Pacman

Historical bug tracker for the Pacman package manager.

The pacman bug tracker has moved to gitlab:
https://gitlab.archlinux.org/pacman/pacman/-/issues

This tracker remains open for interaction with historical bugs during the transition period. Any new bugs reports will be closed without further action.
Tasklist

FS#44424 - [updpkgsums] should respect md5sums=('SKIP')

Attached to Project: Pacman
Opened by (Det) - Wednesday, 01 April 2015, 04:58 GMT
Last edited by Allan McRae (Allan) - Tuesday, 11 October 2016, 10:34 GMT
Task Type Bug Report
Category Scripts & Tools
Status Closed
Assigned To No-one
Architecture All
Severity Low
Priority Normal
Reported Version git
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 2
Private No

Details

'updpkgsums' should not overwrite "md5sums=('SKIP')" with a 'correct' md5sum.
This task depends upon

Closed by  Allan McRae (Allan)
Tuesday, 11 October 2016, 10:34 GMT
Reason for closing:  Won't fix
Additional comments about closing:  Removed from repo
Comment by Gordian Edenhofer (Edenhofer) - Tuesday, 05 April 2016, 11:27 GMT
This bug is really annoying!
Comment by Dave Reisner (falconindy) - Tuesday, 05 April 2016, 12:13 GMT
I assume this is for cases where the source tarball changes but does not change name? Are there other reasons to want this?

How should updpkgsums behave if one were to have the following:

source=(
foo
bar
)
md5sums=(
SKIP
68b329da9893e34099c7d8ad5cb9c940
)

and then the PKGBUILD is changed, to add a new source:

source=(
baz
foo
bar
)

How does updpkgsums know that the SKIP belonged to 'foo' and should be persisted?
Comment by Gordian Edenhofer (Edenhofer) - Tuesday, 05 April 2016, 13:00 GMT
Sure this is indeed a problem but I think it is the duty of the package maintainer to alter the integrity array accordingly when adding source fields.
The attached patch would fix that by preserving "SKIP" in the same field as before. This would make things easier for at least the maintainer of [1] and [2]. Those are just the two I know of and I guess there are many more.

[1] https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=janusvr
[2] https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=spigot
   0001-makepkg-updpkgsums-honor... (1.4 KiB)
Comment by (Det) - Tuesday, 05 April 2016, 13:06 GMT
Sure:

https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=firefox-nightly
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=google-chrome
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=google-chrome-beta
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=google-chrome-dev
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=google-earth
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=plymouth-git
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=skype4pidgin-git
...
Comment by Giovanni Santini (ItachiSan) - Thursday, 19 May 2016, 06:53 GMT
Same here.
The issue is really annoying; the only files that get respected for not updating the hash are .asc keys (from what I saw in my PKGBUILDs)

Loading...