Hacktoberfest 2026: los issues que los mantenedores marcaron para octubre, abiertos y aptos para principiantes. Explorar issues de Hacktoberfest

FunctionManager silently drops tool modules containing dataclasses with postponed annotations

Abierto
  • ¿Algo no está bien?
#681 0 comentarios 0 reacciones 0 asignados Ver en GitHub

@inchang-ing ya está trabajando en esto.

Desde el 29/9/2026.

  • #682 de @inchang-ing — abierto

Evaluación

Dificultad
3/5
Tiempo estimado
1-2 días
Aptitud para principiantes
72/100
Tipo de issue
Error
Claridad
Bien especificado
Estado de actividad
Activo
Stack tecnológico
python
Área
tooling

Línea de trabajo

Start with utils/function_manager.py and reproduce the temporary tool module containing postponed annotations and a dataclass. Run python reproduce.py --fetch --case chatdev-dataclass-tools --variant after from the linked bundle to inspect the seven regression and control tests. Done means increment is available and failed imports clean up or restore module state while existing filtering remains intact.

Escrito por el modelo de indexación a partir del texto del issue.

Descripción

Problem

The dynamic loader creates a module with module_from_spec() and executes it without inserting it into sys.modules. A tool file containing from __future__ import annotations and a dataclass then fails during class creation, because the dataclass decorator needs to resolve the defining module.

FunctionManager catches the import failure and omits the module's functions, so otherwise valid agent tools become unavailable.

Reproduction

import tempfile
from pathlib import Path
from utils.function_manager import FunctionManager

with tempfile.TemporaryDirectory() as directory:
    path = Path(directory) / 'example.py'
    path.write_text('''from __future__ import annotations
from dataclasses import dataclass

@dataclass
class Payload:
    value: int

def increment(value: int) -> int:
    return Payload(value).value + 1
''', encoding='utf-8')
    manager = FunctionManager(Path(directory))
    print(manager.has_function('increment'))  # Actual: False; expected: True
    print(manager.call_function('increment', 41))  # Expected: 42

Proposed change

Register the new module in sys.modules[module_name] before exec_module(). On an import failure, remove the partially initialized module or restore the previous entry, then re-raise into the existing loader error handling. This follows normal import semantics while preserving the current filtering of private and imported functions.

Validation

Source: utils/function_manager.py, current blob 50de030442ec72e8293482320cc37072367c86cd.

Before: 4 failed, 3 passed. After: 7 passed. The complete FunctionManager module was run against real temporary tool files, including postponed-annotation dataclasses, self-referential annotations, ordinary tools, imported/private function filtering, failed-import cleanup, and restoration after a failed reload.

Python 3.13.5, pytest 9.0.2, Linux. No provider calls or full workflow execution were run. Searches for FunctionManager issues and dataclass-related PRs found memory-store additions, not this dynamic-import failure.

Patch and tests

Source-pinned patch and seven regression/control tests. From the bundle root, run python reproduce.py --fetch --case chatdev-dataclass-tools --variant after. The runner verifies the source hash and patch application before executing the tests.

Lenguaje dominante
Python
Estrellas
34.5k
Forks
4.3k
Sin PR fusionados en 30 d

Preparar el entorno

  • Incluye un Dockerfile o un archivo de Docker Compose
  • Sin plantilla de pull request
  • Sin guía de contribución

Primeros pasos

  1. Lee el issue completo y luego la guía de contribución del proyecto.
  2. Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
  3. Haz un fork del repositorio y trabaja en una rama.
  4. Abre un pull request que haga referencia al número del issue.

Más de OpenBMB/ChatDev

Todos los issues de OpenBMB/ChatDev

Issues similares

Más issues de Python

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.