[arch-dev-public] Moving base-devel dependencies to [core]
Hi all,
I notice when using devtools, the following packages from [extra] get pulled into our clean build root:
extra/gc extra/guile extra/libcroco extra/libsysprof-capture extra/libxml2
These are all dependencies of base-devel packages. I'd like to move them to [core] so they are captured by the signoff process.
Thank you for your lack of objection. :)
Allan
On 23/2/22 22:24, Allan McRae via arch-dev-public wrote:
Hi all,
I notice when using devtools, the following packages from [extra] get pulled into our clean build root:
extra/gc extra/guile extra/libcroco
libcroco is dead and gettext should no longer depend on it. So we can remove that from the list.
extra/libsysprof-capture extra/libxml2
These are all dependencies of base-devel packages. I'd like to move them to [core] so they are captured by the signoff process.
Thank you for your lack of objection. :)
On 24/2/22 00:03, Allan McRae via arch-dev-public wrote:
On 23/2/22 22:24, Allan McRae via arch-dev-public wrote:
Hi all,
I notice when using devtools, the following packages from [extra] get pulled into our clean build root:
extra/gc extra/guile extra/libsysprof-capture extra/libxml2
The above have not been moved into [core]. But now I notice...
community/tpm2-tss -> libsecret -> pinentry -> gnupg
So there is a base-devel package depending on a [community] package! This dependency was added the day of my original email. :)
The tpm2-tss package is co-maintained by Filipe and Jonas. When moved to [core], Filipe will lose access to package it (due to only Developers having access to [core]). So Jonas will become sole maintainer.
@Jonas & @Filipe: is this change in maintenance an issue?
Allan
On 2022-03-04 11:02, Allan McRae via arch-dev-public wrote:
The tpm2-tss package is co-maintained by Filipe and Jonas. When moved to [core], Filipe will lose access to package it (due to only Developers having access to [core]). So Jonas will become sole maintainer.
@Jonas & @Filipe: is this change in maintenance an issue?
That's fine by me: tpm2-tss is also an (albeit optional) dependency of systemd for systemd-cryptenroll, so making it go through [testing] seems sensible anyway to reduce breaking people's encrypted disk unlocking.
Best, Jonas
participants (2)
-
Allan McRae
-
Jonas Witschel