Clarify optional (nullable) fields: explicit is better than implicit

We are going to do strict type checking. The default values of fields that clearly contradict the declared types is an error for MyPy and all other type checkers and IDEs.

Remove the implicit behaviour and make nullable fields explicitly declared as such.

Author: Sergey Vasilyev

data_diff/databases/base.py

@@ -815,7 +815,7 @@ def set_timezone_to_utc(self) -> str:
 @dataclass
 class QueryResult:
     rows: list
-    columns: list = None
+    columns: Optional[list] = None
 
     def __iter__(self):
         return iter(self.rows)

data_diff/info_tree.py

@@ -10,13 +10,13 @@
 class SegmentInfo:
     tables: List[TableSegment]
 
-    diff: List[Union[Tuple[Any, ...], List[Any]]] = None
-    diff_schema: Tuple[Tuple[str, type], ...] = None
-    is_diff: bool = None
-    diff_count: int = None
+    diff: Optional[List[Union[Tuple[Any, ...], List[Any]]]] = None
+    diff_schema: Optional[Tuple[Tuple[str, type], ...]] = None
+    is_diff: Optional[bool] = None
+    diff_count: Optional[int] = None
 
     rowcounts: Dict[int, int] = field(default_factory=dict)
-    max_rows: int = None
+    max_rows: Optional[int] = None
 
     def set_diff(self, diff: List[Union[Tuple[Any, ...], List[Any]]], schema: Optional[Tuple[Tuple[str, type]]] = None):
         self.diff_schema = schema

data_diff/joindiff_tables.py

@@ -138,7 +138,7 @@ class JoinDiffer(TableDiffer):
 
     validate_unique_key: bool = True
     sample_exclusive_rows: bool = False
-    materialize_to_table: DbPath = None
+    materialize_to_table: Optional[DbPath] = None
     materialize_all_rows: bool = False
     table_write_limit: int = TABLE_WRITE_LIMIT
     skip_null_keys: bool = False

data_diff/queries/ast_classes.py

@@ -55,7 +55,7 @@ def cast_to(self, to):
 @dataclass
 class Code(ExprNode, Root):
     code: str
-    args: Dict[str, Expr] = None
+    args: Optional[Dict[str, Expr]] = None
 
 
 def _expr_type(e: Expr) -> type:
@@ -216,7 +216,7 @@ def intersect(self, other: "ITable"):
 @dataclass
 class Concat(ExprNode):
     exprs: list
-    sep: str = None
+    sep: Optional[str] = None
 
 
 @dataclass
@@ -293,7 +293,7 @@ class WhenThen(ExprNode):
 @dataclass
 class CaseWhen(ExprNode):
     cases: Sequence[WhenThen]
-    else_expr: Expr = None
+    else_expr: Optional[Expr] = None
 
     @property
     def type(self):
@@ -491,9 +491,9 @@ def schema(self) -> Schema:
 @dataclass
 class Join(ExprNode, ITable, Root):
     source_tables: Sequence[ITable]
-    op: str = None
-    on_exprs: Sequence[Expr] = None
-    columns: Sequence[Expr] = None
+    op: Optional[str] = None
+    on_exprs: Optional[Sequence[Expr]] = None
+    columns: Optional[Sequence[Expr]] = None
 
     @property
     def schema(self) -> Schema:
@@ -534,9 +534,9 @@ def select(self, *exprs, **named_exprs) -> Union[Self, ITable]:
 @dataclass
 class GroupBy(ExprNode, ITable, Root):
     table: ITable
-    keys: Sequence[Expr] = None  # IKey?
-    values: Sequence[Expr] = None
-    having_exprs: Sequence[Expr] = None
+    keys: Optional[Sequence[Expr]] = None  # IKey?
+    values: Optional[Sequence[Expr]] = None
+    having_exprs: Optional[Sequence[Expr]] = None
 
     def __post_init__(self):
         assert self.keys or self.values
@@ -580,15 +580,15 @@ def schema(self) -> Schema:
 
 @dataclass
 class Select(ExprNode, ITable, Root):
-    table: Expr = None
-    columns: Sequence[Expr] = None
-    where_exprs: Sequence[Expr] = None
-    order_by_exprs: Sequence[Expr] = None
-    group_by_exprs: Sequence[Expr] = None
-    having_exprs: Sequence[Expr] = None
-    limit_expr: int = None
+    table: Optional[Expr] = None
+    columns: Optional[Sequence[Expr]] = None
+    where_exprs: Optional[Sequence[Expr]] = None
+    order_by_exprs: Optional[Sequence[Expr]] = None
+    group_by_exprs: Optional[Sequence[Expr]] = None
+    having_exprs: Optional[Sequence[Expr]] = None
+    limit_expr: Optional[int] = None
     distinct: bool = False
-    optimizer_hints: Sequence[Expr] = None
+    optimizer_hints: Optional[Sequence[Expr]] = None
 
     @property
     def schema(self) -> Schema:
@@ -636,8 +636,8 @@ def make(cls, table: ITable, distinct: bool = SKIP, optimizer_hints: str = SKIP,
 @dataclass
 class Cte(ExprNode, ITable):
     table: Expr
-    name: str = None
-    params: Sequence[str] = None
+    name: Optional[str] = None
+    params: Optional[Sequence[str]] = None
 
     @property
     def source_table(self) -> "ITable":
@@ -667,7 +667,7 @@ def resolve_names(source_table, exprs):
 @dataclass(frozen=False, eq=False, order=False)
 class _ResolveColumn(ExprNode, LazyOps):
     resolve_name: str
-    resolved: Expr = None
+    resolved: Optional[Expr] = None
 
     def resolve(self, expr: Expr):
         if self.resolved is not None:
@@ -750,9 +750,9 @@ def type(self) -> Optional[type]:
 class TimeTravel(ITable):
     table: TablePath
     before: bool = False
-    timestamp: datetime = None
-    offset: int = None
-    statement: str = None
+    timestamp: Optional[datetime] = None
+    offset: Optional[int] = None
+    statement: Optional[str] = None
 
 
 # DDL
@@ -767,9 +767,9 @@ def type(self) -> Optional[type]:
 @dataclass
 class CreateTable(Statement):
     path: TablePath
-    source_table: Expr = None
+    source_table: Optional[Expr] = None
     if_not_exists: bool = False
-    primary_keys: List[str] = None
+    primary_keys: Optional[List[str]] = None
 
 
 @dataclass
@@ -787,8 +787,8 @@ class TruncateTable(Statement):
 class InsertToTable(Statement):
     path: TablePath
     expr: Expr
-    columns: List[str] = None
-    returning_exprs: List[str] = None
+    columns: Optional[List[str]] = None
+    returning_exprs: Optional[List[str]] = None
 
     def returning(self, *exprs) -> Self:
         """Add a 'RETURNING' clause to the insert expression.

data_diff/queries/extras.py

@@ -10,7 +10,7 @@
 @dataclass
 class NormalizeAsString(ExprNode):
     expr: ExprNode
-    expr_type: ColType = None
+    expr_type: Optional[ColType] = None
 
     @property
     def type(self) -> Optional[type]:
@@ -20,7 +20,7 @@ def type(self) -> Optional[type]:
 @dataclass
 class ApplyFuncAndNormalizeAsString(ExprNode):
     expr: ExprNode
-    apply_func: Callable = None
+    apply_func: Optional[Callable] = None
 
 
 @dataclass

data_diff/table_segment.py

@@ -112,18 +112,18 @@ class TableSegment:
 
     # Columns
     key_columns: Tuple[str, ...]
-    update_column: str = None
+    update_column: Optional[str] = None
     extra_columns: Tuple[str, ...] = ()
 
     # Restrict the segment
-    min_key: Vector = None
-    max_key: Vector = None
-    min_update: DbTime = None
-    max_update: DbTime = None
-    where: str = None
-
-    case_sensitive: bool = True
-    _schema: Schema = None
+    min_key: Optional[Vector] = None
+    max_key: Optional[Vector] = None
+    min_update: Optional[DbTime] = None
+    max_update: Optional[DbTime] = None
+    where: Optional[str] = None
+
+    case_sensitive: Optional[bool] = True
+    _schema: Optional[Schema] = None
 
     def __post_init__(self):
         if not self.update_column and (self.min_update or self.max_update):