Add EDF

[garrettmflynn]

Add the EDFRecordingInterface to the list of accepted interfaces.

[CodyCBakerPhD]

@garrettmflynn Do you run into any errors regarding file I/O when you reach the final step using this interface? Just checking if it's platform dependent

[garrettmflynn]

I can't get past the GuidedSourceData page because of the following issue:

cannot import name 'Tester' from 'numpy.testing'

[CodyCBakerPhD]

@garrettmflynn Interesting... does that happen if you regenerate a fresh conda environment?

[garrettmflynn]

Ah I see what you got now. For some reason, clicking the Run Conversion button again allowed me to go through.

I'd seen this issue before because I forgot to add io.close() after reading the HTML representation—but I thought I got rid of it.

Oddly enough, I also saw the same issue—though temporary—when I went back to the Source Data and tried to request the metadata again:

Traceback (most recent call last):
  File "/Users/garrettflynn/Documents/Github/nwb-guide/pyflask/apis/apiNeuroConv.py", line 65, in post
    return get_metadata_schema(api.payload.get("source_data"), api.payload.get("interfaces"))
  File "/Users/garrettflynn/Documents/Github/nwb-guide/pyflask/manageNeuroconv/manage_neuroconv.py", line 128, in get_metadata_schema
    converter = instantiate_custom_converter(source_data, interfaces)
  File "/Users/garrettflynn/Documents/Github/nwb-guide/pyflask/manageNeuroconv/manage_neuroconv.py", line 90, in instantiate_custom_converter
    return CustomNWBConverter(source_data)
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/neuroconv/nwbconverter.py", line 65, in __init__
    self.data_interface_objects = {
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/neuroconv/nwbconverter.py", line 66, in <dictcomp>
    name: data_interface(**source_data[name])
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/neuroconv/datainterfaces/ecephys/edf/edfdatainterface.py", line 37, in __init__
    super().__init__(file_path=file_path, verbose=verbose, es_key=es_key)
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/neuroconv/datainterfaces/ecephys/baserecordingextractorinterface.py", line 32, in __init__
    self.recording_extractor = self.get_extractor()(**source_data)
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/spikeinterface/extractors/neoextractors/edf.py", line 33, in __init__
    NeoBaseRecordingExtractor.__init__(
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/spikeinterface/extractors/neoextractors/neobaseextractor.py", line 185, in __init__
    _NeoBaseExtractor.__init__(self, block_index, **neo_kwargs)
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/spikeinterface/extractors/neoextractors/neobaseextractor.py", line 25, in __init__
    self.neo_reader = self.get_neo_io_reader(self.NeoRawIOClass, **neo_kwargs)
  File "/opt/anaconda3/envs/nwb-guide-edf/lib/python3.9/site-packages/spikeinterface/extractors/neoextractors/neobaseextractor.py", line 64, in get_neo_io_reader
    neo_reader.parse_header()
  File "/opt/anaconda3/envs/env-electron-python/lib/python3.9/site-packages/neo/rawio/baserawio.py", line 179, in parse_header
    self._parse_header()
  File "/opt/anaconda3/envs/env-electron-python/lib/python3.9/site-packages/neo/rawio/edfrawio.py", line 70, in _parse_header
    self.edf_reader = EdfReader(self.filename)
  File "pyedflib/_extensions/_pyedflib.pyx", line 146, in pyedflib._extensions._pyedflib.CyEdfReader.__init__
  File "pyedflib/_extensions/_pyedflib.pyx", line 209, in pyedflib._extensions._pyedflib.CyEdfReader.open
  File "pyedflib/_extensions/_pyedflib.pyx", line 181, in pyedflib._extensions._pyedflib.CyEdfReader.check_open_ok
OSError: /Users/garrettflynn/Desktop/GUIDE/ephy_testing_data/edf/edf+C.edf: file has already been opened```

[CodyCBakerPhD]

Moving to draft until further investigation and possible fix is found

[garrettmflynn]

@CodyCBakerPhD Just tested this again. I can run a successful conversion with an .edf file—so we could push ahead with this and transfer to an issue.

Upon further testing, it seems that the conversion script releases the file some time after returning (on the order of 3-5s). So this is a temporary hold-up, and only if the user goes back to the Source Data page immediately after previewing / converting.

[CodyCBakerPhD]

Ohhhh, huh... right you are

Just waiting 4-5 seconds clears it up. How weird