View Source  |  Submit Feedback

Class ydn.db.Solver

An Algorithm class represents an iterative solver by guiding cursor increment for each scanning iteration.


Constructor
ydn.db.algo.AbstractSolver(solver, initializer, finalizer)
Create an algorithm giving functions.

Solver function is an optimizer receiving input argument, input, of some function to optimized and its function output value, output. Solver return an array giving advance to make for next iteration step. At the same time solver must meet given constrain condition. A solver function decide a result and push to output streamers.

Adapter function format result from scanning iteration to solver function.

Parameters:
{!function(input: !Array, output: !Array, constrain: !Array): Array} solver
The solver function to guide next cursor iteration.
{function(iterator_keys: !Array, iterator_values: !Array, streamer_keys: !Array, streamer_values: !Array): {'input': !Array, 'output': !Array, 'constrain': !Array} adapter
Optional. The adapter function format streaming values from the scanner for advancer function.
{function(iterators: !Array, streamers: !Array, on_complete: Function): boolean} initializer
Optional. The initializer function is invoked before beginning of scanning iteration. Return true to wait for on_complete callback.
{function(on_complete: Function): boolean} finalizer
Optional. The finalizer function is invoked at the end of scanning iteration. Return true to wait for on_complete callback.

Concrete Classes
ydn.db.algo.NestedLoop(out, limit) version 4
Create an nested loop join algorithm.

The following example illustrate friendship graph query. friendship object store has only one listed indexed field friend_list. It uses people object store key as out-of-line key. people object store has many attributes, among them is location field. The query is to find list of friends who know me and other_guy and located in 'Singapore'.

var q1 = ydn.db.Cursor.where('friendship', 'friend_list', '=', me);
var q2 = ydn.db.Cursor.where('friendship', 'friend_list', '=', other_guy);
var q3 = ydn.db.Cursor.where('people', 'location', '=', Singapore);
var out = new ydn.db.Streamer(db, 'people');
var join_algo = new ydn.db.algo.NestedJoin(out);
var result = db.scan([q3, q1, q2], join_algo);
result.done(function() {
  out.collect(function(friends) {
    console.log(friends); // should get desire list of friends
  }
});
Parameters:
{(!Array|!{push: Function}|!ydn.db.Streamer)=} out
The output streamers.
{number=} limit
Limit number of results.
ydn.db.algo.SortedMerge(out, limit)
Create an sorted merge join algorithm.

Sorted merge join given input iterators by their reference value. Generally input iterators are index key cursor iterator and hence their reference values are primary keys.

Parameters:
{(!Array|!{push: Function}|!ydn.db.Streamer)=} out
The output streamers.
{number=} limit
Limit number of results.
ydn.db.algo.Zigzag(out, limit)
Create an zigzag merge join algorithm.

Sorted merge join given input iterators by their effective key, which compose a constant prefix and sorted post-fix. Generally input iterators are composite index key cursor iterator of filtered prefix index and sorted postfix index.

Parameters:
{(!Array|!{push: Function}|!ydn.db.Streamer)=} out
The output streamers.
{number=} limit
Limit number of results.