Update tool HESS to 0.0.1+galaxy0 #59
Conversation
| <requirement type="package" version="7.13.0">nbconvert</requirement> | ||
| <requirement type="package" version="1.20.3">wget</requirement> | ||
| </requirements> | ||
| <command detect_errors="exit_code">ipython '$__tool_directory__/${_data_product._selector}.py'</command> |
There was a problem hiding this comment.
As far as this is converted from notebook, which can contain some jupyter magic, using ipython to run the script assures they will be properly treated. It's not a great sacrifice to have an additional requirement, but the logic may be improved a bit to use it only when magic commands are actually there
| @@ -0,0 +1,138 @@ | |||
| <tool id="hess_astro_tool" name="hess_astro_tool" version="0.0.1+galaxy0" profile="23.0"> | |||
| </conditional> | ||
| </inputs> | ||
| <outputs> | ||
| <data label="${tool.name} -> Image picture" name="out_Image_picture" format="auto" from_work_dir="picture_galaxy.output"> |
There was a problem hiding this comment.
I know format="auto" is the most simple solution and is good in the beginning. I just wanted to point out that is would be nice and also faster if we could specify a fixed value here.
There was a problem hiding this comment.
That's the best general way I was able to came up with. As far as tool description is generated from the notebook, and we can't be 100% sure of the outputs format without actually running it. But I'm attaching some logic to populate the metadata file to the original code of the tool
| <has_text text="*** Job finished successfully ***" /> | ||
| </assert_stdout> | ||
| </test> | ||
| </tests> |
There was a problem hiding this comment.
A help section is missing. What about citations?
|
@bgruening this PR is automatically created by the bot oda-hub/oda-bot#40 using nb2galaxy from oda-hub/nb2workflow#102 It's a bit of a work-in-progress, so some parts are missing. The generation of them is mostly implemented, but the corresponding pieces of info need to be added to the original repo by its author (@andriineronov in this case) |
|
@dsavchenko thanks for the explanations, appreciated! |
No description provided.