Instagram download picture.

download-file.js

var fs = require("fs")
var request = require("request")

module.exports = function(url, dest, cb) {
  var file = fs.createWriteStream(dest)
  var sendReq = request.get(url)

  // verify response code
  sendReq.on("response", function(response) {
    if (response.statusCode !== 200) {
      return cb("Response status was " + response.statusCode)
    }
  })

  // check for request errors
  sendReq.on("error", function(err) {
    fs.unlink(dest)

    if (cb) {
      return cb(err.message)
    }
  })

  sendReq.pipe(file)

  file.on("finish", function() {
    file.close(cb) // close() is async, call cb after close completes.
  })

  file.on("error", function(err) {
    // Handle errors
    fs.unlink(dest)

    // Delete the file async. (But we don't check the result)
    if (cb) {
      return cb(err.message)
    }
  })
}

scraper.js

const fs = require("fs")
const request = require("request")
const mkdirp = require("mkdirp")
const ProgressBar = require("progress")
const { get } = require("lodash")
const download = require("./utils/download-file")

const username = process.argv[2]

if (!username) {
  console.log(
    `
You didn't supply an Instagram username!
Run this command like:

node scrape.js INSTAGRAM_USERNAME
              `
  )
  process.exit()
}

// Convert timestamp to ISO 8601.
const toISO8601 = timestamp => new Date(timestamp * 1000).toJSON()

// Create the progress bar
const bar = new ProgressBar(
  `Downloading instagram posts [:bar] :current/:total :elapsed secs :percent`,
  {
    total: 0,
    width: 30,
  }
)

// Create the images directory
mkdirp.sync(`./data/images`)

let posts = []

// Write json
const saveJSON = _ =>
  fs.writeFileSync(`./data/posts.json`, JSON.stringify(posts, "", 2))

const getPosts = maxId => {
  let url = `https://www.instagram.com/${username}/media`
  if (maxId) url += `?max_id=${maxId}`

  request(url, { encoding: `utf8` }, (err, res, body) => {
    if (err) console.log(`error: ${err}`)
    body = JSON.parse(body)
    // Parse posts
    let lastId
    body.items
      .filter(item => item.type === `image`)
      .map(item => {
        // Parse item to a simple object
        return {
          id: get(item, `id`),
          code: get(item, `code`),
          username: get(item, `user.username`),
          avatar: get(item, `user.profile_picture`),
          time: toISO8601(get(item, `created_time`)),
          type: get(item, `type`),
          likes: get(item, `likes.count`),
          comment: get(item, `comments.count`),
          text: get(item, `caption.text`),
          media: get(item, `images.standard_resolution.url`, ``).replace(
            `/s640x640`,
            ``
          ),
          image: `images/${item.code}.jpg`,
        }
      })
      .forEach(item => {
        if (posts.length >= 100) return

        // Download image locally and update progress bar
        bar.total++
        download(item.media, `./data/images/${item.code}.jpg`, _ => bar.tick())

        // Add item to posts
        posts.push(item)

        // Save lastId for next request
        lastId = item.id
      })

    if (posts.length < 100 && get(body, `more_available`)) getPosts(lastId)
    else saveJSON()
  })
}

getPosts()