Anonymized raw can't be saved: mne.io.anonymize_info() doesn't seem to work on the annotations

[adam2392]

Describe the bug

When I instantiate a raw object, attach annotations, and I perform mne.io.anonymize_info() after I get an error. I believe this is due to the fact that anonymization doesn't "alter" the annotations. It works fine if I call anonymization before setting annotations.

Possible solutions:

1. better warning/error message
2. anonymization to alter the annotations - origtime as well

Steps to reproduce

When I run:

# set a measurement date to allow anonymization to run
        raw.set_meas_date(meas_date=datetime.datetime.now(tz=datetime.timezone.utc))
        mne.io.anonymize_info(raw.info)
        raw.save(tmpfpath)

Actual results

I get the following:

  File "/Users/adam2392/Documents/SPORF/mtsmorf/io/run_bids_conversion.py", line 56, in convert_mat_to_bids
    raw.save(tmpfpath)
  File "<decorator-gen-155>", line 21, in save
  File "/Users/adam2392/miniconda3/envs/rf/lib/python3.8/site-packages/mne/io/base.py", line 1345, in save
    _write_raw(fname, self, info, picks, fmt, data_type, reset_range,
  File "/Users/adam2392/miniconda3/envs/rf/lib/python3.8/site-packages/mne/io/base.py", line 1834, in _write_raw
    sk_onsets, sk_ends = _annotations_starts_stops(raw, 'bad_acq_skip')
  File "/Users/adam2392/miniconda3/envs/rf/lib/python3.8/site-packages/mne/annotations.py", line 521, in _annotations_starts_stops
    onsets = _sync_onset(raw, onsets)
  File "/Users/adam2392/miniconda3/envs/rf/lib/python3.8/site-packages/mne/annotations.py", line 495, in _sync_onset
    assert raw.info['meas_date'] == raw.annotations.orig_time

Additional information

Using v0.20

[agramfort]

can you share a short code snippet to replicate the pb?

…

[bloyl]

`anonymize_info` only anonymizes the `meas_info` object. If you want to anonymize the `raw` (both the `meas_info` and `annotations`) you need to call `raw.anonymize()`

…

On Tue, Apr 7, 2020 at 11:36 AM Alexandre Gramfort ***@***.***> wrote: can you share a short code snippet to replicate the pb? > — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#7575 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABKTXHLORAJSJCUG43HNJLDRLNCBRANCNFSM4MDGAPGQ> .

[adam2392]

Awesome. Thanks!