public:lofar_packages

This is an old revision of the document!


Available packages on the LOFAR offline clusternodes

This page gives an overview of available packages, other than the 'standard' ones distributed as part of the Linux distribution, and how to initialize these. See the Login environment page for instructions on activating packages at login time.

Package Description Initialize Update frequency Comments
Casa Browser, viewer use Casa infrequent NRAO development
grid
hdf5 library infrequent Used by other packs, now 1.8.2
hdf5viewer infrequent
LofIm Lofar Imaging use LofIm daily Follow this link
pipeline
Pythonlibs Several useful Python modules use Pythonlibs infrequent Follow this link
wcslib library infrequent Used by other packs, now 4.3



The package LofIm (Lofar Imaging) consists of the following previously known subpackages:

  • casacore
  • casarest
  • pyrap
  • LOFAR
  • ASKAPsoft

All these are build every night in a single go, with source code extracted from each package's source repository. Activating the LofIm package ensures that you will have a consistent set of libraries and executables.

To activate it on commandline use:

> use LofIm 

This will activate the version that was build last night. It will add all settings needed to use the subpackages toyour PATH, PYTHONPATH and LD_LIBRARY_PATH environment variables.

As all of the individual subpackages are in a development phase, we cannot guarantee that each nightly build is bugfree, or even that is has succeeded. Therefore, we keep a buffer of previous builds for a week. This ensures that if a nightly build has failed, you can use a (working) previous build. To activate a previous build on commandline use:

> use LofIm <day> (e.g., use LofIm Wed)

If the problem that caused a build to fail has not been solved in a week's time, you will not be able to use the package until it has been fixed. In due time we will implement a system with a release policy, that will enable you to use 'stable' versions of these libraries.

We have made available several useful Python modules. These can be activated with a single command:

> use Pythonlibs

(see this page for information on how to activate at login time). The Pythonlibs currently contain:

  • Last modified: 2009-11-13 10:43
  • by Arno Schoenmakers