1. diana clarke
  2. sqlalchemy-2422


sqlalchemy-2422 / lib / sqlalchemy / ext / horizontal_shard.py

# ext/horizontal_shard.py
# Copyright (C) 2005-2012 the SQLAlchemy authors and contributors <see AUTHORS file>
# This module is part of SQLAlchemy and is released under
# the MIT License: http://www.opensource.org/licenses/mit-license.php

"""Horizontal sharding support.

Defines a rudimental 'horizontal sharding' system which allows a Session to
distribute queries and persistence operations across multiple databases.

For a usage example, see the :ref:`examples_sharding` example included in 
the source distribution.


from sqlalchemy import exc as sa_exc
from sqlalchemy import util
from sqlalchemy.orm.session import Session
from sqlalchemy.orm.query import Query

__all__ = ['ShardedSession', 'ShardedQuery']

class ShardedQuery(Query):
    def __init__(self, *args, **kwargs):
        super(ShardedQuery, self).__init__(*args, **kwargs)
        self.id_chooser = self.session.id_chooser
        self.query_chooser = self.session.query_chooser
        self._shard_id = None

    def set_shard(self, shard_id):
        """return a new query, limited to a single shard ID.

        all subsequent operations with the returned query will 
        be against the single shard regardless of other state.

        q = self._clone()
        q._shard_id = shard_id
        return q

    def _execute_and_instances(self, context):
        def iter_for_shard(shard_id):
            context.attributes['shard_id'] = shard_id
            result = self._connection_from_session(
            return self.instances(result, context)

        if self._shard_id is not None:
            return iter_for_shard(self._shard_id)
            partial = []
            for shard_id in self.query_chooser(self):

            # if some kind of in memory 'sorting' 
            # were done, this is where it would happen
            return iter(partial)

    def get(self, ident, **kwargs):
        if self._shard_id is not None:
            return super(ShardedQuery, self).get(ident)
            ident = util.to_list(ident)
            for shard_id in self.id_chooser(self, ident):
                o = self.set_shard(shard_id).get(ident, **kwargs)
                if o is not None:
                    return o
                return None

class ShardedSession(Session):
    def __init__(self, shard_chooser, id_chooser, query_chooser, shards=None, 
                 query_cls=ShardedQuery, **kwargs):
        """Construct a ShardedSession.

        :param shard_chooser: A callable which, passed a Mapper, a mapped instance, and possibly a
          SQL clause, returns a shard ID.  This id may be based off of the
          attributes present within the object, or on some round-robin
          scheme. If the scheme is based on a selection, it should set
          whatever state on the instance to mark it in the future as
          participating in that shard.

        :param id_chooser: A callable, passed a query and a tuple of identity values, which
          should return a list of shard ids where the ID might reside.  The
          databases will be queried in the order of this listing.

        :param query_chooser: For a given Query, returns the list of shard_ids where the query
          should be issued.  Results from all shards returned will be combined
          together into a single listing.

        :param shards: A dictionary of string shard names to :class:`~sqlalchemy.engine.base.Engine`

        super(ShardedSession, self).__init__(query_cls=query_cls, **kwargs)
        self.shard_chooser = shard_chooser
        self.id_chooser = id_chooser
        self.query_chooser = query_chooser
        self.__binds = {}
        self.connection_callable = self.connection
        if shards is not None:
            for k in shards:
                self.bind_shard(k, shards[k])

    def connection(self, mapper=None, instance=None, shard_id=None, **kwargs):
        if shard_id is None:
            shard_id = self.shard_chooser(mapper, instance)

        if self.transaction is not None:
            return self.transaction.connection(mapper, shard_id=shard_id)
            return self.get_bind(mapper, 

    def get_bind(self, mapper, shard_id=None, instance=None, clause=None, **kw):
        if shard_id is None:
            shard_id = self.shard_chooser(mapper, instance, clause=clause)
        return self.__binds[shard_id]

    def bind_shard(self, shard_id, bind):
        self.__binds[shard_id] = bind