At work, we already have a custom script to perform regressions for us. I have been tasked with the creation of a wrapper script which will ultimately invoke our pre existing script in a custom fashion and do a bunch of house keeping in and around the executed tests. Essentially what I need to provide are:
- Ability to invoke simulation using custom switches, paths etc
- Provide some form of regression tracking (history of pass fail)
- Some form of basic web interface to view regression results
- Perhaps an automated email if something breaks
Now nothing I have mentioned is new and I am sure there are a lot of ways to approach this.
What I am looking for is some suggestions as to a good way to go about this.
Some solutions that come to mind:
- Build custom python script (most effort)
Extend python's built in Unit Testing framework. Subclass where necessary. This is really the crux of my question. Is this a good solution?
Use some other framework? Jenkins? I have not used jenkins but I've heard good things about this tool. Any thoughts on if this tool would suit here?
Thanks for your help!
Some of you may ask why I don't extend the base simulation script directly. Well, it's a few thousand line Perl monster! I don't speak Perl, nor do I have any intention to start!
