Django中如何从Model1的save触发Model2保存并同步字段

现有可正常运行的Django代码：

class Model1(models.Model):
    name = models.CharField(max_length=50)
    type = models.CharField(max_length=50)    

    class Meta:
        verbose_name_plural = "Model1s"
        ordering = ('sioticket',)

    def save(self, *args, **kwargs):
        super(Model1, self).save(*args, **kwargs)

class Model2(models.Model):
    name = models.ForeignKey(Model1, on_delete=models.CASCADE, null=True, blank=True)
    cpm = models.CharField(max_length=50)

    class Meta:
        verbose_name_plural = "Model2s"
        ordering = ('sioticket',)
    
    def save(self, *args, **kwargs):
        vartype = self.name.type
        self.cpm = vartype
        return super(Model2, self).save(*args, **kwargs)

需求：更新Model1的type字段后，关联的Model2的cpm字段需同步更新。当前创建新Model1并关联Model2时功能正常，但更新Model1的type后，Model2的cpm不会同步。

尝试修改Model1的save方法后触发错误：

def save(self, *args, **kwargs):
    super(Model1, self).save(*args, **kwargs)
    children = self.objects.all()
    for child in children:
        if child._meta.verbose_name_plural=="Model2s":
            child.save()

错误信息：

Manager isn't accessible via Model1 instances

错误原因分析 #

self.objects.all()写法错误：objects是模型类的管理器，只能通过Model1.objects调用，不能通过Model1的实例（self）访问。要获取当前Model1实例关联的所有Model2对象，需使用Django的反向关联管理器。

解决方案 #

方案1：触发关联Model2的save方法 #

因为Model2的外键字段是name，Django默认会为Model1生成反向关联管理器model2_set，用它可获取当前实例关联的所有Model2对象：

class Model1(models.Model):
    name = models.CharField(max_length=50)
    type = models.CharField(max_length=50)    

    class Meta:
        verbose_name_plural = "Model1s"
        ordering = ('sioticket',)

    def save(self, *args, **kwargs):
        super(Model1, self).save(*args, **kwargs)
        # 获取当前Model1关联的所有Model2对象并触发save
        for model2_obj in self.model2_set.all():
            model2_obj.save()

class Model2(models.Model):
    name = models.ForeignKey(Model1, on_delete=models.CASCADE, null=True, blank=True)
    cpm = models.CharField(max_length=50)

    class Meta:
        verbose_name_plural = "Model2s"
        ordering = ('sioticket',)
    
    def save(self, *args, **kwargs):
        if self.name:  # 避免name为null时抛出异常
            self.cpm = self.name.type
        super(Model2, self).save(*args, **kwargs)

方案2：直接批量更新Model2的cpm字段（更高效） #

如果不需要触发Model2的其他save逻辑，直接用批量更新可减少数据库操作：

class Model1(models.Model):
    name = models.CharField(max_length=50)
    type = models.CharField(max_length=50)    

    class Meta:
        verbose_name_plural = "Model1s"
        ordering = ('sioticket',)

    def save(self, *args, **kwargs):
        super(Model1, self).save(*args, **kwargs)
        # 批量更新关联Model2的cpm字段
        self.model2_set.update(cpm=self.type)

class Model2(models.Model):
    name = models.ForeignKey(Model1, on_delete=models.CASCADE, null=True, blank=True)
    cpm = models.CharField(max_length=50)

    class Meta:
        verbose_name_plural = "Model2s"
        ordering = ('sioticket',)
    
    def save(self, *args, **kwargs):
        if self.name:
            self.cpm = self.name.type
        super(Model2, self).save(*args, **kwargs)

优化建议 #

若觉得model2_set名称不够直观，可给Model2的外键字段设置related_name：

class Model2(models.Model):
    name = models.ForeignKey(Model1, on_delete=models.CASCADE, null=True, blank=True, related_name='model2s')
    # 其他字段保持不变

之后在Model1里就可以用self.model2s.all()或self.model2s.update(...)操作关联的Model2对象。

内容的提问来源于stack exchange，提问作者Manuel Sánchez