Description of the problem

For a description of the problem check this pdf.

mne.concatenate_raws(raws) wrongly concatenates raws if the order of the channel names varies across raws.

In my opinion, this is a huge bug that I noticed only accidentally. In my data, I add reference channels when they are missing and the index of this channel varies across my raws. In other words, the order of raw.ch_names is not consistent across my raws.

We should probably prepare the raws within mne.concatenate_raws with

for raw in raws:
    raw.reorder_channels(sorted(raw.ch_names))

I don't think we can expect all mne users to know this and do it themselves. I would have never thought that this function does not check the names of the channels it concatenates.

Steps to reproduce

See jupyter-notebook pdf above.

Link to data

No response

Expected results

mne.concatenate_raws reads in the channel names and only concatenates "ch1" of raw1 with "ch1" of raw2.

Actual results

mne.concatenate_raws concatenates raw1.ch_names[0] with raw2.ch_names[0] without checking whether raw1.ch_names[0] == raw2.ch_names[0].

Additional information

Platform: macOS-11.7.3-x86_64-i386-64bit
Python: 3.11.0 | packaged by conda-forge | (main, Jan 15 2023, 05:44:48) [Clang 14.0.6 ]
Executable: /opt/anaconda3/envs/bids_neu/bin/python
CPU: i386: 8 cores
Memory: 16.0 GB

mne: 1.3.0
numpy: 1.23.4 {OpenBLAS 0.3.20 with 4 threads}
scipy: 1.10.0
matplotlib: 3.7.0 {backend=module://matplotlib_inline.backend_inline}

sklearn: 1.2.1
numba: Not found
nibabel: 5.0.1
nilearn: Not found
dipy: Not found
openmeeg: Not found
cupy: Not found
pandas: 1.5.3
pyvista: Not found
pyvistaqt: Not found
ipyvtklink: Not found
vtk: Not found
qtpy: Not found
ipympl: Not found
pyqtgraph: Not found
pooch: v1.6.0

mne_bids: 0.12
mne_nirs: Not found
mne_features: Not found
mne_qt_browser: Not found
mne_connectivity: Not found
mne_icalabel: Not found