call to uwapm.models() gives error about arlpy HOT 6 CLOSED

The call to pm.models() runs bellhop.exe with an empty env file. This is the expected behavior. But bellhop isn't supposed to terminate with a runtime error. On my machine, I get this back from Bellhop:

$ bellhop.exe test.env 
STOP Fatal Error: See print file for details

So I suspect this is to do with the version of Bellhop or with the compilation options. What version of Bellhop are you using?

Also, can you try running this in a new notebook:

import arlpy.uwapm as pm
import arlpy.plot as plt
import numpy as np

env = pm.create_env2d()
rays = pm.compute_eigenrays(env, model='bellhop')
pm.plot_rays(rays, env=env, width=900)

and see if it works...? This will skip the check for bellhop, and create a env file with proper content and run bellhop on it. It'll help us debug the problem.

from arlpy.

Hi mchitre, thanks for the quick answer! Yes, with a proper environment file, the code executes nicely. So the problem only occurs when the environmental file is empty. I get the same behaviour if I run bellhop.exe with an empty env file in the terminal, independent of your library.

from arlpy.

OK, probably a change in bellhop.exe behavior, or differences across platforms. Let me try and test with the latest version, and I'll update the mechanism to test availability of bellhop in the next release.

In the mean time, I presume you shouldn't be blocked, as you can just add the model = 'bellhop' to your uwapm calls, and it'll work fine without needing the detection.

from arlpy.

Yes, no urgency. I can continue working in the mean time.

Unrelated to this, I was wondering if you have plans to implement 3d environments and execution of bellhop3d?

from arlpy.

I haven't explored the file formats for Bellhop 3D since I don't use it, but they shouldn't be that different. It will likely be useful someday for me, and so I'll be happy to implement it if there's interest by others in using it. If you feel that you'll find value in it, do raise a feature request and I'll attend to it as soon as I get some time.

from arlpy.

Hopefully fixed. I can't test since I don't seem to be able to replicate the issue, but it now uses a non-existent file instead of an empty file, and at least on OSX, that seems to print an error message and gracefully terminate.

from arlpy.