Using Dynaconf results in TypeError: __name__ must be set to a string object error

[ahmadalli]

Hi

If I Add FlaskDynaconf(app) before FlaskInjector(app=app, modules=[init_inject]), it'll result in TypeError: __name__ must be set to a string object

this is the dynaconf project: https://dynaconf.readthedocs.io/en/latest/guides/flask.html

[jstasiak]

Please provide full stack trace and versions of the relevant software you're using, the exception itself provides almost no information.

[ahmadalli]

sorry for lacking info. here's the stack trace:

TypeError: __name__ must be set to a string object
  File "runpy.py", line 193, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "__main__.py", line 45, in <module>
    cli.main()
  File "debugpy/../debugpy/server/cli.py", line 430, in main
    run()
  File "debugpy/../debugpy/server/cli.py", line 317, in run_module
    run_module_as_main(target, alter_argv=True)
  File "runpy.py", line 193, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "__main__.py", line 15, in <module>
    main(as_module=True)
  File "flask/cli.py", line 967, in main
    cli.main(args=sys.argv[1:], prog_name="python -m flask" if as_module else None)
  File "flask/cli.py", line 586, in main
    return super(FlaskGroup, self).main(*args, **kwargs)
  File "click/core.py", line 782, in main
    rv = self.invoke(ctx)
  File "click/core.py", line 1259, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "click/core.py", line 1066, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "click/core.py", line 610, in invoke
    return callback(*args, **kwargs)
  File "click/decorators.py", line 73, in new_func
    return ctx.invoke(f, obj, *args, **kwargs)
  File "click/core.py", line 610, in invoke
    return callback(*args, **kwargs)
  File "flask/cli.py", line 848, in run_command
    app = DispatchingApp(info.load_app, use_eager_loading=eager_loading)
  File "flask/cli.py", line 305, in __init__
    self._load_unlocked()
  File "flask/cli.py", line 83, in find_best_app
    app = call_factory(script_info, app_factory)
  File "flask/cli.py", line 119, in call_factory
    return app_factory()
  File "app.py", line 36, in create_app
    FlaskInjector(app=app, modules=[init_inject])
  File "__init__.py", line 327, in __init__
    process_dict(container, injector)
  File "__init__.py", line 380, in process_dict
    d[key] = wrap_fun(value, injector)
  File "__init__.py", line 66, in wrap_fun
    return wrap_function(fun, injector)
  File "__init__.py", line 94, in wrap_function
    def wrapper(*args: Any, **kwargs: Any) -> Any:
  File "functools.py", line 54, in update_wrapper
    setattr(wrapper, attr, value)

[ahmadalli]

if I run FlaskDynaconf(app) after FlaskInjector(app=app, modules=[init_inject]) everything works fine

[jstasiak]

Thanks. I have to repeat my request for you to provide full stack trace (including full file paths and package names, the trace you pasted is modified; you can mask some private/confidential details but not this much, I'd like to know what I'm dealing with and for the general structure to be intact) and versions of the relevant software (Python, Flask, Flask-Injector, Injector, Flask-Dynaconf).

[ahmadalli]

sorry it took so long to answer, I was busy on other projects. here's an example project I set up: https://github.com/ahmadalli/flask-dynaconf-inject

when I run it with this commands, it fails with the same error:

pipenv shell
export FLASK_APP=microblog
flask run

[vshulgin]

The same for me, using FlaskDynaconf(app) breaks FlaskInjector call with mentioned error

[zeidlerb]

I'm having the same issue. The problem seems to be that the current logic assumes that everything that has the __bindings__ or __call__ attribute should be wrapped. However, @functools.wraps tries to set the __name__ attribute of the wrapping function (among others), which is problematic in this case, since the Dynaconf Configuration can be called, but has no __name__. Consequently, trying to wrap it using functools causes an exception.

Note that as an (admittedly brittle) workaround, you can access app.jinja_env.globals['config'] once before applying Dynaconf , which seems to mask the issue by pinning the cached version of the config to the original Config class of the Flask app, which does not have the __call__ attribute.

Full sample code for reproduction:

from dynaconf import FlaskDynaconf
from flask import Flask
from flask_injector import FlaskInjector


app = Flask(__name__)
FlaskDynaconf(app)
FlaskInjector(app)

$  python --version
Python 3.8.2

requirements.txt

click==7.1.2
dynaconf==3.1.2
Flask==1.1.2
Flask-Injector==0.12.3
injector==0.18.4
itsdangerous==1.1.0
Jinja2==2.11.2
MarkupSafe==1.1.1
typing-extensions==3.7.4.3
Werkzeug==1.0.1

Stacktrace:

Traceback (most recent call last):
  File "C:/Users/User/PycharmProjects/test_me/app.py", line 8, in <module>
    FlaskInjector(app)
  File "C:\Users\User\PycharmProjects\test_me\venv\lib\site-packages\flask_injector\__init__.py", line 327, in __init__
    process_dict(container, injector)
  File "C:\Users\User\PycharmProjects\test_me\venv\lib\site-packages\flask_injector\__init__.py", line 380, in process_dict
    d[key] = wrap_fun(value, injector)
  File "C:\Users\User\PycharmProjects\test_me\venv\lib\site-packages\flask_injector\__init__.py", line 66, in wrap_fun
    return wrap_function(fun, injector)
  File "C:\Users\User\PycharmProjects\test_me\venv\lib\site-packages\flask_injector\__init__.py", line 94, in wrap_function
    def wrapper(*args: Any, **kwargs: Any) -> Any:
  File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\functools.py", line 54, in update_wrapper
    setattr(wrapper, attr, value)
TypeError: __name__ must be set to a string object