-
-
Save arupalan/1e6066b6abb48800d355a25b70f23643 to your computer and use it in GitHub Desktop.
Scraper based on Horseman, written in functional way with Ramda/Folktale.Task
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
/* global $ document */ | |
const Horseman = require('node-horseman'); | |
const cheerio = require('cheerio'); | |
const R = require('ramda'); | |
const phantomjs = require('phantomjs-prebuilt'); | |
const { Task, mapf } = require('../utils/f.js'); | |
// fetch :: string -> Task Error HTML | |
const fetch = (url, commands = {}) => { | |
return new Task((reject, result) => { | |
const horseman = new Horseman({ phantomPath: phantomjs.path }); | |
horseman | |
.viewport(1366, 768) | |
.userAgent('Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.94 Safari/537.36') // eslint-disable-line | |
.open(url) | |
.evaluate((selector) => { | |
if (selector) { | |
$(selector).click(); | |
} | |
}, commands.click) | |
.waitFor((selector) => { | |
if (!selector) { | |
return true; | |
} | |
return $(selector).length ? $(selector).is(':hidden') : true; | |
}, commands.click, true) | |
.evaluate(() => document.documentElement.outerHTML) | |
.catch(error => reject(error)) | |
.then(body => result(body)) | |
.close(); | |
}); | |
}; | |
// window :: HTML -> Task Error $ | |
const window = HTML => cheerio.load(HTML); | |
// dom :: string -> Task Error $ | |
const dom = R.compose(mapf(window), fetch); | |
// sel :: string -> $ -> [DOMNode] | |
const sel = R.curry((selector, $) => $(selector)); | |
// absolutize :: string -> string -> string | |
const absolutize = R.curry((root, href) => { | |
return href.startsWith('http') ? href : `${root}${href}`; | |
}); | |
// toArray :: {} -> [] | |
const toArray = obj => obj.toArray(); | |
// text :: $ -> DOMNode -> string | |
const text = $ => node => $(node).text(); | |
// attr :: string -> $ -> DOMNode -> string | |
const attr = attrubute => $ => node => $(node).attr(attrubute); | |
// selattr -> string -> string -> ($ -> string) | |
const selattr = (selector, attribute) => R.converge(R.call, [ | |
attr(attribute), | |
sel(selector), | |
]); | |
// selarrattr -> string -> string -> ($ -> [string]) | |
const selarrattr = (selector, attribute) => R.converge(R.call, [ | |
R.compose(R.map, attr(attribute)), | |
R.compose(toArray, sel(selector)), | |
]); | |
// seltext -> string -> ($ -> string) | |
const seltext = selector => R.converge(R.call, [ | |
text, | |
sel(selector), | |
]); | |
// selarrtext -> string -> ($ -> [string]) | |
const selarrtext = selector => R.converge(R.call, [ | |
R.compose(R.map, text), | |
R.compose(toArray, sel(selector)), | |
]); | |
module.exports = { | |
fetch, | |
window, | |
dom, | |
selattr, | |
selarrattr, | |
seltext, | |
selarrtext, | |
absolutize, | |
}; |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment