Merge pull request #19301 from mvdbeek/implicit_dataset_as_input

[24.1] Record implicitly converted dataset as input dataset

lib/galaxy/datatypes/data.py

@@ -847,7 +847,11 @@ def convert_dataset(
         job, converted_datasets, *_ = converter.execute(
             trans, incoming=params, set_output_hid=visible, history=history, flush_job=False
         )
+        # We should only have a single converted output, but let's be defensive here
+        n_converted_datasets = len(converted_datasets)
         for converted_dataset in converted_datasets.values():
+            if converted_dataset.extension == "auto" and n_converted_datasets == 1:
+                converted_dataset.extension = target_type
             original_dataset.attach_implicitly_converted_dataset(trans.sa_session, converted_dataset, target_type)
         trans.app.job_manager.enqueue(job, tool=converter)
         if len(params) > 0:

lib/galaxy_test/api/test_tools.py

@@ -2511,6 +2511,24 @@ def test_multi_param_column_nested_list_fails_on_invalid_column(self):
                 exception_raised = e
             assert exception_raised, "Expected invalid column selection to fail job"
 
+    @skip_without_tool("implicit_conversion_format_input")
+    def test_implicit_conversion_input_dataset_tracking(self):
+        with self.dataset_populator.test_history() as history_id:
+            compressed_path = self.test_data_resolver.get_filename("1.fastqsanger.gz")
+            with open(compressed_path, "rb") as fh:
+                dataset = self.dataset_populator.new_dataset(
+                    history_id, content=fh, file_type="fastqsanger.gz", wait=True
+                )
+            outputs = self._run(
+                "Grep1", history_id=history_id, inputs={"data": {"src": "hda", "id": dataset["id"]}}, assert_ok=True
+            )
+            job_details = self.dataset_populator.get_job_details(outputs["jobs"][0]["id"], full=True).json()
+            assert job_details["inputs"]["input"]["id"] != dataset["id"]
+            converted_input = self.dataset_populator.get_history_dataset_details(
+                history_id=history_id, content_id=job_details["inputs"]["input"]["id"]
+            )
+            assert converted_input["extension"] == "fastqsanger"
+
     @skip_without_tool("column_multi_param")
     def test_implicit_conversion_and_reduce(self):
         with self.dataset_populator.test_history() as history_id: