Neuroscan Support

[yjmantilla]

PR Description

PR to add support of .cnt files.

Closes #922

Output of make pep just said:

"Running flake8"
0
"Running pydocstyle"
"Checking MANIFEST.in"
lists of files in version control and sdist match

I ran the tests locally and it said 237 passed (at the end).

Documentation build ended with:

Sphinx-Gallery successfully executed 13 out of 13 files subselected by:

    gallery_conf["filename_pattern"] = '^((?!sgskip).)*$'
    gallery_conf["ignore_pattern"]   = '__init__\\.py'

after excluding 0 files that had previously been run (based on MD5).

embedding documentation hyperlinks...
embedding documentation hyperlinks for auto_examples... [100%] update_bids_datasets.htmlhtml
make[1]: Leaving directory 'Y:/code/mne-bids/doc'
cd doc/ && make view
make[1]: Entering directory 'Y:/code/mne-bids/doc'
make[1]: Leaving directory 'Y:/code/mne-bids/doc'

Let me know if anything else is required, like an specific test for cnt files.

Merge checklist

Maintainer, please confirm the following before merging:

• All comments resolved
• This is not your own PR
• All CIs are happy
• PR title starts with [MRG]
• whats_new.rst is updated
• PR description includes phrase "closes <#issue-number>"

[welcome]

Hello! 👋 Thanks for opening your first pull request here! ❤️ We will try to get back to you soon. 🚴🏽‍♂️

[codecov]

Codecov Report

Merging #924 (f97dd9c) into main (36b21b5) will increase coverage by 0.07%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##             main     #924      +/-   ##
==========================================
+ Coverage   94.71%   94.79%   +0.07%     
==========================================
  Files          23       23              
  Lines        3499     3495       -4     
==========================================
- Hits         3314     3313       -1     
+ Misses        185      182       -3     

Impacted Files	Coverage Δ
mne_bids/config.py	97.46% <ø> (ø)
mne_bids/inspect.py	98.78% <ø> (+1.75%)	⬆️

---

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 36b21b5...f97dd9c. Read the comment docs.

[agramfort]

looks good

@yjmantilla can you add your name in this file

https://github.com/mne-tools/mne-bids/blob/main/doc/whats_new.rst

with a description of the change?

[sappelhoff]

Aaaand could you add a small test please?

[yjmantilla]

@agramfort Sure, im a bit busy but as soon as I can I will do that.

@sappelhoff Yeah I could. Here is a public cnt dataset which may suffice:

https://figshare.com/articles/dataset/The_original_EEG_data_for_driver_fatigue_detection/5202739

Should I add a single or a couple of this cnts file to mne.datasets.testing?

Should I be following an example test? Im guessing it somehow has to do with

mne-bids/mne_bids/tests/test_write.py

Lines 118 to 123 in 4c0755c

	test_converteeg_data = [
	('Persyst', 'BrainVision', 'sub-pt1_ses-02_task-monitor_acq-ecog_run-01_clip2.lay', _read_raw_persyst), # noqa
	('NihonKohden', 'BrainVision', 'MB0400FU.EEG', _read_raw_nihon),
	('Persyst', 'EDF', 'sub-pt1_ses-02_task-monitor_acq-ecog_run-01_clip2.lay', _read_raw_persyst), # noqa
	('NihonKohden', 'EDF', 'MB0400FU.EEG', _read_raw_nihon),
	]

Would it suffice adding the cnt to the testing datasets and append the CNT case to this list?

[agramfort]

maybe look at test functions like test_set or or test_bdf in test_write.py?

…

[sappelhoff]

Would it suffice adding the cnt to the testing datasets and append the CNT case to this list?

In the MNE-testing-data there is a CNT file called scan41_short.cnt. That one will probably suffice for a short/tiny test.

Sure, im a bit busy but as soon as I can I will do that.

Cool, please add info about this PR under "enhancements" then. Note, that you also need to add your name to https://github.com/mne-tools/mne-bids/blob/main/doc/authors.rst in order for the link to your name to work.

maybe look at test functions like test_set or or test_bdf in test_write.py?

agreed, or maybe this one:

mne-bids/mne_bids/tests/test_write.py

Line 1150 in 4c0755c

def test_eegieeg(dir_name, fname, reader, _bids_validate, tmp_path):

[yjmantilla]

Hi @sappelhoff and @agramfort , sorry for the delay. I added the tests and updated what's new.

For the tests I mainly imitated what I saw already implemented and made use of the variables test_eegieeg_data and test_converteeg_data.

I had to add some warning filters for cnt files inspired on:

cnt_ignore_warns = [
    pytest.mark.filterwarnings(
        'ignore:.*Could not parse meas date from the header. Setting to None.'
    ),
    pytest.mark.filterwarnings((
        'ignore:.*Could not define the number of bytes automatically.'
        ' Defaulting to 2.')
    ),
]

found in this mne code

I also had the problem of handedness set to None for the cnt file example. So I had to add also this warning filter:
no_hand='ignore:.*Not setting subject handedness.:RuntimeWarning:mne'

Maybe there is a better way to deal with that.

make pep output:

Running flake8
0
Running pydocstyle
Checking MANIFEST.in
lists of files in version control and sdist match

make test output:

======================= 250 passed in 317.49s (0:05:17) =======================

ci for docs-buikd is failing but maybe this is due to #941 , in correspondence my local build of the docs is also failing with the same error of the ci:

Extension error:
Here is a summary of the problems encountered when running the examples

Unexpected failing examples:
Y:\code\mne-bids\examples\read_bids_datasets.py failed leaving traceback:
Traceback (most recent call last):
  File "Y:\code\mne-bids\examples\read_bids_datasets.py", line 67, in <module>
    openneuro.download(dataset=dataset, target_dir=bids_root,
  File "C:\software\anaconda3\envs\mnedev\lib\site-packages\openneuro\download.py", line 602, in download
    local_tag = _get_local_tag(dataset_id=dataset,
  File "C:\software\anaconda3\envs\mnedev\lib\site-packages\openneuro\download.py", line 495, in _get_local_tag
    raise RuntimeError(f'The existing dataset in the target directory '
RuntimeError: The existing dataset in the target directory appears to be different from the one you requested to download. "DatasetDOI" field in local "dataset_description.json": doi:10.18112/openneuro.ds002778.v1.0.5. Requested dataset: ds002778

[hoechenberger]

I just merged main, hopefully CI will now go green!

[yjmantilla]

Im failing code coverage for some reason because of mne_bids/inspect.py although I didnt touch it. Otherwise it seems good

mne_bids/config.py | 97.46% <ø> (ø) |  
mne_bids/inspect.py | 97.02% <66.66%> (-1.74%) | ⬇️
mne_bids/read.py | 95.84% <100.00%> (+0.02%) | ⬆️
mne_bids/write.py | 96.22% <100.00%> (ø)

[sappelhoff]

thanks @yjmantilla

[agramfort]

@hoechenberger merge if happy

[welcome]

🎉 Congrats on merging your first pull request! 🥳 Looking forward to seeing more from you in the future! 💪

[hoechenberger]

Thanks, @yjmantilla!