Add Dataset Collections to Object Oriented API

bioblend/galaxy/objects/wrappers.py

@@ -447,7 +447,6 @@ def run(self, input_map=None, history='', params=None, import_inputs=False,
         res = self.gi.gi.workflows.run_workflow(self.id, **kwargs)
         # res structure: {'history': HIST_ID, 'outputs': [DS_ID, DS_ID, ...]}
         out_hist = self.gi.histories.get(res['history'])
-        assert set(res['outputs']).issubset(out_hist.dataset_ids)
         outputs = [out_hist.get_dataset(_) for _ in res['outputs']]
 
         if wait:
@@ -626,6 +625,63 @@ def delete(self):
         self.refresh()
 
 
+@six.add_metaclass(abc.ABCMeta)
+class DatasetCollection(Wrapper):
+    """
+    Abstract base class for Galaxy dataset collections.
+    """
+    BASE_ATTRS = Wrapper.BASE_ATTRS + (
+        'state', 'deleted'
+    )
+    POLLING_INTERVAL = 1  # for state monitoring
+
+    @abc.abstractmethod
+    def __init__(self, dsc_dict, container, gi=None):
+        super(DatasetCollection, self).__init__(dsc_dict, gi=gi)
+        object.__setattr__(self, 'container', container)
+
+    def refresh(self):
+        """
+        Re-fetch the attributes pertaining to this object.
+
+        Returns: self
+        """
+        gi_client = getattr(self.gi.gi, self.container.API_MODULE)
+        dsc_dict = gi_client.show_dataset_collection(self.container.id, self.id)
+        self.__init__(dsc_dict, self.container, self.gi)
+        return self
+
+
+class HistoryDatasetCollectionAssociation(DatasetCollection):
+    """
+    Maps to a Galaxy ``HistoryDatasetCollectionAssociation``.
+    """
+    BASE_ATTRS = DatasetCollection.BASE_ATTRS + ('tags', 'visible', 'elements')
+    SRC = 'hdca'
+
+    def __init__(self, dsc_dict, container, gi=None):
+        super(HistoryDatasetCollectionAssociation, self).__init__(
+            dsc_dict, container, gi=gi)
+
+    @property
+    def gi_module(self):
+        return self.gi.histories
+
+    @property
+    def _stream_url(self):
+        base_url = self.gi.gi._make_url(
+            self.gi.gi.histories, module_id=self.container.id, contents=True)
+        return "%s/%s/display" % (base_url, self.id)
+
+    def delete(self):
+        """
+        Delete this dataset collection.
+        """
+        self.gi.gi.histories.delete_dataset_collection(self.container.id, self.id)
+        self.container.refresh()
+        self.refresh()
+
+
 class LibRelatedDataset(Dataset):
     """
     Base class for LibraryDatasetDatasetAssociation and LibraryDataset classes.
@@ -807,6 +863,7 @@ class History(DatasetContainer):
     """
     BASE_ATTRS = DatasetContainer.BASE_ATTRS + ('annotation', 'state', 'state_ids', 'state_details', 'tags')
     DS_TYPE = HistoryDatasetAssociation
+    DSC_TYPE = HistoryDatasetCollectionAssociation
     CONTENT_INFO_TYPE = HistoryContentInfo
     API_MODULE = 'histories'
 
@@ -965,6 +1022,33 @@ def download(self, jeha_id, outf, chunk_size=bioblend.CHUNK_SIZE):
         return self.gi.gi.histories.download_history(
             self.id, jeha_id, outf, chunk_size=chunk_size)
 
+    def new_dataset_collection(self, collection_description):
+        """
+        Create a new dataset collection in the history by providing the dataset ids.
+
+        :type collection_description: bioblend.galaxy.dataset_collections.CollectionDescription
+        :param collection_description: a description of the dataset collection
+
+        :rtype: class:`~.HistoryDatasetCollectionAssociation`
+        :return: the new dataset collection
+        """
+        dataset_collection = self.gi.gi.histories.create_dataset_collection(self.id, collection_description)
+        self.refresh()
+        return self.get_dataset_collection(dataset_collection['id'])
+
+    def get_dataset_collection(self, dsc_id):
+        """
+        Retrieve the dataset collection corresponding to the given id.
+
+        :type dsc_id: str
+        :param dsc_id: dataset collection id
+
+        :rtype: :class:`~.HistoryDatasetCollectionAssociation`
+        :return: the dataset collection corresponding to ``dsc_id``
+        """
+        dsc_dict = self.gi.gi.histories.show_dataset_collection(self.id, dsc_id)
+        return self.DSC_TYPE(dsc_dict, self, gi=self.gi)
+
 
 class Library(DatasetContainer):
     """

tests/TestGalaxyObjects.py

@@ -16,6 +16,7 @@
 import bioblend.galaxy.objects.wrappers as wrappers
 import bioblend.galaxy.objects.galaxy_instance as galaxy_instance
 from bioblend import ConnectionError
+from bioblend.galaxy import dataset_collections as collections
 
 import test_util
 from test_util import unittest
@@ -558,6 +559,15 @@ def __check_dataset(self, hda):
         self.assertEqual(len(self.hist.dataset_ids), 1)
         self.assertEqual(self.hist.dataset_ids[0], hda.id)
 
+    def __check_dataset_collection(self, hdca):
+        self.assertIsInstance(hdca, wrappers.HistoryDatasetCollectionAssociation)
+        self.assertIs(hdca.container, self.hist)
+        self.assertEqual(len(self.hist.dataset_ids), 2)
+        dataset1 = self.hist.get_dataset(self.hist.dataset_ids[0])
+        dataset2 = self.hist.get_dataset(self.hist.dataset_ids[1])
+        self.assertEqual(dataset1.id, hdca.elements[0]['object']['id'])
+        self.assertEqual(dataset2.id, hdca.elements[1]['object']['id'])
+
     def test_import_dataset(self):
         lib = self.gi.libraries.create('test_%s' % uuid.uuid4().hex)
         lds = lib.upload_data(FOO_DATA)
@@ -617,6 +627,29 @@ def test_update(self):
         self.assertEqual(self.hist.annotation, new_annotation)
         self.assertEqual(self.hist.tags, new_tags)
 
+    def test_new_dataset_collection(self):
+        collection_description = self._create_collection_description()
+        dataset_collection = self.hist.new_dataset_collection(collection_description)
+        self.__check_dataset_collection(dataset_collection)
+
+    def test_delete_dataset_collection(self):
+        collection_description = self._create_collection_description()
+        dataset_collection = self.hist.new_dataset_collection(collection_description)
+        dataset_collection.delete()
+        self.assertTrue(dataset_collection.deleted)
+
+    def _create_collection_description(self):
+        dataset1 = self.hist.paste_content(FOO_DATA)
+        dataset2 = self.hist.paste_content(FOO_DATA_2)
+        collection_description = collections.CollectionDescription(
+            name="MyDatasetList",
+            elements=[
+                collections.HistoryDatasetElement(name="sample1", id=dataset1.id),
+                collections.HistoryDatasetElement(name="sample2", id=dataset2.id),
+            ]
+        )
+        return collection_description
+
 
 @test_util.skip_unless_galaxy()
 class TestHDAContents(GalaxyObjectsTestBase):
@@ -704,6 +737,24 @@ def test_new_history(self):
     def test_params(self):
         self.__test(params=True)
 
+    def test_run_workflow_with_dataset_collection(self):
+        wf_file = os.path.join(THIS_DIR, 'data', 'dataset_collection_run.ga')
+        with open(wf_file) as f:
+            wf = self.gi.workflows.import_new(f.read())
+        history_name = "Run Workflow With Dataset Collection"
+        outputhist = self.gi.histories.create(history_name)
+        collection_description = collections.CollectionDescription(
+            name="MyDatasetList",
+            elements=[
+                collections.HistoryDatasetElement(name="sample1", id=self.inputs[0].id),
+                collections.HistoryDatasetElement(name="sample2", id=self.inputs[1].id),
+            ]
+        )
+        dataset_collection = outputhist.new_dataset_collection(collection_description)
+        input_map = {"Input Dataset Collection": dataset_collection}
+        outputs = wf.run(input_map, history_name)
+        self.assertEqual(len(outputs), 2)
+
 
 @test_util.skip_unless_galaxy()
 class TestJob(GalaxyObjectsTestBase):

tests/data/dataset_collection_run.ga

@@ -0,0 +1,65 @@
+{
+    "a_galaxy_workflow": "true", 
+    "annotation": "", 
+    "format-version": "0.1", 
+    "name": "DC Run", 
+    "steps": {
+        "0": {
+            "annotation": "", 
+            "id": 0, 
+            "input_connections": {}, 
+            "inputs": [
+                {
+                    "description": "", 
+                    "name": "Input Dataset Collection"
+                }
+            ], 
+            "label": null, 
+            "name": "Input dataset collection", 
+            "outputs": [], 
+            "position": {
+                "left": 338, 
+                "top": 306
+            }, 
+            "tool_errors": null, 
+            "tool_id": null, 
+            "tool_state": "{\"collection_type\": \"list\", \"name\": \"Input Dataset Collection\"}", 
+            "tool_version": null, 
+            "type": "data_collection_input", 
+            "user_outputs": [], 
+            "uuid": "4a497708-240c-420e-9c7c-a2dc0200476e"
+        }, 
+        "1": {
+            "annotation": "", 
+            "id": 1, 
+            "input_connections": {
+                "input": {
+                    "id": 0, 
+                    "output_name": "output"
+                }
+            }, 
+            "inputs": [], 
+            "label": null, 
+            "name": "Sort", 
+            "outputs": [
+                {
+                    "name": "out_file1", 
+                    "type": "input"
+                }
+            ], 
+            "position": {
+                "left": 676, 
+                "top": 286
+            }, 
+            "post_job_actions": {}, 
+            "tool_errors": null, 
+            "tool_id": "sort1", 
+            "tool_state": "{\"__page__\": 0, \"style\": \"\\\"num\\\"\", \"column\": \"\\\"1\\\"\", \"__rerun_remap_job_id__\": null, \"order\": \"\\\"DESC\\\"\", \"input\": \"null\", \"column_set\": \"[]\"}", 
+            "tool_version": "1.0.3", 
+            "type": "tool", 
+            "user_outputs": [], 
+            "uuid": "3c33333f-08ce-43bf-bc09-fa0b5776845b"
+        }
+    }, 
+    "uuid": "fd34f048-e17a-4e06-85e8-ed391192d284"
+}