Requirements

PyDrizzle and Multidrizzle depend on the following support software (to be installed separately): 

• PyRAF
• STSDAS v3.0 (or later)
• Python 2.2 
• numarray version 0.5

• PyFITS Version 0.7.8
• readgeis Version 1.5
• UPINCD Version 0.3
• deriv.py



---

Installing PyDrizzle

This distribution provides all the Python source code for installing the Multidrizzle and PyDrizzle packages under STSDAS 3.0 (CVS Versions Version_1_2public and Version_4_3public, respectively). This distribution requires that STSDAS 3.0 already be installed. In addition, these packages also require numarray version 0.5 (or later), and the instructions for getting that package are included here. This distribution also includes new versions of supporting Python libraries developed at STScI for running these packages, including PyFITS Version 0.7.8, readgeis, fitsdiff, upincd, and a new numarray-based version of the 'deriv' task. Finally, a new version of the DITHER package script is included to insure that multidrizzle gets defined within that package as well.

Installation of PyDrizzle and Multidrizzle starts by copying the package to a directory which Python can load, either by default or as specified in your PYTHONPATH environment variable. Once PyRAF has been started, the packages should be available for use. This can either be done by updating existing STSDAS 3.0 Python directories, or by installing in a user's personal directory.

For those who can update their IRAF/STSDAS V3.0 directories:

1. [OPTIONAL] For the sake of retaining the old versions until the new version has been installed, rename stsdas$python/pydrizzle and stsdas$python/multidrizzle.
2. Unpack the distribution into the STSDAS v3.0 stsdas$python directory. NOTE: This assumes that symbolic links have been set up to point to both 'pydrizzle' and 'multidrizzle' directories. The links should point to:

              pydrizzle    -> ../pkg/analysis/dither/pydrizzle
                   multidrizzle -> ../pkg/analysis/dither/multidrizzle
   

   As a result, the option to follow symbolic links should be used when un-tarring this distribution:

               tar xhf pydrizzle.tar
   

3. If PyFITS is installed elsewhere, move 'pyfits.py', 'readgeis.py' and 'fitsdiff.py' to the standard location on your system for locally installed Python libraries.
4. Verify that numarray has been upgraded to Version 0.5. This version can be downloaded from the numarray development site.

   NOTE: Upgrading from Version 0.4 requires completely removing ALL compiled libraries and source code for Version 0.4 before Version 0.5 can be successfully installed.

5. Replace the existing stsdas$pkg/analysis/dither/dither.cl script with the new version from this installation. This will setup the IRAF interfaces for Multidrizzle, PyDrizzle, and related tasks.
6. Start PyRAF.

For those who can NOT update their IRAF/STSDAS V3.0 directories:

1. Unpack the distribution into a directory in a location of your choosing. For example: /data/your_machine/python NOTE: The STSDAS distribution assumes that symbolic links \ have been set up to point to the STSDAS 'pydrizzle' and 'multidrizzle' directories located in the dither package. However, for your own installation, you should simply create sub-directories in your directory:

               mkdir pydrizzle
               mkdir multidrizzle
   

   This distribution can then be unpacked simply using:

               
               tar xf pydrizzle.tar
   

2. Install numarray Version 0.5 in a local directory, if system version has not been updated to this version.
3. Start PyRAF.
4. Start STSDAS, but do NOT start the 'dither' package yet!

           --> stsdas
   

5. Update the Python system path to point to your personal version of the PyDrizzle/Multidrizzle code using:

           --> sys.path.insert(1,'/data/your_machine/python')
   

6. Load the dither package now.

            
           --> dither
   

7. Define the IRAF variable mydrizzle to point to the newly installed multidrizzle and pydrizzle IRAF interface code and parameter files.

           --> set mydrizzle = '/data/your_machine/python/'
   

   The environment variable mdrizzle also needs to be defined, as it is used by the *_iraf.py files in this distribution:

           
           --> set mdrizzle = '/data/your_machine/python/'
   

   Finally, re-define the PyDrizzle, Multidrizzle, buildasn, and xytosky tasks to point to your new version:

           
           --> pyexecute('mydrizzle$pydrizzle/pydrizzle_iraf.py',tasknames='pydrizzle')
           --> pyexecute("mydrizzle$pydrizzle/wcsutil_iraf.py",tasknames="xytosky")
           --> pyexecute("mydrizzle$pydrizzle/buildasn_iraf.py",tasknames="buildasn")
           --> pyexecute("mydrizzle$multidrizzle/multidrizzle_iraf.py",tasknames="multidrizzle")
   

   These are based on the commands found in the dither.cl script. The commands given in this instruction can be put into a custom CL script that can be run under PyRAF after loading the 'dither' package to automate this process for future sessions.

You can now verify that your personal version of these tasks has been loaded using:

        --> import pydrizzle, pyfits
        --> pydrizzle
        module 'pydrizzle' from '/data/machine/python/pydrizzle/__init__.py'
        --> pyfits
        module 'pyfits' from '/data/machine/python/pyfits.py'
        --> unlearn pydrizzle
        --> unlearn multidrizzle

Download PyDrizzle and Multidrizzle 

numarray