Faceted search

Sample type definition (note the usage of FacetedSearchFilterBackend and faceted_search_fields).

from elasticsearch_dsl import DateHistogramFacet, RangeFacet
from graphene import Node
from graphene_elastic import ElasticsearchObjectType
from graphene_elastic.filter_backends import (
    FacetedSearchFilterBackend,
    # ...
)

from search_index.documents import Post as PostDocument

class Post(ElasticsearchObjectType):

    class Meta:

        document = PostDocument
        interfaces = (Node,)
        filter_backends = [
            # ...
            FacetedSearchFilterBackend,
            # ...
        ]

        # For `FacetedSearchFilterBackend` backend
        faceted_search_fields = {
            'category': 'category.raw',
            'category_global': {
                'field': 'category.raw',
                # Setting `global` to True, makes the facet global
                'global': True,
            },
            'tags': {
                'field': 'tags.raw',
                'enabled': True,  # Will appear in the list by default
                'global': True,
            },
            'created_at': {
                'field': 'created_at',
                'facet': DateHistogramFacet,
                'options': {
                    'interval': 'year',
                }
            },
            'num_views_count': {
                'field': 'num_views',
                'facet': RangeFacet,
                'options': {
                    'ranges': [
                        ("<10", (None, 10)),
                        ("11-20", (11, 20)),
                        ("20-50", (20, 50)),
                        (">50", (50, None)),
                    ]
                }
            },
        }

Sample GraphQL query:

query {
  allPostDocuments(
        search:{title:{value:"alice"}}
        facets:[category]
  ) {
    facets
    edges {
      node {
        id
        title
        highlight
      }
    }

  }
}

Sample response:

{
  "data": {
    "allPostDocuments": {
      "facets": {
        "tags": {
          "doc_count": 9,
          "aggs": {
            "doc_count_error_upper_bound": 0,
            "sum_other_doc_count": 0,
            "buckets": [
              {
                "key": "photography",
                "doc_count": 7
              },
              {
                "key": "art",
                "doc_count": 6
              },
              {
                "key": "article",
                "doc_count": 5
              },
              {
                "key": "black and white",
                "doc_count": 5
              },
              {
                "key": "package",
                "doc_count": 5
              },
              {
                "key": "models",
                "doc_count": 4
              },
              {
                "key": "programming",
                "doc_count": 4
              }
            ]
          }
        },
        "category": {
          "doc_count": 9,
          "aggs": {
            "doc_count_error_upper_bound": 0,
            "sum_other_doc_count": 0,
            "buckets": [
              {
                "key": "Python",
                "doc_count": 3
              },
              {
                "key": "Model Photography",
                "doc_count": 2
              },
              {
                "key": "Django",
                "doc_count": 1
              },
              {
                "key": "Elastic",
                "doc_count": 1
              },
              {
                "key": "Machine Learning",
                "doc_count": 1
              },
              {
                "key": "MongoDB",
                "doc_count": 1
              }
            ]
          }
        }
      },
      "edges": [
        {
          "node": {
            "id": "UG9zdDpBVWNwVm0wQklwZ2dXbVlJTndOVA==",
            "title": "better Alice must",
            "highlight": {
              "title": [
                "better <b>Alice</b> must"
              ]
            }
          }
        },
        ...
      ]
    }
  }
}

Note, that category appeared in the result because we explicitly requested so (in facets:[category]) and the tags are there because they have been enabled by default (in faceted_search_fields).