Gabriel Selzer

The pyimagej team just ran into this issue today. Has there been any progress on this?

Great, thanks @thewtex!

This is super insightful, thanks so much @bsteffensmeier! My use case involves the creation of arrays that are arbitrary dimensionality; modification is also important for the use case. For now,...

I'll note that yesterday I made a very hacky fix that seemed to make things work for my current use case in napari-imagej. It did, as far as I can...

> In order for this op reduction thing to work, in general, we need to lean on the `ConvertService` layer to take care of coercing the INPUT and BOTH types...

@NicoKiaru this would make it marginally easier to use `gui` mode from macOS (which is not quite the same as `interactive` mode).

I did a little debugging. Turns out that the order of `fail_services` actually *affects the screen size* as reported by [`quartz`](https://github.com/vispy/vispy/blob/f0af63083b25c27bd02ce3778bd9e62096c3ccd1/vispy/util/dpi/_quartz.py#L24). * When `DefaultCommandService` is first, we have a nonzero...

I did some more hacking to your notebook example, which I turned into a script: ```python3 import imagej from scyjava import config # config.add_option('-agentlib:jdwp=transport=dt_socket,server=y,address=5286,suspend=y') #ij = imagej.init("2.5.0", add_legacy=False) imagej._create_jvm("2.5.0", mode='headless',...

See https://github.com/imglib/imglyb/pull/14

> While developing my proposed `MANIFEST.in` before, I tried starting with `exclude *` before `include` lines, but then when testing without it, I got the same result, so I didn't...