What did I do?

This week, I wrote a test for our mismatch_loader script that verifies its functionality using a temporary database and data directory. It worked as expected in Linux but failed on Windows. This was a hurdle, but after further investigation, I discovered that temporary files are handle defferently in Linux and Windows. Linux offers automatic removal of temporary files, but in Windows, we have to remove it manually.

Some of the tests have been failing for quite some time. After discussing with my mentor, I temporarily disabled them.

What’s coming up next week?

For the upcoming week, I plan to:

  • yaml checks for mismatch_relations pull request.
  • enable mismatch feature for remaining parsers.

These YAML checks will validate the mismatch_relations.yml file within the data/** directory. I’m considering creating custom rules to ensure the file contains both the required lists: purls and invalid_vendors.

While introducing the mismatch database, I initially enabled its use only for the python parser. It’s been some time now, so I think I’ll enable it for the rest of the parsers.

Catch you later! More to come next week.