This is a pre-production deployment of Warehouse. Changes made here affect the production instance of PyPI (
Help us improve Python packaging - Donate today!

Python interface for groonga

Project Description

What’s this?

Python interface for groonga fulltext search engine.


  • Python 2.6 or 3.x and later
  • groonga


from pypi:

% pip install pyroonga

from source:

% python install


First, Please run groonga by server mode or daemon mode. see following:

# server mode
% groonga -s DB_PATH_NAME

# daemon mode
% groonga -d DB_PATH_NAME

See groonga --help for more options.

Create Table

from pyroonga import tablebase, Column, Groonga

# create the base class for table definition.
Table = tablebase()

# define the table
class Site(Table):
   title = Column()
   name = Column()

class Blog(Table):
   entry = Column()

# create and bind the groonga connection object
grn = Groonga()

# create the all table on groonga's database

Data loading

data = [Site(_key='key1', title='foo', name='hoge'),
        Site(_key='key2', title='bar', name='fuga'),
        Site(_key='key3', title='baz', name='piyo'),
        Site(_key='key4', title='qux', name='xyzzy')]


The example above is load the data to groonga immediately. Also pseudo commit and rollback are supported:

data1 = [Site(_key='key5', title='Constellation', name='Sagittarius'),
         Site(_key='key6', title='Constellation', name='Pisces')]

# first load, but not load to groonga actually
data = Site.load(data1, immediate=False)

data2 = [Site(_key='key7', title='Constellation', name='Aquarius')]
data.load(data2)  # same as previous

# load data to groonga actually

Or reset loaded data:


Note that loaded data reset is only if set immediate=False.

Query and get data as a mapped object

Get the all data from Site table:

data =

And print the data:

for row in data:
    print(row._id, row._key, row.title)

Fulltext search queries:'foo').all(),'bar').all()

The above example is same as following queries:

select --table Site --match_columns 'title' --query "foo"
select --table Site --match_columns 'title OR name' --query "bar"

For more complex queries using pyroonga.odm.GE:

from pyroonga.odm import GE'foo') | GE('bar')).all()

The above example is same as following query:

select --table Site --match_columns 'title' --query "(foo OR bar)"

And also not use match_columns:'foo').all()'foo', name='bar').all()  # "or" search

The above example is same as following queries:

select --table Site --query "(title:@\"foo\")"
select --table Site --query "(title:@\"foo\" OR name:@\"bar\")"

Conditional search query: == 'bar').all()

Conbination for a condition: > 3) & (Site.title == 'baz')).all()

Limit and offset:

Sortby:   # asc  # desc

Select the output columns:

# get the title and name columns,

# get the all columns


Switch to the drilldown query after the call of drilldown() from select() method chain:

data =

Result of drilldown will be stored to the drilldown attribute of the return value from all() method:

for drilldown in data.drilldown:
    print(drilldown._key, drilldown._nsubrecs)

A sortby() method in example above, It is query option of --sortby. For sortby of drilldown, Please call of sortby() method after the call of drilldown() method:

A sortby() method in example above, It is query option of --drilldown_sortby. Of course, As well as limit() , offset() and output_columns() methods.


N.B. The Groonga’s suggest feature is still in draft.

First, Create table if still not created:

from pyroonga import SuggestTable

grn = Groonga()

Second, Data loading:

import time
from pyroonga import event_query

data = [event_query(time=time.time(), sequence=1, item='e'),
        event_query(time=time.time(), sequence=1, item='en'),
        event_query(time=time.time(), sequence=1, item='eng'),
        event_query(time=time.time(), sequence=1, item='engi'),
        event_query(time=time.time(), sequence=1, item='engin'),
        event_query(time=time.time(), sequence=1, item='engine', type='submit')]

Finally, Querying:

from pyroonga import item_query, SuggestType

query = 'en'
result = item_query.suggest(query).types(SuggestType.complete). \
for r in result.complete:
    print("key is '%s', score is %s" % (r._key, r._score))

See also

More information

Still not written.


pyroonga is licensed under the MIT license.


v0.5.2 (2013-09-17)

  • Multibyte support
  • Add ‘filter’ API
  • Some changes

v0.5.1 (2013-08-17)

  • More support the queries of Groonga
  • Add GroongaRecord and use to mapping it instead of Table sub-class
  • Host and port for connect to Groonga can now specify by constructor of pyroonga.Groonga

v0.5 (2013-07-30)

  • Change package name from pyroonga.orm to pyroonga.odm (Note that incompatible with old releases)
  • Add match_columns and query API
  • Change license to MIT
  • Fix issues#2

v0.4 (2012-03-28)

  • Add suggest

v0.3 (2012-02-17)

  • Add load the data to groonga

v0.2 (2012-02-17)

  • Add ORM
  • Add documentation of basic usage

v0.1 (2012-02-05)

  • First release
Release History

Release History

This version
History Node


History Node


History Node


History Node


History Node


History Node


History Node


Download Files

Download Files

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

File Name & Checksum SHA256 Checksum Help Version File Type Upload Date
pyroonga-0.5.2.tar.gz (40.3 kB) Copy SHA256 Checksum SHA256 Source Sep 17, 2013

Supported By

WebFaction WebFaction Technical Writing Elastic Elastic Search Pingdom Pingdom Monitoring Dyn Dyn DNS Sentry Sentry Error Logging CloudAMQP CloudAMQP RabbitMQ Heroku Heroku PaaS Kabu Creative Kabu Creative UX & Design Fastly Fastly CDN DigiCert DigiCert EV Certificate Rackspace Rackspace Cloud Servers DreamHost DreamHost Log Hosting