ymolliks · March 24, 2022 16:12
diff --git a/medium-to-scribe.js b/medium-to-scribe.js
 // ==UserScript==
 // @name        Medium to Scribe
 // @description Redirects Medium.com articles to scribe.rip
 // @match       *://*/*
 // @run-at       document-start
 // @version     1.1
 // @updateURL   https://gist.githubusercontent.com/samhenrigold/4a082dde823bc3cb62e43a2fc2b12b8e/raw/893a9480a1922e87a5b1cd16c66e405db53e43b6/medium-to-scribe.js
 // @downloadURL   https://gist.githubusercontent.com/samhenrigold/4a082dde823bc3cb62e43a2fc2b12b8e/raw/893a9480a1922e87a5b1cd16c66e405db53e43b6/medium-to-scribe.js
 // ==/UserScript==

 const badHosts = [
    "android.jlelse.eu",
    "betterhumans.coach.me",
    "survivingtomorrow.org",
    "blog.angularindepth.com",
    "blog.bitsrc.io",
    "blog.devartis.com",
    "blog.getambassador.io",
    "blog.hipolabs.com",
    "blog.maddevs.io",
    "blog.prototypr.io",
    "blog.usejournal.com",
    "calia.me",
    "www.cantorsparadise.com",
    "codeburst.io",
    "engineering.opsgenie.com",
    "entrepreneurshandbook.co",
    "hackernoon.com",
    "instagram-engineering.com",
    "itnext.io",
    "levelup.gitconnected.com",
    "medium.com",
    "medium.freecodecamp.org",
    "medium.mybridge.co",
    "proandroiddev.com",
    "productcoalition.com",
    "psiloveyou.xyz",
    "robinhood.engineering",
    "theascent.pub",
    "thebolditalic.com",
    "towardsdatascience.com",
    "ux.shopify.co",
    "uxdesign.cc",
    "uxplanet.org",
    "blog.gds-gov.tech"
 ];

 let subdomain = "";


 // Easy, general match with the list of known Medium hosts. Most of the time, this will be the case.
 if (badHosts.includes(window.location.host)) {
    window.location.href = `https://scribe.rip${window.location.pathname}`;
 } else if (window.location.hostname.indexOf("medium.com") > -1) {
    /*
        If medium.com is somewhere hostname but the first if statement didn't catch it, we assume the hostname is <example>.medium.com.
        Any article URL matching that subdomain pattern can be rewritten as medium.com/<example>/article-title.
    */
    let hostnameClone = window.location.hostname;
    // We need to check if the subdomain contains "www" and remove it before using the "pure" subdomain.
    if (hostnameClone.indexOf('www.') > -1) { hostnameClone.replace('www.', ''); }
    window.location.href = `https://scribe.rip/${hostnameClone.split('.')[0]}${window.location.pathname}`;
 }

 // const debuggingArticles = [
 //     "http://medium.com/finding-insight/cancel-culture-just-cancelled-my-favorite-furniture-b652b2afb088",
 //     "http://medium.com/good-service/introducing-dynamic-route-maps-a8e56dd8a33",
 //     "https://medium.com/hackernoon/removing-that-ugly-focus-ring-and-keeping-it-too-6c8727fefcd2",
 //     "https://michaellong.medium.com/swiftui-vs-jetpack-compose-why-android-wins-hands-down-b5f849b730db"]

 /* 
 ## Digging Deeper

 So I ran into an article on https://blog.gds-gov.tech/ which is hosted by Medium. None of the logic up until this point has checked if a site is Medium but it's not in the URL. So I need to figure out how to check if a particular page is Medium by hunting through the <head>. I'm not sure how to do that without tanking performance.

 I'm trying to find a meta tag or something that is immutable. In other words, a tag where the domain owner (the GDS in this case) cannot change it. I've found this:
    <meta data-rh="true" property="og:site_name" content="Medium">
 but I need to find more examples to see what other sites use this.

 Another consideration: doing the check client-side, if the site is Medium but was not in the big array, asking the user if they'd like to share that addition to the database.

 */
	// ==UserScript==
	// @name Medium to Scribe
	// @description Redirects Medium.com articles to scribe.rip
	// @match :///*
	// @run-at document-start
	// @version 1.1
	// @updateURL https://gist.githubusercontent.com/samhenrigold/4a082dde823bc3cb62e43a2fc2b12b8e/raw/893a9480a1922e87a5b1cd16c66e405db53e43b6/medium-to-scribe.js
	// @downloadURL https://gist.githubusercontent.com/samhenrigold/4a082dde823bc3cb62e43a2fc2b12b8e/raw/893a9480a1922e87a5b1cd16c66e405db53e43b6/medium-to-scribe.js
	// ==/UserScript==

	const badHosts = [
	"android.jlelse.eu",
	"betterhumans.coach.me",
	"survivingtomorrow.org",
	"blog.angularindepth.com",
	"blog.bitsrc.io",
	"blog.devartis.com",
	"blog.getambassador.io",
	"blog.hipolabs.com",
	"blog.maddevs.io",
	"blog.prototypr.io",
	"blog.usejournal.com",
	"calia.me",
	"www.cantorsparadise.com",
	"codeburst.io",
	"engineering.opsgenie.com",
	"entrepreneurshandbook.co",
	"hackernoon.com",
	"instagram-engineering.com",
	"itnext.io",
	"levelup.gitconnected.com",
	"medium.com",
	"medium.freecodecamp.org",
	"medium.mybridge.co",
	"proandroiddev.com",
	"productcoalition.com",
	"psiloveyou.xyz",
	"robinhood.engineering",
	"theascent.pub",
	"thebolditalic.com",
	"towardsdatascience.com",
	"ux.shopify.co",
	"uxdesign.cc",
	"uxplanet.org",
	"blog.gds-gov.tech"
	];

	let subdomain = "";


	// Easy, general match with the list of known Medium hosts. Most of the time, this will be the case.
	if (badHosts.includes(window.location.host)) {
	window.location.href = `https://scribe.rip${window.location.pathname}`;
	} else if (window.location.hostname.indexOf("medium.com") > -1) {
	/*
	If medium.com is somewhere hostname but the first if statement didn't catch it, we assume the hostname is <example>.medium.com.
	Any article URL matching that subdomain pattern can be rewritten as medium.com/<example>/article-title.
	*/
	let hostnameClone = window.location.hostname;
	// We need to check if the subdomain contains "www" and remove it before using the "pure" subdomain.
	if (hostnameClone.indexOf('www.') > -1) { hostnameClone.replace('www.', ''); }
	window.location.href = `https://scribe.rip/${hostnameClone.split('.')[0]}${window.location.pathname}`;
	}

	// const debuggingArticles = [
	// "http://medium.com/finding-insight/cancel-culture-just-cancelled-my-favorite-furniture-b652b2afb088",
	// "http://medium.com/good-service/introducing-dynamic-route-maps-a8e56dd8a33",
	// "https://medium.com/hackernoon/removing-that-ugly-focus-ring-and-keeping-it-too-6c8727fefcd2",
	// "https://michaellong.medium.com/swiftui-vs-jetpack-compose-why-android-wins-hands-down-b5f849b730db"]

	/*
	## Digging Deeper

	So I ran into an article on https://blog.gds-gov.tech/ which is hosted by Medium. None of the logic up until this point has checked if a site is Medium but it's not in the URL. So I need to figure out how to check if a particular page is Medium by hunting through the <head>. I'm not sure how to do that without tanking performance.

	I'm trying to find a meta tag or something that is immutable. In other words, a tag where the domain owner (the GDS in this case) cannot change it. I've found this:
	<meta data-rh="true" property="og:site_name" content="Medium">
	but I need to find more examples to see what other sites use this.

	Another consideration: doing the check client-side, if the site is Medium but was not in the big array, asking the user if they'd like to share that addition to the database.

	*/