michaelgalloy.com

The COMMAND_LINE_ARGS routine was introduced in IDL 6.2, allowing IDL programs to access command line args passed to IDL when starting it. For example, to pass command line arguments into the MYPROGRAM routine, call IDL like below:

$ idl -e "myprogram" -args a b c

Then, in MYPROGRAM, the command line arguments could be retrieved with:

args = command_line_args()

The args string array would contain “a”, “b”, and “c”. To make it more convenient to launch IDL this way, you can create a script that calls IDL in the proper way, passing arguments of the script to arguments listed after the -args in the IDL call. If you name the wrapper script myprogram also, your calls would look like:

$ myprogram a b c

But, suppose you wanted your routine to have command line switches like:

$ myprogram --verbose --header=test.hdr test.dat

Then you had to parse the args received from COMMAND_LINE_ARGS yourself. The mg_options class simplifies this parsing and adds several other conveniences.

Continue reading “Command line options for your IDL program.”

I’ve released mgunit 1.0: get the source distribution or save file distribution. Also check out the project website to join the mgunit mailing list, read a short tutorial, check out development code from Subversion, etc.

Some features of the 1.0 release:

1. Several test runners for outputting test results to various formats are available; output can be sent to stdout, log files, or html files. In addition, there is a GUI test runner that will show results and re-run tests with the push of a button (also recompiling tests before running them).

2. Contains IDL Workbench templates for making it even faster to create new test cases/suites.

3. Keywords to MGUNIT to retrieve the number of passing and failing tests.

Color output to the terminal window when possible.

Command line wrapper so that mgunit can be run from the UNIX command line (get the source distribution to use this ability).

This release also contains several bug fixes.

Before I release mgunit 1.0, I would like to get a some feedback. I am using this for all the testing of my projects, but I would like to make sure it is working for others’ use cases.

1. Several test runners for outputting test results to various formats are available; output can be sent to stdout, log files, or html files. In addition, there is a GUI test runner that will show results and re-run tests with the push of a button (also recompiling tests before running them).

2. Contains IDL Workbench templates for making it even faster to create new test cases/suites.

3. Keywords to mgunit to retrieve the number of passing and failing tests.

4. Color output to the terminal window when possible.

This release also contains several bug fixes.

I recommend getting the source distribution so that you don’t have to do a manual restore: source distribution and save file distribution.

MGunit is a unit testing framework modeled on other unit testing frameworks such as JUnit. The goal is to allow easy creation and reporting of results of tests, but still allow for all needed flexibility. More details and examples after the jump.

Continue reading “Unit testing framework.”