Maarten Breddels

The concatenated doesn't have that, you should do that to the individual dataframes, which are in vxframe.dfs.

> Tried only with 2.6.1, in 3.0.0 I keep getting that C++ error when vaex tries to write the first file. Is that error only present when you close the...

Yeah, I think vaex tries to protect you from doing this, it doesn't want to create dtype=object. Should we allow this under a particular flag maybe?

That's odd, this seems like an installation issue, you can try to remove all vaex packages (check that site-packages contains no vaex directories), upgrade pip, and install vaex again.

What do you expect to see as output btw, just count/missing values? e.g. No statistics?

> Please track this issue for updates. On the right on github you see:  When we implement to feature, you'll be notified.

I think we need to explore the option mentioned in https://github.com/vaexio/vaex/issues/1879#issuecomment-1033691134 first

looks that way.. although our apply should be parallel (multiprocessing), but it's a good point, I'll see if we can have the reverse without too many changes.

Yeah, that was me :) I had to rebase, hopefully it is green now.

green enough, thanks Ralf!