From the changelog:
```
Version 0.7.76, 2015-08-06
+ XML output: line breaks and indents in between attributes for readability
+ Trace feature: XML trace update, only if compiled with trace feature
+ Amazon S3 support (REST API v2), CLI/DLL only and if compiled with
libcurl support
+ FFV1: improved slice analysis (currently activated only with trace
feature and for 1st frame)
x MXF: optimization of the parsing, reading a lot less data (e.g. can be
reduced from 1 GB to 10 MB with some UHD files)
x MXF: wrong frame count with files not having the video stream as the
first stream
x Dolby E in MPEG-TS: "SMPTE ST 302" information was accidentally removed
x MPEG-TS: avoid filling delay from file name, not applicable on MPEG-TS
x MXF: better handling of huge padding blocks, better handling of
descriptors without link to a TrackID
x IMX: streams claiming that they are IMX are actually pure MPEG Video,
probing both IMX and MPEG Video
```
Their license explicitly prohibits distribution and it's unclear whether they allow to download their binary automatically. I've asked to clarify their terms; this may be temporary.
Although I couldn't test this because I'm not using a DE, nobody else
than the one submitting the pull request has commented on this. So if it
should break the icon for other people, nobody would probably start an
assassination because of this and the commit can be easily reverted if
it should break the icon.
Remove patches that no longer apply (integrated upstream I guess), but
forward port the part that adds the missing "include <unistd.h>"[1].
Build and run tested.
[1]: http://sourceforge.net/p/meshlab/bugs/342/
Adds a dependency on PackageKit...
In better news:
- Default to PDF when saving in all cases.
- Always show file format options and use a combo box to fit better into
dialog.
- Detect HP scanners and install drivers using PackageKit if found.
- If SANE detects no scanners but we see USB IDs of Brother, Epson or
Samsung scanners then prompt the user to go to their website to install
drivers.
- Use GUsb instead of GUdev.
- Fix error dialogs not showing.
Currently it errors out at build time with:
/nix/store/HASH-cudatoolkit-6.5.19/usr_include/host_config.h:82:2:
error: #error -- unsupported GNU version! gcc 4.9 and up are not supported!
Instead of downgrading gcc to 4.8, this patch upgrades cuda to 7.0, which
I think is the better choice. (Cuda 7 dropped support for some older graphics
cards, but gained support for newer ones.)
firefoxWrapper: switch to GStreamer 1.0
This definitely fixes the compatibility issues. @peti We should open an issue to figure out what is wrong with audio for that machine.
wkennington@f6c1004 switched Firefox to GStreamer 1.0 by changing its
buildInput *only*, but that is not enough. We need to fix Firefox
wrappers by changing their buildInputs and set GST_PLUGIN_SYSTEM_PATH_1_0
instead of GST_PLUGIN_SYSTEM_PATH.
With above changes playing H.264/MP4 media works in firefoxWrapper and
conkerorWrapper as tested with
http://www.quirksmode.org/html5/tests/video.html and
https://soundcloud.com/immclovin33/synthetix-sundays-53-with-marko-maric-19715
It should help with peti#9247
Reviewed-by: kmicu <kmicu@protonmail.ch>
Tested-by: kmicu <kmicu@protonmail.ch>
This commit removes files that are not referenced in nixpkgs anymore. I
have created this list by finding potential candidates using
find nixpkgs/pkgs -type d -links 2 | parallel -N 1 'out=`grep -L {} top-level/all-packages.nix`; test -n "$out" && echo {}'
and have then searched for those candidates with grep in whole nixpkgs.
Icons no longer missing (fix#5509).
In `*.desktop` files:
- Replaced absolute path to the the store by the program name.
This is so that files can be dragged elsewhere by the user
(e.g.: desktop, bar) and still work after upgrade + garbage
collection and can be shared between machines.
- Replace program name `soffice` by program name `libreoffice`
so that we're sure the desktop file really refers to our
package's binary and not start office or open office.
Add the possibility of building without the help. This build is
not modular and take a really long time to complete so I want
a mean of improving shortcuts without having to rebuild the
whole thing (see #899). A wrapper script is the next step.
Tested (build and ran the program) with `en_US` only and
without the help module.
This fixes some issues with libanjuta. The previous version spitted out hundrets
of error messages whenever gtkpod changed the progressbar. This issue is fixed
in 2.1.5.
Regression introduced by f91dacdd07.
Accidentally thought that it's compiling with XRandR support enabled,
because the cmake output said so:
Looking for XRRQueryExtension in Xrandr - found
Unfortunately, despite this message, the relevant part is:
Looking for XRRNotifyEvent - not found
So, ea4afb7 still holds true and I've added a small comment to avoid
this from happening in the future.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
I don't know what a "mouse keyboard" is, but instead of fixing the
description, let's use the one from the upstream README file, which is
also shorter than what we previously had.
The homepage http://synergy-foss.org/ is outdate since ages, so let's
point to the new site.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Overview of the updated versions:
beta: 45.0.2454.15 -> 45.0.2454.26
dev: 45.0.2454.15 -> 46.0.2471.2
Changes for getting beta and dev channel to build:
* The reference for chrome::FILE_FLASH_PLUGIN doesn't exist anymore in
version 46, because it has been dropped upstream, see the following
review URL:
https://codereview.chromium.org/1255943002
We set the PPAPI Flash path using a command line flag anyway, so it
doesn't hurt us if we don't patch that path (which was an old
artifact from the NSAPI->PPAPI conversion anyway).
Changes for the dev channel only:
* It seems that in the SCM, chrome/test/data/webui/ contains a lot of
files, however they are missing in the tarball.
This has been reported upstream at: https://crbug.com/515917
Our fix is to just not include webui/i18n_process_css_test.html at
all, to avoid the configure (gyp) phase to fail, because we're not
building tests anyway.
All channels built and tested by my Hydra instance at:
https://headcounter.org/hydra/eval/218978
Test reports:
x86: https://headcounter.org/hydra/build/723341/download/1/log.html
x86_64: https://headcounter.org/hydra/build/723342/download/1/log.html
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Some plugin bundles must be unpacked when used in Eclipse. With this
change the plugin manifest is checked for the setting indicating that
unpacking should happen.
Simplifies `buildEclipsePlugin` and `buildEclipseUpdateSite` functions
such that they require only absolutely necessary arguments. Also
add/expand comments slightly.
It does not really make sense to install the plugin packages directly as
they are intended for use with `eclipseWithPlugins`. Therefore it is
best not to present them to users as such.
This function provides functionality common to all Eclipse plugin
builders. In particular, it sets a package name and flags the derivation
as an Eclipse plugin.
I think the previous name comes from the fact that some older versions
were distributed in tarballs named ng-spice-rework-<version>. But now
the "rework" name seems odd; most references I found on the internet
calls the package "ngspice". Fix it.
We track the git version, because haskell-mode does not provide regular
releases except via MELPA, which our Emacs infrastructure does not support yet.
This reverts commit 37ff33196f.
This can't be it. `lcms` and `lcms2` are two entirely different packages.
Furthermore, the dependency is propagated through colord and should be
fixed there, if needed.
