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.
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.
FS#56246 - makepkg: resolve makedepends and depends at the same time
Attached to Project:
Pacman
Opened by Erich Eckner (deepthought) - Sunday, 05 November 2017, 14:34 GMT
Last edited by Eli Schwartz (eschwartz) - Sunday, 05 November 2017, 14:55 GMT
Opened by Erich Eckner (deepthought) - Sunday, 05 November 2017, 14:34 GMT
Last edited by Eli Schwartz (eschwartz) - Sunday, 05 November 2017, 14:55 GMT
|
Detailsmakepkg should resolve makedepends and depends simultanously, or at least makedepends not after depends.
The current issue occures when you have: in the PKGBUILD: depends=(a) makedepends=(b) and the following package: c: provides=(a) b: provides=(a), conflicts=(c) If for whatever reason pacman decides to install c to resolve depends=(a), then the makedepends cannot be resolved. This happens for example, if c is in testing, while b is in extra. I have seen this issue (when building for archlinux32) with openoffice-{still,fresh} and with jre[89]-openjdk (the latter being somewhat different, because they do not "conflict" via PKGBUILD variable, but if jre8 is installed after jre9, the default jre is still 9 and the build will/may fail). |
This task depends upon
Closed by Eli Schwartz (eschwartz)
Sunday, 05 November 2017, 14:55 GMT
Reason for closing: Duplicate
Additional comments about closing: FS#32723
Sunday, 05 November 2017, 14:55 GMT
Reason for closing: Duplicate
Additional comments about closing: