| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Not all set functions respected NULL is empty behaviour, changed
add_set_value signature to return a set instead so it can conform.
Bug: https://bugs.gentoo.org/893424
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
| |
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
| |
Bug: https://bugs.gentoo.org/885801
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
| |
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
| |
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
| |
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
| |
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
| |
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
|
|
|
| |
Allow to reduce the search by having a query atom. This will skip
categories and packages not matching the atom, possibly avoiding lots of
work.
This needs to be exploited from applets where necessary.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
|
|
|
| |
To further hide the difference between trees, add a wrapper around
tree_get_meta, which deals with the underlying storage format. This is
most notably useful when dealing with vdb-based and binpkgs because they
have a specific or more expensive way of retrieval. This can now
properly be hidden from the q applets.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
| |
Technically, this should be enough to implement the Puppet provider for
Gentoo.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
| |
This functionality combined with -F allows usage with e.g. Puppet
providers.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
| |
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
| |
By default just operate on the arch specified by the selected profile.
This is probably going to be useful in the majority of the cases.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
|
| |
Most of the times, rmspace is unnecessary, and doing so, requires a
mutable copy of the data. If the callers call rmspace when necessary,
set can be a bit more efficient.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
|
|
| |
With some help from libq/tree now doing the sorting right (and
contructing an atom exactly once for each package) we can avoid
exploding atoms and greatly benefit in performance. On my setup I've
seen timings half or even more.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
|
| |
since cache was basically a shadow of vdb, and vdb grew too many
non-vdb-like behaviour, renamed to tree such that further features only
have to be implemented once
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
|
| |
- if metadata.xml is absent, don't crash
- if metadata.xml didn't contain an email, don't match as result
Bug: https://bugs.gentoo.org/685132
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
|
| |
Add option to match against maintainer email address listed in
package metadata.xml.
Closes: https://bugs.gentoo.org/685052
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
| |
Bug: https://bugs.gentoo.org/684252
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
|
|
|
|
|
| |
using -p/-c one can restrict to a possibly empty set, just bail on that,
don't crash
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|
|
qcache has its roots in reading the metadata cache, but since this is
standard functionality provided by libq/cache now, all that qcache does
really is things with keywords.
Signed-off-by: Fabian Groffen <grobian@gentoo.org>
|