I have eyelink data (.asc) where I "deidentified" the header by blanking out some data, see this example:
** CONVERTED FROM edf using edfapi 4.2.1.0 Windows standalone Jun 19 2021
** DATE:
** TYPE: EDF_FILE BINARY EVENT SAMPLE TAGGED
** VERSION: EYELINK II 1
** SOURCE: EYELINK CL
** EYELINK II CL v5.15 Jan 24 2018
** CAMERA: Eyelink GL Version 1.2 Sensor=AG7
** SERIAL NUMBER:
** CAMERA_CONFIG:
** RECORDED BY ...
**
- "CONVERTED FROM" was adjusted to no longer name the original data file from which the conversion was performed (as it included a recording date)
- "DATE" was blanked out, as were "SERIAL NUMBER" and "CAMERA_CONFIG"
I have performed "blanking out" in discussion with SR-Research (who actually also supplied a sed based script that I could use on EDF files directly), and I assume therefore that their tools won't have problems with this "blanking out".
However, MNE-Python does throw an error:
[331](file:///C:/Users/stefan/miniconda3/envs/pmetrics_analysis/Lib/_strptime.py:331) found = format_regex.match(data_string)
[332](file:///C:/Users/stefan/miniconda3/envs/pmetrics_analysis/Lib/_strptime.py:332) if not found:
--> [333](file:///C:/Users/stefan/miniconda3/envs/pmetrics_analysis/Lib/_strptime.py:333) raise ValueError("time data %r does not match format %r" %
[334](file:///C:/Users/stefan/miniconda3/envs/pmetrics_analysis/Lib/_strptime.py:334) (data_string, format))
[335](file:///C:/Users/stefan/miniconda3/envs/pmetrics_analysis/Lib/_strptime.py:335) if len(data_string) != found.end():
[336](file:///C:/Users/stefan/miniconda3/envs/pmetrics_analysis/Lib/_strptime.py:336) raise ValueError("unconverted data remains: %s" %
[337](file:///C:/Users/stefan/miniconda3/envs/pmetrics_analysis/Lib/_strptime.py:337) data_string[found.end():])
ValueError: time data '' does not match format '%a %b %d %H:%M:%S %Y'
If I edit my ASC file header to include a "DATE" again (of this form: Mon Jan 01 12:12:12 2024), then everything works.
I think we should be able to deal with blanked out dates, and assign a missing measurement date.
Thoughts? @scott-huberty
I have eyelink data (
.asc) where I "deidentified" the header by blanking out some data, see this example:I have performed "blanking out" in discussion with SR-Research (who actually also supplied a
sedbased script that I could use on EDF files directly), and I assume therefore that their tools won't have problems with this "blanking out".However, MNE-Python does throw an error:
If I edit my ASC file header to include a
"DATE"again (of this form:Mon Jan 01 12:12:12 2024), then everything works.I think we should be able to deal with blanked out dates, and assign a missing measurement date.
Thoughts? @scott-huberty