Ignore unreadable siemens private headers

[hbraunDSP]

Workaround for #812.
dicomwrappers.py will catch and recover from a CSAReadError in csarereader.py, setting csa=None and giving a warning. This allows use on some dicoms with CSA headers that are unreadable by nibabel, either because they are corrupted or not supported yet.

[pep8speaks]

Hello @hbraunDSP, Thank you for updating!

Cheers! There are no style issues detected in this Pull Request. 🍻 To test for issues locally, pip install flake8 and then run flake8 nibabel.

Comment last updated at 2019-11-11 15:57:08 UTC

[effigies]

@hbraunDSP Is this still a WIP? If you can find a test file, that'd be great. Feel free to submit to https://github.com/effigies/nitest-dicom/.

[hbraunDSP]

It should be good to go...I'll be sending a file for testing soon. Thanks for the reminder!

[effigies]

We'll need to update nitest-dicom in https://github.com/nipy/nibabel/tree/master/nibabel-data:

git submodule init
git submodule sync
git submodule update
git -C nibabel-data/nitest-dicom fetch
git -C nibabel-data/nitest-dicom checkout origin/master
git add nibabel-data/nitest-dicom

Then we'll need to add the new file as follows:

nibabel/nibabel/nicom/tests/test_dicomwrappers.py

Lines 40 to 41 in ddf2683

	DATA_FILE_4D_DERIVED = pjoin(get_nibabel_data(), 'nitest-dicom',
	'4d_multiframe_with_derived.dcm')

And test it similar to the following:

nibabel/nibabel/nicom/tests/test_dicomwrappers.py

Lines 628 to 634 in ddf2683

	@dicom_test
	@needs_nibabel_data('nitest-dicom')
	def test_data_derived_shape(self):
	# Test 4D diffusion data with an additional trace volume included
	# Excludes the trace volume and generates the correct shape
	dw = didw.wrapper_from_file(DATA_FILE_4D_DERIVED)
	assert_equal(dw.image_shape, (96, 96, 60, 33))

[codecov]

Codecov Report

Merging #818 into master will increase coverage by 0.26%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #818      +/-   ##
==========================================
+ Coverage    90.1%   90.36%   +0.26%     
==========================================
  Files          96       96              
  Lines       11911    12196     +285     
  Branches     2125     2136      +11     
==========================================
+ Hits        10732    11021     +289     
+ Misses        834      832       -2     
+ Partials      345      343       -2

Impacted Files	Coverage Δ
nibabel/nicom/csareader.py	87.4% <ø> (+2.96%)	⬆️
nibabel/nicom/dicomwrappers.py	90.9% <100%> (+0.65%)	⬆️
nibabel/funcs.py	80.28% <0%> (-0.28%)	⬇️
nibabel/gifti/giftiio.py	100% <0%> (ø)	⬆️
nibabel/keywordonly.py	100% <0%> (ø)	⬆️
nibabel/streamlines/tractogram_file.py	100% <0%> (ø)	⬆️
nibabel/arrayproxy.py	100% <0%> (ø)	⬆️
nibabel/affines.py	100% <0%> (ø)	⬆️
nibabel/imageclasses.py	100% <0%> (ø)	⬆️
nibabel/streamlines/array_sequence.py	100% <0%> (ø)	⬆️
... and 36 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 1d91b72...702515e. Read the comment docs.

[effigies]

@hbraunDSP Just a bump on these tests. If you don't have time, I can make a quick PR to your branch.

[effigies]

See hbraunDSP#2.

[hbraunDSP]

I can't really look at this until Wednesday, sorry. I just merged your PR after a brief glance.

[effigies]

I think this is good as-is, and we're now getting full diff coverage (thanks!). Feel free to open a new PR if there's anything else you'd like to update...