“关系管理器”是一个用于处理“一对多”和“多对多”关联关系的管理器。在以下两种情况用到：
\n“另一边”的关联关系 ForeignKey。即:
from django.db import models\n\nclass Reporter(models.Model):\n # ...\n pass\n\nclass Article(models.Model):\n reporter = models.ForeignKey(Reporter, on_delete=models.CASCADE)\n在上面的例子中，以下方法会在管理器 reporter.article_set 中可用。
双边关系 ManyToManyField
class Topping(models.Model):\n # ...\n pass\n\nclass Pizza(models.Model):\n toppings = models.ManyToManyField(Topping)\n在这个例子中，下文列出的方法在 topping.pizza_set 和 pizza.toppings 中均可用。
将特定的模型对象加入关联对象集合。
\n举例:
\n>>> b = Blog.objects.get(id=1)\n>>> e = Entry.objects.get(id=234)\n>>> b.entry_set.add(e) # Associates Entry e with Blog b.\n上述例子中，由于有 ForeignKey 关联关系，才可以使用 QuerySet.update() 更新数据。这要求对象事先已经被保存在数据库内了。
你可以使用 bulk=False 参数让关系管理器通过调用 e.save() 来执行更新操作。
Using add() with a many-to-many relationship, however, will not\ncall any save() methods (the bulk argument doesn't exist), but\nrather create the relationships using QuerySet.bulk_create(). If you need to execute\nsome custom logic when a relationship is created, listen to the\nm2m_changed signal, which will\ntrigger pre_add and post_add actions.
Using add() on a relation that already exists won't duplicate the\nrelation, but it will still trigger signals.
For many-to-many relationships add() accepts either model instances\nor field values, normally primary keys, as the *objs argument.
Use the through_defaults argument to specify values for the new\nintermediate model instance(s), if\nneeded.
创建一个新对象，保存至数据库，并将其放入关联对象集合内。返回刚创建的对象:
\n>>> b = Blog.objects.get(id=1)\n>>> e = b.entry_set.create(\n... headline='Hello',\n... body_text='Hi',\n... pub_date=datetime.date(2005, 1, 1)\n... )\n\n# No need to call e.save() at this point -- it's already been saved.\nThis is equivalent to (but simpler than):
\n>>> b = Blog.objects.get(id=1)\n>>> e = Entry(\n... blog=b,\n... headline='Hello',\n... body_text='Hi',\n... pub_date=datetime.date(2005, 1, 1)\n... )\n>>> e.save(force_insert=True)\n无需指定定义了模型间关系的关键字参数。在上述例子中，我们并未向 create() 传递参数 blog。Django 知道要将新 Entry 对象的 blog 字段设置为 b。
Use the through_defaults argument to specify values for the new\nintermediate model instance, if\nneeded.
从关系集合中移除特定的对象:
\n>>> b = Blog.objects.get(id=1)\n>>> e = Entry.objects.get(id=234)\n>>> b.entry_set.remove(e) # Disassociates Entry e from Blog b.\nSimilar to add(), e.save() is called in the example above\nto perform the update. Using remove() with a many-to-many\nrelationship, however, will delete the relationships using\nQuerySet.delete() which\nmeans no model save() methods are called; listen to the\nm2m_changed signal if you wish to\nexecute custom code when a relationship is deleted.
For many-to-many relationships remove() accepts either model\ninstances or field values, normally primary keys, as the *objs\nargument.
For ForeignKey objects, this method only\nexists if null=True. If the related field can't be set to None\n(NULL), then an object can't be removed from a relation without\nbeing added to another. In the above example, removing e from\nb.entry_set() is equivalent to doing e.blog = None, and because\nthe blog ForeignKey doesn't have\nnull=True, this is invalid.
For ForeignKey objects, this method accepts\na bulk argument to control how to perform the operation.\nIf True (the default), QuerySet.update() is used.\nIf bulk=False, the save() method of each individual model\ninstance is called instead. This triggers the\npre_save and\npost_save signals and comes at the\nexpense of performance.
For many-to-many relationships, the bulk keyword argument doesn't\nexist.
清空关系集合:
\n>>> b = Blog.objects.get(id=1)\n>>> b.entry_set.clear()\nNote this doesn't delete the related objects -- it just disassociates\nthem.
\nJust like remove(), clear() is only available on\nForeignKeys where null=True and it also\naccepts the bulk keyword argument.
For many-to-many relationships, the bulk keyword argument doesn't\nexist.
Replace the set of related objects:
\n>>> new_list = [obj1, obj2, obj3]\n>>> e.related_set.set(new_list)\nThis method accepts a clear argument to control how to perform the\noperation. If False (the default), the elements missing from the\nnew set are removed using remove() and only the new ones are added.\nIf clear=True, the clear() method is called instead and the\nwhole set is added at once.
For ForeignKey objects, the bulk\nargument is passed on to add() and remove().
For many-to-many relationships, the bulk keyword argument doesn't\nexist.
Note that since set() is a compound operation, it is subject to\nrace conditions. For instance, new objects may be added to the database\nin between the call to clear() and the call to add().
For many-to-many relationships set() accepts a list of either model\ninstances or field values, normally primary keys, as the objs\nargument.
Use the through_defaults argument to specify values for the new\nintermediate model instance(s), if\nneeded.