Core Markdown JS code based/ripped off on Pinboard Particular bookmarklet

mdProCore.js

(function() {

/******************* begin configuration options ***************************/

// Change `read` to true to invoke the promptless, self-closing
// version of the bookmarklet.
var readlater = false;
var appUrl = null;
// when set to true selected text is quoted using markdown quote syntax
var quoteSelectionAsMarkdown = true;
// when this text appears in title or description they are added as tags
var tagKeywords = {
  javascript:'javascript',
  js:'javascript',
  python:'python',
  ios:'ios',
  youtube:'video',
  vimeo:'video',
  video:'video',
  books:'book',
  book:'book'
};

// this matches domain names to special selectors for the title
var titleTweaks = {
  "github.com":".entry-title .js-current-repository"
};

// this matches domain names to special selectors for the title
var descriptionTweaks = {
  "www.kickstarter.com":".short-blurb"
};

// limit long titles and descriptions, mostly to avoid 'HTTP/1.0 414 Request URI too long'
var textLengthLimit = 1000;

/********************* begin code ********************************************/

// reduce a string to some canonical representation
// right now this just picks a case but could get really complicated if need be
// see: http://stackoverflow.com/questions/227950/programatic-accent-reduction-in-javascript-aka-text-normalization-or-unaccentin
// some people like stack overflow straighten their curly quotes
var normalize = function(string) {
  return string.toLowerCase();
};

var elementText = function(el) {
  return el ? el.textContent.trim().replace(/\s+/g,' ') : null;
};

var normalizedDocumentTitle = normalize(document.title);

// used as tes
var isSubtitle = function(string) {
  if(string) {
    return normalizedDocumentTitle.indexOf(normalize(string)) !== -1;
  }
  else {
    return false;
  }
};

// loops over a node list and applies a function
// returning the first value that is non-null
var selectFromNodeList = function(nodeList,func,thisObj) {
  thisObj = thisObj || window;
  var l = nodeList.length;
  var result;
  for(var i=0;i<l;++i) {
    result = func.call(thisObj,nodeList[i]);
    if(result !== null) {
      return result;
    }
  }
  return null;
};

var getTitle = function() {
  var url = location.href;
  var host = location.hostname;
  var e;
  if(host in titleTweaks) {
    e = document.querySelector(titleTweaks[host]);
    if(e) {
      return elementText(e);
    }
  }
  var documentTitle = document.title;
  e = document.querySelector("meta[property='og:title']");
  if(e) {
    documentTitle = e.content.trim().replace(/\s+/g,' ');
  }

  // hEntry microformat
  if(selectFromNodeList(document.getElementsByClassName('hentry'), function() {return true;})) {
    var htitle = document.querySelector(
        '.hentry .entry-title'
      );
    if(htitle) {
      return elementText(htitle);
    }
  }

  // method 1 - look for link to self with text that is contained in title

  var a_text = selectFromNodeList(document.getElementsByTagName('A'), function(a) {
    if(a.href === url) {
      a_text = elementText(a);
      if(isSubtitle(a_text)) {
        return a_text;
      }
    }
    return null;
  });
  if(a_text) {
    return a_text;
  }

  // method 2 - look at header tags and see if it matches part of title
  var headerTags = ['h1','h2','h3','h4','h5','h6'];
  var headerTitle;
  for(var j=0;j<headerTags.length;++j) {
    selectFromNodeList(document.getElementsByTagName(headerTags[j]), function(h) {
      var h_text = elementText(h);
      if(isSubtitle(h_text) && (!headerTitle || h_text.length > headerTitle.length)) {
        headerTitle = h_text;
      }
      return null;
    });
  }
  if(headerTitle) {
    return headerTitle;
  }

  // method 3 - just return the title
  return documentTitle;
};

var getTags = function(text) {
  text = normalize(text);
  var tags = [];
  var re;
  for(var keyword in tagKeywords) {
    re = keyword instanceof RegExp ? keyword : new RegExp("\\b"+keyword+"\\b","i");
    if(re.test(text)) {
      tags.push(tagKeywords[keyword]);
    }
  }
  return tags;
};

var getMetaDescription = function() {
  var e;
  e = document.querySelector("meta[name='description']");
  if(e) {
    return e.content.trim().replace(/\s+/g,' ');
  }
  e = document.querySelector("meta[property='og:description']");
  if(e) {
    return e.content.trim().replace(/\s+/g,' ');
  }
  return "";
};

var getDescription = function() {
  var text;
  // Grab the text selection (if any) and quote it
  if('' !== (text = String(document.getSelection()))) {
    if(quoteSelectionAsMarkdown) {
      text = text.trim().split("\n").map(function(s) {return "> "+s;}).join("\n");
    }
  }

  var host = location.hostname;
  var e;
  if(host in descriptionTweaks) {
    e = document.querySelector(descriptionTweaks[host]);
    if(e) {
      return elementText(e);
    }
  }

  if(!text) {
    text = getMetaDescription();
  }
  return text;
};

// Assembles default form pre-fill arguments.
var url = location.href;
var title = getTitle();
var description = getDescription();
// remove if title is trailing or leading
var ix = description.indexOf(title);
if(ix === 0) {
  description = description.substring(title.length).trim();
}
else if(ix === description.length-title.length) {
  description = description.substring(0,ix).trim();
}

var tags = getTags(document.title+" "+description+" "+getMetaDescription());

if(textLengthLimit > 0) {
  title = title.substring(0, textLengthLimit);
  description = description.substring(0, textLengthLimit);  
}

var args = [
  'url=', encodeURIComponent(url),
  '&title=', encodeURIComponent(title),
  '&description=', encodeURIComponent(description),
  '&tags=', encodeURIComponent(tags.join(" "))
];


var md  = '## » [' + title + '](' + url + ')\n\n' + description;

/********************* Outputs Options ********************************************/

// Bookmarklet
 prompt("Enter Ctrl+C to copy this markdown:", md);

//Ulysses
var mdURI = encodeURIComponent(md);

//Copied App iOS
// Return true if the clipping can be reformatted withe this script.

// function canFormat(clipping) {
//     return clipping.title != null
// }

// // Process data and return a new/updated clipping object or a string.
// function format(clipping) {
//     return '## » [' + clipping.title + '](' + clipping.url + ')\n\n' + clipping.text
// }

// window.location.href = "ulysses://x-callback-url/"
//                         + "new-sheet?text="
//                         + mdURI
//                         + "&group=5typos"

// Copied macOS
//copied = Application('Copied')
// copied.save('Text to save')

// // If readlater mode, add the auto-close parameter and read-later flag:
// if(readlater) {
//   args = args.concat([
//     '&later=', 'yes',
//     '&jump=', 'close'
//   ]);
// }
// if(appUrl) {
//   args = args.concat([
//     '&x-source=Safari',
//     '&x-success=',encodeURIComponent(location.href),
//     '&x-cancel=',encodeURIComponent(location.href)
//   ]);
//   window.location = appUrl+args.join('');
// }
// else {
//   var pin = open('http://pinboard.in/add?'+args.join(''), 'Pinboard', 'toolbar=no,width=610,height=350');

//   // Send the window to the background if readlater mode.
//   if(readlater) {
//     pin.blur();
//   }
// }

})();