Setup¶

$ pip install txes2

$ git clone git@github.com:lextoumbourou/txes2.git
$ cd txes2
$ python setup.py install

./bin/unit

vagrant up --provision

./bin/integration

Usage Example¶

from pprint import pprint

from twisted.internet import reactor
from twisted.internet.defer import inlineCallbacks

from txes2 import ElasticSearch


@inlineCallbacks
def example():
    index = 'main'
    doc_type = 'person'

    ########################
    # Setup connection
    ########################
    es = ElasticSearch('127.0.0.1:9200')

    ########################
    # Index a document
    ########################
    document = {'name': 'Travis Bickle', 'tagline': 'You talking to me?'}
    yield es.index(document, doc_type=doc_type, index=index, id=123)

    ########################
    # Retrieve a document
    ########################
    result = yield es.get(index=index, doc_type=doc_type, id=123)
    pprint(result)

    ########################
    # Refresh an index
    ########################
    yield es.refresh(index)

    ########################
    # Perform a query
    ########################
    query = {'query': {'match': {'name': 'Travis'}}}
    results = yield es.search(query, doc_type=doc_type, index=index)
    pprint(results)


if __name__ == '__main__':
    df = example()
    df.addErrback(lambda err: err.printTraceback())
    df.addCallback(lambda _: reactor.stop())
    reactor.run()

API¶

class txes2.ElasticSearch(servers='127.0.0.1:9200', timeout=30, bulk_size=400, discover=True, retry_time=10, discovery_interval=300, default_indexes=None, autorefresh=False, *args, **kwargs)¶

A PyES-like ElasticSearch client.

Parameters:

servers – either a single ES server URL or list of servers. If you don’t provide a scheme (eg https://) then the request will use HTTP by default. timeout (int) – connection timeout in seconds. bulk_size (int) – how much bulk data to accumulate before indexing (when indexing in bulk). retry_time (int) – frequency in seconds for retrying broken ES nodes. discover (bool) – if True, will autodiscover ES nodes at connection time. discovery_interval (bool) – time in seconds between node discovery after initial discovery, set to False to skip. default_indexes (list) – list of indexes to use by default when querying ES. autorefresh (bool) – should we perform index autorefresh. persistent (bool) – use persistent connection. http_auth (tuple) – optional http auth tuple in the form (‘username’, ‘password’). pool (HTTPConnectionPool) – optionally pass in HTTPConnectionPool instance to use for connection pool.

add_alias(alias, indices)¶

Add an alias to point to a set of indices.

analyze(text, index, analyzer=None)¶

Perform analysis on textual input.

change_aliases(*commands)¶

Change the aliases stored.

A command is a tuple of ([“add”|”remove”], index, alias).

You may specify multiple commands as additional arguments.

close_index(index)¶

Close an index.

cluster_health(level='cluster', wait_for_status=None, wait_for_relocating_shards=None, wait_for_nodes=None, timeout=30)¶

Check the current cluster health.

cluster_nodes(nodes=None)¶

The cluster nodes info API.

cluster_state(metrics=None, indices=None, **kwargs)¶

Retrieve the cluster state.

Parameters:	metric – a list of metrics for filtering results (see ES docs). indices – a list of indicies for filtering results (see ES docs).

collect_info()¶

Collect info about the connection and fill the info dictionary.

count(query, indexes=None, doc_types=None, **params)¶

Execute a query against one or more indices & get the hit count.

create_index(index, settings=None)¶

Create an index with the optional settings dict.

create_index_if_missing(index, settings=None)¶

Create an index with the optional settings dict.

Doesn’t fail when index already exists.

create_river(river, river_name=None)¶

Create a river.

delete(index, doc_type, id, bulk=False, **query_params)¶

Delete a document based on its id.

delete_alias(alias, indices)¶

Delete an alias.

delete_by_query(indexes, doc_types, query, **query_params)¶

Delete documents from one or more indexes/types from query.

delete_index(index)¶

Deletes an index.

delete_index_if_exists(index)¶

Deletes an index if it exists.

delete_mapping(index, doc_type, **query_params)¶

Delete a document type from a specific index.

delete_river(river, river_name=None)¶

Delete a river.

flush_bulk(forced=False)¶

Wait to process all pending operations.

force_bulk()¶

Force executing of all bulk data.

get(index, doc_type, id, fields=None, routing=None, **query_params)¶

Get a typed document from an index based on its id.

get_alias(alias)¶

Return a list of indices pointed to by a given alias.

Raises IndexMissionException if the alias does not exist.

get_indices(include_aliases=False)¶

Retrieve a dict where each key is the index name and each value is another dict containing the following properties:

• num_docs: Number of documents in the index or alias.

• alias_for: Only present for an alias: holds a list of indices

  which this is an alias for. Requires the include_aliases param to be True.

Example:

{u'website': {'num_docs': 10837},

u’website_alias’: {

‘alias_for’: [u’website’], ‘num_docs’: 10837}}

get_mapping(doc_type=None, indexes=None)¶

Get the mapping definition

index(doc, index, doc_type, id=None, parent=None, force_insert=None, bulk=False, version=None, **query_params)¶

Index a dict into an index.

mget(ids, index=None, doc_type=None, **query_params)¶

Get multiples documents based on id.

ids can be:

list of tuples: (index, type, id) list of ids: index and doc_type are required

more_like_this(index, doc_type, id, **query_params)¶

Execute a “more like this” query against one or more fields.

open_index(index)¶

Open an index.

optimize(indexes=None, wait_for_merge=False, max_num_segments=None, only_expunge_deletes=False, refresh=True, flush=True)¶

Optimize one or more indices.

partial_update(index, doc_type, id, doc=None, script=None, script_file=None, params=None, upsert=None, **query_params)¶

Partially update a document with a script.

If script is passed in, script_file is ignored. For script_file to work, ES >= 1.4.5 is required as per: https://github.com/elastic/elasticsearch/issues/10007

put_mapping(doc_type, mapping, indexes=None)¶

Register mapping definition for a specific type.

scan(query, indexes=None, doc_type=None, scroll_timeout='10m', **params)¶

Start a scan eventually returning a Scroller.

search(query, indexes=None, doc_type=None, **params)¶

Execute a search against one or more indices.

servers¶

Return a list of servers available for connections.

set_alias(alias, indices)¶

Set and alias (possibly removing what it already points to).

status(indexes=None)¶

Retrieve the status of one or more indices.

update_settings(index, settings)¶

Update settings of an index.

Contributing¶