Use a file lock so that if multiple subjects are processed in parallel, the output dataset_description.json will only be written once.
Also fix the docker tag in the metadata, which referenced fmriprep.
I chose to make the lock file visible (ie, not starting with a .) because I figure it's best if it's visible just in case it causes problems and needs to be manually cleaned up. Happy to change this though
Changes proposed in this pull request
Use a file lock so that if multiple subjects are processed in parallel, the output
dataset_description.json
will only be written once.Also fix the docker tag in the metadata, which referenced fmriprep.
I chose to make the lock file visible (ie, not starting with a .) because I figure it's best if it's visible just in case it causes problems and needs to be manually cleaned up. Happy to change this though