Re: [pacman-dev] [PATCH] Add information on how an installed package was validated

2012-02-19 Thread Dan McGee
On Sat, Feb 18, 2012 at 12:42 AM, Allan McRae al...@archlinux.org wrote: When installing a package, store information on which validation method was used and output this on pacman -Qi operations. e.g. Validated By   : SHA256 Sum Possible values are Unknown, None, MD5 Sum, SHA256 Sum,

[pacman-dev] Strange (critical) bug with checkspace

2012-02-19 Thread Nagy Gabor
Hi! I've run into the following bug: $ df -h Filesystem Size Used Avail Use% Mounted on rootfs 44G 30G 13G 71% / /dev501M 0 501M 0% /dev /run503M 276K 503M 1% /run /dev/sda344G 30G 13G 71% / none503M 0 503M 0%

Re: [pacman-dev] Strange (critical) bug with checkspace

2012-02-19 Thread Nagy Gabor
Well, I did some good old printf debugging: pacman simply thinks that /etc/... files will be installed to /e. I think there is a string comparison bug somewhere in the code, but I am too tired to find it now (and fully read diskspace.c). NG

Re: [pacman-dev] Strange (critical) bug with checkspace

2012-02-19 Thread Dan McGee
On Sun, Feb 19, 2012 at 9:25 PM, Nagy Gabor ng...@bibl.u-szeged.hu wrote: Hi! I've run into the following bug: $ df -h Filesystem      Size  Used Avail Use% Mounted on rootfs           44G   30G   13G  71% / /dev            501M     0  501M   0% /dev /run            503M  276K  503M   1%

[pacman-dev] [PATCH] diskspace: ensure we match only full path components

2012-02-19 Thread Dan McGee
If one had a mountpoint at '/e' (don't ask), a file being installed to '/etc' would map to it incorrectly. Ensure we do more than just prefix matching on paths by doing some more sanity checks once the simple strncmp() call succeeds. Signed-off-by: Dan McGee d...@archlinux.org ---

[pacman-dev] [PATCH 1/3 v2] Add information on how an installed package was validated

2012-02-19 Thread Allan McRae
When installing a package, store information on which validation method was used and output this on pacman -Qi operations. e.g. Validated By : SHA256 Sum Possible values are Unknown, None, MD5 Sum, SHA256 Sum, Signature. Signed-off-by: Allan McRae al...@archlinux.org --- Changed

[pacman-dev] [PATCH 2/3] Remove _alpm_csum

2012-02-19 Thread Allan McRae
The enum alpm_pkgvalidation_t is essentially a more generic version of _alpm_csum, so use it instead. Signed-off-by: Allan McRae al...@archlinux.org --- lib/libalpm/be_package.c |4 ++-- lib/libalpm/package.c|2 +- lib/libalpm/sync.c |2 +- lib/libalpm/util.c |6

[pacman-dev] [PATCH 3/3] Simplify package validation info output for sync packages

2012-02-19 Thread Allan McRae
Unify the output for local and sync packages by only printing a list of possible validation types for sync packages. This also has the advantage of not printing the very long sha256 checksum which line wrapped on a standard width terminal. Signed-off-by: Allan McRae al...@archlinux.org ---

[pacman-dev] [PATCH] Give better error messages on database locking failures

2012-02-19 Thread Dan McGee
This was noted when trying to perform an operation on a pacman database on a read-only file system. Print the actual underlying errno string, and only show the you can remove message if the lock file actually exists. Before: $ pacman -Su error: failed to init transaction (unable to lock

[pacman-dev] [PATCH] Don't check for diskspace when using --dbonly

2012-02-19 Thread Dan McGee
It doesn't make much sense, as we aren't going to be manipulating any package files other than local database entries. Signed-off-by: Dan McGee d...@archlinux.org --- lib/libalpm/sync.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/lib/libalpm/sync.c

Re: [pacman-dev] [PATCH] Don't check for diskspace when using --dbonly

2012-02-19 Thread Allan McRae
On 20/02/12 15:49, Dan McGee wrote: It doesn't make much sense, as we aren't going to be manipulating any package files other than local database entries. I'd say great minds think alike... but they are both you so that proves nothing!