Anders Hovmöller
Anders Hovmöller
Aha. I tried copying over pkg_resources from my main install over the one inside pip/_vendor, but I didn't see any difference in speed :/
@csdummi I have a PR that does this. It helps somewhat.
https://github.com/pypa/pip/pull/6346
@csdummi I'd say no, not at this time at least. But I do have some ideas for that at https://github.com/boxed/p but it's way too early for that.
@CSDUMMI I think both oclif and your code have the wrong idea, where git has the right idea (for this purpose at least). OCLIF is the more egregious to me...
I would be happy to accept such a PR for sure.
Cool! Yea that's great! Timeout for mutants that introduce infinite loops isn't implemented yet. I just disabled break->continue mutation and similar to avoid the scenario as far as possible without...
Memory limit is an interesting idea. You could set it at like 20% over the current memory, as the forked process should have loaded 100% of what is needed from...
Yea you're right. It should be x% over of the peak memory usage during the clean run, which is probably hard to capture correctly. I'm used to working with systems...
Ooh. That is some dark magic 🤣 Cool!