I am making a query in django that ends up hitting the database 4 times. I want to be sure that there's no more efficient way of pulling the data than I already am doing.
I have a model:
class Film(models.Model):
studio = models.ForeignKey(Studio)
name = models.CharField(max_length=30)
rating = models.ForeignKey(Rating)
actor = models.ManyToManyField('Actor', blank=True, related_name='actor', db_table=u'moviedb_film_actor')
staff = models.ManyToManyField(Staff, blank=True, through='Association')
My view is here:
def film_by_id(request, id):
# Look up the car name (and raise a 404 if it can't be found).
object = get_object_or_404(film.objects.select_related(), id__iexact=id)
template_name = 'Film_by_id.html',
association_objects = Association.objects.select_related(depth=1).filter(Q(film__name=object.name))
source_objects = Source.objects.filter(film__name=object.name)
object.association_objects = association_objects
object.trim_objects = trim_objects
return render_to_response(
template_name,
{"object" : object},
context_instance = RequestContext(request))
And finally, my template:
Make: {{ object.studio }}<br>
Rating: {{ object.rating }}<br><br>
<u>--Actors--</u><br>
{% for actor in object.actor.all %}
{{ actor.name }}<br>
{% endfor %}
<br>
<u>--Staff-开发者_Python百科-</u><br>
{% for item in object.association_objects %}
{{ item.staff.name }} - {{ item.get_type_display }}<br>
{% endfor %}
<u>--Source--</u><br>
<ul>
{% for item in object.source_objects %}
<li>{{ item }}
{% endfor %}
</ul>
The debug toolbar indicates that I'm hitting the database 4 times. The problem sees to be the ManytoMany fields and the "reverse select_related" I'm trying to execute with Source.
So the initial data pull hits once
Referencing Actor hits once Referencing Association_objects hits once Referencing Source hits onceSo overall, is there a way to reduce the database hits? Specifically, is there a better way to pull the many-to-many relations without creating more simultaneous database queries?
You could eliminate the lookups for Association
and Source
if you were to use a ForeignKey
relationship between them. Then select_related
could pull these as well.
If you really, deeply care about optimizing the ManyToManyField I would suggest caching the result of the lookup.
精彩评论