When accessing model fields that were not loaded as part of the original query,\nDjango will fetch that field’s data from the database. You can customize the\nbehavior of this fetching with a fetch mode, making it more efficient or\neven blocking it.
\nUse QuerySet.fetch_mode() to set the fetch mode for model\ninstances fetched by a QuerySet:
from django.db import models\n\nbooks = Book.objects.fetch_mode(models.FETCH_PEERS)\nFetch modes apply to:
\nForeignKey fields
OneToOneField fields and their reverse accessors
Fields deferred with QuerySet.defer() or QuerySet.only()
Django copies the fetch mode of an instance to any related objects it fetches,\nso the mode applies to a whole tree of relationships, not just the top-level\nmodel in the initial QuerySet. This copying is also done in related\nmanagers, even though fetch modes don’t affect such managers” queries.
Django provides three fetch modes. We’ll explain them below using these models:
\nfrom django.db import models\n\n\nclass Author(models.Model): ...\n\n\nclass Book(models.Model):\n author = models.ForeignKey(Author, on_delete=models.CASCADE)\n ...\n…and this loop:
\nfor book in books:\n print(book.author.name)\n…where books is a QuerySet of Book instances using some fetch mode.
Fetches the missing field for the current instance only. This is the default\nmode.
\nUsing FETCH_ONE for the above example would use:
1 query to fetch books
N queries, where N is the number of books, to fetch the missing author\nfield
…for a total of 1+N queries. This query pattern is known as the «N+1 queries\nproblem» because it often leads to performance issues when N is large.
\nFetches the missing field for the current instance and its «peers»—instances\nthat came from the same initial QuerySet. The behavior of this mode is\nbased on the assumption that if you need a field for one instance, you probably\nneed it for all instances in the same batch, since you’ll likely process them\nall identically.
Using FETCH_PEERS for the above example would use:
1 query to fetch books
1 query to fetch all missing author fields for the batch of books
…for a total of 2 queries. The batch query makes this mode a lot more efficient\nthan FETCH_ONE and is similar to an on-demand call to\nQuerySet.prefetch_related() or\nprefetch_related_objects(). Using FETCH_PEERS can\nreduce most cases of the «N+1 queries problem» to two queries without\nmuch effort.
The «peer» instances are tracked in a list of weak references, to avoid\nmemory leaks where some peer instances are discarded.
\nRaises a FieldFetchBlocked exception.
Using FETCH_RAISE for the above example would raise an exception at the\naccess of book.author, like:
FieldFetchBlocked("Fetching of Primary.value blocked.")\nThis mode can prevent unintentional queries in performance-critical\nsections of code.
\nSet the default fetch mode for a model class with a\ncustom manager that overrides get_queryset():
from django.db import models\n\n\nclass BookManager(models.Manager):\n def get_queryset(self):\n return super().get_queryset().fetch_mode(models.FETCH_PEERS)\n\n\nclass Book(models.Model):\n title = models.TextField()\n author = models.ForeignKey("Author", on_delete=models.CASCADE)\n\n objects = BookManager()\n