Merge branch 'release_24.0' into release_24.1

Author: nsoranzo

.github/workflows/pr-title-update.yml

@@ -2,26 +2,28 @@ name: Update PR title
 
 on:
   pull_request_target:
-    types: [opened, edited]
-    branches:
-      - "release_**"
+    types: [opened, edited, reopened]
 
 jobs:
   update-title:
+    if: github.event.action != 'edited' || github.event.changes.base.ref.from != ''
     runs-on: ubuntu-latest
     permissions:
       pull-requests: write
     steps:
-      - uses: actions/checkout@v4
       - name: Update PR title
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           PR_NUMBER: ${{ github.event.pull_request.number }}
           TARGET_BRANCH: "${{ github.base_ref }}"
           PR_TITLE: "${{ github.event.pull_request.title }}"
+          REPO: "${{ github.repository }}"
         run: |
-          VERSION=$(echo $TARGET_BRANCH | grep -oP '\d+\.\d+')
-          if [[ -n "$VERSION" && ! "$PR_TITLE" =~ ^\[$VERSION\] ]]; then
-            NEW_TITLE="[$VERSION] $PR_TITLE"
-            gh pr edit $PR_NUMBER --title "$NEW_TITLE"
+          VERSION=$(echo $TARGET_BRANCH | grep -oP '^release_\K\d+.\d+$' || true)
+          NEW_TITLE=$(echo "$PR_TITLE" | sed -E "s/\[[0-9]+\.[0-9]+\] //")
+          if [[ -n "$VERSION" ]]; then
+            NEW_TITLE="[$VERSION] $NEW_TITLE"
+          fi
+          if [[ "$NEW_TITLE" != "$PR_TITLE" ]]; then
+            gh pr edit $PR_NUMBER --repo "$REPO" --title "$NEW_TITLE"
           fi

lib/galaxy/managers/workflows.py

@@ -598,7 +598,7 @@ def read_workflow_from_path(self, app, user, path, allow_in_directory=None) -> m
         import_options = ImportOptions()
         import_options.deduplicate_subworkflows = True
         as_dict = python_to_workflow(as_dict, galaxy_interface, workflow_directory=None, import_options=import_options)
-        raw_description = RawWorkflowDescription(as_dict, path)
+        raw_description = RawWorkflowDescription(as_dict)
         created_workflow = self.build_workflow_from_raw_description(trans, raw_description, WorkflowCreateOptions())
         return created_workflow.workflow
 
@@ -937,8 +937,9 @@ def to_format_2(wf_dict, **kwds):
         return wf_dict
 
     def _sync_stored_workflow(self, trans, stored_workflow):
-        workflow_path = stored_workflow.from_path
-        self.store_workflow_to_path(workflow_path, stored_workflow, stored_workflow.latest_workflow, trans=trans)
+        if trans.user_is_admin:
+            workflow_path = stored_workflow.from_path
+            self.store_workflow_to_path(workflow_path, stored_workflow, stored_workflow.latest_workflow, trans=trans)
 
     def store_workflow_artifacts(self, directory, filename_base, workflow, **kwd):
         modern_workflow_path = os.path.join(directory, f"{filename_base}.gxwf.yml")

lib/galaxy/model/__init__.py

@@ -2630,7 +2630,7 @@ class PostJobAction(Base, RepresentById):
     workflow_step_id: Mapped[Optional[int]] = mapped_column(ForeignKey("workflow_step.id"), index=True)
     action_type: Mapped[str] = mapped_column(String(255))
     output_name: Mapped[Optional[str]] = mapped_column(String(255))
-    action_arguments: Mapped[Optional[bytes]] = mapped_column(MutableJSONType)
+    _action_arguments: Mapped[Optional[bytes]] = mapped_column(MutableJSONType)
     workflow_step: Mapped[Optional["WorkflowStep"]] = relationship(
         back_populates="post_job_actions",
         primaryjoin=(lambda: WorkflowStep.id == PostJobAction.workflow_step_id),
@@ -2643,6 +2643,18 @@ def __init__(self, action_type, workflow_step=None, output_name=None, action_arg
         self.workflow_step = workflow_step
         ensure_object_added_to_session(self, object_in_session=workflow_step)
 
+    @property
+    def action_arguments(self):
+        if self.action_type in ("HideDatasetAction", "DeleteIntermediatesAction") and self._action_arguments is True:
+            # Fix up broken workflows resulting from imports with gxformat2 <= 0.20.0
+            return {}
+        else:
+            return self._action_arguments
+
+    @action_arguments.setter
+    def action_arguments(self, value: Dict[str, Any]):
+        self._action_arguments = value
+
 
 class PostJobActionAssociation(Base, RepresentById):
     __tablename__ = "post_job_action_association"
@@ -6900,11 +6912,6 @@ class HistoryDatasetCollectionAssociation(
         primaryjoin=copied_from_history_dataset_collection_association_id == id,
         remote_side=[id],
         uselist=False,
-        back_populates="copied_to_history_dataset_collection_association",
-    )
-    copied_to_history_dataset_collection_association = relationship(
-        "HistoryDatasetCollectionAssociation",
-        back_populates="copied_from_history_dataset_collection_association",
     )
     implicit_input_collections: Mapped[List["ImplicitlyCreatedDatasetCollectionInput"]] = relationship(
         primaryjoin=(

lib/galaxy/model/store/__init__.py

@@ -999,14 +999,16 @@ def _import_collection_copied_associations(
             # sense.
             hdca_copied_from_sinks = object_import_tracker.hdca_copied_from_sinks
             if copied_from_object_key in object_import_tracker.hdcas_by_key:
-                hdca.copied_from_history_dataset_collection_association = object_import_tracker.hdcas_by_key[
-                    copied_from_object_key
-                ]
+                source_hdca = object_import_tracker.hdcas_by_key[copied_from_object_key]
+                if source_hdca is not hdca:
+                    # We may not have the copied source, in which case the first included HDCA in the chain
+                    # acts as the source, so here we make sure we don't create a cycle.
+                    hdca.copied_from_history_dataset_collection_association = source_hdca
             else:
                 if copied_from_object_key in hdca_copied_from_sinks:
-                    hdca.copied_from_history_dataset_collection_association = object_import_tracker.hdcas_by_key[
-                        hdca_copied_from_sinks[copied_from_object_key]
-                    ]
+                    source_hdca = object_import_tracker.hdcas_by_key[hdca_copied_from_sinks[copied_from_object_key]]
+                    if source_hdca is not hdca:
+                        hdca.copied_from_history_dataset_collection_association = source_hdca
                 else:
                     hdca_copied_from_sinks[copied_from_object_key] = dataset_collection_key
 
@@ -1072,7 +1074,7 @@ def attach_workflow_step(imported_object, attrs):
             for step_attrs in invocation_attrs["steps"]:
                 imported_invocation_step = model.WorkflowInvocationStep()
                 imported_invocation_step.workflow_invocation = imported_invocation
-                ensure_object_added_to_session(imported_invocation, session=self.sa_session)
+                ensure_object_added_to_session(imported_invocation_step, session=self.sa_session)
                 attach_workflow_step(imported_invocation_step, step_attrs)
                 restore_times(imported_invocation_step, step_attrs)
                 imported_invocation_step.action = step_attrs["action"]
@@ -1926,12 +1928,14 @@ def __init__(
         self.export_files = export_files
         self.included_datasets: Dict[model.DatasetInstance, Tuple[model.DatasetInstance, bool]] = {}
         self.dataset_implicit_conversions: Dict[model.DatasetInstance, model.ImplicitlyConvertedDatasetAssociation] = {}
-        self.included_collections: List[Union[model.DatasetCollection, model.HistoryDatasetCollectionAssociation]] = []
+        self.included_collections: Dict[
+            Union[model.DatasetCollection, model.HistoryDatasetCollectionAssociation],
+            Union[model.DatasetCollection, model.HistoryDatasetCollectionAssociation],
+        ] = {}
         self.included_libraries: List[model.Library] = []
         self.included_library_folders: List[model.LibraryFolder] = []
         self.included_invocations: List[model.WorkflowInvocation] = []
         self.collection_datasets: Set[int] = set()
-        self.collections_attrs: List[Union[model.DatasetCollection, model.HistoryDatasetCollectionAssociation]] = []
         self.dataset_id_to_path: Dict[int, Tuple[Optional[str], Optional[str]]] = {}
 
         self.job_output_dataset_associations: Dict[int, Dict[str, model.DatasetInstance]] = {}
@@ -2298,8 +2302,7 @@ def export_collection(
     def add_dataset_collection(
         self, collection: Union[model.DatasetCollection, model.HistoryDatasetCollectionAssociation]
     ) -> None:
-        self.collections_attrs.append(collection)
-        self.included_collections.append(collection)
+        self.included_collections[collection] = collection
 
     def add_implicit_conversion_dataset(
         self,
@@ -2362,7 +2365,7 @@ def to_json(attributes):
 
         collections_attrs_filename = os.path.join(export_directory, ATTRS_FILENAME_COLLECTIONS)
         with open(collections_attrs_filename, "w") as collections_attrs_out:
-            collections_attrs_out.write(to_json(self.collections_attrs))
+            collections_attrs_out.write(to_json(self.included_collections.values()))
 
         conversions_attrs_filename = os.path.join(export_directory, ATTRS_FILENAME_CONVERSIONS)
         with open(conversions_attrs_filename, "w") as conversions_attrs_out:
@@ -2383,12 +2386,12 @@ def to_json(attributes):
             #
 
             # Get all jobs associated with included HDAs.
-            jobs_dict: Dict[str, model.Job] = {}
+            jobs_dict: Dict[int, model.Job] = {}
             implicit_collection_jobs_dict = {}
 
             def record_job(job):
-                if not job:
-                    # No viable job.
+                if not job or job.id in jobs_dict:
+                    # No viable job or job already recorded.
                     return
 
                 jobs_dict[job.id] = job
@@ -2414,10 +2417,11 @@ def record_associated_jobs(obj):
                     )
                 job_hda = hda
                 while job_hda.copied_from_history_dataset_association:  # should this check library datasets as well?
+                    # record job (if one exists) even if dataset was copied
+                    # copy could have been created manually through UI/API or using database operation tool,
+                    # in which case we have a relevant job to export.
+                    record_associated_jobs(job_hda)
                     job_hda = job_hda.copied_from_history_dataset_association
-                if not job_hda.creating_job_associations:
-                    # No viable HDA found.
-                    continue
 
                 record_associated_jobs(job_hda)
 

lib/galaxy/tools/__init__.py

@@ -3786,14 +3786,13 @@ def produce_outputs(self, trans, out_data, output_collections, incoming, history
         final_sorted_identifiers = [
             element.element_identifier for element in elements1 if element.element_identifier in old_elements2_dict
         ]
-        # Raise Exception if it is empty
         if len(final_sorted_identifiers) == 0:
             # Create empty collections:
             output_collections.create_collection(
-                next(iter(self.outputs.values())), "output1", elements={}, propagate_hda_tags=False
+                self.outputs["output1"], "output1", elements={}, propagate_hda_tags=False
             )
             output_collections.create_collection(
-                next(iter(self.outputs.values())), "output2", elements={}, propagate_hda_tags=False
+                self.outputs["output2"], "output2", elements={}, propagate_hda_tags=False
             )
             return
 
@@ -3811,7 +3810,7 @@ def output_with_selected_identifiers(old_elements_dict, output_label):
             self._add_datasets_to_history(history, new_elements.values())
             # Create collections:
             output_collections.create_collection(
-                next(iter(self.outputs.values())), output_label, elements=new_elements, propagate_hda_tags=False
+                self.outputs[output_label], output_label, elements=new_elements, propagate_hda_tags=False
             )
 
         # Create outputs:

lib/galaxy/tools/harmonize_two_collections_list.xml

@@ -171,6 +171,37 @@
             <output_collection name="output1" type="list" count="0"/>
             <output_collection name="output2" type="list" count="0"/>
         </test>
+        <!-- test that collection types get propagated to outputs correctly -->
+        <test>
+            <param name="input1">
+                <collection type="list">
+                    <element name="element_1" value="simple_line.txt" />
+                </collection>
+            </param>
+            <param name="input2">
+                <collection type="list:paired">
+                    <element name="element_1">
+                        <collection type="paired">
+                            <element name="forward" value="1.fastqsanger" ftype="fastqsanger" />
+                            <element name="reverse" value="1.fastqsanger" ftype="fastqsanger" />
+                         </collection>
+                    </element>
+                </collection>
+            </param>
+            <output_collection name="output1" type="list" count="1">
+                <element name="element_1">
+                    <assert_contents>
+                        <has_text_matching expression="^This is a line of text.\n$" />
+                    </assert_contents>
+                </element>
+            </output_collection>
+            <output_collection name="output2" type="list:paired" count="1">
+                <element name="element_1">
+                    <element name="forward" file="1.fastqsanger" ftype="fastqsanger" />
+                    <element name="reverse" file="1.fastqsanger" ftype="fastqsanger" />
+                </element>
+            </output_collection>
+        </test>
     </tests>
     <help><![CDATA[
 

test/integration/test_workflow_tasks.py

@@ -17,6 +17,7 @@
 from galaxy_test.base import api_asserts
 from galaxy_test.base.api import UsesCeleryTasks
 from galaxy_test.base.populators import (
+    DatasetCollectionPopulator,
     DatasetPopulator,
     RunJobsSummary,
     WorkflowPopulator,
@@ -27,6 +28,7 @@
 
 class TestWorkflowTasksIntegration(PosixFileSourceSetup, IntegrationTestCase, UsesCeleryTasks, RunsWorkflowFixtures):
     dataset_populator: DatasetPopulator
+    dataset_collection_populator: DatasetCollectionPopulator
     framework_tool_and_types = True
 
     @classmethod
@@ -37,6 +39,7 @@ def handle_galaxy_config_kwds(cls, config):
     def setUp(self):
         super().setUp()
         self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
+        self.dataset_collection_populator = DatasetCollectionPopulator(self.galaxy_interactor)
         self.workflow_populator = WorkflowPopulator(self.galaxy_interactor)
         self._write_file_fixtures()
 
@@ -124,6 +127,47 @@ def test_export_import_invocation_with_step_parameter(self):
             invocation_details = self._export_and_import_workflow_invocation(summary, use_uris)
             self._rerun_imported_workflow(summary, invocation_details)
 
+    def test_export_import_invocation_with_copied_hdca_and_database_operation_tool(self):
+        with self.dataset_populator.test_history() as history_id:
+            self.dataset_collection_populator.create_list_in_history(history_id=history_id, wait=True).json()
+            new_history = self.dataset_populator.copy_history(history_id=history_id).json()
+            copied_collection = self.dataset_populator.get_history_collection_details(new_history["id"])
+            workflow_id = self.workflow_populator.upload_yaml_workflow(
+                """class: GalaxyWorkflow
+inputs:
+  input:
+    type: collection
+    collection_type: list
+steps:
+  extract_dataset:
+    tool_id: __EXTRACT_DATASET__
+    in:
+      input:
+        source: input
+"""
+            )
+            inputs = {"input": {"src": "hdca", "id": copied_collection["id"]}}
+            workflow_request = {"history": f"hist_id={new_history['id']}", "inputs_by": "name", "inputs": inputs}
+            invocation = self.workflow_populator.invoke_workflow_raw(
+                workflow_id, workflow_request, assert_ok=True
+            ).json()
+            invocation_id = invocation["id"]
+            self.workflow_populator.wait_for_invocation_and_jobs(history_id, workflow_id, invocation_id)
+            jobs = self.workflow_populator.get_invocation_jobs(invocation_id)
+            summary = RunJobsSummary(
+                history_id=history_id,
+                workflow_id=workflow_id,
+                invocation_id=invocation["id"],
+                inputs=inputs,
+                jobs=jobs,
+                invocation=invocation,
+                workflow_request=workflow_request,
+            )
+            imported_invocation_details = self._export_and_import_workflow_invocation(summary)
+            original_contents = self.dataset_populator.get_history_contents(new_history["id"])
+            contents = self.dataset_populator.get_history_contents(imported_invocation_details["history_id"])
+            assert len(contents) == len(original_contents) == 5
+
     def _export_and_import_workflow_invocation(
         self, summary: RunJobsSummary, use_uris: bool = True, model_store_format="tgz"
     ) -> Dict[str, Any]: