Django Admin超慢选择单个记录

HenryM 发表于 Dev

亨利·M

我有一个Pick模型，当我在Django Admin中选择一条记录时，需要花一些时间（大约20秒）来检索记录（大约70k）。但是，当我尝试创建/保存记录时，它很快。我已经添加了索引以尝试加快检索速度，但是我真的不知道该怎么做

class Pick (models.Model):
    team = models.ForeignKey(Team, on_delete=models.CASCADE, null=True, blank=True)
    user = models.ForeignKey(settings.AUTH_USER_MODEL, on_delete=models.CASCADE, db_index=True)
    contest_entry=models.ForeignKey(ContestEntry, on_delete=models.CASCADE, db_index=True)
    game_round=models.ForeignKey(GameRound, on_delete=models.CASCADE, db_index=True)

    objects = DataFrameManager()

    def __str__(self):
        return f'%s %s' % (self.contest_entry, self.game_round)

    class Meta:
        unique_together = ['user', 'contest_entry', 'game_round']
        ordering = ['contest_entry','game_round','user','team']
        index_together = [["user", "contest_entry", 'game_round'],]
        indexes = [
            models.Index(fields=['user', 'contest_entry', 'game_round']),
            models.Index(fields=['game_round'], name='game_round_idx'),
            models.Index(fields=['contest_entry'], name='contest_entry_idx'),
        ]

class PickAdmin(admin.ModelAdmin):
    model = Pick
    list_filter= (
        ('user', RelatedDropdownFilter),
        ('contest_entry__contest', RelatedDropdownFilter),
        ('game_round', RelatedDropdownFilter)
    )

    def get_queryset(self, request):
        return super().get_queryset(request).select_related('user','contest_entry','game_round')
admin.site.register(Pick, PickAdmin)

我如何在这里提高性能？

bdoubleu

关键是raw_id_fields在admin类上指定。<select>外键关系的默认小部件可能会导致检索和呈现所有选项的大量开销。

class PickAdmin(admin.ModelAdmin):
    model = Pick
    list_filter= (
        ('user', RelatedDropdownFilter),
        ('contest_entry__contest', RelatedDropdownFilter),
        ('game_round', RelatedDropdownFilter)
    )
    raw_id_fields = ('user', 'contest_entry', 'game_round',)

    def get_queryset(self, request):
        return super().get_queryset(request).select_related('user','contest_entry','game_round')

本文收集自互联网，转载请注明来源。

如有侵权，请联系 [email protected] 删除。