| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
| |
There's no obvious reason why we should restrict this in the generator.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
Fixes https://github.com/hughsie/appstream-glib/issues/264
|
|
|
|
| |
Nothing uses it, and it's not API or ABI stable.
|
|
|
|
|
|
|
|
|
| |
This fixes, among other things, appdata validation when the version goes
from 9.5 to 10, or 0.9 to 1.
This partially fixes https://github.com/hughsie/appstream-glib/issues/270
We're still failing to correctly compare "Build 9.5" to "Build 10".
|
|
|
|
| |
Firmware should now be distributed using the LVFS and consumed by fwupd.
|
|
|
|
|
|
|
| |
Two problems here:
* Don't force an unknown icon to CACHED when reading from a desktop file
* Don't try to convert an icon to a CACHED icon when it's already CACHED.
|
| |
|
|
|
|
|
|
|
| |
This allows us to support desktop-less components that specify a cached icon.
The icon search code has been split into a new plugin that is run after the
appdata and desktop plugins have added the AsbApp to the list of available apps.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
If the guessing fails, we fall back to the arch/rpm/deb-derived name but we
don't want to show weird things in the logs.
Fixes https://github.com/hughsie/appstream-glib/issues/228, many thanks to
Andrew Gunnerson for the initial patch.
|
|
|
|
|
| |
... to match with what the shell extensions plugin in gnome-software
does.
|
|
|
|
|
| |
Padding is no longer being used as it makes the software center look bad.
Lets make the application author look bad instead so they can fix the icon.
|
|
|
|
|
| |
Sometimes it returns a return code 25088 (which is 0x6200) which isn't an error
code in rwpng.h -- we can't really trust pngquant so just log the warning.
|
|
|
|
|
| |
... and convert the appstream ID to be compatible with what gnome-tweaks
uses (@'s replaced with _ and no .shell-extension at the end).
|
|
|
|
| |
Fixes https://github.com/hughsie/appstream-glib/issues/223
|
| |
|
|
|
|
| |
Languages may have been pre-added based on the information in `metainfo.xml` files, so just because we didn’t auto-detect a language doesn’t mean that none are present. Fixes #207.
|
| |
|
|
|
|
| |
Fixes https://github.com/hughsie/appstream-glib/issues/203
|
|
|
|
| |
Fixes an extraction issue when using OpenSUSE.
|
| |
|
|
|
|
| |
Fixes a build error with the mcomix package in openSUSE.
|
| |
|
|
|
|
| |
Fixes https://github.com/hughsie/appstream-glib/issues/190
|
|
|
|
|
|
|
|
|
| |
Any XML parser should ignore the extra tags or attributes, and so it serves no
purpose to artificailly hide these when building metadata for older versions of
the specification.
This also fixes a bug on i386 where 0.6 != 0.6 and I didn't want to use fabs()
all over the place for basically no reason.
|
|
|
|
|
|
|
|
|
| |
There’s a precondition check for (search != NULL), so the function will
hit an assertion failure before this branch.
Coverity CID: 1454189
Signed-off-by: Philip Withnall <philip@tecnocode.co.uk>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Calling g-ir-scanner when cross-compiling with bitbake/OpenEmbedded
fails because it calls the wrong ld (the one from the host). More work
will be needed in meson.bbclass and/or gobject-introspection.bbclass
to make it work.
In the meantime, having an option to turn of introspection is useful
perhaps also in other cases where the extra work is not needed. For
example, fwupd works fine also when it is off.
The name of the new meson option matches the --disable-introspection
that is used by some autotools-based projects.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
|
|
|
|
| |
This fixes the build for appstream-glib on Mageia.
|
|
|
|
| |
These components are allowed to have no icon since b5b4a855.
|
|
|
|
|
|
|
|
|
|
| |
In 0.41.0 meson introduced it's own 'dist' target and so we don't need to
define this ourselves anymore. Use a version compare to avoid depending on a
very new meson.
Also, fix up some underlinking which was somehow allowed before.
Fixes: https://github.com/hughsie/appstream-glib/issues/170
|
|
|
|
|
| |
This is better done client side. Add a build flag so we can still build metadata
for old versions of Fedora.
|
|
|
|
| |
I guess this was missed in the move to meson.
|
|
|
|
| |
Now that we're not threaded we can provide a sane error.
|
| |
|
|
|
|
| |
Without the CacheID it's no longer useful.
|
|
|
|
| |
This leads to just too many crashes; librpm is just not thread safe enough to do this.
|
|
|
|
|
|
|
|
|
| |
The results when using the partial update were not always the same as doing a
fresh run (some components depend on other components) and even I'm not using
this functionality anymore when running appstream-builder on the whole Fedora,
achive, even across on all branches.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1426166
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
This would have helped some head scratching this morning...
|
| |
|