WIP: ENH: Implement command run-bidsapp

datalad_neuroimaging/__init__.py

@@ -19,5 +19,12 @@
             'BIDS2Scidata',
             'bids2scidata',
         ),
+        (
+            'datalad_neuroimaging.bidsapp',
+            'BidsApp',
+            'run-bidsapp',
+            'run_bidsapp',
+        )
+
     ]
 )

datalad_neuroimaging/bidsapp.py

@@ -0,0 +1,98 @@
+import logging
+from os.path import join as opj
+from datalad.interface.base import build_doc, Interface
+from datalad.support.param import Parameter
+from datalad.distribution.dataset import datasetmethod, EnsureDataset, \
+    require_dataset, resolve_path, Dataset
+from datalad.interface.utils import eval_results
+from datalad.support.constraints import EnsureStr
+from datalad.support.constraints import EnsureNone
+from datalad.support.exceptions import InsufficientArgumentsError
+from datalad.coreapi import run
+from datalad.interface.results import get_status_dict
+
+
+lgr = logging.getLogger("datalad.cbbsimaging.bidsapp")
+
+
+@build_doc
+class BidsApp(Interface):
+    """Run a BIDSApp on the dataset
+    """
+
+    _params_ = dict(
+        dataset=Parameter(
+            args=("-d", "--dataset"),
+            doc="""dataset to run the BIDSApp on""",
+            constraints=EnsureDataset() | EnsureNone()),
+        cmd=Parameter(
+            args=("cmd",),
+            doc="""Preconfigured BIDSApp to run. The name corresponds to
+            datalad.neuroimaging.bidsapp.NAME configuration.""",
+            metavar="CMD",
+            constraints=EnsureStr() | EnsureNone())
+        )
+
+    @staticmethod
+    @datasetmethod(name='run_bidsapp')
+    @eval_results
+    def __call__(cmd, dataset=None):
+
+        dataset = require_dataset(dataset, check_installed=True,
+                                  purpose="run BIDS App")
+
+        container_name = \
+            dataset.config.get("datalad.neuroimaging.bidsapp.%s.container-name"
+                               % cmd)
+        if not container_name:
+            raise ValueError("Missing configuration for "
+                             "datalad.neuroimaging.bidsapp.%s.container-name"
+                             % cmd)
+
+        container_ds = Dataset(opj(dataset.path, ".git",
+                                   "environment", container_name))
+        if not container_ds.is_installed():
+            source = \
+                dataset.config.get("datalad.neuroimaging.bidsapp.%s.container-url"
+                                   % cmd)
+            if not source:
+                raise ValueError("Missing configuration for "
+                                 "datalad.neuroimaging.bidsapp.%s.container-url"
+                                 % cmd)
+
+            lgr.info("Installing BIDSApp %s from %s", container_name, source)
+            for r in dataset.install(path=container_ds.path, source=source, result_xfm=None, return_type='generator'):
+
+                if r['type'] == 'dataset':
+                    container_ds = Dataset(r['path'])
+                yield r
+
+        image_name = \
+            container_ds.config.get("datalad.neuroimaging.bidsapp.container")
+        if image_name is None:
+            image_name = ""
+        else:
+            if not container_ds.repo.file_has_content(image_name):
+                container_ds.get(image_name)
+
+        image_exec = \
+            container_ds.config.get("datalad.neuroimaging.bidsapp.exec")
+        if not image_exec:
+            raise ValueError("Missing configuration for "
+                             "datalad.neuroimaging.bidsapp.exec in %s" %
+                             container_ds.path)
+
+        cmd_call = \
+            dataset.config.get("datalad.neuroimaging.bidsapp.%s.exec" % cmd)
+        if not cmd_call:
+            raise ValueError("Missing configuration for "
+                             "datalad.neuroimaging.bidsapp.%s.exec" % cmd)
+
+        run_cmd = "{exec_} {img} {call}".format(exec_=image_exec,
+                                                img="" if image_name is None
+                                                else opj(container_ds.path,
+                                                         image_name),
+                                                call=cmd_call)
+
+        for r in dataset.run(run_cmd, message="Run of bidsapp %s" % cmd):
+            yield r