Skip to main content

Index and search records using ElasticSearch.

Project description

https://img.shields.io/travis/Kinto/kinto-elasticsearch.svg https://img.shields.io/pypi/v/kinto-elasticsearch.svg https://coveralls.io/repos/Kinto/kinto-elasticsearch/badge.svg?branch=master

kinto-elasticsearch forwards the records to ElasticSearch and provides a /search endpoint to query the indexed data.

Install

pip install kinto-elasticsearch

Setup

In the Kinto settings:

kinto.includes = kinto_elasticsearch
kinto.elasticsearch.hosts = localhost:9200

By default, ElasticSearch is smart and indices are not refreshed on every change. You can force this (with a certain drawback in performance):

kinto.elasticsearch.force_refresh = true

By default, indices names are prefixed with kinto-. You change this with:

kinto.elasticsearch.index_prefix = myprefix

Run ElasticSearch

Running a local install of ElasticSearch on localhost:9200 with Docker is pretty straightforward:

sudo docker run -p 9200:9200 elasticsearch

It is also be installed manually on Ubuntu with:

sudo apt-get install elasticsearch

And more information is available in the official docs.

Usage

Create a new record:

$ echo '{"data": {"note": "kinto"}}' | http POST http://localhost:8888/v1/buckets/example/collections/notes/records --auth token:alice-token

It should now be possible to search for it using the ElasticSearch API.

For example, using a quick querystring search:

$ http "http://localhost:8888/v1/buckets/example/collections/notes/search?q=note:kinto"--auth token:alice-token

Or an advanced search using request body:

$ echo '{"query": {"match_all": {}}}' | http POST http://localhost:8888/v1/buckets/example/collections/notes/search --auth token:alice-token
HTTP/1.1 200 OK
Access-Control-Expose-Headers: Retry-After, Content-Length, Alert, Backoff
Content-Length: 333
Content-Type: application/json; charset=UTF-8
Date: Wed, 20 Jan 2016 12:02:05 GMT
Server: waitress

{
    "_shards": {
        "failed": 0,
        "successful": 5,
        "total": 5
    },
    "hits": {
        "hits": [
            {
                "_id": "453ff779-e967-4b08-99b9-5c16af865a67",
                "_index": "example-assets",
                "_score": 1.0,
                "_source": {
                    "id": "453ff779-e967-4b08-99b9-5c16af865a67",
                    "last_modified": 1453291301729,
                    "note": "kinto"
                },
                "_type": "example-assets"
            }
        ],
        "max_score": 1.0,
        "total": 1
    },
    "timed_out": false,
    "took": 20
}

Custom index mapping

By default, ElasticSearch infers the data types from the indexed records.

But it’s possible to define the index mappings (ie. schema) from the collection metadata, in the index:schema property:

$ echo '{
  "data": {
    "index:schema": {
      "properties": {
        "id": {"type": "keyword"},
        "last_modified": {"type": "long"},
        "build": {
          "properties": {
              "date": {"type": "date", "format": "strict_date"},
              "id": {"type": "keyword"}
          }
        }
      }
    }
  }
}' | http PATCH "http://localhost:8888/v1/buckets/blog/collections/builds" --auth token:admin-token --verbose

Refer to ElasticSearch official documentation for more information about mappings.

See also, domapping a CLI tool to convert JSON schemas to ElasticSearch mappings.

Running the tests

$ make tests

Changelog

0.3.0 (2017-09-12)

New features

  • Add StatsD timer to measure E/S indexation (fixes #54)

  • Add a kinto-reindex command to reindex existing collections of records (fixes #56)

0.2.1 (2017-06-14)

Bug fixes

  • Fix the number of results when specified in query (ref #45)

0.2.0 (2017-06-13)

Bug fixes

  • Limit the number of results returned by default (fixes #45)

  • Fix crash on search parse exceptions (fixes #44)

0.1.0 (2017-05-26)

New features

  • Flush indices when server is flushed (fixes #4)

  • Perform insertions and deletion in bulk for better efficiency (fixes #5)

  • Add setting to force index refresh on change (fixes #6)

  • Add heartbeat (fixes #3)

  • Delete indices when buckets and collections are deleted (fixes #21)

  • Support quick search from querystring (fixes #34)

  • Return details about invalid queries in request body (fixes #23)

  • Support defining mapping from the index:schema property in the collection metadata (ref #8)

Bug fixes

  • Only index records if the storage transaction is committed (fixes #15)

  • Do not allow to search if no read permission on collection or bucket (fixes #7)

  • Fix empty results response when plugin was enabled after collection creation (ref #20)

Internal changes

  • Create index when collection is created (fixes #27)

0.0.1 (2017-05-22)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kinto-elasticsearch-0.3.0.tar.gz (15.0 kB view hashes)

Uploaded Source

Built Distribution

kinto_elasticsearch-0.3.0-py2.py3-none-any.whl (12.8 kB view hashes)

Uploaded Python 2 Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page