NB: As of GDAL version 1.11.2 the image has been renamed from homme/gdal to
geodata/gdal.
This is an Ubuntu derived image containing the Geospatial Data Abstraction Library (GDAL) compiled with a broad range of drivers. The build process is based on that defined in the GDAL TravisCI tests.
Each branch in the git repository corresponds to a supported GDAL version
(e.g. 1.11.2) with the master branch following GDAL master. These branch names
are reflected in the image tags on the Docker Index (e.g. branch 1.11.2
corresponds to the image geodata/gdal:1.11.2).
Running the container without any arguments will by default output the GDAL version string as well as the supported raster and vector formats:
docker run geodata/gdal
The following command will open a bash shell in an Ubuntu based environment with GDAL available:
docker run -t -i geodata/gdal /bin/bash
You will most likely want to work with data on the host system from within the
docker container, in which case run the container with the -v option. Assuming
you have a raster called test.tif in your current working directory on your
host system, running the following command should invoke gdalinfo on
test.tif:
docker run -v $(pwd):/data geodata/gdal gdalinfo test.tif
This works because the current working directory is set to /data in the
container, and you have mapped the current working directory on your host to
/data.
GDAL will be run under user nobody in the container, so if editing or creating
files the appropriate permissions for that user must be applied to the directory
pointed to by $(pwd). Alternatively pass the --user option to docker run.
Note that the image tagged latest, GDAL represents the latest code at the
time the image was built. If you want to include the most up-to-date commits
then you need to build the docker image yourself locally along these lines:
docker build -t geodata/gdal:local git://github.com/geo-data/gdal-docker/