Last active
October 6, 2022 13:05
-
-
Save martian17/0bbd3117d7df9ece8a1b8d18ff305d04 to your computer and use it in GitHub Desktop.
scrape contacts
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
/* | |
init commands to run | |
$ npm init | |
$ npm install jsdom | |
$ npm install node-fetch | |
*/ | |
const {JSDOM} = require("jsdom"); | |
const fetch = (()=>{let m = import("node-fetch");return async (...args)=>await (await m).default(...args);})(); | |
let scrapeContacts = async function(url){ | |
let dom = new JSDOM(await (await fetch(url)).text()); | |
let document = dom.window.document; | |
let contacts = document.getElementById("ctlMailList_lstList"); | |
let cards = contacts.querySelectorAll(".CompanyBox"); | |
let infos = []; | |
for(let card of cards){ | |
let lines = [[]]; | |
for(let e of card.childNodes){ | |
if(e.nodeName === "DIV"){ | |
lines[lines.length-1].push(e); | |
lines.push([]); | |
}else if(e.nodeName === "BR"){ | |
lines.push([]); | |
}else{ | |
lines[lines.length-1].push(e); | |
} | |
} | |
lines = lines.filter(l=>l.length>0) | |
.map(l=>l.map(e=>e.textContent.trim()).join(" ").trim()); | |
let info = { | |
general:lines[0], | |
company:lines[1], | |
name:lines[2], | |
title:lines[3] | |
} | |
let rest = lines.slice(4); | |
//extract email/phone/address | |
let address = []; | |
let ctx = 0; | |
for(let i = 0; i < rest.length; i++){ | |
let l = rest[i]; | |
if(ctx === 0){ | |
if(l.match(/^Email\:/) || l.match(/^Phone\:/)){ | |
ctx = 1; | |
i--; | |
continue; | |
} | |
address.push(l); | |
}else{ | |
if(l.match(/^Email\:/)){ | |
info.email = l.split(":")[1].trim(); | |
}else if(l.match(/^Phone\:/)){ | |
info.phone = l.split(":")[1].trim(); | |
}else{ | |
console.log("unexpected line:",l); | |
} | |
} | |
} | |
info.address = address.join("\n"); | |
infos.push(info); | |
} | |
return infos | |
}; | |
let main = async function(){ | |
console.log(await scrapeContacts("https://www.puc.texas.gov/industry/electric/directories/brk/report_brk.aspx?ID=BRSQL01DB1245872600005")); | |
}; | |
main(); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment