in particular, look at this:
for subject_id in subject_ids: # pragma: no cover src_url = ('http://openfmri.s3.amazonaws.com/tarballs/' 'ds117_R0.1.1_sub%03d_raw.tgz' % subject_id) tar_fname = op.join(data_path, repo + '.tgz') target_dir = op.join(data_path, repo) if not op.exists(target_dir): if not op.exists(tar_fname): _fetch_file(url=src_url, file_name=tar_fname, print_destination=True, resume=True, timeout=10.) tf = tarfile.open(tar_fname) print('Extracting files. This may take a while ...') tf.extractall(path=data_path) shutil.move(op.join(data_path, 'ds117'), target_dir) os.remove(tar_fname)
if more than one subject is defined in
subject_ids, this will have no impact ... because before downloading, there is always the check to only do that if
target_dir does not yet exist.
HOWEVER, target dir will always exist after the first subject has been downloaded.
I prepared the final report already so that @agramfort and @jasmainak have a chance to read it before I submit.
If you guys are okay with it, I'll make the evaluation soon
If I should make some changes, please let me know :-)
@sappelhoff do you plan to do the fetcher before GSoC finishes?
I am a bit worried that it will not get finished ... and when my contract at the institute starts again, I will be expected to do lots of data recording and analysis instead of open source shenanigans :(