minimongo 0.2.5

Major features

• No schema declaration

  Minimongo has no schema declaration. You do not pre-declare the names and types of your fields. Minimongo takes a minimalist and flexible approach to schemas. You can set any value on any Minimongo-derived object.

• Declared database & collection names.

  Miminogo allows you to progromaticaly declare your database and collection names.

  You’ll only need to put your collection name in one place, and Minimongo provides classmethods for accessing the DB and Collection names programatically. This means you can easily switch DB or Collection names without having to change all your code.

• Automatic MongoDB Connection mangagement and connection pooling.

  Minimongo automatically connects to your database for you, and has its own simple Connection pool. Connections are persistent and last the lifetime of your application.

• Friendly support for DBRef types.

  Minimongo can easily generate fully-specified (DB name included) DBRef objects, and store these into fields. Fetching via DBRef is simplified as well.

• Uses Pymongo_’s native query syntax.

  Query methods are passed directly to Pymongo’s API. Minimongo provides very few custom methods, and delegates nearly all operations directly to Pymongo. This means as features are added to Pymongo, they will be automatically accessible via Minimongo.

• Easy Index creation & management.

  Indices for a given collection can be specified at declaration time. This results in automatic calls to pymongo’s ensure_index() function at the time your program starts up. Therefore, all the proper indices for your collections are always in place.

• Easily extensible.

  Minimongo-derived objects can be easily extended to add new functionality.

• Easy object creation via initalization from Dicts.

Example

Here’s a very brief example of creating an object, querying for it, modifying a field, and then saving it back again.:

from minimongo import Model, MongoCollection

class MyCollection(Model):
    # Here, we specify the database and collection names.
    # A connection to your DB is automatically created.
    mongo = MongoCollection(database='test', collection='minimongo.example')

    # Now, we programatically declare what indices we want.
    # The arguments to the Index constructor are identical to the args to
    # pymongo's ensure_index function.
    indices = (Index('x'),)

if __name__ == '__main__':
    # Create & save an object, and return a local inmemory copy of it:
    obj = MyCollection({'x': 1, 'y': 2}).save()

    # Find that object again, loading it into memory:
    res = MyCollection.collection.find_one({'x': 1})

    # Change a field value, and save it back to the DB.
    res.other = 'some data'
    res.save()

TODOs & Upcoming features:

• Per-object configuration directives. Read-only, Rigid (no schema change allowed after read), Type-Rigid (allow changing values, but not changing types), etc.

• Support for automatic DBRef field dereferencing via wrapper types.

• Better support for SON and field ordering. Right now, most things are Python dict, which means that ordering is not defined.

• Delta modification tracking so that when you call save(), it doesn’t send the whole document back to the server to modify one field.

• Support for the mongodb atomic operations like $inc, $push, $pull, etc via native Python primitives.

• Better support for nested Model objects. (Right now, nested data must be of a native Python type, not of another Model).

Feedback welcome!

Please email github@slacy.com with comments, suggestions, or comment via http://github.com/slacy/minimongo

Changes