:thumbsup:

Ah, I had forgotten that python3 added this. My brain still clings to the global/locals (+closure) scope model in py2.

https://stackoverflow.com/questions/45292479/name-binding-in-except-clause-deleted-after-the-clause

It's not technically implemented as a scope though. Granted, from a practical perspective, the difference is irrelevant.

This PEP intends to resolve this issue by adding a cleanup semantic to except clauses in Python 3 whereby the target name is deleted at the end of the except suite.

There are similar behavior changes with list comprehensions and generator expressions. Those are apperently implemented with function scopes.

Kind of a style nit, but having task_error shift from boolean to exception bugs me...

https://github.com/mikem23/koji-playground/commits/pagure/pr/845

Also, I'm trying to remember why we deferred raising the exception until after the loop.

This code has been refactored a few times and I think it used to matter, but I don't see how it does not. We get an error, we save the exception object, we break, and then we re-raise if we got the error. Looks like the flow would be the same if we just moved that bit into the exception handler, except we'd get a cleaner exception because we could just use a bare raise.

Or maybe I'm missing something in my haste...

I was going through history and that behaviour is there from beginning of git, so not sure what is there. I'm also not able to reinvent the reason, so I've tried to remove it. I've also added small optimization to not check for same result twice - it shouldn't hurt.

koji/tasks.py:393:76: F841 local variable 'ex' is assigned to but never used

Looks good otherwise

removed

Commit 77ae912f fixes this pull-request