Let's say we have a large database, few milions row, one-to-many relationship models with some text field, and we want to search some keywords.
Traditional way will be real pain and slow, I know. So let's do something smart and enjoy lightning-fast execution with Full Text Search.
Suppose we have an app named main and it's models look like this
from django.db import models class Parent(models.Model): title = models.CharField(max_length=255) class Child(models.Model): parent = models.ForeignKey(Parent, on_delete=models.CASCADE, related_name="children") content = models.TextField(null=True, blank=True)
For example, we want to search keyword in childs content, and we want to do it quick! You can do like this, but that not quick enough for me. Maybe because I have too damn much data. But there are other way. First we need to add this line to settings.py
settings = [ # ... "django.contrib.postgres", # ... ]
Then we add SearchVectorField to the model
search_vector = SearchVectorField(null=True, blank=True)
Finally we can create index
from django.contrib.postgres.search import SearchVectorField from django.contrib.postgres.indexes import GinIndex class Child(models.Model): content = models.TextField(null=True, blank=True) search_vector = SearchVectorField(null=True, blank=True) class …