| Commit message (Collapse) | Author | Age |
|
|
|
| |
this point
|
|
|
|
|
| |
* Skip running scriptlet when chroot fails - to prevent issues in the host
filesystem
|
|
|
|
|
| |
NOTE that we do NOT want symlinked *files* resolved, only paths (i.e.
/usr/lib/perl5/site_perl/*)
|
|
|
|
| |
package. Regression from 2.9.8.
|
| |
|
|
|
|
| |
* Fix typo in pacman.c
|
|
|
|
|
|
| |
Giovanni Scafora <linuxmania@gmail.com>
* Added pactest test for -Se operation
James Rosten <seinfeld90@gmail.com>
|
|
|
|
| |
file:////....: Unsupported scheme.
|
| |
|
| |
|
|
|
|
|
|
| |
* More significant error messages when -Qo fails
* Potential fix for reported error that pacman indicates corrupt packages
when space is full on the cache dir
|
|
|
|
|
| |
Alessio 'mOLOk' Bolognino <themolok.ml@gmail.com>
* Make makeworld compatable with makepkg lacking -w option.
|
|
|
|
| |
* pacman.8: add documentation for -Se option.
|
|
|
|
| |
(Is this useful to anyone?)
|
|
|
|
| |
user does not install all required dependencies, makepkg will now bail.
|
| |
|
| |
|
| |
|
|
|
|
| |
* correct _alpm_rmrf usage with regard to symlinks and directories
|
|
|
|
|
|
| |
* Two new pactest test cases
* Fix some "required by" / "requires" output by unifying all "dependency types"
to a single value (PM_DEP_TYPE_DEPEND)
|
|
|
|
|
|
|
|
| |
Giovanni Scafora <linuxmania@gmail.com>
* Many minor issues mentioned by Scott Horowitz <stonecrest@gmail.com>.
- extra 'done' text in --noprogressbar output
- missing flags in the pacman.8 manpage
- unclear text in --noconfirm help description
|
|
|
|
|
| |
* Updated Italian translation
Giovanni Scafora <linuxmania@gmail.com>
|
|
|
|
| |
* Add missing help line for --sync --list.
|
| |
|
| |
|
| |
|
|
|
|
| |
Nagy Gabor <ngaba@petra.hos.u-szeged.hu>
|
|
|
|
| |
Владимир Байраковский <4rayven@gmail.com>
|
| |
|
|
|
|
| |
* Updated fr.po courtesy of solsTiCe d'Hiver <solstice.dhiver@laposte.net>.
|
|
|
|
|
|
| |
issues, this should fix it. Progress bars now go from displaying, to showing
percent only, to not displaying at all. Changed unsigned -> signed to
prevent wraparound errors in integer comparison.
|
|
|
|
| |
Fix repo-add filename addition (use basename)
|
|
|
|
| |
work (it was downloading files as .package-name) at all.
|
| |
|
| |
|
|
|
|
| |
(and thus not copying pertinent data w.r.t. reading package files)
|
|
|
|
| |
'required by' field.
|
|
|
|
|
|
|
|
| |
duplicated.
* Updated the util Makefile.am to link with the proper libalpm.la.
* Fixed bitmasking issues in be_files.c and db.h.
* Rankmirrors updates from James Rosten (with some cleaning up of my own).
KeyboardInterrupts are now handled gracefully.
|
| |
|
| |
|
|
|
|
|
|
| |
(Because we are in a string freeze, I did not add "to the transaction" to
the original string but kept it in your translation)
Nagy Gabor <ngaba@petra.hos.u-szeged.hu>
|
|
|
|
|
|
|
| |
the alpm strcmp operation which takes void* references.
* We had this great visibility patch, but never actually took advantage of
it. Added the right compile flag to make it work and added some more
SYMEXPORTs where necessary to have a successful compile.
|
|
|
|
|
| |
Вам Зачем <4rayven@gmail.com>
* Updated manpage to document --cachedir and --dbpath options.
|
|
|
|
|
|
| |
Giovanni Scafora <linuxmania@gmail.com>
* long -> float conversion for package size output (which still may be the
wrong size, this needs to be looked at)
|
| |
|
|
|
|
|
|
|
|
|
| |
- Check EACCES in can_remove_file, to verify - we're only checking for a
read-only FS here... if write permissions fail, we can still remove it
- Switched the message when creating DB directories from an error to a debug
message
* Fix the dep tests once and for all (it wasn't working for provides=()
packages)
|
| |
|
|
|
|
| |
exactly what they think they are.
|
|
|
|
|
|
|
|
|
|
| |
Giovanni Scafora <linuxmania@gmail.com>
* Changed default SRCDEST in makepkg.conf. Two reasons:
1) consistency with PKGDEST in /home/packages
2) building packages as root is NOT recommended, so no reason to ever
encourage this use by caching source in a dir with root permissions
Yes, we may all do this differently and build our packages in other
directories. But that is why they are variables and can be set anyway.
|
| |
|