Marius Gedminas

I've a failing test that reproduces this: ``` diff diff --git a/zilch/tests/test_store.py b/zilch/tests/test_store.py index 8606f65..d05ecac 100644 --- a/zilch/tests/test_store.py +++ b/zilch/tests/test_store.py @@ -33,18 +33,19 @@ class TestEventRecord(TestStore): with patch('zilch.client.send') as mock_send:...

@anarcat your code is still accessing updated_parsed even when published_parsed exists. I think you want this instead: ```python timestamp = item.get('published_parsed') or orig ``` or, if you intend to keep...

TBH I would make sure my CI build environment is not broken by doing a `sudo apt-get install -y python3-venv` instead of trying to work around it. (I also don't...

I don't know what to say. All other changes in 0.46 were trivial cleanups/refactorings and the switch from Travis CI to GitHub Actions. I wonder if it would be useful...

Huh, it looks like you're stracing pre-commit rather than check-manifest. I was hoping to see the clone()/execve()/wait() from check-manifest invoking the actual sdist build thing (python setup.py sdist or the...

> I am pretty curious what you could tell me about the above strace outputs. You see pre-commit reading check-manifest's output from a pipe (fd 4). The timestamp indicates when...

Heh, yes, strace files are something. I wrote strace-process-tree so I wouldn't have to interpret them myself. Now strace-process-tree doesn't like this format, but after some careful massaging with vim...

This is a bug, or, more precisely, a missing feature: check-manifest currently processes exclude directives (such as `prune`) and ignores include directives. I would like to fix it some day.

> I assume that this is nontrivial and not something that is an appropriate contribution for a first time contributor to the project. But, if I was selfishly interested in...

This is a very good idea!