Tal Yarkoni

@christian-oreilly that's correct, there's currently no such functionality. I think it would be a great addition though, and would be happy to review a PR. It could probably be a...

`bids/utils.py` mostly contains very general utilities that are (or could be) useful package-wide. There's actually some stuff in there that should probably move into a `bids/layout/utils.py` module (well, really just...

@oesteban @effigies note that automatic flattening would also solve the problem in #682 for free, as "run" would now be "runDescription", and hence would never be scanned in the offending...

> What about KeyA.KeyB to make it clear that it's going into an object? Fine with me. I agree that adds clarity. > Do we care that much about those...

> Finally, it seems that querying is working fine - as you said in the other thread, we really need to fix the test, not the tool :D. The current...

Okay, on revisiting the various threads, I guess we're still kind of stuck. @oesteban's patch in #684 fixes the specific problem that spawned this whole discussion (in #681), but it...

Thanks for opening this @Remi-Gau; I think this kind of thing is definitely in scope for pybids, and it would be great to support it. It's worth noting that there...

Hey, unrelated to the above thread, doesn't this implementation run counter to the suggestion previously that we handle as much of the densification logic as possible in the (implicit) `to_dense`...

It doesn't necessarily have to live in `to_dense`, but it does need to abstracted out of `Convolve`, because there are other transformations that will need to deal with the same...

Merging this as-is seems fine. The refactor will probably be pretty straightforward, and if any problems arise, they'll likely be coming from integration with 0.8 (though I think it should...