Allow setting tags on targets & contents in data fetch API.

lib/galaxy/managers/collections.py

@@ -46,11 +46,11 @@ def __init__(self, app):
         self.tag_manager = tags.GalaxyTagManager(app.model.context)
         self.ldda_manager = lddas.LDDAManager(app)
 
-    def precreate_dataset_collection_instance(self, trans, parent, name, structure, implicit_inputs=None, implicit_output_name=None):
+    def precreate_dataset_collection_instance(self, trans, parent, name, structure, implicit_inputs=None, implicit_output_name=None, tags=[]):
         # TODO: prebuild all required HIDs and send them in so no need to flush in between.
         dataset_collection = self.precreate_dataset_collection(structure, allow_unitialized_element=implicit_output_name is not None)
         instance = self._create_instance_for_collection(
-            trans, parent, name, dataset_collection, implicit_inputs=implicit_inputs, implicit_output_name=implicit_output_name, flush=False
+            trans, parent, name, dataset_collection, implicit_inputs=implicit_inputs, implicit_output_name=implicit_output_name, flush=False, tags=tags
         )
         return instance
 
@@ -148,7 +148,14 @@ def _create_instance_for_collection(self, trans, parent, name, dataset_collectio
             log.exception(message)
             raise MessageException(message)
 
-        tags = self._append_tags(dataset_collection_instance, implicit_inputs, tags)
+        # Tags may be coming in as a dictionary or tag model objects if copying them from other
+        # existing Galaxy objects or as a list of strings if the tags are coming from user supplied
+        # values.
+        if tags and isinstance(tags, list):
+            assert implicit_inputs is None, implicit_inputs
+            tags = self.tag_manager.add_tags_from_list(trans.user, dataset_collection_instance, tags)
+        else:
+            tags = self._append_tags(dataset_collection_instance, implicit_inputs, tags)
         return self.__persist(dataset_collection_instance, flush=flush)
 
     def create_dataset_collection(self, trans, collection_type, element_identifiers=None, elements=None,

lib/galaxy/tools/actions/upload.py

@@ -121,7 +121,8 @@ def _precreate_fetched_hdas(trans, history, target, outputs):
         uploaded_dataset = Bunch(
             type='file', name=name, file_type=file_type, dbkey=dbkey
         )
-        data = upload_common.new_upload(trans, '', uploaded_dataset, library_bunch=None, history=history)
+        tag_list = item.get("tags", [])
+        data = upload_common.new_upload(trans, '', uploaded_dataset, library_bunch=None, history=history, tag_list=tag_list)
         outputs.append(data)
         item["object_id"] = data.id
 
@@ -136,11 +137,12 @@ def _precreate_fetched_collection_instance(trans, history, target, outputs):
     if not name:
         return
 
+    tags = target.get("tags", [])
     collections_service = trans.app.dataset_collections_service
     collection_type_description = collections_service.collection_type_descriptions.for_collection_type(collection_type)
     structure = UninitializedTree(collection_type_description)
     hdca = collections_service.precreate_dataset_collection_instance(
-        trans, history, name, structure=structure
+        trans, history, name, structure=structure, tags=tags
     )
     outputs.append(hdca)
     # Following flushed needed for an ID.

lib/galaxy/tools/actions/upload_common.py

@@ -274,11 +274,16 @@ def __new_library_upload(trans, cntrller, uploaded_dataset, library_bunch, state
     return ldda
 
 
-def new_upload(trans, cntrller, uploaded_dataset, library_bunch=None, history=None, state=None):
+def new_upload(trans, cntrller, uploaded_dataset, library_bunch=None, history=None, state=None, tag_list=[]):
     if library_bunch:
-        return __new_library_upload(trans, cntrller, uploaded_dataset, library_bunch, state)
+        upload_target_dataset_instance = __new_library_upload(trans, cntrller, uploaded_dataset, library_bunch, state)
     else:
-        return __new_history_upload(trans, uploaded_dataset, history=history, state=state)
+        upload_target_dataset_instance = __new_history_upload(trans, uploaded_dataset, history=history, state=state)
+
+    if tag_list:
+        trans.app.tag_handler.add_tags_from_list(trans.user, upload_target_dataset_instance, tag_list)
+
+    return upload_target_dataset_instance
 
 
 def get_uploaded_datasets(trans, cntrller, params, dataset_upload_inputs, library_bunch=None, history=None):

lib/galaxy/tools/data_fetch.py

@@ -95,6 +95,7 @@ def _resolve_src(item):
         dbkey = item.get("dbkey", "?")
         requested_ext = item.get("ext", "auto")
         info = item.get("info", None)
+        tags = item.get("tags", [])
         object_id = item.get("object_id", None)
         link_data_only = upload_config.link_data_only
         if "link_data_only" in item:
@@ -146,6 +147,8 @@ def _resolve_src(item):
             rval["info"] = info
         if object_id is not None:
             rval["object_id"] = object_id
+        if tags:
+            rval["tags"] = tags
         return rval
 
     elements = elements_tree_map(_resolve_src, items)

lib/galaxy/tools/parameters/output_collect.py

@@ -407,7 +407,7 @@ def populate_collection_elements(self, collection, root_collection_builder, file
             dataset_name = fields_match.name or designation
 
             link_data = discovered_file.match.link_data
-
+            tag_list = discovered_file.match.tag_list
             dataset = self.create_dataset(
                 ext=ext,
                 designation=designation,
@@ -417,6 +417,7 @@ def populate_collection_elements(self, collection, root_collection_builder, file
                 filename=filename,
                 metadata_source_name=metadata_source_name,
                 link_data=link_data,
+                tag_list=tag_list,
             )
             log.debug(
                 "(%s) Created dynamic collection dataset for path [%s] with element identifier [%s] for output [%s] %s",
@@ -473,6 +474,7 @@ def create_dataset(
         library_folder=None,
         link_data=False,
         primary_data=None,
+        tag_list=[],
     ):
         app = self.app
         sa_session = self.sa_session
@@ -493,6 +495,10 @@ def create_dataset(
             metadata_source = self.inp_data[metadata_source_name]
 
         sa_session.flush()
+
+        if tag_list:
+            app.tag_handler.add_tags_from_list(self.job.user, primary_data, tag_list)
+
         # Move data from temp location to dataset location
         if not link_data:
             app.object_store.update_from_file(primary_data.dataset, file_name=filename, create=True)
@@ -873,6 +879,10 @@ def visible(self):
     def link_data(self):
         return bool(self.as_dict.get("link_data_only", False))
 
+    @property
+    def tag_list(self):
+        return self.as_dict.get("tags", [])
+
     @property
     def object_id(self):
         return self.as_dict.get("object_id", None)

test/api/test_dataset_collections.py

@@ -189,12 +189,13 @@ def test_enforces_unique_names(self):
         self._assert_status_code_is(create_response, 400)
 
     def test_upload_collection(self):
-        elements = [{"src": "files", "dbkey": "hg19", "info": "my cool bed"}]
+        elements = [{"src": "files", "dbkey": "hg19", "info": "my cool bed", "tags": ["name:data1", "group:condition:treated", "machine:illumina"]}]
         targets = [{
             "destination": {"type": "hdca"},
             "elements": elements,
             "collection_type": "list",
             "name": "Test upload",
+            "tags": ["name:collection1"]
         }]
         payload = {
             "history_id": self.history_id,
@@ -204,10 +205,16 @@ def test_upload_collection(self):
         self.dataset_populator.fetch(payload)
         hdca = self._assert_one_collection_created_in_history()
         self.assertEquals(hdca["name"], "Test upload")
+        hdca_tags = hdca["tags"]
+        assert len(hdca_tags) == 1
+        assert "name:collection1" in hdca_tags
         assert len(hdca["elements"]) == 1, hdca
         element0 = hdca["elements"][0]
         assert element0["element_identifier"] == "4.bed"
-        assert element0["object"]["file_size"] == 61
+        dataset0 = element0["object"]
+        assert dataset0["file_size"] == 61
+        dataset_tags = dataset0["tags"]
+        assert len(dataset_tags) == 3, dataset0
 
     def test_upload_nested(self):
         elements = [{"name": "samp1", "elements": [{"src": "files", "dbkey": "hg19", "info": "my cool bed"}]}]

test/api/test_tools_upload.py

@@ -227,10 +227,15 @@ def test_upload_dbkey(self):
             datasets = run_response.json()["outputs"]
             assert datasets[0].get("genome_build") == "hg19", datasets[0]
 
-    def test_fetch_dbkey(self):
+    def test_fetch_metadata(self):
         table = ONE_TO_SIX_WITH_SPACES
-        details = self._upload_and_get_details(table, api='fetch', dbkey="hg19")
+        details = self._upload_and_get_details(table, api='fetch', dbkey="hg19", info="cool upload", tags=["name:data", "group:type:paired-end"])
         assert details.get("genome_build") == "hg19"
+        assert details.get("misc_info") == "cool upload", details
+        tags = details.get("tags")
+        assert len(tags) == 2, details
+        assert "group:type:paired-end" in tags
+        assert "name:data" in tags
 
     def test_upload_multiple_files_1(self):
         with self.dataset_populator.test_history() as history_id: