Use Solr's API from Python with the tedious parts taken care of.
The options out there seem to require learning a whole new Api and object hirearchy. But building query URLs by hand and keeping track of start and rows was tedious and error prone. This library provides fluent interface (JQuery style) that mimics Solr parameter names. It also takes car of pulling data behind the scenes.
Supports python 2 and 3 (most likely 2.6+)
Some day SimpleSolr will be a pip install away. But for know:
$ wget https://raw2.github.com/dvbalen/simple-solr/master/SimpleSolr.py
And place SimpleSolr.py somewhere in your python library path
Create a solr object an iterate over a query:
>>> import SimpleSolr
>>>
>>> local_core0 = SimpleSolr.select('http://localhost:8983','core0')
>>> for row in local_core0.q(*:*):
... print(row)
...
{'_version_': 1456990903946182656,
'address_s': '920 Disc Drive Scotts Valley, CA 95066',
'compName_s': 'Maxtor Corporation',
'id': 'maxtor'}
{'_version_': 1456990903946182657,
'address_s': '105 Challenger Rd. Ridgefield Park, NJ 07660-0511',
'compName_s': 'Samsung Electronics Co. Ltd.',
'id': 'samsung'}
{'_version_': 1456990903947231232,
'address_s': '381 Brea Canyon Road Walnut, CA 91789-0708',
'compName_s': 'ViewSonic Corp',
'id': 'viewsonic'}
>>>Get the Solr query:
>>> str(local_core0.q('*:*').start(9))
'http://localhost:8983/solr/core0/select?q=%2A%3A%2A&start=9&wt=json'
>>> Solr queries are just iterators:
>>> ids_q = local_core.q('*:*').fl('id')
>>> ids = [ r['id'] for id in ids_q ]
>>> import pprint
>>> pprint.pprint(ids)
['adata',
'apple',
'asus',
'ati',
'belkin',
'canon',
'corsair',
'dell',
'maxtor',
'samsung',
'viewsonic']
>>>Leave an issue or contact me @dvbalen (twitter)
- Package
- Test suite
- Facet query support
- Less common parameters (dismax,edismax,etc)
- Join
- examples