BrainVision annotation parsing is needlessly complicated

It used to be BV files would neatly map stim codes to integers: 'Stimulus/S 11' would become 11 and so on. Now, with annotations, we have essentially random assignment; what 'S 11' becomes is essentially unpredictable.

Now it is trivial to write a small function to get neater mappings, but 1. if you work with BV files, you will probably write that same function every single time you import your data, which is another 5 or so lines of essentially boilerplate, and 2. this presents a significant barrier of entry to newcomers.

Of course, it could be quite easy: least-surprise default mapping BV markers to integer codes in the way you’d expect, i.e., 'S 11' becomes 11.

Or am I missing something?

@cbrnr @massich

[jona-sassenhagen]

I will implement it as such!

[jona-sassenhagen]

Yes, and that function, or something like it, should have been the default for mapping markers to event code ints.

I will go about that, unless @massich or somebody else wants to.