This section describes how to install the pipeline in development mode, for members of the GPI team and Gemini staff.
This method requires you have an IDL license. If you do not have an IDL license, please refer to this page for installing compiled executables instead.
The GPI Pipeline souce code is developed using a Subversion version control repository hosted at the SETI Institute. GPI team members should see the GPI Computer Account Setup document for instructions to obtain an account on this repository.
If you are not familiar with version control, you may wish to see this intro to version control.
Once you have that account, in a directory of your choosing (preferably somewhere inside your $IDL_PATH) execute the commands:
>>> svn checkout https://repos.seti.org/gpi/pipeline --username <yourusername> >>> svn checkout https://repos.seti.org/gpi/external --username <yourusername>
The above commands will download the GPI data pipeline, GPItv viewer, and a directory containing external dependencies of the code, for instance the Goddard IDL library, the Coyote library, etc. You may already have copies of many of these external routines in your own IDL library, in which case you can delete the excess copies from this directory if you so desire.
All code has been tested extensively with the external program versions in the repository. Error-free operation is not guaranteed for other versions of these libraries. The usual caveats about name collisions between different versions of IDL routines apply. If you have old versions of e.g. the Goddard IDL library functions in your $IDL_PATH, you may encounter difficulties. We suggest placing the data pipeline code first in your $IDL_PATH.
Mac OS and Linux
On Mac OS and Linux, you will likely want to add the pipeline\scripts subdirectory to your shell’s $PATH.
Proceed now to Configuring the Pipeline.