Merge pull request #18494 from mvdbeek/fix_invalid_id_error_message

[24.1] Increase API robustness to invalid requests, improve compressed data serving

lib/galaxy/datatypes/data.py

@@ -484,16 +484,17 @@ class Data(metaclass=DataMeta):
 
     def _serve_binary_file_contents_as_text(self, trans, data, headers, file_size, max_peek_size):
         headers["content-type"] = "text/html"
-        return (
-            trans.fill_template_mako(
-                "/dataset/binary_file.mako",
-                data=data,
-                file_contents=open(data.get_file_name(), "rb").read(max_peek_size),
-                file_size=util.nice_size(file_size),
-                truncated=file_size > max_peek_size,
-            ),
-            headers,
-        )
+        with open(data.get_file_name(), "rb") as fh:
+            return (
+                trans.fill_template_mako(
+                    "/dataset/binary_file.mako",
+                    data=data,
+                    file_contents=fh.read(max_peek_size),
+                    file_size=util.nice_size(file_size),
+                    truncated=file_size > max_peek_size,
+                ),
+                headers,
+            )
 
     def _serve_file_contents(self, trans, data, headers, preview, file_size, max_peek_size):
         from galaxy.datatypes import images
@@ -502,16 +503,17 @@ class Data(metaclass=DataMeta):
         if not preview or isinstance(data.datatype, images.Image) or file_size < max_peek_size:
             return self._yield_user_file_content(trans, data, data.get_file_name(), headers), headers
 
-        # preview large text file
-        headers["content-type"] = "text/html"
-        return (
-            trans.fill_template_mako(
-                "/dataset/large_file.mako",
-                truncated_data=open(data.get_file_name(), "rb").read(max_peek_size),
-                data=data,
-            ),
-            headers,
-        )
+        with compression_utils.get_fileobj(data.get_file_name(), "rb") as fh:
+            # preview large text file
+            headers["content-type"] = "text/html"
+            return (
+                trans.fill_template_mako(
+                    "/dataset/large_file.mako",
+                    truncated_data=fh.read(max_peek_size),
+                    data=data,
+                ),
+                headers,
+            )
 
     def display_data(
         self,

lib/galaxy/datatypes/tabular.py

@@ -193,14 +193,15 @@ class TabularData(Text):
                 return open(dataset.get_file_name(), mode="rb"), headers
             else:
                 headers["content-type"] = "text/html"
-                return (
-                    trans.fill_template_mako(
-                        "/dataset/large_file.mako",
-                        truncated_data=open(dataset.get_file_name()).read(max_peek_size),
-                        data=dataset,
-                    ),
-                    headers,
-                )
+                with compression_utils.get_fileobj(dataset.get_file_name(), "rb") as fh:
+                    return (
+                        trans.fill_template_mako(
+                            "/dataset/large_file.mako",
+                            truncated_data=fh.read(max_peek_size),
+                            data=dataset,
+                        ),
+                        headers,
+                    )
         else:
             column_names = "null"
             if dataset.metadata.column_names:

lib/galaxy/managers/hdas.py

@@ -311,7 +311,10 @@ class HDAManager(
 
         truncated = preview and os.stat(file_path).st_size > MAX_PEEK_SIZE
         with get_fileobj(file_path) as fh:
-            hda_data = fh.read(MAX_PEEK_SIZE)
+            try:
+                hda_data = fh.read(MAX_PEEK_SIZE)
+            except UnicodeDecodeError:
+                raise exceptions.RequestParameterInvalidException("Cannot generate text preview for dataset.")
         return truncated, hda_data
 
     # .... annotatable

lib/galaxy/tools/parameters/basic.py

@@ -2053,7 +2053,8 @@ def src_id_to_item(
         item = sa_session.get(src_to_class[value["src"]], decoded_id)
     except KeyError:
         raise ValueError(f"Unknown input source {value['src']} passed to job submission API.")
-    assert item
+    if not item:
+        raise ValueError("Invalid input id passed to job submission API.")
     item.extra_params = {k: v for k, v in value.items() if k not in ("src", "id")}
     return item
 

lib/galaxy/web/framework/middleware/statsd.py

@@ -28,8 +28,13 @@ class StatsdMiddleware:
         start_time = time.time()
         req = self.application(environ, start_response)
         dt = int((time.time() - start_time) * 1000)
-        page = environ.get("controller_action_key", None) or environ.get("PATH_INFO", "NOPATH").strip("/").replace(
-            "/", "."
+        page = (
+            environ.get("controller_action_key", None)
+            or environ.get("PATH_INFO", "NOPATH")
+            .strip("/")
+            .replace("/", ".")
+            .encode("ascii", errors="replace")
+            .decode()
         )
         self.galaxy_stasd_client.timing(page, dt)
         try:

lib/galaxy/workflow/modules.py

@@ -746,7 +746,8 @@ class SubWorkflowModule(WorkflowModule):
                         # This can happen when importing workflows with missing tools.
                         # We can't raise an exception here, as that would prevent loading
                         # the workflow.
-                        log.error(
+                        # This is also listed when opening such a workflow in the workflow editor.
+                        log.warning(
                             f"Workflow output '{workflow_output['output_name']}' defined, but not listed among data outputs"
                         )
                         continue