Enrico Guiraud

it's really executed every time in my case, because `self._at_once` is `False` EDIT: `self._at_once` is `False` here because `sync()` calls `prepare_new_status()` without using the `at_once()` context manager, here: https://github.com/pimutils/vdirsyncer/blob/54a90aa5dd34fb3583965535be82b0a552f18c7d/vdirsyncer/sync/__init__.py#L143-L146 A...

Alright, I'll give it a go after I create another calendar for this purpose, because problem number 3. [here](https://github.com/pimutils/vdirsyncer/issues/818#issuecomment-1461072201) makes me wary of running `vdirsyncer sync` to completion on my...

The patch proposed at https://github.com/pimutils/vdirsyncer/pull/1051 seems to fix points 1. and 2.

> It would be useful to see both versions and see what changed. > Do any of your entries have more than one PRODID field? This seems to be triggered...

> you created this on the phone, vdirsyncer downloaded it, and the next sync it was trying to create it? yes, the next sync it updated it as above. for...

Hi, due to https://github.com/Gilnaa/globwalk/issues/28 mentioned above passing a path starting with `./` or `../` to `Tera::new` is completely broken. It does not look like https://github.com/Gilnaa/globwalk/issues/28 is going to be fixed...

> without breaking any existing code I am not sure how to guarantee that (I am not sure globwalk and globset behave identically for all inputs). Something simpler to check...

> we could check if the input starts with ./ or ../ and return an error > That would still be a breaking change though. Only for users that expect...

I'm perfectly sure those weirdos exist (hyrum's law and all that) :smile: but you don't necessarily have to keep them happy :P

A PR that makes Tera::new error out is up. Locally it produced no test failures. For the record I still think that doing what 99.9% of users would expect out...