fix: deal with case sensitivity on file extension derivation

Author: georgeRobertson

src/dve/core_engine/backends/base/contract.py

@@ -339,7 +339,7 @@ def read_raw_entities(
             reader_metadata = contract_metadata.reader_metadata[entity_name]
             extension = "." + (
                 get_file_suffix(resource) or ""
-            )  # Already checked that extension supported.
+            ).lower()  # Already checked that extension supported.
 
             reader_config = reader_metadata[extension]
             reader_type = get_reader(reader_config.reader)

src/dve/pipeline/utils.py

@@ -47,7 +47,7 @@ def load_config(
 
 def load_reader(dataset: Dataset, model_name: str, file_extension: str):
     """Loads the readers for the diven feed, model name and file extension"""
-    reader_config = dataset[model_name].reader_config[f".{file_extension}"]
+    reader_config = dataset[model_name].reader_config[f".{file_extension.lower()}"]
     reader = _READER_REGISTRY[reader_config.reader](**reader_config.kwargs_)
     return reader
 

tests/features/books.feature

@@ -5,7 +5,7 @@ Feature: Pipeline tests using the books dataset
     introduces more complex transformations that require aggregation.
 
     Scenario: Validate complex nested XML data (spark)
-        Given I submit the books file nested_books.xml for processing
+        Given I submit the books file nested_books.XML for processing
         And A spark pipeline is configured with schema file 'nested_books.dischema.json'
         And I add initial audit entries for the submission
         Then the latest audit record for the submission is marked with processing status file_transformation
@@ -32,7 +32,7 @@ Feature: Pipeline tests using the books dataset
             | number_warnings          | 0     |
 
     Scenario: Validate complex nested XML data (duckdb)
-        Given I submit the books file nested_books.xml for processing
+        Given I submit the books file nested_books.XML for processing
         And A duckdb pipeline is configured with schema file 'nested_books_ddb.dischema.json'
         And I add initial audit entries for the submission
         Then the latest audit record for the submission is marked with processing status file_transformation

tests/test_core_engine/test_engine.py

@@ -99,8 +99,8 @@ def test_dummy_books_run(self, spark, temp_dir: str):
             with test_instance:
                 _, errors_uri = test_instance.run_pipeline(
                     entity_locations={
-                        "header": get_test_file_path("books/nested_books.xml").as_posix(),
-                        "nested_books": get_test_file_path("books/nested_books.xml").as_posix(),
+                        "header": get_test_file_path("books/nested_books.XML").as_posix(),
+                        "nested_books": get_test_file_path("books/nested_books.XML").as_posix(),
                     }
                 )