Wayback Machine Image Fixer

Attempts to fix broken images by replacing them with working timestamps based on JSON results

当前为 2016-01-02 提交的版本,查看 最新版本

您需要先安装一个扩展,例如 篡改猴Greasemonkey暴力猴,之后才能安装此脚本。

You will need to install an extension such as Tampermonkey to install this script.

您需要先安装一个扩展,例如 篡改猴暴力猴,之后才能安装此脚本。

您需要先安装一个扩展,例如 篡改猴Userscripts ,之后才能安装此脚本。

您需要先安装一款用户脚本管理器扩展,例如 Tampermonkey,才能安装此脚本。

您需要先安装用户脚本管理器扩展后才能安装此脚本。

(我已经安装了用户脚本管理器,让我安装!)

您需要先安装一款用户样式管理器扩展,比如 Stylus,才能安装此样式。

您需要先安装一款用户样式管理器扩展,比如 Stylus,才能安装此样式。

您需要先安装一款用户样式管理器扩展,比如 Stylus,才能安装此样式。

您需要先安装一款用户样式管理器扩展后才能安装此样式。

您需要先安装一款用户样式管理器扩展后才能安装此样式。

您需要先安装一款用户样式管理器扩展后才能安装此样式。

(我已经安装了用户样式管理器,让我安装!)

// ==UserScript==
// @name          Wayback Machine Image Fixer
// @namespace     DoomTay
// @description   Attempts to fix broken images by replacing them with working timestamps based on JSON results
// @include       http://web.archive.org/web/*
// @include       https://web.archive.org/web/*
// @exclude       /\*/
// @exclude       *.jpg
// @exclude       *.jpeg
// @exclude       *.png
// @exclude       *.gif
// @exclude       *.bmp
// @version       1.3.0
// @grant         GM_xmlhttpRequest

// ==/UserScript==

var pics = document.images;
var backgrounds = document.querySelectorAll("[background]");

function specialUses(pic)
{
	pic.image = pic.src || relativeToAbsolute(pic.background);
	var specialCases = [{domain:"northarc.com/images/unsorted/",replacements:[pic.image.replace("thumb.","tn_"),pic.image.replace("thumb.","")],maxDimensions:{width:100,height:80}},
		{domain:"*",replacements:["data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACH5BAEAAAAALAAAAAABAAEAAAICRAEAOw=="],condition:pic.image.indexOf("spacer.gif") > -1 || pic.image.indexOf("blank.gif") > -1},
		{domain:"tinypic.com/",replacements:[pic.image.replace("/i","/oi")]}];
	var filteredCases = {};
	filteredCases.replacements = [];
	for(var c = 0; c < specialCases.length; c++)
	{
		var additionalCondition = specialCases[c].condition !== undefined ? specialCases[c].condition : true;
		if((pic.image.indexOf(specialCases[c].domain) > -1 || specialCases[c].domain == "*") && additionalCondition)
		{
			filteredCases.replacements = filteredCases.replacements.concat(specialCases[c].replacements);
			if(specialCases[c].maxDimensions) filteredCases.maxDimensions = specialCases[c].maxDimensions;
		}
	}
	return filteredCases;
}

var timestamp = /web\/(\d{1,14})/.exec(window.location.href)[1];

function replaceImage(target, altURLs)
{
	var possibleUses = specialUses(target);
	var URLGuesses = [];
	if(altURLs && altURLs.length > 0) URLGuesses = altURLs;
	else if(possibleUses.replacements.length > 0) URLGuesses = possibleUses.replacements;
	else URLGuesses = [target.src || relativeToAbsolute(target.background)];
	
	var testSet = [];
	for(var p = 0; p < URLGuesses.length; p++)
	{
		if(URLGuesses[p].indexOf("http") > -1) testSet[p] = APITest(URLGuesses[p]);
		else if(URLGuesses[p].indexOf("data:") > -1)
		{
			testSet[p] = Promise.resolve(URLGuesses[p]);
			break;
		}
	}
	Promise.all(testSet).then(function(value) {
		for(var v = 0; v < value.length; v++)
		{
			if(value[v] !== null)
			{
				if(possibleUses.maxDimensions) switchWithResize(value[v], target, possibleUses.maxDimensions.width, possibleUses.maxDimensions.height);
				else changeImage(value[v], target);
				return;
			}
		}
		//Try and "expose" image links that are unclickable due to the image not loading
		if(target.alt === "" && target.width === 0 && target.parentNode.nodeName === "A" && target.src)
		{
			//Changing the source is pretty hacky, but it's the only way I can think of to turn "invisible" image links into something clickable
			target.src = target.src.substring(target.src.lastIndexOf("http"));
			target.width = 25;
			target.height = 25;
		}
	});
	
	function APITest(replacement)
	{
		var subPromise = new Promise(function(resolve,reject) {
			var originalURL = replacement.substring(replacement.lastIndexOf("http"));
			GM_xmlhttpRequest({
			url: "http://archive.org/wayback/available?url=" + originalURL + "&timestamp=" + timestamp,
			method: "GET",
			headers: {"Accept": "application/json"},
			onload: function(response) {
				if(response.status == 503) reject(response.statusText + " for " + originalURL);
				else if(JSON.parse(response.responseText).archived_snapshots.closest !== undefined) resolve(JSON.parse(response.responseText).archived_snapshots.closest.url);
				else resolve(null);
			}
			});
		}).catch(function(e) { return APITest(replacement); });
		return subPromise;
	}
	
	function switchWithResize(url, target, width, height)
	{
		var oldDimensions = {width:width, height: height};
		changeImage(url, target);
		target.onload = function() {
			var aspectRatio = target.naturalWidth / target.naturalHeight;
			if(target.naturalWidth >= target.naturalHeight)
			{
				target.width = oldDimensions.width;
				target.height = oldDimensions.width / aspectRatio;
			}
			else if(target.naturalWidth <= target.naturalHeight)
			{
				target.height = oldDimensions.height;
				target.width = oldDimensions.height * aspectRatio;
			}
		};
	}
	
	function changeImage(url, target)
	{
		if(target.src) target.src = url;
		else if(target.background) target.background = url;
	}
}

function relativeToAbsolute(bgURL)
{
	var img = new Image();
	img.src = bgURL;
	return img.src;
}

function evaluateImage(pic)
{
	GM_xmlhttpRequest({
		url: pic.src,
		method: "GET",
		onload: function(response) {
			//Going off of response code is unreliable. Sometimes an image will return a status code of 200 even though it would redirect to an error page should you view the image directly, so we're looking at content type instead
			if(response.responseHeaders.indexOf("Content-Type: text/html") > -1)
			{
				//This might be a case where if you were visit the image directly, you would be redirected elsewhere. This attempts to catch that and replace the pic's src with where it would take you.
				var doc = document.implementation.createHTMLDocument("Possible Replacement");
				doc.documentElement.innerHTML = response.responseText;
				if(doc.getElementsByClassName("impatient")[0])
				{
					GM_xmlhttpRequest({
						url: doc.getElementsByClassName("impatient")[0].firstChild.href,
						method: "HEAD",
						onload: function(response) {
							//Going off of response code is unreliable. Sometimes an image will return a status code of 200 even though it would redirect to an error page should you view the image directly, so we're looking at content type instead
							if(response.responseHeaders.indexOf("Content-Type: text/html") == -1)
							{
								replaceImage(pic,[doc.getElementsByClassName("impatient")[0].firstChild.href]);
							}
							else
							{
								replaceImage(pic);
							}
						}
					});
				}
				else if(response.status != 403)
				{
					replaceImage(pic);
				}
			}
		}
	});
}

for(var i = 0; i < pics.length; i++)
{
	//Skip over stuff related to the Wayback Machine toolbar and data URIs
	if((document.getElementById("wm-ipp") && document.getElementById("wm-ipp").contains(pics[i])) || pics[i].src.indexOf("data:") > -1) continue;
	if(pics[i].src.indexOf("ttp://") === 0) pics[i].src = "web.archive.org/web/" + timestamp + "/h" + pics[i].src;
	evaluateImage(pics[i]);
}

for(var b = 0; b < backgrounds.length; b++)
{
	evaluateImage(backgrounds[b]);
}