flumeview-reduce

A flumeview into a reduce function. Stream append-only log data in the order the log is written into a reduce function to calculate a state.

Example

var FlumeLog = require('flumelog-offset')
var codec = require('flumecodec')
var Flume = require('flumedb')
var Reduce = require('flumeview-reduce')

//statistics exports a reduce function that calculates
//mean, stdev, etc!
var statistics = require('statistics')

//initialize a flumelog with a codec.
//this example uses flumelog-offset, but any flumelog is valid.
var log = FlumeLog(file, 1024*16, codec.json) //use any flume log

//attach the reduce function.
var db = Flume(log)
  .use('stats', Reduce(
    1,                                    // version
    statistics,                           // reducer
    function (data) { return data.value } // map
  ))

db.append({value: 1}, function (err) {

  db.stats.get(function (err, stats) {
    console.log(stats) // => {mean: 1, stdev: 0, count: 1, sum: 1, ...}
  })
})

FlumeViewReduce(version, reduce, map?, codec?, initialState?) => FlumeView

construct a flumeview from this reduce function. version should be a number, and must be provided. If you make a breaking change to either reduce or map then increment version and the view will be rebuilt.

map is optional. If map is applied, then each item in the log is passed to map and then if the returned value is not null, it is passed to reduce.

var _value = map(value)
if(_value != null)
  state = reduce(state, _value, seq)

using a map function is useful, because it enables efficiently streaming the realtime changes in the state to a remote client.

then, pass the flumeview to db.use(name, flumeview) and you'll have access to the flumeview methods on db[name]...

codec (optional) - specify the codec to use in the event your log uses the filesystem. initialState (optional) - declare an initial state for your reducer. This will be ignored if a persisted state is found.

db[name].get(cb)

get the current state of the reduce. This will wait until the view is up to date, if necessary.

db[name].stream({live: boolean}) => PullSource

Creates a pull-stream whose:

  • first value is the current state of the view,
  • following values are not the view state, but the new values (they're had your map applied, but the reducer hasn't been applied yet).

This is a light-weight for a remote client to keep up to date with the view - get a snapshot, and then update it themselves. This way we don't need to send a massive view every time there's a new log entry.

var db = Flume(log)
  .use('stats', Reduce(2, myReducer, myMap))

var viewState // this is our view we're calculating remotely
pull(
  db.stats.stream({live:true}),
  pull.drain(function(value) {
    if (!view) viewState = value      // store the current snapshot (the first value)
    else myReducer(viewState, value)  // update the snapshot use reducer + mapped values

    console.log(value)                // do something with 
  }
)

db.append(
  // ... some code that adds new code to the log, triggering stream.
)

Stores

flumeview-reduce currently includes several store implementations, this is how the actual data is persisted. the current implementations are

  • 'flumeview-reduce/store/fs' - store in a file.
  • 'flumeview-reduce/store/local-storage' - localStorage, in a browser
  • 'flumeview-reduce/store/remote' - a meta store that keeps a local copy of a remote view.

to set a store, you must set up flumeview-reduce via the lower level dependency injection api.

var createReduce = require('flumeview-reduce/inject')

var Reduce = createReduce(Store)

//then use Reduce normally

var view = db.use('view', Reduce(version, reduce, map)) //etc

//since remote is most interesting

var Remote = require('flumeview-reduce/store/remote')
function get (opts, cb) {
  //call the get method on the remote copy of the flumeview
  view.get(opts, cb)
}
var RemoteReduce = createReduce(Remote(get, Store, codec))

var remoteView = _db.use('view', Reduce(version, reduce, map)) //etc
//make sure you pass the exact same reduce and map functions to the remote view!

License

MIT

results matching ""

    No results matching ""