[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gentoo-dev
Subject:    Re: [gentoo-dev] Upstream remote-id types in package metadata
From:       Ulrich Mueller <ulm () gentoo ! org>
Date:       2022-05-25 9:23:59
Message-ID: u4k1eklsg () gentoo ! org
[Download RAW message or body]


>>>>> On Sun, 22 May 2022, Ulrich Mueller wrote:

>>>>> On Sun, 22 May 2022, Hanno Böck wrote:
>> I'm not sure about Google code.

>> While it's no longer an active site, it is still online in an
>> archived state. We maintain plenty of packages that have no active
>> upstream, and having a reference to an unmaintained previous upstream
>> which still allows downloading the code and the repo archive seems
>> like a good thing.

> The same is true for gitorious, but we have dropped those remote-ids
> from the tree nevertheless:
> https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=f8fd6bd07efee4d36a1babf55d6e69c7cb4a93d4

> However, I think that your point is valid. So the basic question is
> whether we should keep dead upstreams in that list, for archival
> purposes? If the answer is yes, then consequently we should also keep
> gitorious (and maybe revert above commit?).

For gitorious, I went through all packages where that remote-id was
dropped in the above mentioned commit. These packages were either
last-rited, or moved to different hosting. So restoring gitorious in
package metadata makes no sense for either of them.

Also, https://gitorious.org/ has a security certificate that expired in
early 2019.

Unless I see any objections here, I'll drop gitorious from the XML
schema and the DTD tomorrow.

Ulrich

["signature.asc" (application/pgp-signature)]

[prev in list] [next in list] [prev in thread] [next in thread]