Last active
September 15, 2018 15:14
-
-
Save rergw/2a91baa6f36447fc0d5195e10ee17ea3 to your computer and use it in GitHub Desktop.
General purpose scraper adapted to angel.co
This file contains 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
/* | |
General purpose scraper adapted to angel.co. | |
Does not support pagination. | |
Usage: | |
1. For other pages change `map` and `item`. | |
2. Copy entire code into browser console | |
3. Results are copied to clipboard and can be pasted on a spreadsheet. | |
*/ | |
JSONresults = [] | |
TSVresults = [] | |
map = { | |
name: '.startup-link', | |
URL: ['.startup-link', 'href'], | |
title: '.collapsed-title', | |
compensation: '.collapsed-compensation', | |
tags: '.collapsed-tags', | |
active: '.tag.active', | |
applicants: '.tag.applicants', | |
locations: '.tag.locations', | |
employees: '.tag.employees' | |
} | |
item = '.header-info' | |
TSVresults.push(Object.keys(map).join("\t")) | |
document | |
.querySelectorAll(item) | |
.forEach(function(node){ | |
result = {} | |
for (name in map) { | |
map[name] = Array.isArray(map[name]) ? map[name] : [map[name]] | |
selector = map[name][0] | |
method = map[name][1] || 'innerText' | |
result[name] = node.querySelector(selector)[method] | |
} | |
JSONresults.push(result) | |
TSVresults.push(Object.values(result).join("\t")) | |
}) | |
TSVresults = TSVresults.join("\n") | |
copy(TSVresults) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment