python
type_error
ai_generated
true
AttributeError: Mock object has no attribute 'foo' when using spec=MyClass
ID: python/unittest-mock-spec-attribute
80%Fix Rate
84%Confidence
0Evidence
2025-10-08First Seen
Version Compatibility
| Version | Status | Introduced | Deprecated | Notes |
|---|---|---|---|---|
| 3.9 | active | — | — | — |
Root Cause
Creating a mock with spec=MyClass restricts the mock to only have attributes that exist on MyClass, and accessing a non-existent attribute raises an AttributeError.
generic中文
使用 spec=MyClass 创建模拟对象会限制模拟仅具有 MyClass 上存在的属性,访问不存在的属性会引发 AttributeError。
Workarounds
-
80% success
Add the missing attribute to MyClass or use a mock with spec=MyClass and then add the attribute dynamically: mock.foo = MagicMock()
-
85% success
Use autospec=True instead of spec to automatically create a mock that matches the class's interface.
Dead Ends
Common approaches that don't work:
-
50% fail
Using spec_set=True instead of spec still raises the same error for non-existent attributes.
-
60% fail
Removing the spec parameter entirely allows any attribute but may hide API mismatches.