Omero tree fixes

[will-moore]

Fixes a couple of issues from #9

• No filtering by default group or user
• Alphabetical ordering
• Also caches childCount so we don't have to request this from the server multiple times.

To test:

$ omero login
# idr.openmicroscopy.org, public, public
$ napari_omero

The biggest issue now is that we try to load the whole depth of the tree up-front, instead of waiting for the user to expand each node. This means that with IDR, you are loading forever and never reach the whole tree. But the top of the tree loads OK:

[codecov-commenter]

Codecov Report

Merging #10 into master will not change coverage. The diff coverage is 0.00%.

@@          Coverage Diff           @@
##           master     #10   +/-   ##
======================================
  Coverage    0.00%   0.00%           
======================================
  Files          11      11           
  Lines         703     724   +21     
======================================
- Misses        703     724   +21     

Impacted Files	Coverage Δ
src/napari_omero/widgets/gateway.py	0.00% <0.00%> (ø)
src/napari_omero/widgets/main.py	0.00% <0.00%> (ø)
src/napari_omero/widgets/tree_model.py	0.00% <0.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 70065ef...216aeec. Read the comment docs.

[tlambert03]

oh awesome. thank you!

The biggest issue now is that we try to load the whole depth of the tree up-front, instead of waiting for the user to expand each node.

Yeah that was a short-term hack. I'll follow up and fix the _populate_tree method to be lazy again, using fetchMore and canFetchMore (which you may have seen were commented out for now). I did want to ask though, I feel like expanding a node (fetching children for just that node) is still much slower than it is in the webclient. Any thoughts there? Is there any prefetching going on in the webclient to make it snappy?

[tlambert03]

I've also left out screens for now... since I have no experience with them

[will-moore]

Using those getObjects() queries might be a bit faster. I noticed that the listChildren() calls are also loading annotations. Also the caching of childCount should help a little. Can't think of anything else just now.

[will-moore]

Working well. Thanks 👍

[tlambert03]

there's enough goodness here to merge. we can address further improvements to the tree later