Working on adding entity_message

[rcali21]

WIP-Entity message in BIDS Prov

[rcali21]

Hi @djarecka,

Couple questions....

When I have the following JN test:

shelly = pydra.ShellCommandTask(name='shelly', in_file=file, executable=cmd, audit_flags=AuditFlag.PROV, messengers=PrintMessenger())
print(shelly.inputs.in_file)

I am unable to retrieve the in_file from the print statement and instead I'm left with "Inputs has no attribute 'in_file'" even though I define it as a path to a file in a previous line. I will need this in order to fill the 'at_location' field in entity by getting the abs. path of the in_file.

Additionally, you will see that I have added self.audit_message again at the end of audit.py. I'm assuming there is a more elegant way to add another message to self.audit_message.

[codecov]

Codecov Report

Base: 76.77% // Head: 76.81% // Increases project coverage by +0.04% 🎉

Coverage data is based on head (19beeba) compared to base (4b9df08).
Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #593      +/-   ##
==========================================
+ Coverage   76.77%   76.81%   +0.04%     
==========================================
  Files          19       19              
  Lines        4349     4357       +8     
  Branches     1193     1194       +1     
==========================================
+ Hits         3339     3347       +8     
  Misses        821      821              
  Partials      189      189              

Flag	Coverage Δ
unittests	76.72% <100.00%> (+0.04%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
pydra/engine/audit.py	92.50% <100.00%> (+0.83%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[rcali21]

Wondering if we should strip the filename and just return dir path. 1cc91ef

[djarecka]

Hi @djarecka,

Couple questions....

When I have the following JN test:

shelly = pydra.ShellCommandTask(name='shelly', in_file=file, executable=cmd, audit_flags=AuditFlag.PROV, messengers=PrintMessenger())
print(shelly.inputs.in_file)

I am unable to retrieve the in_file from the print statement and instead I'm left with "Inputs has no attribute 'in_file'" even though I define it as a path to a file in a previous line. I will need this in order to fill the 'at_location' field in entity by getting the abs. path of the in_file.

I don't see input_spec in your example, so in_file is not treated as an input, we talked about it last time

[rcali21]

Will add accompanying pytest for d0770b4 later today.

[djarecka]

I think your current tests stopped working, it might be good to figure out this first

[djarecka]

you fixed the tests, nice! you should add more tests for entity now, I also left some comments, we can discuss them later today if something is not clear

[rcali21]

@djarecka What are your thoughts on adding a test that uses a task that requires file manipulation? Would be helpful for me in further development. Is it best to use a nifti file living on the repo like in nipype, or should we just create a scratch .txt file and then delete it each run?

[djarecka]

@rcali21 - thanks for fixing most of the issues, but I'm not sure why you just removed the second file from the input in test test_audit_shellcommandtask_file. As we discussed that is a good test that checks what happens when you have multiple files as an input. It just needed to be fixed

[rcali21]

@rcali21 - thanks for fixing most of the issues, but I'm not sure why you just removed the second file from the input in test test_audit_shellcommandtask_file. As we discussed that is a good test that checks what happens when you have multiple files as an input. It just needed to be fixed

Resolved in 662ade1.

[djarecka]

Thanks @rcali21 for fixing the test! I know we were talking about adding output in this PR, but I will merge this now, and output can be done in a separate PR. Also, the connection of input and output to the task will have to be added