Eve-Elastic

Elasticsearch data layer for eve rest framework


License
GPL-3.0+
Install
pip install Eve-Elastic==7.0.1

Documentation

Eve-Elastic

https://travis-ci.org/petrjasek/eve-elastic.png?branch=elastic7

Eve-Elastic is elastic data layer for eve REST framework.

Features

  • fulltext search
  • filtering via elasticsearch filter dsl
  • facets support
  • aggregations support
  • elasticsearch mapping generator for schema

License

Eve-Elastic is GPLv3 licensed.

Supported elastic versions

It supports elastic 7 versions.

Install

$ pip install Eve-Elastic

Usage

Set elastic as your eve data layer.

import eve
from eve_elastic import Elastic

app = eve.Eve(data=Elastic)
app.run()

Config

There are 2 options for Eve-Elastic taken from app.config:

  • ELASTICSEARCH_URL (default: 'http://localhost:9200/') - this can be either single url or list of urls
  • ELASTICSEARCH_INDEX - (default: 'eve')
  • ELASTICSEARCH_INDEXES - (default: {}) - resource to index mapping
  • ELASTICSEARCH_FORCE_REFRESH - (default: True) - force index refresh after every modification
  • ELASTICSEARCH_AUTO_AGGREGATIONS - (default: True) - return aggregates on every search if configured for resource

Query params

Eve-Elastic supports eve like queries via where param which work as term filter.

On top of this, there is a predefined query_string query which does fulltext search.

$ curl http://localhost:5000/items?q=foo&df=name
  • q - query (default: *)
  • df - default field (default: _all)

Filtering

For more sophisticated filtering, you can use filter query param which will be used as filter for the query, using elastic filter dsl.

Facets

To add a facets support for specific resource, add facets into its datasource:

DOMAIN = {
    'contacts': {
        'datasource':
            'facets': {
                'urgency': {'terms': {'field': 'urgency'}},
                'versioncreated': {'date_histogram': {'field': 'versioncreated', 'interval': 'hour'}}
            }
        }
    }

You will find more info about facets in elasticsearch docs.