Home

Awesome

node-warc

Parse Web Archive (WARC) files or create WARC files using

Run npm install node-warc or yarn add node-warc to ge started

npm Package

Documentation

Full documentation available at n0tan3rd.github.io/node-warc

Parsing

Using async iteration

Requires node 10 or greater

const fs = require('fs')
const zlib = require('zlib')
// recordIterator only exported if async iteration on readable streams is available
const { recordIterator } = require('node-warc')

async function iterateRecords (warcStream) {
  for await (const record of recordIterator(warcStream)) {
    console.log(record)
  }
}

iterateRecords(
  fs.createReadStream('<path-to-gzipd-warcfile>').pipe(zlib.createGunzip())
).then(() => {
  console.log('done')
})

Or using one of the parsers

for await (const record of new AutoWARCParser('<path-to-warcfile>')) {
    console.log(record)
}

Using Stream Transform

const fs = require('fs')
const { WARCStreamTransform } = require('node-warc')

fs
  .createReadStream('<path-to-warcfile>')
  .pipe(new WARCStreamTransform())
  .on('data', record => {
    console.log(record)
  })

Both .warc and .warc.gz

const { AutoWARCParser } = require('node-warc')

const parser = new AutoWARCParser('<path-to-warcfile>')
parser.on('record', record => { console.log(record) })
parser.on('done', () => { console.log('finished') })
parser.on('error', error => { console.error(error) })
parser.start()

Only gzip'd warc files

const { WARCGzParser } = require('node-warc')

const parser = new WARCGzParser('<path-to-warcfile>')
parser.on('record', record => { console.log(record) })
parser.on('done', () => { console.log('finished') })
parser.on('error', error => { console.error(error) })
parser.start()

Only non gzip'd warc files

const { WARCGzParser } = require('node-warc')

const parser = new WARCParser('<path-to-gzipd-warcfile>')
parser.on('record', record => { console.log(record) })
parser.on('done', () => { console.log('finished') })
parser.on('error', error => { console.error(error) })
parser.start()

WARC Creation

Environment

Examples

Using chrome-remote-interface

const CRI = require('chrome-remote-interface')
const { RemoteChromeWARCWriter, RemoteChromeCapturer } = require('node-warc')

;(async () => {
  const client = await CRI()
  await Promise.all([
    client.Page.enable(),
    client.Network.enable(),
  ])
  const cap = new RemoteChromeCapturer(client.Network)
  cap.startCapturing()
  await client.Page.navigate({ url: 'http://example.com' });
  // actual code should wait for a better stopping condition, eg. network idle
  await client.Page.loadEventFired()
  const warcGen = new RemoteChromeWARCWriter()
  await warcGen.generateWARC(cap, client.Network, {
    warcOpts: {
      warcPath: 'myWARC.warc'
    },
    winfo: {
      description: 'I created a warc!',
      isPartOf: 'My awesome pywb collection'
    }
  })
  await client.close()
})()

Using chrome-remote-interface-extra

const { CRIExtra, Events, Page } = require('chrome-remote-interface-extra')
const { CRIExtraWARCGenerator, CRIExtraCapturer } = require('node-warc')

;(async () => {
  let client
  try {
    // connect to endpoint
    client = await CRIExtra({ host: 'localhost', port: 9222 })
    const page = await Page.create(client)
    const cap = new CRIExtraCapturer(page, Events.Page.Request)
    cap.startCapturing()
    await page.goto('https://example.com', { waitUntil: 'networkIdle' })
    const warcGen = new CRIExtraWARCGenerator()
    await warcGen.generateWARC(cap, {
      warcOpts: {
        warcPath: 'myWARC.warc'
      },
      winfo: {
        description: 'I created a warc!',
        isPartOf: 'My awesome pywb collection'
      }
    })
  } catch (err) {
    console.error(err)
  } finally {
    if (client) {
      await client.close()
    }
  }
})()

Using Puppeteer

const puppeteer = require('puppeteer')
const { Events } = require('puppeteer')
const { PuppeteerWARCGenerator, PuppeteerCapturer } = require('node-warc')

;(async () => {
  const browser = await puppeteer.launch()
  const page = await browser.newPage()
  const cap = new PuppeteerCapturer(page, Events.Page.Request)
  cap.startCapturing()
  await page.goto('http://example.com', { waitUntil: 'networkidle0' })
  const warcGen = new PuppeteerWARCGenerator()
  await warcGen.generateWARC(cap, {
    warcOpts: {
      warcPath: 'myWARC.warc'
    },
    winfo: {
      description: 'I created a warc!',
      isPartOf: 'My awesome pywb collection'
    }
  })
  await page.close()
  await browser.close()
})()

Note

The generateWARC method used in the preceding examples is helper function for making the WARC generation process simple. See its implementation for a full example of WARC generation using node-warc

Or see one of the crawler implementations provided by Squidwarc.