Django haystack for advanced search

1494 views

                  from haystack import indexes
from blog.models import Post


class PostIndex(indexes.SearchIndex, indexes.Indexable):
    text = indexes.CharField(document=True, use_template=True)
    title = indexes.CharField(model_attr='title')
    author = indexes.CharField(model_attr='author__username')
    published_at = indexes.DateTimeField(model_attr='published_at')

    def get_model(self):
        return Post

    def index_queryset(self, using=None):
        """Only index published posts."""
        return self.get_model().objects.filter(status='published')

                  from haystack.query import SearchQuerySet
from django.shortcuts import render


def search_view(request):
    query = request.GET.get('q', '')

    if query:
        results = SearchQuerySet().filter(content=query).load_all()

        # Faceting
        results = results.facet('author')

        # Highlighting
        results = results.highlight()
    else:
        results = []

    return render(request, 'search/results.html', {
        'results': results,
        'query': query
    })

Haystack provides unified search across backends (Elasticsearch, Solr, Whoosh). I define search indexes mapping models to searchable fields. It handles full-text search, faceting, and highlighting. The SearchQuerySet API is similar to Django's ORM. I use SearchIndex classes to specify indexed fields and document preparation. For autocomplete, I use edge n-grams. Haystack supports real-time indexing via signals or batch updates. This enables Google-like search in Django apps without writing backend-specific code.

Priya Sharma

More from Priya Sharma

Django haystack for advanced search

0 Comments

More from Priya Sharma