import requests

# Cluster health
# GET _cluster/health
response = requests.get('http://localhost:9200/_cluster/health')
response.json() # dict

# Node info
# GET _nodes/stats
response = requests.get('http://localhost:9200/_nodes/stats')
response.json()
            

Create

# create index
response = requests.put('http://localhost:9200/tutorial_curl')
response.json()

# create index with a mapping
headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'mappings': {
        'properties': {
            'Firstname': {
                'type': 'keyword',
            },
            'Lastname': {
                'type': 'long',
            },
        },
    },
}

response = requests.put('http://localhost:9200/tutorial_curl', headers=headers, json=json_data)
            

Insert

# insert document
headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'Firstname': 'Lin',
    'Lastname': 'Chen',
}

response = requests.post('http://localhost:9200/tutorial_curl/_doc', headers=headers, json=json_data)

# insert a document with an id
headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'Firstname': 'Lin',
    'Lastname': 'Chen',
}

response = requests.post('http://localhost:9200/tutorial_curl/_doc/1', headers=headers, json=json_data)
            

Read

# read document
response = requests.get('http://localhost:9200/tutorial_curl/_doc/1')
response.json()
            

Update

# update document
headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'doc': {
        'Firstname': 'Unknown',
        'Lastname': 'Unknown',
    },
}

response = requests.post('http://localhost:9200/tutorial_curl/_update/1', headers=headers, json=json_data)
            

Delete

# delete index
response = requests.delete('http://localhost:9200/tutorial_curl')
response.json()

# delete document
response = requests.delete('http://localhost:9200/tutorial_curl/_doc/1')

# delete by query
headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'query': {
        'match': {
            'Firstname': {
                'query': 'Lin',
            },
        },
    },
}

response = requests.post('http://localhost:9200/tutorial_curl/_delete_by_query', headers=headers, json=json_data)
            

Insert

import requests

headers = {
    'Content-Type': 'application/json',
}

# \n is mandatory at the end of each line
data = '{"index": {"_index": "baseline_1", "_id":3}}\n{"Absolute vorticity": 0.0001307,"Apparent temperature": 232.5}\n{"index": {"_index": "baseline_1", "_id":4}}\n{"Absolute vorticity": 0.0001307,"Apparent temperature": 233.6}\n'

response = requests.post('http://localhost:9200/_bulk', headers=headers, data=data)
            

Delete

import requests

headers = {
    # Already added when you pass json= but not when you pass data=
    # 'Content-Type': 'application/json',
}

params = {
    'conflicts': 'proceed',
}

json_data = {
    'query': {
        'match_all': {},
    },
}

response = requests.post('http://localhost/index_name/_delete_by_query', params=params, headers=headers, json=json_data)
            

# by default, elasticsearch search the first 10,000 hits and return 10 of them
headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'query': {
        'range': {
            'date': {
                'gte': '2017-05-28T00:00:00.000-04:00',
                'lt': '2017-12-26T00:00:00.000-05:00',
            },
        },
    },
}

response = requests.get('http://localhost:9200/news_headlines2/_search', headers=headers, json=json_data)
response.json()['hits']['hits']
            

# search all the matched documents and return a specific number of them
headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'query': {
        'range': {
            'date': {
                'gte': '2017-05-28T00:00:00.000-04:00',
                'lt': '2017-12-26T00:00:00.000-05:00',
            },
        },
    },
    "track_total_hits": True,
    "size": 5
}

response = requests.get('http://localhost:9200/news_headlines2/_search', headers=headers, json=json_data)
response.json()['hits']['hits']
            

headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'aggs': {
        'by_category': {
            'terms': {
                'field': 'category',
                'size': 100,
            },
        },
    },
}

response = requests.get('http://localhost:9200/news_headlines/_search', headers=headers, json=json_data)
response.json()['aggregations']['by_category']['buckets']
            

paging less than 10,000 hits

headers = {
    # Already added when you pass json=
    # 'Content-Type': 'application/json',
}

json_data = {
    'query': {
        'range': {
            'date': {
                'gte': '2017-05-28T00:00:00.000-04:00',
                'lt': '2017-12-26T00:00:00.000-05:00',
            },
        },
    },
    "track_total_hits": True,
    "from": 9000,
    "size": 2
}

response = requests.get('http://localhost:9200/news_headlines2/_search', headers=headers, json=json_data)
            

Paging more than 10,000 hits

# step 1, create a pit id
# Point in Time (pit) is a lightweight view of the request results
params = {
    'keep_alive': '1m',
}

response = requests.post('http://localhost:9200/news_headlines2/_pit', params=params)
pit_id = response.json()['id']
            

# step 2, get the first page
import requests

headers = {
    # Already added when you pass json= but not when you pass data=
    # 'Content-Type': 'application/json',
}

json_data = {
    'size': 10,
    'query': {
        'range': {
            'date': {
                'gte': '2017-05-28T00:00:00.000-04:00',
                'lt': '2017-12-26T00:00:00.000-05:00',
            },
        },
    },
    'pit': {
        'id': pit_id,
        'keep_alive': '1m',
    },
    'sort': [
        {
            'date': 'asc',
        },
    ],
}

response = requests.get('http://localhost:9200/_search', headers=headers, json=json_data)
response.json()['hits']['hits']
            

# step 3, get the next page using the last hit’s sort values in the previous page
import requests

headers = {
    # Already added when you pass json= but not when you pass data=
    # 'Content-Type': 'application/json',
}

json_data = {
    'size': 1000,
    'query': {
        'range': {
            'date': {
                'gte': '2017-05-28T00:00:00.000-04:00',
                'lt': '2017-12-26T00:00:00.000-05:00',
            },
        },
    },
    'pit': {
        'id': pit_id,
        'keep_alive': '1m',
    },
    'sort': [
        {
            'date': 'asc',
        },
    ],
    'search_after': [
        1496016000000, 67206
    ],
}

response = requests.get('http://localhost:9200/_search', headers=headers, json=json_data)
response.json()['hits']['hits']
            

Point in time API

Paginate search results

Convert curl commands to Python, JavaScript, PHP, R, Go, Rust, Elixir, Java, MATLAB, Dart, CFML, Ansible URI, Strest or JSON