LinkBot ver.UserScript.user.js

// ==UserScript==
// @name         LinkBot ver.UserScript
// @namespace    https://github.com/PaeP3nguin/LinkBot
// @version      0.6
// @description   Removed linkBot from Chrome Web store. https://chrome.google.com/webstore/detail/chnfcfcbnhloogdohcmjogkklghefofm
// @author       ten9miq
// @match        http*://*/*
// @icon         data:image/png;base64,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
// @updateURL    https://gist.github.com/53JIlLenWe11/03ad50188de8b335b899fe7495f1ab1a/raw/LinkBot%2520ver.UserScript.user.js
// @downloadURL https://gist.github.com/53JIlLenWe11/03ad50188de8b335b899fe7495f1ab1a/raw/LinkBot%2520ver.UserScript.user.js
// @grant        none
// ==/UserScript==
const URL_REGEX = /\b(?:(?:h?ttps?|ftp|):\/\/)?((?:(?:[01]?\d?\d|2[0-4]\d|25[0-5])(?:\.(?:[01]?\d?\d|2[0-4]\d|25[0-5])){3}|(?:[a-z\u00a1-\uffff\d]+-)*[a-z\u00a1-\uffff\d]+(?:\.(?:[a-z\u00a1-\uffff\d]+--?)*[a-z\u00a1-\uffff\d]+)*\.(?:com?|net|org|edu|gov|cc|in(?:fo)?|io|bi(?:z|d)|mobi|tv|bz|fm|am|me|ly|gl|gdn?|do(?:wnload)?|tw|us|tk|cf|cn|de|uk|ru|nl|eu|br|au|fr|it|pl|jp|ws|ca|ws|es|ch|be|im|pr|pw|gs|nu|ie|is|mn|mp|nz|rs|sh|vg|lu|ug|xn--[a-z\u00a1-\uffff\d-]{4,59}|xyz|top?|wang|win|cl(?:ub|ick)|li(?:nk)?|vip|online|science|engineering|si(?:te)?|racing|date|bar|chat|website|social|life|lol|ai|group|space|town|pro|love|host|fyi|zone|estate|moe|world|work|lgbt|church))(?::\d{2,5})?(?:[\/?#]\S*[a-z\u00a1-\uffff\d=])?)\b/gi;
const EMAIL_REGEX = /\b[\w\u00a1-\uffff!#$%&'*+/=?^`{|}~-]+(?:\.[\w\u00a1-\uffff!#$%&'*+/=?^`{|}~-]+)*@(?:(?:[01]?\d?\d|2[0-4]\d|25[0-5])(?:\.(?:[01]?\d?\d|2[0-4]\d|25[0-5])){3}|(?:[a-z\u00a1-\uffff\d]+-)*[a-z\u00a1-\uffff\d]+(?:\.(?:[a-z\u00a1-\uffff\d]+-)*[a-z\u00a1-\uffff\d]+)*(?:\.(?:[a-z\u00a1-\uffff]{2,})))[a-z\u00a1-\uffff\d]?\b/gi;
const SUBREDDIT_REGEX = /(?:\b|\/)(r\/[a-z0-9][a-z0-9_]{2,29})\b/gi;
// Temporary placeholder for potentially conflicting email substitution
const TEMP_CHAR = '\uFFFF';
const TEMP_CHAR_REGEX = /\uFFFF/gi;

const range = document.createRange();
const parse = range.createContextualFragment.bind(range);
// A collection of tags to not replace text inside of
const EXCLUDED_TAGS = {
    // Already clickable
    A: true,
    BUTTON: true,
    OPTION: true,
    // May cause issues if we replace things
    IFRAME: true,
    NOSCRIPT: true,
    SCRIPT: true,
    STYLE: true,
    META: true,
    EMBED: true,
    // Better UX if we don't, tags may be user input or contain HTML
    CITE: true,
    TITLE: true,
    TEXTAREA: true,
    INPUT: true,
    FORM: true,
    PRE: false,
    // Every example site seems to use this...
    // CODE: true,
    H1: true,
};

(function() {
    'use strict';
    recursiveLink(document.body);
    const observerOptions = {
        subtree: true,
        characterData: true,
        childList: true
    };
    // Watch for DOM changes
    const observer = new MutationObserver(function(mutations) {
        mutations.forEach(function(m) {
            if (!shouldLinkParents(m.target) || !shouldLink(m.target)) {
                return;
            }
            var linkCount = 0;
            if (m.type === "characterData") {
                // Actual text node itself changed
                linkCount += linkSingleNode(m.target);
            } else if (m.type === "childList") {
                // Added or removed stuff somewhere
                for (var i = 0, l = m.addedNodes.length; i < l; i++) {
                    var child = m.addedNodes[i];
                    if (shouldLink(child)) {
                        linkCount += linkSingleNode(child);
                    }
                }
            }
            if (linkCount > 0) {
                // Stop and restart watching so we don't see mutations that we're causing
                // and allow webpage JS to run, which prevents infinite loops
                observer.disconnect();
                setTimeout(function() {
                    observer.observe(document.body, observerOptions);
                }, 0);
            }
        });
    });
    observer.observe(document.body, observerOptions);
})();


// Convert all links under a root node, returns number of links found
function recursiveLink(root) {
    // Initialize a TreeWalker to start looking at text from the root node
    const walker = document.createTreeWalker(root, NodeFilter.SHOW_ALL, {
        acceptNode: nodeFilter
    }, false);
    let prev;
    let node = walker.nextNode();
    let linkCount = 0;
    while (node !== null) {
        // Advance the walker past the current node to prevent
        // seeing the same text nodes twice due to our own DOM changes
        prev = node;
        node = walker.nextNode();
        linkCount += linkTextNode(prev);
    }
    return linkCount;
}

// Filter for TreeWalker to determine which nodes to examine
function nodeFilter(node) {
    switch (node.nodeType) {
        case Node.TEXT_NODE:
            // Skip if text is too short to be a link
            // Shortest possible link is something like g.co
            if (node.data.trim().length <= 4) {
                return NodeFilter.FILTER_SKIP;
            }
            return NodeFilter.FILTER_ACCEPT;
        case Node.ELEMENT_NODE:
            // Skip node and all descendants of an editable node
            if (isNodeEditable(node)) {
                return NodeFilter.FILTER_REJECT;
            }
            // Skip node and all descendants of any excluded tags
            if (EXCLUDED_TAGS[node.tagName]) {
                return NodeFilter.FILTER_REJECT;
            }
            // Pass by boring old non-text nodes
            return NodeFilter.FILTER_SKIP;
        default:
            // What are you????
            return NodeFilter.FILTER_SKIP;
    }
}

// Tests whether a node is contentEditable
function isNodeEditable(node) {
    return node.isContentEditable || node.contentEditable === "true";
}

// Link a single node based on its nodeType, returns number of nodes found
function linkSingleNode(node) {
    if (node.nodeType === Node.TEXT_NODE) {
        // Skip if text is too short to be a link
        // Shortest possible link is something like g.co
        if (node.data.trim().length <= 4) {
            return 0;
        }
        return linkTextNode(node);
    } else if (node.nodeType === Node.ELEMENT_NODE) {
        // Skip node and all descendants of an editable node
        if (isNodeEditable(node)) {
            return 0;
        }
        return recursiveLink(node);
    }
    return 0;
}

// Find links in a text node. Returns number of links found
function linkTextNode(node) {
    // Email saving variables and functions
    const emails = [];
    let i = 0;
    let urlCount = 0;
    // Save the text to compare with later
    let oldText = node.data;
    let newText = oldText;
    // Save emails and replace with a temporary, noncharacter Unicode character
    // We'll put the emails back in later
    // Why? Because otherwise the part after the @ sign will be recognized and replaced as a URL!
    newText = newText.replace(EMAIL_REGEX, function(email) {
        emails.push('<a href="mailto:' + email + '">' + email + '</a>');
        return TEMP_CHAR;
    });
    // Replace URLs with links
    newText = newText.replace(URL_REGEX, function(match, part) {
        urlCount++;
        return '<a href="//' + part + '">' + match + '</a>';
    });
    if ( window.location.hostname === "www.reddit.com") {
        newText = newText.replace(SUBREDDIT_REGEX, function(match, part) {
            urlCount++;
            return '<a href="//www.reddit.com/' + part + '">' + match + '</a>';
        });
    }
    if (emails.length) {
        // Put emails back in, if any
        newText = newText.replace(TEMP_CHAR_REGEX, function() {
            return emails[i++];
        });
    }
    if (newText !== oldText) {
        // If we successfully added any links, insert into DOM
        node.replaceWith(parse(newText));
    }
    return i + urlCount;
}

// Returns false if any of the parents of a node should not be linkified.
function shouldLinkParents(node) {
    let parent = node.parentNode;
    while (parent !== null) {
        if (shouldLink(parent)) {
            parent = parent.parentNode;
        } else {
            return false;
        }
    }
    return true;
}

// Returns true if we should link the node.
function shouldLink(node) {
    if (EXCLUDED_TAGS[node.tagName] || isNodeEditable(node)) {
        return false;
    } else {
        return true;
    }
}