Allow users to set work-dir when using bidsmapper (GitHub issue #276)

Author: marcelzwiers

bidscoin/bidscoiner.py

@@ -25,7 +25,7 @@
 from bidscoin.plugins import EventsParser
 
 
-def bidscoiner(sourcefolder: str, bidsfolder: str, participant: list=(), force: bool=False, bidsmap: str= 'bidsmap.yaml', cluster: str= '') -> None:
+def bidscoiner(sourcefolder: str, bidsfolder: str, participant: list=(), force: bool=False, bidsmap: str= 'bidsmap.yaml', cluster: str= '', workdir: str='') -> None:
     """
     Main function that processes all the subjects and session in the sourcefolder and uses the
     bidsmap.yaml file in bidsfolder/code/bidscoin to cast the data into the BIDS folder.
@@ -36,7 +36,7 @@ def bidscoiner(sourcefolder: str, bidsfolder: str, participant: list=(), force:
     :param force:        If True, participant will be processed, regardless of existing folders in the bidsfolder. Otherwise, existing folders will be skipped
     :param bidsmap:      The name of the bidsmap YAML-file. If the bidsmap pathname is just the base name (i.e. no "/" in the name) then it is assumed to be located in the current directory or in bidsfolder/code/bidscoin
     :param cluster:      Use the DRMAA library to submit the bidscoiner jobs to a high-performance compute (HPC) cluster with DRMAA native specifications for submitting bidscoiner jobs to the HPC cluster. See cli/_bidscoiner() for default
-    :return:             Nothing
+    :param workdir:      Working directory for temporary unpacking of zipped or DICOMDIR data. Defaults to the system temporary folder if not specified
     """
 
     # Input checking & defaults
@@ -263,7 +263,7 @@ def bidscoiner(sourcefolder: str, bidsfolder: str, participant: list=(), force:
         for session in sessions:
 
             # Unpack the data in a temporary folder if it is tarballed/zipped and/or contains a DICOMDIR file
-            sesfolders, unpacked = unpack(session, bidsmap.options.get('unzip',''))
+            sesfolders, unpacked = unpack(session, bidsmap.options.get('unzip',''), workdir)
             for sesfolder in sesfolders:
 
                 # Run the bidscoiner plugins

bidscoin/bidsmapper.py

@@ -24,7 +24,7 @@
 _, uptodate, versionmessage = check_version()
 
 
-def bidsmapper(sourcefolder: str, bidsfolder: str, bidsmap: str, template: str, plugins: list, subprefix: str, sesprefix: str, unzip: str, store: bool=False, automated: bool=False, force: bool=False, no_update: bool=False) -> BidsMap:
+def bidsmapper(sourcefolder: str, bidsfolder: str, bidsmap: str, template: str, plugins: list, subprefix: str, sesprefix: str, unzip: str, store: bool=False, automated: bool=False, force: bool=False, no_update: bool=False, workdir: str='') -> BidsMap:
     """
     Main function that processes all the subjects and session in the sourcefolder and that generates a fully filled-in bidsmap.yaml
     file in bidsfolder/code/bidscoin. Folders in sourcefolder are assumed to contain a single dataset.
@@ -39,8 +39,9 @@ def bidsmapper(sourcefolder: str, bidsfolder: str, bidsmap: str, template: str,
     :param unzip:        Wildcard pattern to select tar/zip-files in the session folder. Leave empty to use the bidsmap value
     :param store:        If True, the provenance samples will be stored
     :param automated:    The bidseditor will not be launched if True
-    :param no_update:    Do not update any sub-/ses-prefixes in or prepend the sourcefolder name to the <<filepath:regex>> expression
     :param force:        If True, the previous bidsmap and logfiles will be deleted
+    :param no_update:    Do not update any sub-/ses-prefixes in or prepend the sourcefolder name to the <<filepath:regex>> expression
+    :param workdir:      Working directory for temporary unpacking of zipped or DICOMDIR data. Defaults to the system temporary folder if not specified
     :return:             The new bidsmap
     """
 
@@ -118,7 +119,7 @@ def bidsmapper(sourcefolder: str, bidsfolder: str, bidsmap: str, template: str,
 
             # Unpack the data in a temporary folder if it is tarballed/zipped and/or contains a DICOMDIR file
             LOGGER.info(f"Mapping: {session} (subject {n}/{len(subjects)})")
-            sesfolders, unpacked = unpack(session, unzip)
+            sesfolders, unpacked = unpack(session, unzip, workdir)
             for sesfolder in sesfolders:
                 if store:
                     bidsmap_new.store = {'source': sesfolder.parent.parent.parent.parent if unpacked else rawfolder.parent,

bidscoin/cli/_bidscoiner.py

@@ -34,5 +34,6 @@ def get_parser():
     parser.add_argument('-b','--bidsmap',     help='The study bidsmap file with the mapping heuristics. If the bidsmap filename is just the base name (i.e. no "/" in the name) then it is assumed to be located in the current directory or in bidsfolder/code/bidscoin. Default: bidsmap.yaml', metavar='NAME', default='bidsmap.yaml')
     parser.add_argument('-f','--force',       help='Process all subjects, regardless of existing subject folders in the bidsfolder. Otherwise these subject folders will be skipped', action='store_true')
     parser.add_argument('-c','--cluster',     help='Use the DRMAA library to submit the bidscoiner jobs to a high-performance compute (HPC) cluster. You can add an opaque DRMAA argument with native specifications for your HPC resource manager (NB: Use quotes and include at least one space character to prevent premature parsing -- see examples)', metavar='SPECS', nargs='?', const='-l walltime=00:30:00,mem=4gb', type=str)
+    parser.add_argument('-w','--workdir',     help='Optional working directory for temporary unpacking of zipped or DICOMDIR data. Defaults to the system temporary folder if not specified', default='')
 
     return parser

bidscoin/cli/_bidsmapper.py

@@ -44,6 +44,7 @@ def get_parser() -> argparse.ArgumentParser:
     parser.add_argument('-s','--store',       help='Store newly discovered data samples in the bidsfolder/code/provenance folder (useful for editing e.g. zipped or DICOMDIR datasets)', action='store_true')
     parser.add_argument('-a','--automated',   help='Save the automatically generated bidsmap to disk and without interactively tweaking it with the bidseditor', action='store_true')
     parser.add_argument('-f','--force',       help='Discard the previously saved bidsmap and log file, instead of reusing them (use this option for a fresh start)', action='store_true')
+    parser.add_argument('-w','--workdir',     help='Optional working directory for temporary unpacking of zipped or DICOMDIR data. Defaults to the system temporary folder if not specified', default='')
     parser.add_argument('--no-update',        help="Do not update any sub-/ses-prefixes in or prepend the sourcefolder name to the <<filepath:regex>> expression that extracts the subject/session labels. This is normally done to make the extraction more robust, but could cause problems for certain use cases", action='store_true')
 
     return parser