In oh-bugimporters, bugimporters/trac.py checks if the remote bug tracker seems
to be the "old" kind that requires extra scraping to get all the metadata.
The simple way to fix this is to make sure that this case gets covered in the
unit tests, and look at the old mysite/customs/bugimporters/trac.py code to see
how things were done there, and make corresponding changes.
The other way to fix it would be to rewrite the old_trac handling so that it
doesn't store the RSS-type data in the Django database. Instead, it could just
retain a local HTTP cache of the RSS feed contents, and re-parse that every time
it refreshes a bug stored in a old_trac-type tracker. (Hope that makes sense. If
not, I'm happy to explain this idea. It's not the fastest way to fix the
problem, anyway.)
Comment by paulproteus:
This commit removed that functionality -- https://github.com/openhatch/oh-bugimporters/commit/425f5f6d5ab3113ac5928a697f2d19ae7c569ef3 -- because it was crashing.
The simple way to fix this is to make sure that this case gets covered in the unit tests, and look at the old mysite/customs/bugimporters/trac.py code to see how things were done there, and make corresponding changes.
The other way to fix it would be to rewrite the old_trac handling so that it doesn't store the RSS-type data in the Django database. Instead, it could just retain a local HTTP cache of the RSS feed contents, and re-parse that every time it refreshes a bug stored in a old_trac-type tracker. (Hope that makes sense. If not, I'm happy to explain this idea. It's not the fastest way to fix the problem, anyway.)
Comment by paulproteus:
Status: chatting Nosy List: berryp, jwm, paulproteus Priority: bug Imported from roundup ID: 721 (view archived page) Last modified: 2012-05-02.22:51:13
ehashman commented at 2014-08-25T20:08:19Z:
Related: #942