Django ORM查询,不同的值并加入同一个表

时间:2017-07-13 11:02:52

标签: mysql django

我希望根据sitehide = 0中的不同值返回所有列,并按created的日期对其进行排序。我知道只有PostgresSQL支持使用指定字段名称的distinct()调用,但我正在运行MySQL。我有一个有效的SQL查询(它可能效率不高)但不确定如何将其转换为Django ORM。

models.py

from django.db import models
from django.utils import timezone

# Create your models here.

class Results(models.Model):
    user_ip = models.GenericIPAddressField(unpack_ipv4=True)
    site_ip = models.GenericIPAddressField(unpack_ipv4=True)
    site = models.URLField()
    reason = models.CharField(max_length=50)
    hide = models.BooleanField(default=False)
    created = models.DateTimeField(default=timezone.now)

    def __str__(self):
        return self.site

表格结构:

mysql > SHOW CREATE TABLE results\G
*************************** 1. row ***************************
       Table: results
Create Table: CREATE TABLE `results` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `user_ip` char(39) NOT NULL,
  `site_ip` char(39) NOT NULL,
  `site` varchar(200) NOT NULL,
  `reason` varchar(50) NOT NULL,
  `hide` tinyint(1) NOT NULL,
  `created` datetime(6) NOT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=836 DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

SQL查询:

SELECT * FROM 
(
    SELECT site, MAX(created) created 
    FROM results 
    GROUP BY site 
    ORDER BY MAX(created) DESC 
    LIMIT 10
) _d 
JOIN results USING (site, created) 
ORDER BY _d.created DESC

2 个答案:

答案 0 :(得分:0)

有一个模块允许您对Django模型进行分组,并且仍然可以在结果中使用QuerySet:https://github.com/kako-nawao/django-group-by

例如:

from django_group_by import GroupByMixin

class ResultsQuerySet(QuerySet, GroupByMixin):
    pass

class Results(Model):
    # your model
class GroupedResultsListView(ListView):
    template_name = 'xxx/results.html'
    model = Results

    def get_queryset(self):
        return Results.objects.group_by('site').annotate(
            max_created=Max('created')).order_by(
            'created').distinct()
        # order by 'max_created' might also work

'XXX / results.html'

<ul>
{% for result in object_list %}
    <li>
        <h2>{{ result.site }}</td>
        <p>{{ result.max_created }}</p>
    </li>
{% endfor %}
</ul>

annotate / aggregate基本Django查询的不同之处在于使用了相关字段的属性,例如result.site。您还可以在一个查询中按多个属性进行分组,例如

Results.objects.group_by('site', 'user_ip')

如果您需要已组合在一起的实例的PK,请添加以下注释:

.annotate(pks=ArrayAgg('id'))

注意:ArrayAgg是Postgres特定的功能,可从Django 1.9开始提供:https://docs.djangoproject.com/en/1.10/ref/contrib/postgres/aggregates/#arrayagg

答案 1 :(得分:0)

我有一个解决方法,但我想知道是否有更好的方法,因为这需要两个数据库命中:

{{1}}