Cache type_object_type() (#19514)

ilevkivskyi · web-flow · commit a8d2f1334b48 · 2025-07-27T22:49:16.000+01:00
This gives almost 4% performance boost (Python 3.12, compiled). Note
there is an old bug in `type_object_type()`, we treat not ready types as
`Any` without deferring, I disable caching in this case.

Unfortunately, using this in fine-grained mode is tricky, essentially I
have three options:
* Use some horrible hacks to invalidate cache when needed
* Add (expensive) class target dependency from `__init__`/`__new__`
* Only allow constructor caching during initial load, but disable it in
fine-grained increments

I decided to choose the last option. I think it has the best balance
complexity/benefits.
diff --git a/mypy/checker.py b/mypy/checker.py
@@ -397,6 +397,7 @@ def __init__(
         self.is_stub = tree.is_stub
         self.is_typeshed_stub = tree.is_typeshed_file(options)
         self.inferred_attribute_types = None
+        self.allow_constructor_cache = True
 
         # If True, process function definitions. If False, don't. This is used
         # for processing module top levels in fine-grained incremental mode.
@@ -500,12 +501,16 @@ def check_first_pass(self) -> None:
                         )
 
     def check_second_pass(
-        self, todo: Sequence[DeferredNode | FineGrainedDeferredNode] | None = None
+        self,
+        todo: Sequence[DeferredNode | FineGrainedDeferredNode] | None = None,
+        *,
+        allow_constructor_cache: bool = True,
     ) -> bool:
         """Run second or following pass of type checking.
 
         This goes through deferred nodes, returning True if there were any.
         """
+        self.allow_constructor_cache = allow_constructor_cache
         self.recurse_into_functions = True
         with state.strict_optional_set(self.options.strict_optional), checker_state.set(self):
             if not todo and not self.deferred_nodes:
diff --git a/mypy/checker_shared.py b/mypy/checker_shared.py
@@ -137,6 +137,7 @@ class TypeCheckerSharedApi(CheckerPluginInterface):
     module_refs: set[str]
     scope: CheckerScope
     checking_missing_await: bool
+    allow_constructor_cache: bool
 
     @property
     @abstractmethod
diff --git a/mypy/nodes.py b/mypy/nodes.py
@@ -3022,6 +3022,7 @@ class is generic then it will be a type constructor of higher kind.
         "dataclass_transform_spec",
         "is_type_check_only",
         "deprecated",
+        "type_object_type",
     )
 
     _fullname: str  # Fully qualified name
@@ -3178,6 +3179,10 @@ class is generic then it will be a type constructor of higher kind.
     # The type's deprecation message (in case it is deprecated)
     deprecated: str | None
 
+    # Cached value of class constructor type, i.e. the type of class object when it
+    # appears in runtime context.
+    type_object_type: mypy.types.FunctionLike | None
+
     FLAGS: Final = [
         "is_abstract",
         "is_enum",
@@ -3236,6 +3241,7 @@ def __init__(self, names: SymbolTable, defn: ClassDef, module_name: str) -> None
         self.dataclass_transform_spec = None
         self.is_type_check_only = False
         self.deprecated = None
+        self.type_object_type = None
 
     def add_type_vars(self) -> None:
         self.has_type_var_tuple_type = False
diff --git a/mypy/semanal_infer.py b/mypy/semanal_infer.py
@@ -31,6 +31,7 @@ def infer_decorator_signature_if_simple(
     """
     if dec.var.is_property:
         # Decorators are expected to have a callable type (it's a little odd).
+        # TODO: this may result in wrong type if @property is applied to decorated method.
         if dec.func.type is None:
             dec.var.type = CallableType(
                 [AnyType(TypeOfAny.special_form)],
@@ -47,6 +48,8 @@ def infer_decorator_signature_if_simple(
     for expr in dec.decorators:
         preserve_type = False
         if isinstance(expr, RefExpr) and isinstance(expr.node, FuncDef):
+            if expr.fullname == "typing.no_type_check":
+                return
             if expr.node.type and is_identity_signature(expr.node.type):
                 preserve_type = True
         if not preserve_type:
diff --git a/mypy/server/update.py b/mypy/server/update.py
@@ -1025,10 +1025,12 @@ def key(node: FineGrainedDeferredNode) -> int:
     # We seem to need additional passes in fine-grained incremental mode.
     checker.pass_num = 0
     checker.last_pass = 3
-    more = checker.check_second_pass(nodes)
+    # It is tricky to reliably invalidate constructor cache in fine-grained increments.
+    # See PR 19514 description for details.
+    more = checker.check_second_pass(nodes, allow_constructor_cache=False)
     while more:
         more = False
-        if graph[module_id].type_checker().check_second_pass():
+        if graph[module_id].type_checker().check_second_pass(allow_constructor_cache=False):
             more = True
 
     if manager.options.export_types:
diff --git a/mypy/typeops.py b/mypy/typeops.py
@@ -11,6 +11,7 @@
 from collections.abc import Iterable, Sequence
 from typing import Any, Callable, TypeVar, cast
 
+from mypy.checker_state import checker_state
 from mypy.copytype import copy_type
 from mypy.expandtype import expand_type, expand_type_by_instance
 from mypy.maptype import map_instance_to_supertype
@@ -145,6 +146,15 @@ def type_object_type(info: TypeInfo, named_type: Callable[[str], Instance]) -> P
     where ... are argument types for the __init__/__new__ method (without the self
     argument). Also, the fallback type will be 'type' instead of 'function'.
     """
+    allow_cache = (
+        checker_state.type_checker is not None
+        and checker_state.type_checker.allow_constructor_cache
+    )
+
+    if info.type_object_type is not None:
+        if allow_cache:
+            return info.type_object_type
+        info.type_object_type = None
 
     # We take the type from whichever of __init__ and __new__ is first
     # in the MRO, preferring __init__ if there is a tie.
@@ -167,7 +177,15 @@ def type_object_type(info: TypeInfo, named_type: Callable[[str], Instance]) -> P
     init_index = info.mro.index(init_method.node.info)
     new_index = info.mro.index(new_method.node.info)
 
-    fallback = info.metaclass_type or named_type("builtins.type")
+    if info.metaclass_type is not None:
+        fallback = info.metaclass_type
+    elif checker_state.type_checker:
+        # Prefer direct call when it is available. It is faster, and,
+        # unfortunately, some callers provide bogus callback.
+        fallback = checker_state.type_checker.named_type("builtins.type")
+    else:
+        fallback = named_type("builtins.type")
+
     if init_index < new_index:
         method: FuncBase | Decorator = init_method.node
         is_new = False
@@ -189,7 +207,10 @@ def type_object_type(info: TypeInfo, named_type: Callable[[str], Instance]) -> P
                     is_bound=True,
                     fallback=named_type("builtins.function"),
                 )
-                return class_callable(sig, info, fallback, None, is_new=False)
+                result: FunctionLike = class_callable(sig, info, fallback, None, is_new=False)
+                if allow_cache:
+                    info.type_object_type = result
+                return result
 
         # Otherwise prefer __init__ in a tie. It isn't clear that this
         # is the right thing, but __new__ caused problems with
@@ -199,12 +220,19 @@ def type_object_type(info: TypeInfo, named_type: Callable[[str], Instance]) -> P
     # Construct callable type based on signature of __init__. Adjust
     # return type and insert type arguments.
     if isinstance(method, FuncBase):
+        if isinstance(method, OverloadedFuncDef) and not method.type:
+            # Do not cache if the type is not ready. Same logic for decorators is
+            # achieved in early return above because is_valid_constructor() is False.
+            allow_cache = False
         t = function_type(method, fallback)
     else:
         assert isinstance(method.type, ProperType)
         assert isinstance(method.type, FunctionLike)  # is_valid_constructor() ensures this
         t = method.type
-    return type_object_type_from_function(t, info, method.info, fallback, is_new)
+    result = type_object_type_from_function(t, info, method.info, fallback, is_new)
+    if allow_cache:
+        info.type_object_type = result
+    return result
 
 
 def is_valid_constructor(n: SymbolNode | None) -> bool:
@@ -865,8 +893,8 @@ def function_type(func: FuncBase, fallback: Instance) -> FunctionLike:
         if isinstance(func, FuncItem):
             return callable_type(func, fallback)
         else:
-            # Broken overloads can have self.type set to None.
-            # TODO: should we instead always set the type in semantic analyzer?
+            # Either a broken overload, or decorated overload type is not ready.
+            # TODO: make sure the caller defers if possible.
             assert isinstance(func, OverloadedFuncDef)
             any_type = AnyType(TypeOfAny.from_error)
             dummy = CallableType(
@@ -1254,6 +1282,8 @@ def get_protocol_member(
     if member == "__call__" and class_obj:
         # Special case: class objects always have __call__ that is just the constructor.
 
+        # TODO: this is wrong, it creates callables that are not recognized as type objects.
+        # Long-term, we should probably get rid of this callback argument altogether.
         def named_type(fullname: str) -> Instance:
             return Instance(left.type.mro[-1], [])
 
