Clemens Brunner

Sure, but it would be nice to retain some control over libxdf after the PR gets merged. What I mean is that ideally we would be in the collaborators group...

Yes, it's me :smile:. I'm not saying we want exclusive control over libxdf. Once it's in there, all collaborators should be able to perform merges in that folder. It's just...

Any update on this? The C++ library is now fully integrated in SigViewer and working nicely. @Yida-Lin would contribute it to this repo. Should he go ahead and make a...

Indeed, libxdf by @Yida-Lin is tailored towards SigViewer, which means that it might not be ideal for a generic XDF reader library. However, I'm not even sure if a C...

Is there something official from Brain Products that's open source? I'd also consider BioSig (if it has a BV reader, not sure).

> +1 for biosig, if many people use it No idea if many people still use it, but is has always been a reader with support for a lot of...

How do you want to handle this? I'd put no restrictions on type and description. Instead, I would write markers exactly as passed to the function (when passed a list...

Yes exactly. We might even consider splitting this into two parameters: - `events`: only accepts 2D arrays and has (some of) our conversions in place - `markers`: the list of...

If we disable all checks regarding marker type and description when passing a list of dicts, we run into the problem that many (all?) tests currently use events passed as...

Here's something to consider: pybv can already export arbitrary markers as type `Comment`, so maybe there is no need for such a big change? Coming from MNE, the private `_export_mne_raw()`...