Basics

The basics

First of all, MongoLite relied heavily on pymongo. As such, all pymongo API is exposed. If you don't find what you want in the Mongokit document, please take a look at pymongo's one. All the pymongo API is exposed via connection, database and collection so Connection, Database and Collection are wrapper around pymongo objects.

Simple Example

Let's create our first object. First of all we need to instanciate a connection:

from mongolite import Connection
connection = Connection()

Now we can create a document:

from mongolite import Document

@connection.register
class BlogPost(Document):
    __database__ = "tutorial"
    __collection__ = "blogposts"
    skeleton = {
        'title':unicode,
        'body':unicode,
        'author':unicode,
        'date_creation':datetime.datetime,
    }
    optional = {
        'tags': [unicode],
        'rank':int,
    }
    default_values = {'rank':0, 'date_creation':datetime.datetime.utcnow}

Let's explain step by step. Our document is register to the connection via the decorator @connection.register.

You can specify the database and the collection related to the document via the attribute __database__ and __collection__. Note that if you want to specify the __database__ you should also specify the __collection__ attribute.

the structure

Now let's see the document structure. By convention, all fields which are required should be declared into the skeleton part. Optional fields are set into optional.

This is handy for two things :

1. first, it is a good documentation to see what fields are always in the database and thoses who might not be present.
2. second, when a document is instanciate, a skeleton is generated based on fields present into the skeleton.

In summary, while fields in skeleton should be present in the document, fields in optional attribut are not generated by default. It aims to be for documentation only...

creating a document

All document are registered against a connection so to access to a document just use the connection :

>>> blogpost = connection.BlogPost()

You can to use another database or collection than thoses set into __database__ and __collection__:

>>> blogpost = connection.other_database.other_collection.BlogPost()

It is also possible to access the documents from the database::

>>> blogpost = connection.other_database.BlogPost() # this will use __collection__ as collection name

The blogpost object is a dict instance which have a skeleton generated:

>>> blogpost
{'body': None, 'title': None, 'date_creation': datetime.datetime(...), 'author': None}

Note that only fields present into skeleton are set. date_creation has a default value specified into default_values.

FAQ

I wan't to register my Document dynamically (or my python version doesn't support decorators)

You can do it like this:

connection.register([Document1, Document2, Document3])

My database/collection is dynamic. How can I do ?

If the database or the collection changes dynamically, then don't set the __database__ or __collection__ attribute. The following example assumes that the database is dynamic and __collection__ is set:

>>> doc = connection[my_dynamic_database].MyDocument()

See next how to query document