[FIX] Add CE agent to output figure filename templates

[jdkent]

supercedes #479
addresses nipreps/fmriprep#2033

[pull-assistant]

Best reviewed: commit by commit

---

Optimal code review plan

     replace ce with ceagent

     test ceagent

     add ceagent as needed_entity for tests

     ensure one file per reportlet

     change dir to direction

     change expected size of output

     ensure titles have the correct keys

     add the required entities to figures.json

     ensure all the figures are copied over

Powered by Pull Assistant. Last update 4ae4176 ... eb1c4bd. Read the comment docs.

[effigies]

Does this have any API effect?

[jdkent]

No, I do not believe so. This does not impact the recently added --bids-filter, and makes figures.json naming consistent with fmriprep.yml.

[effigies]

Okay. @oesteban Is this consistent with your understanding as well?

[oesteban]

Yup, I also think this should not affect the API

[effigies]

Actually, since #479 isn't merged, let's just make this the main PR, and we'll merge into master after the next release. (Assuming that this is API neutral and safe to put in a bug-fix.)

[codecov]

Codecov Report

Merging #482 into maint/1.1.x will decrease coverage by 5.17%.
The diff coverage is 100.00%.

@@               Coverage Diff               @@
##           maint/1.1.x     #482      +/-   ##
===============================================
- Coverage        62.93%   57.76%   -5.18%     
===============================================
  Files               41       40       -1     
  Lines             5005     4991      -14     
  Branches           727      726       -1     
===============================================
- Hits              3150     2883     -267     
- Misses            1704     1980     +276     
+ Partials           151      128      -23     

Flag	Coverage Δ
#documentation	?
#reportlettests	?
#travis	57.76% <100.00%> (-0.03%)	⬇️
#unittests	?

Impacted Files	Coverage Δ
niworkflows/reports/core.py	72.09% <100.00%> (-0.48%)	⬇️
niworkflows/func/util.py	25.00% <0.00%> (-62.50%)	⬇️
niworkflows/anat/ants.py	12.15% <0.00%> (-57.46%)	⬇️
niworkflows/anat/freesurfer.py	39.13% <0.00%> (-52.18%)	⬇️
niworkflows/anat/skullstrip.py	30.00% <0.00%> (-50.00%)	⬇️
niworkflows/interfaces/itk.py	26.92% <0.00%> (-12.31%)	⬇️
niworkflows/interfaces/fixes.py	41.17% <0.00%> (-11.77%)	⬇️
niworkflows/interfaces/bids.py	80.15% <0.00%> (-10.32%)	⬇️
niworkflows/interfaces/ants.py	57.81% <0.00%> (-7.82%)	⬇️
niworkflows/interfaces/utils.py	37.76% <0.00%> (-4.99%)	⬇️
... and 1 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2bf5430...eb1c4bd. Read the comment docs.

[effigies]

I've tested with a synthetic dataset where I switched from run-* to ce-*, and fmriprep 20.0.4 + this PR works as expected.

Given @bbfrederick's comments, I'm going to go ahead and merge.

Thanks!

[bbfrederick]

Works for me. I'll wait until the dust settles, sync up with the patched release version of niworkflows, and then put in a PR for the BIDS_NAME thing. AFAIK it doesn't actually cause any problems at the moment, but it is wrong and it might bite somebody later if they decide to use it, so it's not currently urgent.