Question

在我类似REST的API中，我有一种机制允许用户自定义模型（添加自定义字段并添加自定义模型）。

这依赖于一组描述自定义的模型。

以下是我的代码的简化版本，以显示概念。

+ myproject
   + customization
      . models.py

class CustomModel(django.db.models.Model):
    # […]

class CustomField(django.db.models.Model):
    # […]

和

+ myproject
    + entities
        . models.py

class RootEntityModel(django.db.models.Model):
    name = CharField(…)
    # […]

from myproject import customization
from myproject.customization.models import CustomField 

# Here apply the custom filters on the root entity type
for custom_field in CustomField.objects.filter(entity__name='RootEntityModel').all():
    customize.build_field_and_apply(custom_field, RootEntityModel)

# Here instantiate custom models, etc
# […]

在Django 1.6中，我没有任何问题让一切正常。

现在，转换到Django 1.11，我做了一些清理工作，主要是因为在应用程序的根软件包中导入模型是不可能的。（另见this answer）。

我已经解除了我所拥有的每个AppRegistryNotReady例外情况，但对于myproject.entity.models模块中的例外情况。实际上，要构建我的实体，我必须查询CustomField和CustomModel表，但正如我目前正在构建我的模型一样，Django禁止我这样做。

我陷入了这种循环依赖。我们欢迎任何对变通方法的想法。

编辑：跟踪堆栈跟踪

  […]
  File "D:/Code/django-1.11/django/__init__.py", line 27, in setup
    apps.populate(settings.INSTALLED_APPS)
  File "D:/Code/django-1.11/django/apps/registry.py", line 108, in populate
    app_config.import_models()
  File "D:/Code/django-1.11/django/apps/config.py", line 202, in import_models
    self.models_module = import_module(models_module_name)
  File "C:/Python27/lib/importlib/__init__.py", line 37, in import_module
    __import__(name)
  File "D:/Code/resterserver/myproject/entities/models/__init__.py", line 369, in <module>
    for custom_field in CustomField.objects.filter(entity__name='RootEntityModel').all():
  File "D:/Code/django-1.11/django/db/models/manager.py", line 85, in manager_method
    return getattr(self.get_queryset(), name)(*args, **kwargs)
  File "D:/Code/django-1.11/django/db/models/query.py", line 784, in filter
    return self._filter_or_exclude(False, *args, **kwargs)
  File "D:/Code/django-1.11/django/db/models/query.py", line 802, in _filter_or_exclude
    clone.query.add_q(Q(*args, **kwargs))
  File "D:/Code/django-1.11/django/db/models/sql/query.py", line 1250, in add_q
    clause, _ = self._add_q(q_object, self.used_aliases)
  File "D:/Code/django-1.11/django/db/models/sql/query.py", line 1276, in _add_q
    allow_joins=allow_joins, split_subq=split_subq,
  File "D:/Code/django-1.11/django/db/models/sql/query.py", line 1154, in build_filter
    lookups, parts, reffed_expression = self.solve_lookup_type(arg)
  File "D:/Code/django-1.11/django/db/models/sql/query.py", line 1034, in solve_lookup_type
    _, field, _, lookup_parts = self.names_to_path(lookup_splitted, self.get_meta())
  File "D:/Code/django-1.11/django/db/models/sql/query.py", line 1331, in names_to_path
    if field.is_relation and not field.related_model:
  File "D:/Code/django-1.11/django/utils/functional.py", line 35, in __get__
    res = instance.__dict__[self.name] = self.func(instance)
  File "D:/Code/django-1.11/django/db/models/fields/related.py", line 115, in related_model
    apps.check_models_ready()
  File "D:/Code/django-1.11/django/apps/registry.py", line 132, in check_models_ready
    raise AppRegistryNotReady("Models aren't loaded yet.")
django.core.exceptions.AppRegistryNotReady: Models aren't loaded yet.

Answer 1

我不是直接在我的models.py模块中添加自定义模型，而是通过覆盖其AppConfig方法在ready()对象中执行此操作。

这是工作代码存根：

+ myproject
    + entities
        . apps.py

from django.apps.config import AppConfig

class EntitiesAppConfig(AppConfig):
    name = 'myproject.entities'
    label = 'entities'
    verbose_name = 'MyProject Entities app'
    def ready(self):
        super(EntitiesAppConfig, self).ready()

        # HERE i can do my customizations
        # ...

从django 1.6到1.11：AppRegistryNotReady异常

1 个答案: