Django 按中间模型查询集合顺序_

我正在尝试创建一个中间模型，其中

created

字段记录以下关系的时间

模型如下：

class Profile(models.Model):
    user = models.OneToOneField(User, related_name='profile')
    realname = models.CharField(max_length=20, verbose_name='真實姓名', blank=True)
    nickname = models.CharField(max_length=20, verbose_name='暱稱/顯示名稱')

## the intermediate model is connected through following field  ##
    followings = models.ManyToManyField('self', through='FollowShip',
                                        related_name='followers', symmetrical=False,
                                        through_fields=('profile_from', 'profile_to'))

    total_followers = models.IntegerField(default=0)

    def __str__(self):
        return 'Profile of User:{}'.format(self.user.username)


class FollowShip(models.Model):
    profile_from = models.ForeignKey(Profile, related_name='follow_from_set')
    profile_to = models.ForeignKey(Profile, related_name='follow_to_set')
    created = models.DateField(auto_now_add=True, db_index=True)

    def __str__(self):
        return "{} follows {}".format(self.profile_from, self.profile_to)

    class Meta:
        unique_together = ('profile_from', 'profile_to')

它很好用。但是，当我试图通过

FollowShip

模型的

created

字段访问QuerySet订单时。错误发生了

在

manage.py shell中

：

>>> user.profile.followings.order_by('followship__created')
Traceback (most recent call last):
  File "<console>", line 1, in <module>
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/query.py", line 232, in __repr__
    data = list(self[:REPR_OUTPUT_SIZE + 1])
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/query.py", line 256, in __iter__
    self._fetch_all()
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/query.py", line 1087, in _fetch_all
    self._result_cache = list(self.iterator())
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/query.py", line 54, in __iter__
    results = compiler.execute_sql()
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/compiler.py", line 824, in execute_sql
    sql, params = self.as_sql()
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/compiler.py", line 369, in as_sql
    extra_select, order_by, group_by = self.pre_sql_setup()
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/compiler.py", line 47, in pre_sql_setup
    order_by = self.get_order_by()
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/compiler.py", line 293, in get_order_by
    field, self.query.get_meta(), default_order=asc))
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/compiler.py", line 553, in find_ordering_name
    field, targets, alias, joins, path, opts = self._setup_joins(pieces, opts, alias)
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/compiler.py", line 586, in _setup_joins
    pieces, opts, alias)
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/query.py", line 1402, in setup_joins
    names, opts, allow_many, fail_on_missing=True)
  File "/Users/young/Desktop/env/strayvoice/lib/python3.5/site-packages/django/db/models/sql/query.py", line 1327, in names_to_path
    "Choices are: %s" % (name, ", ".join(available)))
django.core.exceptions.FieldError: Cannot resolve keyword 'followship' into field. Choices are: follow_from_set, follow_to_set, followers, followings, id, nickname, realname, total_followers, user, user_id

---

没有任何错误的线索。

使用

p。从集合中跟随。按（'created'）排序。

user.profile.following

表示

user.profile

遵循的配置文件。

user.profile.followers

指跟随

user.profile

的个人资料
因此

user.profile.followers

和

user.profile.followers

都将返回

profile

的查询集，它们只能按

profile

的字段排序，例如user、realname、昵称

如果您想查找后面跟着

use.profile

和order\u by

created

的配置文件，您可以使用：

result = []
for profile_id in user.profile.follow_from_set.order_by('created').values('profile_to'):
    result.append(Profile.objects.get(id=profile_id))

如果有以下内容，您可以通过以下方法优化sql：

profile_ids = user.profile.follow_from_set.order_by('created').values('profile_to')
profiles = Profile.objects.filter(id__in=profile_ids)

最后，按照profile_id的顺序重新排列profile

result = []
result_dict = {}
for profile in profiles:
    result_dict[profile.id] = profile
for id in profile_ids:
    result.append(result_dict[id])
return result

---

也许有很多更好的方法来解决这个问题。有谁能做出更好的解决方案吗？

相同的错误消息。看起来m2m关系指向“自我”，中间模型的行为与指向不同模型的行为不同。请参阅本文的最后一部分。我想我可以像文档中所说的那样查询Person.objects.filter（…group\uu name='The披头士'，…membership\uu date\u joined\uu gt=date（1961,1,1））。你的方法行得通。但要做到这一点，我必须多次将点击添加到数据库中，代价是巨大的。queryset

Profile.objects.filter（followers\uu昵称\uu包含=“”）

就像您推荐的那样。在这里，您可能需要类似于

Profile.objects.filter（followers\u from\u set\u created\u lt=datetime.date（2016,1,1））

或

Profile.objects.filter（followers\u created\u lt=datetime.date（2016,1,1））

（两者都不起作用）似乎访问中介模型的唯一方法就像您通过

follow\u from\u set

或

follow\u to\u set

所说的那样。使用第二种方法，这是可以接受的。sql是

'选择“myuser\u followship.”从“myuser\u followship”中选择“profile\u to\u id”，其中是“myuser\u followship”。“profile\u FROM\u id”=1个“myuser\u followship”顺序。“创建”ASC LIMIT 21'

并

选择“myuser\u profile”。“id”，“myuser\u profile”。“user\u id”，“myuser\u profile”，“myuser\u profile”，“myuser\u profile”，“myuser\u profile”，“realname”，“myuser\u profile”，“myuser\u profile”，“myuser\u profile”，“昵称”，“myuser\u pro。“myuser\u profile”中的“total\u followers”和“myuser\u profile”中的“total\u followship”和“myuser\u followship”中的“total\u followship”和“myuser\u profile\u followship”中的“to\u id”选择U0。profile\u FROM\u id”=1）LIMIT 21'

。我不太清楚

LIMIT 21代表什么。
result = []
result_dict = {}
for profile in profiles:
    result_dict[profile.id] = profile
for id in profile_ids:
    result.append(result_dict[id])
return result