Stef Smeets

> Without a `setup.py` there is no way to do a dev install (AFAIK) This is also what I found. `pip` cannot do an editable install without a `setup.py`, so...

Yep, that's what I did 👍

Turns out the gmsh22 reader is passing tuples `('line', array([...]))` instead of instances of `CellBlock`, which causes the issue.

Ran into the same issue. It helps to set `forceNiceScale: true` in y-axis through `apex_config`. i.e. in the OP: ```yaml apex_config: yaxis: tickAmount: 5 forceNiceScale: true ```

This is an optimizer for the pattern matching, which is different altogether. I think it could be useful, because from my understanding the orientations from pattern matching are quite discrete...

@jameshgrn @vijaysm Thanks for reviewing Nanomesh and the feedback! I have addressed the spelling mistakes, clarified the text and update the contributing guidelines. I addressed the more in-depth comments and...

Cheers, I just accepted the PR, letme know if there is anything else I can do :-)

Hi @prashjha Thanks for the useful suggestions, I tried to incorporate all of them in the final draft. I created a new release: https://github.com/hpgem/nanomesh/tree/0.9.1 and uploaded it to Zenodo: https://zenodo.org/record/7157382...

Hi @rcomer , I just noticed that a `nanmedian` function exists in [dask](https://docs.dask.org/en/latest/array-api.html#dask.array.nanmedian). Would this be a way to make the median operation lazy in iris?

This is indeed a bug, see shelx documentation here: https://shelx.uni-goettingen.de/shelxl_html.php Seems easy enough to fix, I'll have a look.