Set the field 'TaskName' in the json file for multiecho func data

[pvelasco]

Fixes #417

[codecov]

Codecov Report

Merging #420 into master will increase coverage by 0.07%.
The diff coverage is 92.3%.

@@            Coverage Diff             @@
##           master     #420      +/-   ##
==========================================
+ Coverage   74.71%   74.79%   +0.07%     
==========================================
  Files          35       35              
  Lines        2757     2773      +16     
==========================================
+ Hits         2060     2074      +14     
- Misses        697      699       +2

Impacted Files	Coverage Δ
heudiconv/convert.py	80% <92.3%> (+0.59%)	⬆️
heudiconv/heuristics/reproin.py	81.96% <0%> (-0.16%)	⬇️

---

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 756fd80...7a5f13e. Read the comment docs.

[yarikoptic]

Thank you for pushing better support for multiecho forward!
I think for this PR we need

• take the code which embeds task label out from embed_nifti, and move into a function like you did
  • add a dedicated unit test for it
• remove min_meta option from that embed_nifti and just do not bother even using it in embed_metadata_from_dicoms if min_meta is set
• call that new function in case of bids for all those sidecar files (regardless of their number), thus retaining that warning as I have mentioned

[pvelasco]

• take the code which embeds task label out from embed_nifti, and move into a function like you did

Do you want this function in convert.py, dicoms.py or utils.py?

[yarikoptic]

• take the code which embeds task label out from embed_nifti, and move into a function like you did

Do you want this function in convert.py, dicoms.py or utils.py?

has nothing to do with DICOMs per se, will be used in convert, so let's stick it there for now

[pvelasco]

• remove min_meta option from that embed_nifti and just do not bother even using it in embed_metadata_from_dicoms if min_meta is set

Do we need to call embed_metadata_from_dicoms at all if min_meta is set?
I mean, in convert.py, I can call add_taskname_to_infofile, and then skip the if statements related embed_metadata_from_dicoms if min_meta is set...

[yarikoptic]

Do we need to call embed_metadata_from_dicoms at all if min_meta is set?
I mean, in convert.py, I can call add_taskname_to_infofile, and then skip the if statements related embed_metadata_from_dicoms if min_meta is set...

good observation. That embed_nifti still has the same code as now in add_taskname_to_infofile, and that is (as far as I see it) the only piece which is used if min_meta, so I will strip it away from there (after all it is not embedding any metadata from dicoms ;)), and then indeed we could avoid calling embed_metadata_from_dicoms if minmeta. That with_prov handling wouldn't be run, but if we are not embedding (min_meta is set) I think it is ok. I will push this RFing shortly after making piece of mind (and tests) with it ;)

[yarikoptic]

actually -- I will just merge this one, and then follow up with a new PR on top