fs paramater specification causing problem in load_physio

[me-pic]

Expected Behavior

In the generate_physio function, when load_physio is called, it uses the fs specified in the parameters of generate_physio. However, for '.phys' files, this causes problem since the fs information can be retrieved from the file directly. So having fs=None (default value) lead to a fs value of NOTHING (enum type) after loading the file...

See also discussion here.

Actual Behavior

Steps to Reproduce the Problem

1.
2.
3.

Specifications

- Python version:
- peakdet version:
- Platform:

Possible solution

Since the fs parameters should still be specified if loading any file format that is not .phys, I would propose to create another mode to separate those different format rather than grouping everything under mode="physio"

[me-pic]

@smoia wdyt ?

[smoia]

I didn't have a look at the code yet, is it possible that passing arguments more carefully will solve the problem?

Also, yes, fs should be in a 'phys' file, but they should also be in the json file in BIDS format, and I don't remember us supporting anything else than that?

Worth adding a check in the affected function rather than duplicating code, but maybe you have another solution in mind I'm not 100% understanding.

[me-pic]

Should we then constrain the possible file extension here to only .phys here ?

[smoia]

Ah I see. Then yes, please separate the .physio from the other formats, creating another I/O function for those. In those, we can make sure that fs is given while loading somehow.
Unrelated, here's some code for input of those formats:

    if isinstance(data, str):
        if os.path.exists(data):
            ext = os.path.splitext(data)[-1]

            if ext == '.gz':
                ext == os.path.splitext(os.path.splitext(data)[0])[-1]
                    
            if ext == '.tsv':
                data = np.genfromtxt(data, delimiter='\t')
            elif ext == '.csv':
                data = np.genfromtxt(data, delimiter=',')
            else:
                data = np.genfromtxt(data)
        else:
            raise IOError(f'Cannot find {data}')

You can also use reading functions form phys2bids on top of that!

[me-pic]

Thank you for that block of code, will integrate that !