mongomoron 0.3

Vanilla wrapper around pymongo

Motivation

This lib is intended to be some surrogate of SqlAlchemy for no-sql database, particularly for mongo.

I checked up MongoAlchemy (seems to be dead), MongoEngine and some smaller libs but seems no one of them fits.

Basically I need builders for query filters and aggregation pipelines. However schema is not needed, so any table can have any fields and any operation can be applied to any field.

For example, instead of

cursor = db.my_table.aggregate([{
    '$match': {
        '$expr': {
            '$and': [{
                '$eq': ['$col', 'col value']
            }, {
                '$in': [
                    '$details.key',
                    ['key 1', 'key 2', 'key 3']
                ]
            }]
        }
    },
}, ...])

I wish to write

p = aggregate(my_table)
    .match(my_table.col == 'col value' and \
        my_table.details.key in ['key 1', 'key 2', 'key 3'])
    ....
cursor = conn.execute(p)

Advantages:

• more clear and readable code;
• can use IDE's autocomplete;
• ability to implement shortcuts for commonly used operations.

Naming

Because names MongoAlchemy and NoSqlAlchemy were not vacant, I chose mongomoron because it was first what came to mind, and because of that lib is for morons like me for whom dealing with mongo's json is too difficult.