Adam Williamson

yeah, I don't think marking coverable things as uncoverable is a good strategy. covering them, or reducing the relevant target until they can be covered, seem like better options.

How did we do it before everything in `t/` was fully covered? Would seem most logical to go back to that. Anyway, I kinda like the idea of mocking qemu...

blender also uses `_PyArg_Parser` extensively: https://projects.blender.org/blender/blender/search?q=_PyArg_Parser . I am not enough of a C coder to know how easy it would be to port all those uses to the public...

@vstinner on a naive grep it looks like it should be enough, but blender at least also uses `_PySet_NextEntry`, so I can't get a complete compile done to verify unless...

This at least makes the test failures go away. It makes Beaker behave the same on Python 3.13 as it did before. But...I'm not sure it's the right thing to...

Oh, note on the change from `dumbdbm` to `dbm.dumb` - there is no `dumbdbm` in current Python 3. `dbm.dumb` has been its correct name for the whole Python 3 series....

It's the second block, where it changes `log.debug("Couln't load pickle data, creating new storage")` to `log.debug OSError`. That's a syntax error; `log.debug` is a function, not a statement, `log.debug OSError`...

I've got a zillion other things to do, so no. I mean, just post a bug that shows the before and after of running pyupgrade on that file. Anyone with...

so, looking on the x86_64 worker host, where 2621079 is the pid of a `/usr/bin/isotovideo: backend` process, I get: ``` [adamwill@openqa-x86-worker01 ~][PROD]$ grep ^SigBlk /proc/2621079/task/*/status /proc/2621079/task/2621079/status:SigBlk: 0000000000000000 /proc/2621079/task/2621142/status:SigBlk: 0000000000014000 /proc/2621079/task/2621143/status:SigBlk:...

Here's a dumb thing I rigged up: ``` [adamwill@openqa-a64-worker04 ~][PROD-IAD2]$ for i in `ps aux | grep isotovideo | grep backend | cut -d" " -f 2`; do echo "PID...