RF: extracted functionality into a dedicated populate_aggregated_jsons

PyCharm also introduced minor code reformattings to avoid too long lines

Author: yarikoptic

heudiconv/bids.py

@@ -69,16 +69,37 @@ def populate_bids_templates(path, defaults={}):
         "TODO: Provide description for the dataset -- basic details about the "
         "study, possibly pointing to pre-registration (if public or embargoed)")
 
+    populate_aggregated_jsons(path)
+
+
+def populate_aggregated_jsons(path):
+    """Aggregate across the entire BIDS dataset .json's into top level .json's
+
+    Top level .json files would contain only the fields which are
+    common to all subject[/session]/type/*_modality.json's.
+
+    ATM aggregating only for *_task*_bold.json files. Only the task- and
+    OPTIONAL _acq- field is retained within the aggregated filename.  The other
+    BIDS _key-value pairs are "aggregated over".
+
+    Parameters
+    ----------
+    path: str
+      Path to the top of the BIDS dataset
+    """
     # TODO: collect all task- .json files for func files to
     tasks = {}
     # way too many -- let's just collect all which are the same!
     # FIELDS_TO_TRACK = {'RepetitionTime', 'FlipAngle', 'EchoTime',
     #                    'Manufacturer', 'SliceTiming', ''}
     for fpath in find_files('.*_task-.*\_bold\.json', topdir=path,
-                        exclude_vcs=True, exclude="/\.(datalad|heudiconv)/"):
+                            exclude_vcs=True,
+                            exclude="/\.(datalad|heudiconv)/"):
         #
-        # According to BIDS spec I think both _task AND _acq (may be more? _rec, _dir, ...?) should be retained?
-        # TODO: if we are to fix it, then old ones (without _acq) should be removed first
+        # According to BIDS spec I think both _task AND _acq (may be more?
+        # _rec, _dir, ...?) should be retained?
+        # TODO: if we are to fix it, then old ones (without _acq) should be
+        # removed first
         task = re.sub('.*_(task-[^_\.]*(_acq-[^_\.]*)?)_.*', r'\1', fpath)
         json_ = load_json(fpath)
         if task not in tasks:
@@ -97,8 +118,10 @@ def populate_bids_templates(path, defaults={}):
         if not op.lexists(events_file):
             lgr.debug("Generating %s", events_file)
             with open(events_file, 'w') as f:
-                f.write("onset\tduration\ttrial_type\tresponse_time\tstim_file\tTODO -- fill in rows and add more tab-separated columns if desired")
-
+                f.write(
+                    "onset\tduration\ttrial_type\tresponse_time\tstim_file"
+                    "\tTODO -- fill in rows and add more tab-separated "
+                    "columns if desired")
     # extract tasks files stubs
     for task_acq, fields in tasks.items():
         task_file = op.join(path, task_acq + '_bold.json')
@@ -110,7 +133,7 @@ def populate_bids_templates(path, defaults={}):
         # to be populated from the file if it already exists
         placeholders = {
             "TaskName": ("TODO: full task name for %s" %
-                           task_acq.split('_')[0].split('-')[1]),
+                         task_acq.split('_')[0].split('-')[1]),
             "CogAtlasID": "TODO",
         }
         if op.lexists(task_file):