hyperlog-reduce v0.0.1

hyperlog-reduce

Implement an async reduce function over a hyperlog.

This was inspired by @dominictarr's flumedb and clean approach to materialized views. This module implements something akin to a flumeview-reduce over a hyperlog.

Usage

Create a sum reduce function over a hyperlog of numbers:

var reduce = require('hyperlog-reduce')
var memdb = require('memdb')
var hyperlog = require('hyperlog')

var log = hyperlog(memdb(), {valueEncoding: 'json'})

var sum = reduce({
  db: memdb(),
  log: log,

  // create the materialized view default value
  init: function () {
    return 0
  },

  // runs on each node in the hyperlog
  reduce: function (node, sum, next) {
    sum += node.value.amount
    next(null, sum)
  }
})

log.append({amount: 3}, function () {
  log.append({amount: 6}, function () {
    log.append({amount: 1}, function () {
    })
  })
})

// blocks until the index is caught up
sum.get(function (err, value) {
  console.log('sum is', value)
})

This will output

sum is 10

API

var reduce = require('hyperlog-reduce')

var reducer = reduce(opts)

Create an async reducer over a hyperlog. opts is required and must have the following key/values:

• db: a levelup instance that the reducer's index will be stored in
• log: a hyperlog instance
• init: a function that returns the initial value of the reducer
• reduce: a function of the form function (node, accum, next), where node is a hyperlog node, accum is the current value of the reduce function, and next is a function to be called when the reduce function has completed. It expects arguments of the form next(err, newAccumValue).

reducer.get(done)

Asynchronously retrieves the current value of the reduce function. If nodes are still being processed, done isn't called until the index catches up.

done is a callback of the form function (err, value) { ... }.

Install

With npm installed, run

$ npm install hyperlog-reduce

License

ISC