swamidass

That's a neat trick with the Zarr store! That aspect is certainly a better way to do this than I have set up currently, and I'll move over to that...

Okay, there is some more information that it worth noting, that turns out to be quite important. I did a test reading the data into memory. Using your code, that...

This is really helpful. And the solution is fairly obscure, so I'm glad we looked into it. I wouldn't have stumbled into that easily. I think I have a way...

So I just released a beta version: 0.1b0. You can see the key code here: https://github.com/swamidasslab/tiffslide-xarray/blob/main/tiffslide_xarray/data.py Turns out the zarr backed approach is the right way to go. By tweaking...

As for attributes and coords, you can see in that file how I add them. It is fairly straightforward. They don't have to be in the zarr store. Instead, I...

As for a comparison with approach you initially offered: ``` # via zarr_engine dataset1 = xarray.open_zarr( tiffslide.open_slide(fn).zarr_group.store, consolidated=False, mask_and_scale=False, ) ``` This is what we gain from the tiffslide wrapping...

> Thanks for implementing the changes in your library already! I'd be happy to integrate xarray support directly into tiffslide, and if you have time, it would be great to...

It is possible the name should be changed to "Instantaneous Bottleneck".

Where is the most up to date version now?

Just being able to do server-side rendering would be very useful, both of depictions (particularly SVGs), and also (and likely easier) for molecule canonization.