gstreamer/markdown/frequently-asked-questions/developing.md

Developing applications with GStreamer

How do I compile programs that use GStreamer?

This depends all a bit on what your development environment and target operating system is. The following is mostly aimed at Linux/unix setups.

GStreamer uses the pkg-config utility to provide applications with the right compiler and linker flags. pkg-config is a standard build tool that is widely used in unix systems to locate libraries and retrieve build settings. If you're already familiar with it, then you're basically set.

If you're not familiar with pkg-config, to compile and link a small one-file program, pass the --cflags and --libs arguments to pkg-config. The following should be sufficient for a gstreamer-only program:

$ libtool --mode=link gcc `pkg-config --cflags --libs gstreamer-1.0` -o myprog myprog.c

If your application also used GTK+ 3.0, you could use

$ libtool --mode=link gcc `pkg-config --cflags --libs gstreamer-1.0 gtk+-3.0` -o myprog myprog.c

Those are back-ticks (on the same key with the tilde on US keyboards), not single quotes.

For bigger projects, you should integrate pkg-config use in your Makefile, or with autoconf using the pkg.m4 macro (providing PKG_CONFIG_CHECK).

How do I develop against an uninstalled GStreamer copy?

It is possible to develop and compile against an uninstalled copy of GStreamer and its plugins, for example, against git checkouts. This enables you to test the latest version of GStreamer without interfering with your system-wide installation. See the Building from source using meson documentation.

How can I use GConf to get the system-wide defaults?

GStreamer used to have GConf-based elements but these were removed in 2011, after GConf itself was deprecated in favor of GSettings.

If what you want is automatic audio/video sinks, consider using the autovideosink and autoaudiosink elements.

How do I debug these funny shell scripts that libtool makes?

When you link a program against uninstalled GStreamer using libtool, funny shell scripts are made to modify your shared object search path and then run your program. For instance, to debug gst-launch, try:

libtool --mode=execute gdb /path/to/gst-launch

If this does not work, you're probably using a broken version of libtool.

If you build GStreamer using the Meson build system, libtool will not be used and this is not a problem. You can run gdb, valgrind or any debugging tools directly on the binaries Meson creates in the build directory.

Why is mail traffic so low on gstreamer-devel?

Our main arenas for coordination and discussion are IRC and Gitlab, not the mailing lists. Join us in #gstreamer on irc.freenode.net. There is also a webchat interface. For larger picture questions or getting more input from more people, a mail to the gstreamer-devel mailing list is never a bad idea, however.

What kind of versioning scheme does GStreamer use?

For public releases, GStreamer uses a standard MAJOR.MINOR.MICRO version scheme. If the release consists of mostly bug fixes or incremental changes, the MICRO version is incremented. If the release contains big changes, the MINOR version is incremented. A change in the MAJOR version indicates incompatible API or ABI changes, which happens very rarely (the last one dates back to 2012). This is also known as semantic versioning.

Even MINOR numbers indicate stable releases: 1.0.x, 1.2.x, 1.4.x, 1.6.x, 1.8.x, and 1.10.x are our stable release series. Odd MINOR numbers are used for unstable development releases and prereleases which should only be used temporarily for testing; your help in testing these tarballs and packages is very much appreciated!

During the development cycle, GStreamer also uses a fourth or NANO number. If this number is 1, then it's a git development version. Any tarball or package that has a nano number of 1 is made from git and thus not supported. Additionally, if you didn't get this package or tarball from the GStreamer team, don't have high hopes on it doing whatever you want it to do.

What is the coding style for GStreamer code?

Basically, the core and almost all plugin modules use K&R with 2-space indenting. Just follow what's already there and you'll be fine. We only require code files to be indented, header may be indented manually for better readability. Please use spaces for indenting, not tabs, even in header files.

Individual plugins in gst-plugins-* or plugins that you want considered for addition to these modules should use the same style. It's easier if everything is consistent. Consistency is, of course, the goal.

One way to make sure you are following our coding style is to run your code (remember, only the *.c files, not the headers) through GNU Indent using the following options:

indent \
  --braces-on-if-line \
  --case-brace-indentation0 \
  --case-indentation2 \
  --braces-after-struct-decl-line \
  --line-length80 \
  --no-tabs \
  --cuddle-else \
  --dont-line-up-parentheses \
  --continuation-indentation4 \
  --honour-newlines \
  --tab-size8 \
  --indent-level2

There is also a gst-indent script in the GStreamer core source tree in the tools directory which wraps GNU Indent and uses the right options.

The easiest way to get the indenting right is probably to develop against a git checkout. The local git commit hook will ensure correct indentation.

Comments should be in /* ANSI C comment style */ and code should generally be compatible with ANSI C89, so please declare all variables at the beginning of the block, etc.

Merge requests should ideally be made against git master or a recent release. Please don't send patches to the mailing list. They will likely get lost there.

See How to submit patches for more details.

How do I get my translations included?

I have translated one of the module .po files into a new language. How do I get it included?

GStreamer translations are uniformly managed through the Translation Project. There are some instructions on how to join the Translation Project team and submit new translations at http://translationproject.org/html/translators.html.

New translations submitted via the Translation Project are merged periodically into git by the maintainers by running make download-po in the various modules when preparing a new release.

We don't merge new translations or translation fixes directly, everything must go via the Translation Project.