Tool Development: How to handle related datasets for a tool?

I am trying to create tools for Automatic Image Registration (AIR) package. Initially, I developed python script which calls align_warp, reslice and softmean tools from AIR package. My problem is when I test my tool with planemo, these tools don’t get some related data files. The command line for align_warp tool is as follows:

align_warp anatomy.img reference.img output.warp -m 12

Along with the files in the command line, align_warp also needs anatomy.hdr and reference.hdr header files which must reside in the same directory as anatomy.img and reference.img.

The problem is when I run test with Planemo, the anatomy.img and reference.img are copied in a temp directory with different names. Even if I add anatomy.hdr and reference.hdr as input parameters, align_warp doesn’t find them because they were renamed.

planemo t align_warp.xml

What I need is Galaxy should copy these files to a temp directory without renaming them and without removing their extensions.

Can anyone please help me with this?

Hi @mmhliton
I am facing the same issue. Have you found a solution?

Thanks

This can be caused by a number of reason, if you shared source code for your tool that’d help. In general you cannot expect the name of the input to be guaranteed. If your software requires it you can make a symlink like in https://github.com/galaxyproject/tools-iuc/blob/main/tools/red/red.xml#L13. But if you wrote the python script change it and do not load your inputs by name.

2 Likes