fix: add user feedback message for unsupported file type

src/dve/core_engine/models.py

@@ -82,7 +82,9 @@ def _ensure_just_file_stem(
     @property
     def file_name_with_ext(self):
         """Return file name with extension."""
-        return f"{self.file_name}.{self.file_extension}"
+        if self.file_extension:
+            return f"{self.file_name}.{self.file_extension}"
+        return self.file_name
 
     @classmethod
     def from_metadata_file(cls, submission_id: str, metadata_uri: Location):

src/dve/pipeline/pipeline.py

@@ -215,10 +215,10 @@ def write_file_to_parquet(
 
         for model_name, model in models.items():
             self._logger.info(f"Transforming {model_name} to stringified parquet")
-            reader: BaseFileReader = load_reader(
-                dataset, model_name, ext, self.backend_reader_kwargs
-            )
             try:
+                reader: BaseFileReader = load_reader(
+                    dataset, model_name, ext, self.backend_reader_kwargs
+                )
                 if not entity_type:
                     reader.write_parquet(
                         reader.read_to_py_iterator(
@@ -241,6 +241,7 @@ def write_file_to_parquet(
                         f"{out}{model_name}",
                     )
             except MessageBearingError as exc:
+                self._logger.error(f"Unable to process {model_name}", exc_info=exc)
                 errors.extend(exc.messages)
 
         return list(dict.fromkeys(errors))  # remove any duplicate errors

src/dve/pipeline/utils.py

@@ -11,9 +11,11 @@
 import dve.core_engine.backends.implementations.duckdb  # pylint: disable=unused-import
 import dve.core_engine.backends.implementations.spark  # pylint: disable=unused-import
 import dve.parser.file_handling as fh
+from dve.core_engine.backends.exceptions import MessageBearingError
 from dve.core_engine.backends.readers import _READER_REGISTRY
 from dve.core_engine.configuration.v1 import SchemaName, V1EngineConfig, _ModelConfig
 from dve.core_engine.loggers import get_logger
+from dve.core_engine.message import FeedbackMessage
 from dve.core_engine.type_hints import URI, SubmissionResult
 from dve.metadata_parser.model_generator import JSONtoPyd
 
@@ -52,7 +54,31 @@ def load_reader(
     backend_reader_kwargs: Optional[dict[str, Any]] = None,
 ):
     """Loads the readers for the diven feed, model name and file extension"""
-    reader_config = dataset[model_name].reader_config[f".{file_extension.lower()}"]
+    try:
+        reader_config = dataset[model_name].reader_config[f".{file_extension.lower()}"]
+    except KeyError as exc:
+        if file_extension:
+            err_msg = (
+                f"The supplied file extension `{file_extension}`"
+                +f" is not a supported file format for {model_name}."
+            )
+        else:
+            err_msg = "No supplied file extension. Unable to parse file without a file extension."
+
+        raise MessageBearingError(
+            f"The file extension provided ({file_extension}) is not supported for this collection.",
+            messages=[
+                FeedbackMessage(
+                    entity=model_name,
+                    record=None,
+                    failure_type="submission",
+                    error_location="Whole File",
+                    error_code="InvalidFileExtension",
+                    error_message=err_msg,
+                )
+            ],
+        ) from exc
+
     reader = _READER_REGISTRY[reader_config.reader](
         **reader_config.kwargs_, **backend_reader_kwargs if backend_reader_kwargs else {}
     )

tests/features/planets.feature

@@ -43,7 +43,9 @@ Feature: Pipeline tests using the planets dataset
         And I add initial audit entries for the submission
         Then the latest audit record for the submission is marked with processing status file_transformation
         When I run the file transformation phase
-        Then the latest audit record for the submission is marked with processing status failed
+        Then the latest audit record for the submission is marked with processing status error_report
+        When I run the error report phase
+        Then An error report is produced
 
     Scenario: Handle a file with duplicated extension provided (spark)
         Given I submit the planets file planets.csv.csv for processing

tests/test_pipeline/test_pipeline_utils.py

@@ -0,0 +1,36 @@
+from dve.core_engine.backends.exceptions import MessageBearingError
+from dve.core_engine.configuration.v1 import _ModelConfig, _ReaderConfig
+from dve.pipeline.utils import load_reader
+
+import pytest
+
+
+class TestLoadReader:
+    test_model_config = _ModelConfig(
+        fields={"test": "str"},
+        reporting_fields=["test"],
+        key_field="test",
+        reader_config={
+            ".csv": _ReaderConfig(reader="TestCsvReader"),
+        }
+    )
+
+    def test_invalid_load_reader_with_file_ext(self):
+        with pytest.raises(MessageBearingError) as exc_info:
+            load_reader(
+                {"test": self.test_model_config},
+                "test_model",
+                "jpeg"
+            )
+
+        assert exc_info.value.messages[0].error_message == "The supplied file extension `jpeg` is not a supported file format for test_model."
+
+    def test_invalid_load_reader_missing_file_ext(self):
+        with pytest.raises(MessageBearingError) as exc_info:
+            load_reader(
+                {"test": self.test_model_config},
+                "test_model",
+                ""
+            )
+
+        assert exc_info.value.messages[0].error_message == "No supplied file extension. Unable to parse file without a file extension."