be working on a future version of fvwm instead - every new feature
> gets into this future version, they take no patches for old version.
> I tried to upstream the patch back then (2018?).

I still don't really like this (carrying patches for a package, when
upstream won't take them, and they are for a feature in pkgsrc that is
not already well supported, so that one cannot say fvwm2 in pkgsrc is
broken for not being reproducible), but as long as there are only a few
packages with this treatment, I don't think it's reasonable to object -
so I will decline to object.

I think it's much more important to have the ability to do reproducible
builds for simple packages somehow integrated/documented in pkgsrc
infrastructure.  Perhaps that happened while I wasn't paying attention.

Home | Main Index | Thread Index | Old Index