Facebook friends scraper

fb-friends-scraper.js

// Works as of 03/08/2017 - selectors may change on Facebook's end

// Useful for grabbing a list of friends and associated lists of their friends from a single fb user's "friends" page.
// I would have used FB's Graph API, but it doesn't provide any means of getting the same data.
// Must have a plugin in the browser to ignore X-Frame-Options header in order for pages to work within iframes.
// I'm using (https://chrome.google.com/webstore/detail/ignore-x-frame-headers/gleekbfjekiniecknbkamfmkohkpodhe) for Chrome.
// Saves output to json files which can later be assembled into a big json file or entered directly into a database.

// To run the script, get the aforementioned plugin, go to a fb user's page, click friends, then paste this in dev tools.
// There's a runaway memory problem somewhere, but rather than diagnose and fix it, I just restart the script when it crashes. 
// It will crash periodically. Just delete the friends that have been completed and start again.
// When the JSON string of friends pops up initially (line 113), copy it and paste into sublime text.
// If it crashes again, note the last friend that completed and delete those friends that have been done already. Load the page again 
// and enter `friends = {...}` from sublime text. Then comment out lines 108-113 and paste this script in again.
// Eventually your downloads folder will fill up with a list of json files that you can later concatenate.

(function(e) {
    e.save = function(t, n) {
        if (!t) {
            e.error("Console.save: No data");
            return
        }
        if (!n) n = "console.json";
        if (typeof t === "object") { t = JSON.stringify(t) }
        var r = new Blob([t], { type: "text/json" }),
            i = document.createEvent("MouseEvents"),
            s = document.createElement("a");
        s.download = n;
        s.href = window.URL.createObjectURL(r);
        s.dataset.downloadurl = ["text/json", s.download, s.href].join(":");
        i.initMouseEvent("click", true, false, window, 0, 0, 0, 0, 0, false, false, false, false, 0, null);
        s.dispatchEvent(i)
    }
})(console)

window.keepGoing = function(){
    var stopgap = iframe.contentDocument.createElement('div');
    stopgap.classList.add("mbm");
    stopgap.classList.add("_5vf");
    stopgap.classList.add("sectionHeader");
    stopgap.classList.add("_4khu");
    iframe.contentWindow.document.body.appendChild(stopgap);
}

window.friendGen = function*(f) {
    keys = Object.keys(f);
    for (var i = 0; i < keys.length; i++) {
        key = keys[i];
        yield [key, f[key]];
    }
};

window.getFriendsOfFriend = function(fItem) {
    var url = fItem[0],
        name = fItem[1];
    console.log("Processing " + name);
    window.iframe = document.createElement('iframe');
    iframe.style.width = "100%";
    iframe.style.height = "1080px";
    iframe.src = url + "&sk=friends";
    iframe.id = "friends-iframe";
    document.body.appendChild(iframe);
    iframe = document.getElementById("friends-iframe")
    window.scrollTo(0, document.body.scrollHeight);
    iframe.onload = function(event) {
        var f;
        console.log("iframe loaded");
        window.linkCount = 0
        iframe.contentWindow.onscroll = function() {
            iframe.contentWindow.setInterval(function() {
                iframe.contentWindow.scrollTo(0, iframe.contentDocument.body.scrollHeight);
                iframe.contentWindow.setTimeout(function() {
                    if (iframe.contentDocument.querySelectorAll(".fsl.fwb.fcb a").length === linkCount) {
                        if (linkCount <= 20) {
                            console.log("Only " + linkCount + " friends found for " + name);
                            keepGoing();
                        }
                    }
                }, 1000);
            }, 800);
            if (iframe.contentDocument.querySelectorAll(".mbm._5vf.sectionHeader._4khu").length == 0) {
                console.log("iframe scrolling");
                window.linkCount = iframe.contentDocument.querySelectorAll(".fsl.fwb.fcb a").length;
            } else {
                iframe.contentWindow.friends = {};
                l = iframe.contentDocument.querySelectorAll(".fsl.fwb.fcb a");
                for (var i = 0; i < l.length; i++) {
                    iframe.contentWindow.friends[l[i].href] = l[i].innerText;
                }
                console.log("Saving " + name + "'s friends");
                f = {};
                f[url] = {};
                f[url][name] = iframe.contentWindow.friends;
                console.save(f, name + ".json");
                document.body.removeChild(iframe);
                getFriendsOfFriend(fgen.next().value);
            }
        }
        iframe.contentWindow.setTimeout(function() { iframe.contentWindow.scrollTo(0, iframe.contentDocument.body.scrollHeight) }, 1000);
    };
};

window.onscroll = function() {
    if (document.querySelectorAll(".mbm._5vf.sectionHeader._4khu").length == 0) {
        console.log("scrolling");
        setTimeout(function() { window.scrollTo(0, document.body.scrollHeight) }, 800);
    } else {
        window.friends = {};
        var l = document.querySelectorAll(".fsl.fwb.fcb a");
        for (var i = 0; i < l.length; i++) {
            friends[l[i].href] = l[i].innerText;
        }
        console.log(JSON.stringify(friends));
        window.fgen = friendGen(friends);
        window.onscroll = function(){};
        getFriendsOfFriend(fgen.next().value);
    }
}

window.scrollTo(0, document.body.scrollHeight);