Refactored create_schema() into database_types

Author: erezsh

data_diff/__main__.py

@@ -11,7 +11,8 @@
 
 from .utils import remove_password_from_url, safezip, match_like
 from .diff_tables import TableDiffer, DEFAULT_BISECTION_THRESHOLD, DEFAULT_BISECTION_FACTOR
-from .table_segment import create_schema, TableSegment
+from .table_segment import TableSegment
+from .databases.database_types import create_schema
 from .databases.connect import connect
 from .parse_time import parse_time_before_now, UNITS_STR, ParseError
 from .config import apply_config_from_file

data_diff/databases/database_types.py

@@ -1,17 +1,20 @@
+import logging
 import decimal
 from abc import ABC, abstractmethod
 from typing import Sequence, Optional, Tuple, Union, Dict, List
 from datetime import datetime
 
 from runtype import dataclass
 
-from data_diff.utils import ArithAlphanumeric, ArithUUID, CaseAwareMapping
+from data_diff.utils import ArithAlphanumeric, ArithUUID, CaseAwareMapping, CaseInsensitiveDict, CaseSensitiveDict
 
 
 DbPath = Tuple[str, ...]
 DbKey = Union[int, str, bytes, ArithUUID, ArithAlphanumeric]
 DbTime = datetime
 
+logger = logging.getLogger("databases")
+
 
 class ColType:
     supported = True
@@ -129,6 +132,8 @@ class UnknownColType(ColType):
 
 
 class AbstractDatabase(ABC):
+    name: str
+
     @abstractmethod
     def quote(self, s: str):
         "Quote SQL name (implementation specific)"
@@ -270,3 +275,15 @@ def _normalize_table_path(self, path: DbPath) -> DbPath:
 
 
 Schema = CaseAwareMapping
+
+
+def create_schema(db: AbstractDatabase, table_path: DbPath, schema: dict, case_sensitive: bool) -> CaseAwareMapping:
+    logger.debug(f"[{db.name}] Schema = {schema}")
+
+    if case_sensitive:
+        return CaseSensitiveDict(schema)
+
+    if len({k.lower() for k in schema}) < len(schema):
+        logger.warning(f'Ambiguous schema for {db}:{".".join(table_path)} | Columns = {", ".join(list(schema))}')
+        logger.warning("We recommend to disable case-insensitivity (set --case-sensitive).")
+    return CaseInsensitiveDict(schema)

data_diff/table_segment.py

@@ -4,29 +4,17 @@
 
 from runtype import dataclass
 
-from .utils import ArithString, split_space, CaseAwareMapping, CaseInsensitiveDict, CaseSensitiveDict
+from .utils import ArithString, split_space
 
 from .databases.base import Database
-from .databases.database_types import DbPath, DbKey, DbTime, Native_UUID, Schema
+from .databases.database_types import DbPath, DbKey, DbTime, Native_UUID, Schema, create_schema
 from .sql import Select, Checksum, Compare, Count, TableName, Time, Value
 
 logger = logging.getLogger("table_segment")
 
 RECOMMENDED_CHECKSUM_DURATION = 10
 
 
-def create_schema(db: Database, table_path: DbPath, schema: dict, case_sensitive: bool) -> CaseAwareMapping:
-    logger.debug(f"[{db.name}] Schema = {schema}")
-
-    if case_sensitive:
-        return CaseSensitiveDict(schema)
-
-    if len({k.lower() for k in schema}) < len(schema):
-        logger.warning(f'Ambiguous schema for {db}:{".".join(table_path)} | Columns = {", ".join(list(schema))}')
-        logger.warning("We recommend to disable case-insensitivity (remove --any-case).")
-    return CaseInsensitiveDict(schema)
-
-
 @dataclass
 class TableSegment:
     """Signifies a segment of rows (and selected columns) within a table