Skip to content


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
a small, expressive orm -- supports postgresql, mysql and sqlite
Python HTML Other

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.



  • a small orm
  • written in python
  • provides a lightweight querying interface over sql
  • uses sql concepts when querying, like joins and where clauses
  • supports sqlite and mysql


# a simple query selecting a user

# get the staff and super users
editors = | Q(is_superuser=True))

# get tweets by editors

# how many active users are there?

# paginate the user table and show me page 3 (users 41-60)'username', 'asc')).paginate(3, 20)

# order users by number of tweets{
    User: ['*'],
    Tweet: [Count('id', 'num_tweets')]
}).group_by('id').join(Tweet).order_by(('num_tweets', 'desc'))

check the documentation for more examples.

specific question? come hang out in the #peewee channel on


peewee began when I was working on a small app in flask and found myself writing lots of queries and wanting a very simple abstraction on top of the sql. I had so much fun working on it that I kept adding features. My goal has always been, though, to keep the implementation incredibly simple. I've made a couple dives into django's orm but have never come away with a deep understanding of its implementation. peewee is small enough that its my hope anyone with an interest in orms will be able to understand the code without too much trouble.

model definitions and schema creation

smells like django:

import peewee

class Blog(peewee.Model):
    title = peewee.CharField()

    def __unicode__(self):
        return self.title

class Entry(peewee.Model):
    title = peewee.CharField(max_length=50)
    content = peewee.TextField()
    pub_date = peewee.DateTimeField()
    blog = peewee.ForeignKeyField(Blog)

    def __unicode__(self):
        return '%s: %s' % (, self.title)

gotta connect:

>>> from peewee import database
>>> database.connect()

create some tables:

>>> Blog.create_table()
>>> Entry.create_table()

foreign keys work like django's

>>> b = Blog(title="Peewee's Big Adventure")
>>> e = Entry(title="Greatest movie ever?", content="YES!", blog=b)
<Blog: Peewee's Big Adventure>
>>> for e in b.entry_set:
...     print e.title
Greatest movie ever?


queries come in 4 flavors (select/update/insert/delete).

there's the notion of a query context which is the model being selected or joined on:'username', 'asc'))

since User is the model being selected, the where clause and the order_by will pertain to attributes on the User model. User is the current query context when the .where() and .order_by() are evaluated.

an example using joins:'pub_date', 'desc')).join(

this will select non-deleted tweets from active users. the first .where() and .order_by() occur when Tweet is the current query context. As soon as the join is evaluated, User becomes the query context and so the following where() pertains to the User model.

now with q objects

for users familiar with django's orm, I've implemented OR queries and complex query nesting using similar notation:
    Q(is_superuser = True) |
    Q(is_staff = True)
    (Q(a='A') | Q(b='B')) &
    (Q(c='C') | Q(d='D'))

# generates something like:
# SELECT * FROM some_obj
# WHERE ((a = "A" OR b = "B") AND (c = "C" OR d = "D"))

using sqlite

import peewee

database = peewee.SqliteDatabase('my.db')

class BaseModel(peewee.Model):
    class Meta:
        database = database

class Blog(BaseModel):
    creator = peewee.CharField()
    name = peewee.CharField()

class Entry(BaseModel):
    creator = peewee.CharField()
    name = peewee.CharField()

using mysql

you can now use MySQL:

import peewee

database = peewee.MySQLDatabase('my_db', user='root')

class BaseModel(peewee.Model):
    class Meta:
        database = database

# ... same as above sqlite example ...
Something went wrong with that request. Please try again.