refactor: Address review comments

Author: stevenhsd

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

@@ -117,7 +117,7 @@ def read_to_entity_type(
         """
         if entity_name == Iterator[dict[str, Any]]:
             return self.read_to_py_iterator(resource, entity_name, schema)  # type: ignore
-        
+
         self.raise_if_not_sensible_file(resource, entity_name)
 
         try:
@@ -141,12 +141,12 @@ def write_parquet(
 
         """
         raise NotImplementedError(f"write_parquet not implemented in {self.__class__}")
-    
+
     @staticmethod
     def _check_likely_text_file(resource: URI) -> bool:
         """Quick sense check of file to see if it looks like text
-           - not 100% full proof, but hopefully enough to weed out most
-           non-text files"""
+        - not 100% full proof, but hopefully enough to weed out most
+        non-text files"""
         with open_stream(resource, "rb") as fle:
             start_chunk = fle.read(4096)
         # check for BOM character - utf-16 can contain NULL bytes
@@ -156,8 +156,10 @@ def _check_likely_text_file(resource: URI) -> bool:
         if b"\x00" in start_chunk:
             return False
         return True
-    
-    def raise_if_not_sensible_file(self, resource: URI, entity_name:str):
+
+    def raise_if_not_sensible_file(self, resource: URI, entity_name: str):
+        """Sense check that the file is a text file. Raise error if doesn't
+        appear to be the case."""
         if not self._check_likely_text_file(resource):
             raise MessageBearingError(
                 "The submitted file doesn't appear to be text",
@@ -168,7 +170,7 @@ def raise_if_not_sensible_file(self, resource: URI, entity_name:str):
                         failure_type="submission",
                         error_location="Whole File",
                         error_code="MalformedFile",
-                        error_message=f"The submitted resource doesn't seem to be a valid text file",
+                        error_message="The resource doesn't seem to be a valid text file",
                     )
                 ],
             )

src/dve/core_engine/backends/implementations/duckdb/readers/csv.py

@@ -16,7 +16,7 @@
     get_duckdb_type_from_annotation,
 )
 from dve.core_engine.backends.implementations.duckdb.types import SQLType
-from dve.core_engine.backends.implementations.duckdb.utilities import check_csv_header_expected
+from dve.core_engine.backends.readers.utilities import check_csv_header_expected
 from dve.core_engine.backends.utilities import get_polars_type_from_annotation
 from dve.core_engine.message import FeedbackMessage
 from dve.core_engine.type_hints import URI, EntityName
@@ -25,7 +25,14 @@
 
 @duckdb_write_parquet
 class DuckDBCSVReader(BaseFileReader):
-    """A reader for CSV files"""
+    """A reader for CSV files including the ability to compare the passed model
+    to the file header, if it exists.
+
+    field_check: flag to compare submitted file header to the accompanying pydantic model
+    field_check_error_code: The error code to provide if the file header doesn't contain
+                            the expected fields
+    field_check_error_message: The error message to provide if the file header doesn't contain
+                               the expected fields"""
 
     # TODO - the read_to_relation should include the schema and determine whether to
     # TODO - stringify or not
@@ -54,14 +61,11 @@ def __init__(
     def perform_field_check(
         self, resource: URI, entity_name: str, expected_schema: type[BaseModel]
     ):
+        """Check that the header of the CSV aligns with the provided model"""
         if not self.header:
             raise ValueError("Cannot perform field check without a CSV header")
 
-        if missing := check_csv_header_expected(
-            resource,
-            expected_schema,
-            self.delim
-        ):
+        if missing := check_csv_header_expected(resource, expected_schema, self.delim):
             raise MessageBearingError(
                 "The CSV header doesn't match what is expected",
                 messages=[
@@ -71,7 +75,7 @@ def perform_field_check(
                         failure_type="submission",
                         error_location="Whole File",
                         error_code=self.field_check_error_code,
-                        error_message=f"{self.field_check_error_message} - missing fields: {missing}",
+                        error_message=f"{self.field_check_error_message} - missing fields: {missing}",  # pylint: disable=line-too-long
                     )
                 ],
             )
@@ -171,9 +175,14 @@ class DuckDBCSVRepeatingHeaderReader(PolarsToDuckDBCSVReader):
     """
 
     def __init__(
-        self, non_unique_header_error_code: Optional[str] = "NonUniqueHeader", *args, **kwargs
+        self,
+        *args,
+        non_unique_header_error_code: Optional[str] = "NonUniqueHeader",
+        non_unique_header_error_message: Optional[str] = None,
+        **kwargs,
     ):
         self._non_unique_header_code = non_unique_header_error_code
+        self._non_unique_header_message = non_unique_header_error_message
         super().__init__(*args, **kwargs)
 
     @read_function(DuckDBPyRelation)
@@ -200,6 +209,8 @@ def read_to_relation(  # pylint: disable=unused-argument
                         failure_type="submission",
                         error_message=(
                             f"Found {no_records} distinct combination of header values."
+                            if not self._non_unique_header_message
+                            else self._non_unique_header_message
                         ),
                         error_location=entity_name,
                         category="Bad file",

src/dve/core_engine/backends/implementations/duckdb/utilities.py

@@ -1,15 +1,8 @@
 """Utility objects for use with duckdb backend"""
 
 import itertools
-from typing import Optional
-
-from pydantic import BaseModel
 
 from dve.core_engine.backends.base.utilities import _split_multiexpr_string
-from dve.core_engine.backends.exceptions import MessageBearingError
-from dve.core_engine.message import FeedbackMessage
-from dve.core_engine.type_hints import URI
-from dve.parser.file_handling import open_stream
 
 
 def parse_multiple_expressions(expressions) -> list[str]:
@@ -46,15 +39,3 @@ def multiexpr_string_to_columns(expressions: str) -> list[str]:
     """
     expression_list = _split_multiexpr_string(expressions)
     return expr_array_to_columns(expression_list)
-
-def check_csv_header_expected(
-    resource: URI,
-    expected_schema: type[BaseModel],
-    delimiter: Optional[str] = ",",
-    quote_char: str = '"') -> set[str]:
-    """Check the header of a CSV matches the expected fields"""
-    with open_stream(resource) as fle:
-        header_fields = fle.readline().rstrip().replace(quote_char,"").split(delimiter)
-    expected_fields = expected_schema.__fields__.keys()
-    return set(expected_fields).difference(header_fields)
-    

src/dve/core_engine/backends/readers/utilities.py

@@ -0,0 +1,21 @@
+"""General utilities for file readers"""
+
+from typing import Optional
+
+from pydantic import BaseModel
+
+from dve.core_engine.type_hints import URI
+from dve.parser.file_handling.service import open_stream
+
+
+def check_csv_header_expected(
+    resource: URI,
+    expected_schema: type[BaseModel],
+    delimiter: Optional[str] = ",",
+    quote_char: str = '"',
+) -> set[str]:
+    """Check the header of a CSV matches the expected fields"""
+    with open_stream(resource) as fle:
+        header_fields = fle.readline().rstrip().replace(quote_char, "").split(delimiter)
+    expected_fields = expected_schema.__fields__.keys()
+    return set(expected_fields).difference(header_fields)

src/dve/pipeline/foundry_ddb_pipeline.py

@@ -109,6 +109,7 @@ def error_report(
             self._logger.exception(exc)
             sub_stats = None
             report_uri = None
+            submission_status = submission_status if submission_status else SubmissionStatus()
             submission_status.processing_failed = True
             dump_processing_errors(
                 fh.joinuri(self.processed_files_path, submission_info.submission_id),

tests/test_core_engine/test_backends/test_implementations/test_duckdb/test_ddb_utils.py

@@ -1,14 +1,8 @@
-import tempfile
-import datetime as dt
-from pathlib import Path
-from uuid import uuid4
-from pydantic import BaseModel, create_model
 import pytest
 
 from dve.core_engine.backends.implementations.duckdb.utilities import (
     expr_mapping_to_columns,
     expr_array_to_columns,
-    check_csv_header_expected,
 )
 
 
@@ -60,43 +54,3 @@ def test_expr_array_to_columns(expressions: dict[str, str], expected: list[str])
     observed = expr_array_to_columns(expressions)
     assert observed == expected
 
-
-@pytest.mark.parametrize(
-    ["header_row", "delim", "schema", "expected"],
-    [
-        (
-            "field1,field2,field3",
-            ",",
-            {"field1": (str, ...), "field2": (int, ...), "field3": (float, 1.2)},
-            set(),
-        ),
-        (
-            "field2,field3,field1",
-            ",",
-            {"field1": (str, ...), "field2": (int, ...), "field3": (float, 1.2)},
-            set(),
-        ),
-        (
-            "str_field|int_field|date_field|",
-            ",",
-            {"str_field": (str, ...), "int_field": (int, ...), "date_field": (dt.date, dt.date.today())},
-            {"str_field","int_field","date_field"},
-        ),
-        (
-            '"str_field"|"int_field"|"date_field"',
-            "|",
-            {"str_field": (str, ...), "int_field": (int, ...), "date_field": (dt.date, dt.date.today())},
-            set(),
-        ),
-        
-    ],
-)
-def test_check_csv_header_expected(
-    header_row: str, delim: str, schema: type[BaseModel], expected: set[str]
-):
-    mdl = create_model("TestModel", **schema)
-    with tempfile.TemporaryDirectory() as tmpdir:
-        fle = Path(tmpdir).joinpath(f"test_file_{uuid4().hex}.csv")
-        fle.open("w+").write(header_row)
-        res = check_csv_header_expected(fle.as_posix(), mdl, delim)
-    assert res == expected

tests/test_core_engine/test_backends/test_readers/test_ddb_json.py

@@ -57,7 +57,7 @@ def test_ddb_json_reader_all_str(temp_json_file):
     expected_fields = [fld for fld in mdl.__fields__]
     reader = DuckDBJSONReader()
     rel: DuckDBPyRelation = reader.read_to_entity_type(
-        DuckDBPyRelation, uri, "test", stringify_model(mdl)
+        DuckDBPyRelation, uri.as_posix(), "test", stringify_model(mdl)
     )
     assert rel.columns == expected_fields
     assert dict(zip(rel.columns, rel.dtypes)) == {fld: "VARCHAR" for fld in expected_fields}
@@ -68,7 +68,7 @@ def test_ddb_json_reader_cast(temp_json_file):
     uri, data, mdl = temp_json_file
     expected_fields = [fld for fld in mdl.__fields__]
     reader = DuckDBJSONReader()
-    rel: DuckDBPyRelation = reader.read_to_entity_type(DuckDBPyRelation, uri, "test", mdl)
+    rel: DuckDBPyRelation = reader.read_to_entity_type(DuckDBPyRelation, uri.as_posix(), "test", mdl)
 
     assert rel.columns == expected_fields
     assert dict(zip(rel.columns, rel.dtypes)) == {
@@ -82,7 +82,7 @@ def test_ddb_csv_write_parquet(temp_json_file):
     uri, _, mdl = temp_json_file
     reader = DuckDBJSONReader()
     rel: DuckDBPyRelation = reader.read_to_entity_type(
-        DuckDBPyRelation, uri, "test", stringify_model(mdl)
+        DuckDBPyRelation, uri.as_posix(), "test", stringify_model(mdl)
     )
     target_loc: Path = uri.parent.joinpath("test_parquet.parquet").as_posix()
     reader.write_parquet(rel, target_loc)

tests/test_core_engine/test_backends/test_readers/test_utilities.py

@@ -0,0 +1,55 @@
+import datetime as dt
+from pathlib import Path
+import tempfile
+from uuid import uuid4
+
+import pytest
+from pydantic import BaseModel, create_model
+
+from dve.core_engine.backends.readers.utilities import check_csv_header_expected
+
+@pytest.mark.parametrize(
+    ["header_row", "delim", "schema", "expected"],
+    [
+        (
+            "field1,field2,field3",
+            ",",
+            {"field1": (str, ...), "field2": (int, ...), "field3": (float, 1.2)},
+            set(),
+        ),
+        (
+            "field2,field3,field1",
+            ",",
+            {"field1": (str, ...), "field2": (int, ...), "field3": (float, 1.2)},
+            set(),
+        ),
+        (
+            "str_field|int_field|date_field|",
+            ",",
+            {"str_field": (str, ...), "int_field": (int, ...), "date_field": (dt.date, dt.date.today())},
+            {"str_field","int_field","date_field"},
+        ),
+        (
+            '"str_field"|"int_field"|"date_field"',
+            "|",
+            {"str_field": (str, ...), "int_field": (int, ...), "date_field": (dt.date, dt.date.today())},
+            set(),
+        ),
+        (
+            'str_field,int_field,date_field\n',
+            ",",
+            {"str_field": (str, ...), "int_field": (int, ...), "date_field": (dt.date, dt.date.today())},
+            set(),
+        ),
+        
+    ],
+)
+def test_check_csv_header_expected(
+    header_row: str, delim: str, schema: type[BaseModel], expected: set[str]
+):
+    mdl = create_model("TestModel", **schema)
+    with tempfile.TemporaryDirectory() as tmpdir:
+        fle = Path(tmpdir).joinpath(f"test_file_{uuid4().hex}.csv")
+        fle.open("w+").write(header_row)
+        res = check_csv_header_expected(fle.as_posix(), mdl, delim)
+    assert res == expected