Thin Python connector for Elasticsearch.
Project description
Thin Python client for Elasticsearch
This is a little wrapper over the requests module that helps you do your thing with Elasticsearch. But not too much. For each request, you need to specify the following, where they are applicable:
desired request method
index, type, ID
“suffix” (which might be the _search, _bulk, etc at the end of the URL)
a dictionary or a string with the data you want to submit
Why would I want that?
Two reasons:
flexibility - the point of this client is to keep up with the many features constantly added to Elasticsearch. You can customize your index operations and queries as much as you need
documentation - once you get the hang of the few options particular to this client, you only have to look at the documentation on Elasticsearch itself to access its features
If this is not what you want, I would suggest to look at pyes or pyelasticsearch
Features and usage
There are a few files in there at the moment:
slimes.py - the module itself
tests_slimes.py - unit tests
Here are the main features:
Converts your dict to JSON (you can disable that for stuff like bulks)
#!/usr/bin/env python import slimes my_requester = slimes.Requester(["localhost:9200"]) #define the document testdoc = {"foo": 2345, "bar": True} #post the document my_requester.request(method="post", myindex="testindex", mytype="testtype", mydata=testdoc)
Can get a list of nodes, for redundancy
#!/usr/bin/env python import slimes my_requester = slimes.Requester(["localhost:9200","my_other_server:9200"])
Loads the JSON results for you
If something goes wrong (eg: non 2XX status code), you’ll get an exception. Otherwise, you’ll get a dictionary with the results from ES:
#!/usr/bin/env python import slimes my_requester = slimes.Requester(["localhost:9200"]) myquery = {\ "query": {\ "term": {\ "age": 28\ }\ }\ } results = my_requester.request(method="post", myindex="testindex", mytype="testtype", mysuffix="_search", mydata=myquery) print "Number of hits is %d" % result['hits']['total']
Can receive parameters at will
So you can index with the “create” parameter, like this:
#!/usr/bin/env python import slimes my_requester = slimes.Requester(["localhost:9200"]) testdoc={"foo": "bar"} my_requester.request(method="put", myindex="testindex", mytype="testtype", myID=1, mysuffix="?op_type=create", mydata=testdoc)
Or, you can put the parameter(s) as a dictionary:
#!/usr/bin/env python import slimes my_requester = slimes.Requester(["localhost:9200"]) testdoc={"foo": "bar"} my_requester.request(method="put", myindex="testindex", mytype="testtype", myID=1, myparams={"op_type":"create"}, mydata=testdoc)
And you can also specify a timeout:
#!/usr/bin/env python import slimes my_requester = slimes.Requester(["localhost:9200"]) #give it 10 seconds to complete, otherwise give Exception my_requester.request(mytimeout=10)
TODOes
Enhancements will be about the transport. Stuff like: - use thrift - get all cluster nodes
Stay tuned :)
Bugs
Please feel free to report them on github
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.