From d16b15f3eb7d5753b60fbc7907e85ca996674bdb Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 10 Feb 2013 17:34:34 +0000
Subject: [PATCH 01/39] begin support for better txt output

---
 src/node/handler/ExportHandler.js |  78 ++++-
 src/node/utils/ExportTxt.js       | 477 ++++++++++++++++++++++++++++++
 2 files changed, 543 insertions(+), 12 deletions(-)
 create mode 100644 src/node/utils/ExportTxt.js

diff --git a/src/node/handler/ExportHandler.js b/src/node/handler/ExportHandler.js
index 1b7fcc26d..8ff5bc488 100644
--- a/src/node/handler/ExportHandler.js
+++ b/src/node/handler/ExportHandler.js
@@ -20,6 +20,7 @@
 
 var ERR = require("async-stacktrace");
 var exporthtml = require("../utils/ExportHtml");
+var exporttxt = require("../utils/ExportTxt");
 var exportdokuwiki = require("../utils/ExportDokuWiki");
 var padManager = require("../db/PadManager");
 var async = require("async");
@@ -48,22 +49,75 @@ exports.doExport = function(req, res, padId, type)
   res.attachment(padId + "." + type);
 
   //if this is a plain text export, we can do this directly
+  // We have to over engineer this because tabs are stored as attributes and not plain text
+
   if(type == "txt")
   {
-    padManager.getPad(padId, function(err, pad)
-    {
-      ERR(err);
-      if(req.params.rev){
-        pad.getInternalRevisionAText(req.params.rev, function(junk, text)
-        {
-          res.send(text.text ? text.text : null);
-        });
-      }
-      else
+    var txt;
+    var randNum;
+    var srcFile, destFile;
+
+    async.series([
+      //render the txt document
+      function(callback)
       {
-        res.send(pad.text());
+        exporttxt.getPadTXTDocument(padId, req.params.rev, false, function(err, _txt)
+        {
+          if(ERR(err, callback)) return;
+          txt = _txt;
+          callback();
+        });
+      },
+      //decide what to do with the txt export
+      function(callback)
+      {
+        //if this is a txt export, we can send this from here directly
+        res.send(txt);
+        callback("stop");
+      },
+      //send the convert job to abiword
+      function(callback)
+      {
+        //ensure html can be collected by the garbage collector
+        txt = null;
+
+        destFile = tempDirectory + "/eplite_export_" + randNum + "." + type;
+        abiword.convertFile(srcFile, destFile, type, callback);
+      },
+      //send the file
+      function(callback)
+      {
+        res.sendfile(destFile, null, callback);
+      },
+      //clean up temporary files
+      function(callback)
+      {
+        async.parallel([
+          function(callback)
+          {
+            fs.unlink(srcFile, callback);
+          },
+          function(callback)
+          {
+            //100ms delay to accomidate for slow windows fs
+            if(os.type().indexOf("Windows") > -1)
+            {
+              setTimeout(function()
+              {
+                fs.unlink(destFile, callback);
+              }, 100);
+            }
+            else
+            {
+              fs.unlink(destFile, callback);
+            }
+          }
+        ], callback);
       }
-    });
+    ], function(err)
+    {
+      if(err && err != "stop") ERR(err);
+    })
   }
   else if(type == 'dokuwiki')
   {
diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
new file mode 100644
index 000000000..99f6085e3
--- /dev/null
+++ b/src/node/utils/ExportTxt.js
@@ -0,0 +1,477 @@
+/**
+ * Copyright 2009 Google Inc.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS-IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+
+var async = require("async");
+var Changeset = require("ep_etherpad-lite/static/js/Changeset");
+var padManager = require("../db/PadManager");
+var ERR = require("async-stacktrace");
+var Security = require('ep_etherpad-lite/static/js/security');
+var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
+function getPadPlainText(pad, revNum)
+{
+  var atext = ((revNum !== undefined) ? pad.getInternalRevisionAText(revNum) : pad.atext());
+  var textLines = atext.text.slice(0, -1).split('\n');
+  var attribLines = Changeset.splitAttributionLines(atext.attribs, atext.text);
+  var apool = pad.pool();
+
+  var pieces = [];
+  for (var i = 0; i < textLines.length; i++)
+  {
+    var line = _analyzeLine(textLines[i], attribLines[i], apool);
+    if (line.listLevel)
+    {
+      var numSpaces = line.listLevel * 2 - 1;
+      var bullet = '*';
+      pieces.push(new Array(numSpaces + 1).join(' '), bullet, ' ', line.text, '\n');
+    }
+    else
+    {
+      pieces.push(line.text, '\n');
+    }
+  }
+
+  return pieces.join('');
+}
+
+function getPadTXT(pad, revNum, callback)
+{
+  var atext = pad.atext;
+  var html;
+  async.waterfall([
+  // fetch revision atext
+
+
+  function (callback)
+  {
+    if (revNum != undefined)
+    {
+      pad.getInternalRevisionAText(revNum, function (err, revisionAtext)
+      {
+        if(ERR(err, callback)) return;
+        atext = revisionAtext;
+        callback();
+      });
+    }
+    else
+    {
+      callback(null);
+    }
+  },
+
+  // convert atext to html
+
+
+  function (callback)
+  {
+    html = getTXTFromAtext(pad, atext);
+    callback(null);
+  }],
+  // run final callback
+
+
+  function (err)
+  {
+    if(ERR(err, callback)) return;
+    callback(null, html);
+  });
+}
+
+exports.getPadTXT = getPadTXT;
+exports.getTXTFromAtext = getTXTFromAtext;
+
+function getTXTFromAtext(pad, atext, authorColors)
+{
+  var apool = pad.apool();
+  var textLines = atext.text.slice(0, -1).split('\n');
+  var attribLines = Changeset.splitAttributionLines(atext.attribs, atext.text);
+
+  var tags = ['h1', 'h2', 'strong', 'em', 'u', 's'];
+  var props = ['heading1', 'heading2', 'bold', 'italic', 'underline', 'strikethrough'];
+  var anumMap = {};
+  var css = "";
+
+  props.forEach(function (propName, i)
+  {
+    var propTrueNum = apool.putAttrib([propName, true], true);
+    if (propTrueNum >= 0)
+    {
+      anumMap[propTrueNum] = i;
+    }
+  });
+
+  function getLineTXT(text, attribs)
+  {
+    var propVals = [false, false, false];
+    var ENTER = 1;
+    var STAY = 2;
+    var LEAVE = 0;
+
+    // Use order of tags (b/i/u) as order of nesting, for simplicity
+    // and decent nesting.  For example,
+    // <b>Just bold<b> <b><i>Bold and italics</i></b> <i>Just italics</i>
+    // becomes
+    // <b>Just bold <i>Bold and italics</i></b> <i>Just italics</i>
+    var taker = Changeset.stringIterator(text);
+    var assem = Changeset.stringAssembler();
+    var openTags = [];
+
+    var urls = _findURLs(text);
+
+    var idx = 0;
+
+    function processNextChars(numChars)
+    {
+      if (numChars <= 0)
+      {
+        return;
+      }
+
+      var iter = Changeset.opIterator(Changeset.subattribution(attribs, idx, idx + numChars));
+      idx += numChars;
+
+      while (iter.hasNext())
+      {
+        var o = iter.next();
+        var propChanged = false;
+        Changeset.eachAttribNumber(o.attribs, function (a)
+        {
+          if (a in anumMap)
+          {
+            var i = anumMap[a]; // i = 0 => bold, etc.
+            if (!propVals[i])
+            {
+              propVals[i] = ENTER;
+              propChanged = true;
+            }
+            else
+            {
+              propVals[i] = STAY;
+            }
+          }
+        });
+        for (var i = 0; i < propVals.length; i++)
+        {
+          if (propVals[i] === true)
+          {
+            propVals[i] = LEAVE;
+            propChanged = true;
+          }
+          else if (propVals[i] === STAY)
+          {
+            propVals[i] = true; // set it back
+          }
+        }
+        // now each member of propVal is in {false,LEAVE,ENTER,true}
+        // according to what happens at start of span
+        if (propChanged)
+        {
+          // leaving bold (e.g.) also leaves italics, etc.
+          var left = false;
+          for (var i = 0; i < propVals.length; i++)
+          {
+            var v = propVals[i];
+            if (!left)
+            {
+              if (v === LEAVE)
+              {
+                left = true;
+              }
+            }
+            else
+            {
+              if (v === true)
+              {
+                propVals[i] = STAY; // tag will be closed and re-opened
+              }
+            }
+          }
+
+          var tags2close = [];
+
+          for (var i = propVals.length - 1; i >= 0; i--)
+          {
+            if (propVals[i] === LEAVE)
+            {
+              //emitCloseTag(i);
+              tags2close.push(i);
+              propVals[i] = false;
+            }
+            else if (propVals[i] === STAY)
+            {
+              //emitCloseTag(i);
+              tags2close.push(i);
+            }
+          }
+          
+          for (var i = 0; i < propVals.length; i++)
+          {
+            if (propVals[i] === ENTER || propVals[i] === STAY)
+            {
+              emitOpenTag(i);
+              propVals[i] = true;
+            }
+          }
+          // propVals is now all {true,false} again
+        } // end if (propChanged)
+        var chars = o.chars;
+        if (o.lines)
+        {
+          chars--; // exclude newline at end of line, if present
+        }
+        
+        var s = taker.take(chars);
+        
+        //removes the characters with the code 12. Don't know where they come 
+        //from but they break the abiword parser and are completly useless
+        s = s.replace(String.fromCharCode(12), "");
+        
+        assem.append(_encodeWhitespace(Security.escapeHTML(s)));
+      } // end iteration over spans in line
+      
+      var tags2close = [];
+      for (var i = propVals.length - 1; i >= 0; i--)
+      {
+        if (propVals[i])
+        {
+          tags2close.push(i);
+          propVals[i] = false;
+        }
+      }
+      
+    } // end processNextChars
+    if (urls)
+    {
+      urls.forEach(function (urlData)
+      {
+        var startIndex = urlData[0];
+        var url = urlData[1];
+        var urlLength = url.length;
+        processNextChars(startIndex - idx);
+        console.warn(url);
+        // assem.append('<a href="' + Security.escapeHTMLAttribute(url) + '">');
+        assem.append(url);
+        processNextChars(urlLength);
+        // assem.append('</a>');
+      });
+    }
+    processNextChars(text.length - idx);
+
+    return _processSpaces(assem.toString());
+  } // end getLineHTML
+  var pieces = [css];
+
+  // Need to deal with constraints imposed on HTML lists; can
+  // only gain one level of nesting at once, can't change type
+  // mid-list, etc.
+  // People might use weird indenting, e.g. skip a level,
+  // so we want to do something reasonable there.  We also
+  // want to deal gracefully with blank lines.
+  // => keeps track of the parents level of indentation
+  var lists = []; // e.g. [[1,'bullet'], [3,'bullet'], ...]
+  for (var i = 0; i < textLines.length; i++)
+  {
+    var line = _analyzeLine(textLines[i], attribLines[i], apool);
+    var lineContent = getLineTXT(line.text, line.aline);
+    if(line.listTypeName == "bullet"){
+      lineContent = "* " + lineContent; // add a bullet
+    }
+    if(line.listLevel > 0){
+      for (var j = line.listLevel - 1; j >= 0; j--){
+        pieces.push('\t');
+      }
+      if(line.listTypeName == "number"){
+        pieces.push(line.listLevel + ". ");
+        // This is bad because it doesn't truly reflect what the user
+        // sees because browsers do magic on nested <ol><li>s
+      }
+      pieces.push(lineContent, '\n');
+    }else{
+      console.warn(line);
+      pieces.push(lineContent, '\n');
+    }
+
+    // I'm not too keen about using teh HTML export filters here, they could cause real pain in the future
+    // I'd suggest supporting getLineTXTForExport
+    var lineContentFromHook = hooks.callAllStr("getLineHTMLForExport", 
+    {
+      line: line,
+      apool: apool,
+      attribLine: attribLines[i],
+      text: textLines[i]
+    }, " ", " ", "");
+    if (lineContentFromHook)
+    {
+      pieces.push(lineContentFromHook, '');
+    } 
+    else 
+    {
+      // pieces.push(lineContent, '\n');
+    }
+  }
+
+  return pieces.join('');
+}
+
+function _analyzeLine(text, aline, apool)
+{
+  var line = {};
+
+  // identify list
+  var lineMarker = 0;
+  line.listLevel = 0;
+  if (aline)
+  {
+    var opIter = Changeset.opIterator(aline);
+    if (opIter.hasNext())
+    {
+      var listType = Changeset.opAttributeValue(opIter.next(), 'list', apool);
+      if (listType)
+      {
+        lineMarker = 1;
+        listType = /([a-z]+)([12345678])/.exec(listType);
+        if (listType)
+        {
+          line.listTypeName = listType[1];
+          line.listLevel = Number(listType[2]);
+        }
+      }
+    }
+  }
+  if (lineMarker)
+  {
+    line.text = text.substring(1);
+    line.aline = Changeset.subattribution(aline, 1);
+  }
+  else
+  {
+    line.text = text;
+    line.aline = aline;
+  }
+
+  return line;
+}
+
+exports.getPadTXTDocument = function (padId, revNum, noDocType, callback)
+{
+  padManager.getPad(padId, function (err, pad)
+  {
+    if(ERR(err, callback)) return;
+
+    getPadTXT(pad, revNum, function (err, html)
+    {
+      if(ERR(err, callback)) return;
+      callback(null, html);
+    });
+  });
+}
+
+function _encodeWhitespace(s) {
+  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c)
+  {
+    return "&#" +c.charCodeAt(0) + ";"
+  });
+}
+
+// copied from ACE
+function _processSpaces(s)
+{
+  var doesWrap = true;
+  if (s.indexOf("<") < 0 && !doesWrap)
+  {
+    // short-cut
+    return s.replace(/ /g, '&nbsp;');
+  }
+  var parts = [];
+  s.replace(/<[^>]*>?| |[^ <]+/g, function (m)
+  {
+    parts.push(m);
+  });
+  if (doesWrap)
+  {
+    var endOfLine = true;
+    var beforeSpace = false;
+    // last space in a run is normal, others are nbsp,
+    // end of line is nbsp
+    for (var i = parts.length - 1; i >= 0; i--)
+    {
+      var p = parts[i];
+      if (p == " ")
+      {
+        if (endOfLine || beforeSpace) parts[i] = '&nbsp;';
+        endOfLine = false;
+        beforeSpace = true;
+      }
+      else if (p.charAt(0) != "<")
+      {
+        endOfLine = false;
+        beforeSpace = false;
+      }
+    }
+    // beginning of line is nbsp
+    for (var i = 0; i < parts.length; i++)
+    {
+      var p = parts[i];
+      if (p == " ")
+      {
+        parts[i] = '&nbsp;';
+        break;
+      }
+      else if (p.charAt(0) != "<")
+      {
+        break;
+      }
+    }
+  }
+  else
+  {
+    for (var i = 0; i < parts.length; i++)
+    {
+      var p = parts[i];
+      if (p == " ")
+      {
+        parts[i] = '&nbsp;';
+      }
+    }
+  }
+  return parts.join('');
+}
+
+
+// copied from ACE
+var _REGEX_WORDCHAR = /[\u0030-\u0039\u0041-\u005A\u0061-\u007A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u1FFF\u3040-\u9FFF\uF900-\uFDFF\uFE70-\uFEFE\uFF10-\uFF19\uFF21-\uFF3A\uFF41-\uFF5A\uFF66-\uFFDC]/;
+var _REGEX_SPACE = /\s/;
+var _REGEX_URLCHAR = new RegExp('(' + /[-:@a-zA-Z0-9_.,~%+\/\\?=&#;()$]/.source + '|' + _REGEX_WORDCHAR.source + ')');
+var _REGEX_URL = new RegExp(/(?:(?:https?|s?ftp|ftps|file|smb|afp|nfs|(x-)?man|gopher|txmt):\/\/|mailto:)/.source + _REGEX_URLCHAR.source + '*(?![:.,;])' + _REGEX_URLCHAR.source, 'g');
+
+// returns null if no URLs, or [[startIndex1, url1], [startIndex2, url2], ...]
+
+function _findURLs(text)
+{
+  _REGEX_URL.lastIndex = 0;
+  var urls = null;
+  var execResult;
+  while ((execResult = _REGEX_URL.exec(text)))
+  {
+    urls = (urls || []);
+    var startIndex = execResult.index;
+    var url = execResult[0];
+    urls.push([startIndex, url]);
+  }
+
+  return urls;
+}
+

From a378f48c00fca7b78d888e3fd5957668fbff8811 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 10 Feb 2013 17:39:02 +0000
Subject: [PATCH 02/39] remove console warns

---
 src/node/utils/ExportTxt.js | 2 --
 1 file changed, 2 deletions(-)

diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 99f6085e3..d9ee708f9 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -261,7 +261,6 @@ function getTXTFromAtext(pad, atext, authorColors)
         var url = urlData[1];
         var urlLength = url.length;
         processNextChars(startIndex - idx);
-        console.warn(url);
         // assem.append('<a href="' + Security.escapeHTMLAttribute(url) + '">');
         assem.append(url);
         processNextChars(urlLength);
@@ -300,7 +299,6 @@ function getTXTFromAtext(pad, atext, authorColors)
       }
       pieces.push(lineContent, '\n');
     }else{
-      console.warn(line);
       pieces.push(lineContent, '\n');
     }
 

From a67a0950dd05bbe494ede5dfd696fcc47dcc9107 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 10 Feb 2013 19:21:27 +0000
Subject: [PATCH 03/39] stop urls being encoded, not sure about other security
 implications here...

---
 src/node/utils/ExportTxt.js | 40 ++-----------------------------------
 1 file changed, 2 insertions(+), 38 deletions(-)

diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index d9ee708f9..462583d3a 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -129,8 +129,6 @@ function getTXTFromAtext(pad, atext, authorColors)
     var assem = Changeset.stringAssembler();
     var openTags = [];
 
-    var urls = _findURLs(text);
-
     var idx = 0;
 
     function processNextChars(numChars)
@@ -239,7 +237,8 @@ function getTXTFromAtext(pad, atext, authorColors)
         //from but they break the abiword parser and are completly useless
         s = s.replace(String.fromCharCode(12), "");
         
-        assem.append(_encodeWhitespace(Security.escapeHTML(s)));
+        // assem.append(_encodeWhitespace(Security.escapeHTML(s)));
+        assem.append(_encodeWhitespace(s));
       } // end iteration over spans in line
       
       var tags2close = [];
@@ -253,20 +252,6 @@ function getTXTFromAtext(pad, atext, authorColors)
       }
       
     } // end processNextChars
-    if (urls)
-    {
-      urls.forEach(function (urlData)
-      {
-        var startIndex = urlData[0];
-        var url = urlData[1];
-        var urlLength = url.length;
-        processNextChars(startIndex - idx);
-        // assem.append('<a href="' + Security.escapeHTMLAttribute(url) + '">');
-        assem.append(url);
-        processNextChars(urlLength);
-        // assem.append('</a>');
-      });
-    }
     processNextChars(text.length - idx);
 
     return _processSpaces(assem.toString());
@@ -452,24 +437,3 @@ function _processSpaces(s)
 // copied from ACE
 var _REGEX_WORDCHAR = /[\u0030-\u0039\u0041-\u005A\u0061-\u007A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u1FFF\u3040-\u9FFF\uF900-\uFDFF\uFE70-\uFEFE\uFF10-\uFF19\uFF21-\uFF3A\uFF41-\uFF5A\uFF66-\uFFDC]/;
 var _REGEX_SPACE = /\s/;
-var _REGEX_URLCHAR = new RegExp('(' + /[-:@a-zA-Z0-9_.,~%+\/\\?=&#;()$]/.source + '|' + _REGEX_WORDCHAR.source + ')');
-var _REGEX_URL = new RegExp(/(?:(?:https?|s?ftp|ftps|file|smb|afp|nfs|(x-)?man|gopher|txmt):\/\/|mailto:)/.source + _REGEX_URLCHAR.source + '*(?![:.,;])' + _REGEX_URLCHAR.source, 'g');
-
-// returns null if no URLs, or [[startIndex1, url1], [startIndex2, url2], ...]
-
-function _findURLs(text)
-{
-  _REGEX_URL.lastIndex = 0;
-  var urls = null;
-  var execResult;
-  while ((execResult = _REGEX_URL.exec(text)))
-  {
-    urls = (urls || []);
-    var startIndex = execResult.index;
-    var url = execResult[0];
-    urls.push([startIndex, url]);
-  }
-
-  return urls;
-}
-

From 626ee97669767ea0f18fc41a99591211c35238ec Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 10 Feb 2013 19:36:46 +0000
Subject: [PATCH 04/39] kinda brutal way of stopping plugins being able to pass
 *s instead of attributes

---
 src/node/db/PadManager.js   | 4 ++--
 src/node/utils/ExportTxt.js | 4 ++++
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/src/node/db/PadManager.js b/src/node/db/PadManager.js
index 5e0af4643..2be9da369 100644
--- a/src/node/db/PadManager.js
+++ b/src/node/db/PadManager.js
@@ -146,12 +146,12 @@ exports.getPad = function(id, text, callback)
   else
   {
     pad = new Pad(id);
-    
+
     //initalize the pad
     pad.init(text, function(err)
     {
       if(ERR(err, callback)) return;
-      
+      console.warn(pad); 
       globalPads.set(id, pad);
       callback(null, pad);
     });
diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 462583d3a..c236df479 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -236,6 +236,10 @@ function getTXTFromAtext(pad, atext, authorColors)
         //removes the characters with the code 12. Don't know where they come 
         //from but they break the abiword parser and are completly useless
         s = s.replace(String.fromCharCode(12), "");
+
+        // remove * from s, it's just not needed on a blank line..  This stops
+        // plugins from being able to display * at the beginning of a line
+        s = s.replace("*", "");
         
         // assem.append(_encodeWhitespace(Security.escapeHTML(s)));
         assem.append(_encodeWhitespace(s));

From bcf9c23b4e9b52bd86e4b7936b713d71c7cd3e93 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 10 Feb 2013 19:38:16 +0000
Subject: [PATCH 05/39] dont use HTML filter hooks on txt export

---
 src/node/utils/ExportTxt.js | 18 ------------------
 1 file changed, 18 deletions(-)

diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index c236df479..9451fd6e8 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -290,24 +290,6 @@ function getTXTFromAtext(pad, atext, authorColors)
     }else{
       pieces.push(lineContent, '\n');
     }
-
-    // I'm not too keen about using teh HTML export filters here, they could cause real pain in the future
-    // I'd suggest supporting getLineTXTForExport
-    var lineContentFromHook = hooks.callAllStr("getLineHTMLForExport", 
-    {
-      line: line,
-      apool: apool,
-      attribLine: attribLines[i],
-      text: textLines[i]
-    }, " ", " ", "");
-    if (lineContentFromHook)
-    {
-      pieces.push(lineContentFromHook, '');
-    } 
-    else 
-    {
-      // pieces.push(lineContent, '\n');
-    }
   }
 
   return pieces.join('');

From 28f6d50011abd7c12b2b3948b14bd547d2ef4aee Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 10 Feb 2013 22:14:05 +0000
Subject: [PATCH 06/39] remove console warn

---
 src/node/db/PadManager.js | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/node/db/PadManager.js b/src/node/db/PadManager.js
index 2be9da369..7d546fc71 100644
--- a/src/node/db/PadManager.js
+++ b/src/node/db/PadManager.js
@@ -151,7 +151,6 @@ exports.getPad = function(id, text, callback)
     pad.init(text, function(err)
     {
       if(ERR(err, callback)) return;
-      console.warn(pad); 
       globalPads.set(id, pad);
       callback(null, pad);
     });

From 60ef5f210ae47529325abf0968966ddf8348bfa7 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 10 Feb 2013 23:41:14 +0000
Subject: [PATCH 07/39] remove duplicate code to the best of my ability right
 now

---
 src/node/utils/ExportHtml.js |   7 +-
 src/node/utils/ExportTxt.js  | 145 +++--------------------------------
 2 files changed, 15 insertions(+), 137 deletions(-)

diff --git a/src/node/utils/ExportHtml.js b/src/node/utils/ExportHtml.js
index 069194880..74b6546f1 100644
--- a/src/node/utils/ExportHtml.js
+++ b/src/node/utils/ExportHtml.js
@@ -92,6 +92,7 @@ function getPadHTML(pad, revNum, callback)
 
 exports.getPadHTML = getPadHTML;
 exports.getHTMLFromAtext = getHTMLFromAtext;
+exports.getPadPlainText = getPadPlainText;
 
 function getHTMLFromAtext(pad, atext, authorColors)
 {
@@ -542,6 +543,8 @@ function _analyzeLine(text, aline, apool)
   return line;
 }
 
+exports._analyzeLine = _analyzeLine;
+
 exports.getPadHTMLDocument = function (padId, revNum, noDocType, callback)
 {
   padManager.getPad(padId, function (err, pad)
@@ -584,10 +587,9 @@ function _encodeWhitespace(s) {
     return "&#" +c.charCodeAt(0) + ";"
   });
 }
+exports._encodeWhitespace = _encodeWhitespace;
 
 // copied from ACE
-
-
 function _processSpaces(s)
 {
   var doesWrap = true;
@@ -651,6 +653,7 @@ function _processSpaces(s)
   return parts.join('');
 }
 
+exports._processSpaces = _processSpaces;
 
 // copied from ACE
 var _REGEX_WORDCHAR = /[\u0030-\u0039\u0041-\u005A\u0061-\u007A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u1FFF\u3040-\u9FFF\uF900-\uFDFF\uFE70-\uFEFE\uFF10-\uFF19\uFF21-\uFF3A\uFF41-\uFF5A\uFF66-\uFFDC]/;
diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 9451fd6e8..edcccfd77 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -21,32 +21,12 @@ var padManager = require("../db/PadManager");
 var ERR = require("async-stacktrace");
 var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
-function getPadPlainText(pad, revNum)
-{
-  var atext = ((revNum !== undefined) ? pad.getInternalRevisionAText(revNum) : pad.atext());
-  var textLines = atext.text.slice(0, -1).split('\n');
-  var attribLines = Changeset.splitAttributionLines(atext.attribs, atext.text);
-  var apool = pad.pool();
-
-  var pieces = [];
-  for (var i = 0; i < textLines.length; i++)
-  {
-    var line = _analyzeLine(textLines[i], attribLines[i], apool);
-    if (line.listLevel)
-    {
-      var numSpaces = line.listLevel * 2 - 1;
-      var bullet = '*';
-      pieces.push(new Array(numSpaces + 1).join(' '), bullet, ' ', line.text, '\n');
-    }
-    else
-    {
-      pieces.push(line.text, '\n');
-    }
-  }
-
-  return pieces.join('');
-}
+var getPadPlainText = require('./ExportHtml').getPadPlainText;
+var _processSpaces = require('./ExportHtml')._processSpaces;
+var _analyzeLine = require('./ExportHtml')._analyzeLine;
+var _encodeWhitespace = require('./ExportHtml')._encodeWhitespace;
 
+// This is slightly different than the HTML method as it passes the output to getTXTFromAText
 function getPadTXT(pad, revNum, callback)
 {
   var atext = pad.atext;
@@ -77,7 +57,7 @@ function getPadTXT(pad, revNum, callback)
 
   function (callback)
   {
-    html = getTXTFromAtext(pad, atext);
+    html = getTXTFromAtext(pad, atext); // only this line is different to the HTML function
     callback(null);
   }],
   // run final callback
@@ -91,8 +71,10 @@ function getPadTXT(pad, revNum, callback)
 }
 
 exports.getPadTXT = getPadTXT;
-exports.getTXTFromAtext = getTXTFromAtext;
 
+
+// This is different than the functionality provided in ExportHtml as it provides formatting
+// functionality that is designed specifically for TXT exports
 function getTXTFromAtext(pad, atext, authorColors)
 {
   var apool = pad.apool();
@@ -294,45 +276,7 @@ function getTXTFromAtext(pad, atext, authorColors)
 
   return pieces.join('');
 }
-
-function _analyzeLine(text, aline, apool)
-{
-  var line = {};
-
-  // identify list
-  var lineMarker = 0;
-  line.listLevel = 0;
-  if (aline)
-  {
-    var opIter = Changeset.opIterator(aline);
-    if (opIter.hasNext())
-    {
-      var listType = Changeset.opAttributeValue(opIter.next(), 'list', apool);
-      if (listType)
-      {
-        lineMarker = 1;
-        listType = /([a-z]+)([12345678])/.exec(listType);
-        if (listType)
-        {
-          line.listTypeName = listType[1];
-          line.listLevel = Number(listType[2]);
-        }
-      }
-    }
-  }
-  if (lineMarker)
-  {
-    line.text = text.substring(1);
-    line.aline = Changeset.subattribution(aline, 1);
-  }
-  else
-  {
-    line.text = text;
-    line.aline = aline;
-  }
-
-  return line;
-}
+exports.getTXTFromAtext = getTXTFromAtext;
 
 exports.getPadTXTDocument = function (padId, revNum, noDocType, callback)
 {
@@ -354,72 +298,3 @@ function _encodeWhitespace(s) {
     return "&#" +c.charCodeAt(0) + ";"
   });
 }
-
-// copied from ACE
-function _processSpaces(s)
-{
-  var doesWrap = true;
-  if (s.indexOf("<") < 0 && !doesWrap)
-  {
-    // short-cut
-    return s.replace(/ /g, '&nbsp;');
-  }
-  var parts = [];
-  s.replace(/<[^>]*>?| |[^ <]+/g, function (m)
-  {
-    parts.push(m);
-  });
-  if (doesWrap)
-  {
-    var endOfLine = true;
-    var beforeSpace = false;
-    // last space in a run is normal, others are nbsp,
-    // end of line is nbsp
-    for (var i = parts.length - 1; i >= 0; i--)
-    {
-      var p = parts[i];
-      if (p == " ")
-      {
-        if (endOfLine || beforeSpace) parts[i] = '&nbsp;';
-        endOfLine = false;
-        beforeSpace = true;
-      }
-      else if (p.charAt(0) != "<")
-      {
-        endOfLine = false;
-        beforeSpace = false;
-      }
-    }
-    // beginning of line is nbsp
-    for (var i = 0; i < parts.length; i++)
-    {
-      var p = parts[i];
-      if (p == " ")
-      {
-        parts[i] = '&nbsp;';
-        break;
-      }
-      else if (p.charAt(0) != "<")
-      {
-        break;
-      }
-    }
-  }
-  else
-  {
-    for (var i = 0; i < parts.length; i++)
-    {
-      var p = parts[i];
-      if (p == " ")
-      {
-        parts[i] = '&nbsp;';
-      }
-    }
-  }
-  return parts.join('');
-}
-
-
-// copied from ACE
-var _REGEX_WORDCHAR = /[\u0030-\u0039\u0041-\u005A\u0061-\u007A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u1FFF\u3040-\u9FFF\uF900-\uFDFF\uFE70-\uFEFE\uFF10-\uFF19\uFF21-\uFF3A\uFF41-\uFF5A\uFF66-\uFFDC]/;
-var _REGEX_SPACE = /\s/;

From 0b5c948549a03e858302bb9740529741c3be4e22 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Tue, 12 Feb 2013 19:45:46 +0000
Subject: [PATCH 08/39] Move code from Html export to a Helper file

---
 src/node/utils/ExportHelper.js | 139 +++++++++++++++++++++++++++++++
 src/node/utils/ExportHtml.js   | 144 +--------------------------------
 src/node/utils/ExportTxt.js    |   8 +-
 3 files changed, 147 insertions(+), 144 deletions(-)
 create mode 100644 src/node/utils/ExportHelper.js

diff --git a/src/node/utils/ExportHelper.js b/src/node/utils/ExportHelper.js
new file mode 100644
index 000000000..030b0dc75
--- /dev/null
+++ b/src/node/utils/ExportHelper.js
@@ -0,0 +1,139 @@
+/**
+ * Helpers for export requests
+ */
+
+/*
+ * 2011 Peter 'Pita' Martischka (Primary Technology Ltd)
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS-IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+var async = require("async");
+var Changeset = require("ep_etherpad-lite/static/js/Changeset");
+var padManager = require("../db/PadManager");
+var ERR = require("async-stacktrace");
+var Security = require('ep_etherpad-lite/static/js/security');
+var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
+
+exports.getPadPlainText = function(pad, revNum){
+  var atext = ((revNum !== undefined) ? pad.getInternalRevisionAText(revNum) : pad.atext());
+  var textLines = atext.text.slice(0, -1).split('\n');
+  var attribLines = Changeset.splitAttributionLines(atext.attribs, atext.text);
+  var apool = pad.pool();
+
+  var pieces = [];
+  for (var i = 0; i < textLines.length; i++){
+    var line = _analyzeLine(textLines[i], attribLines[i], apool);
+    if (line.listLevel){
+      var numSpaces = line.listLevel * 2 - 1;
+      var bullet = '*';
+      pieces.push(new Array(numSpaces + 1).join(' '), bullet, ' ', line.text, '\n');
+    }
+    else{
+      pieces.push(line.text, '\n');
+    }
+  }
+
+  return pieces.join('');
+}
+
+// copied from ACE
+exports._processSpaces = function(s){
+  var doesWrap = true;
+  if (s.indexOf("<") < 0 && !doesWrap){
+    // short-cut
+    return s.replace(/ /g, '&nbsp;');
+  }
+  var parts = [];
+  s.replace(/<[^>]*>?| |[^ <]+/g, function (m){
+    parts.push(m);
+  });
+  if (doesWrap){
+    var endOfLine = true;
+    var beforeSpace = false;
+    // last space in a run is normal, others are nbsp,
+    // end of line is nbsp
+    for (var i = parts.length - 1; i >= 0; i--){
+      var p = parts[i];
+      if (p == " "){
+        if (endOfLine || beforeSpace) parts[i] = '&nbsp;';
+        endOfLine = false;
+        beforeSpace = true;
+      }
+      else if (p.charAt(0) != "<"){
+        endOfLine = false;
+        beforeSpace = false;
+      }
+    }
+    // beginning of line is nbsp
+    for (var i = 0; i < parts.length; i++){
+      var p = parts[i];
+      if (p == " "){
+        parts[i] = '&nbsp;';
+        break;
+      }
+      else if (p.charAt(0) != "<"){
+        break;
+      }
+    }
+  }
+  else
+  {
+    for (var i = 0; i < parts.length; i++){
+      var p = parts[i];
+      if (p == " "){
+        parts[i] = '&nbsp;';
+      }
+    }
+  }
+  return parts.join('');
+}
+
+
+exports._analyzeLine = function(text, aline, apool){
+  var line = {};
+
+  // identify list
+  var lineMarker = 0;
+  line.listLevel = 0;
+  if (aline){
+    var opIter = Changeset.opIterator(aline);
+    if (opIter.hasNext()){
+      var listType = Changeset.opAttributeValue(opIter.next(), 'list', apool);
+      if (listType){
+        lineMarker = 1;
+        listType = /([a-z]+)([12345678])/.exec(listType);
+        if (listType){
+          line.listTypeName = listType[1];
+          line.listLevel = Number(listType[2]);
+        }
+      }
+    }
+  }
+  if (lineMarker){
+    line.text = text.substring(1);
+    line.aline = Changeset.subattribution(aline, 1);
+  }
+  else{
+    line.text = text;
+    line.aline = aline;
+  }
+  return line;
+}
+
+
+exports._encodeWhitespace = function(s){
+  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c){
+    return "&#" +c.charCodeAt(0) + ";"
+  });
+}
diff --git a/src/node/utils/ExportHtml.js b/src/node/utils/ExportHtml.js
index 74b6546f1..a54f566c2 100644
--- a/src/node/utils/ExportHtml.js
+++ b/src/node/utils/ExportHtml.js
@@ -21,31 +21,10 @@ var padManager = require("../db/PadManager");
 var ERR = require("async-stacktrace");
 var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
-function getPadPlainText(pad, revNum)
-{
-  var atext = ((revNum !== undefined) ? pad.getInternalRevisionAText(revNum) : pad.atext());
-  var textLines = atext.text.slice(0, -1).split('\n');
-  var attribLines = Changeset.splitAttributionLines(atext.attribs, atext.text);
-  var apool = pad.pool();
-
-  var pieces = [];
-  for (var i = 0; i < textLines.length; i++)
-  {
-    var line = _analyzeLine(textLines[i], attribLines[i], apool);
-    if (line.listLevel)
-    {
-      var numSpaces = line.listLevel * 2 - 1;
-      var bullet = '*';
-      pieces.push(new Array(numSpaces + 1).join(' '), bullet, ' ', line.text, '\n');
-    }
-    else
-    {
-      pieces.push(line.text, '\n');
-    }
-  }
-
-  return pieces.join('');
-}
+var getPadPlainText = require('./ExportHelper').getPadPlainText
+var _processSpaces = require('./ExportHelper')._processSpaces;
+var _analyzeLine = require('./ExportHelper')._analyzeLine;
+var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
 function getPadHTML(pad, revNum, callback)
 {
@@ -92,7 +71,6 @@ function getPadHTML(pad, revNum, callback)
 
 exports.getPadHTML = getPadHTML;
 exports.getHTMLFromAtext = getHTMLFromAtext;
-exports.getPadPlainText = getPadPlainText;
 
 function getHTMLFromAtext(pad, atext, authorColors)
 {
@@ -504,47 +482,6 @@ function getHTMLFromAtext(pad, atext, authorColors)
   return pieces.join('');
 }
 
-function _analyzeLine(text, aline, apool)
-{
-  var line = {};
-
-  // identify list
-  var lineMarker = 0;
-  line.listLevel = 0;
-  if (aline)
-  {
-    var opIter = Changeset.opIterator(aline);
-    if (opIter.hasNext())
-    {
-      var listType = Changeset.opAttributeValue(opIter.next(), 'list', apool);
-      if (listType)
-      {
-        lineMarker = 1;
-        listType = /([a-z]+)([12345678])/.exec(listType);
-        if (listType)
-        {
-          line.listTypeName = listType[1];
-          line.listLevel = Number(listType[2]);
-        }
-      }
-    }
-  }
-  if (lineMarker)
-  {
-    line.text = text.substring(1);
-    line.aline = Changeset.subattribution(aline, 1);
-  }
-  else
-  {
-    line.text = text;
-    line.aline = aline;
-  }
-
-  return line;
-}
-
-exports._analyzeLine = _analyzeLine;
-
 exports.getPadHTMLDocument = function (padId, revNum, noDocType, callback)
 {
   padManager.getPad(padId, function (err, pad)
@@ -581,79 +518,6 @@ exports.getPadHTMLDocument = function (padId, revNum, noDocType, callback)
   });
 }
 
-function _encodeWhitespace(s) {
-  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c)
-  {
-    return "&#" +c.charCodeAt(0) + ";"
-  });
-}
-exports._encodeWhitespace = _encodeWhitespace;
-
-// copied from ACE
-function _processSpaces(s)
-{
-  var doesWrap = true;
-  if (s.indexOf("<") < 0 && !doesWrap)
-  {
-    // short-cut
-    return s.replace(/ /g, '&nbsp;');
-  }
-  var parts = [];
-  s.replace(/<[^>]*>?| |[^ <]+/g, function (m)
-  {
-    parts.push(m);
-  });
-  if (doesWrap)
-  {
-    var endOfLine = true;
-    var beforeSpace = false;
-    // last space in a run is normal, others are nbsp,
-    // end of line is nbsp
-    for (var i = parts.length - 1; i >= 0; i--)
-    {
-      var p = parts[i];
-      if (p == " ")
-      {
-        if (endOfLine || beforeSpace) parts[i] = '&nbsp;';
-        endOfLine = false;
-        beforeSpace = true;
-      }
-      else if (p.charAt(0) != "<")
-      {
-        endOfLine = false;
-        beforeSpace = false;
-      }
-    }
-    // beginning of line is nbsp
-    for (var i = 0; i < parts.length; i++)
-    {
-      var p = parts[i];
-      if (p == " ")
-      {
-        parts[i] = '&nbsp;';
-        break;
-      }
-      else if (p.charAt(0) != "<")
-      {
-        break;
-      }
-    }
-  }
-  else
-  {
-    for (var i = 0; i < parts.length; i++)
-    {
-      var p = parts[i];
-      if (p == " ")
-      {
-        parts[i] = '&nbsp;';
-      }
-    }
-  }
-  return parts.join('');
-}
-
-exports._processSpaces = _processSpaces;
 
 // copied from ACE
 var _REGEX_WORDCHAR = /[\u0030-\u0039\u0041-\u005A\u0061-\u007A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u1FFF\u3040-\u9FFF\uF900-\uFDFF\uFE70-\uFEFE\uFF10-\uFF19\uFF21-\uFF3A\uFF41-\uFF5A\uFF66-\uFFDC]/;
diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index edcccfd77..0f3b1a634 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -21,10 +21,10 @@ var padManager = require("../db/PadManager");
 var ERR = require("async-stacktrace");
 var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
-var getPadPlainText = require('./ExportHtml').getPadPlainText;
-var _processSpaces = require('./ExportHtml')._processSpaces;
-var _analyzeLine = require('./ExportHtml')._analyzeLine;
-var _encodeWhitespace = require('./ExportHtml')._encodeWhitespace;
+var getPadPlainText = require('./ExportHelper').getPadPlainText;
+var _processSpaces = require('./ExportHelper')._processSpaces;
+var _analyzeLine = require('./ExportHelper')._analyzeLine;
+var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
 // This is slightly different than the HTML method as it passes the output to getTXTFromAText
 function getPadTXT(pad, revNum, callback)

From da246d183daa1a5a012937f0dc0f54ca96873553 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Tue, 12 Feb 2013 19:47:53 +0000
Subject: [PATCH 09/39] Correct license header

---
 src/node/utils/ExportTxt.js | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 0f3b1a634..30673a981 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -1,5 +1,9 @@
 /**
- * Copyright 2009 Google Inc.
+ * TXT export
+ */
+
+/*
+ * 2013 John McLear
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -14,7 +18,6 @@
  * limitations under the License.
  */
 
-
 var async = require("async");
 var Changeset = require("ep_etherpad-lite/static/js/Changeset");
 var padManager = require("../db/PadManager");

From e855bafdf9e5d6217d929e1f4d1933e673e3acac Mon Sep 17 00:00:00 2001
From: Manuel Knitza <manuel.knitza@uni.kn>
Date: Tue, 12 Feb 2013 21:47:40 +0100
Subject: [PATCH 10/39] Update src/node/hooks/express/apicalls.js

---
 src/node/hooks/express/apicalls.js | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/src/node/hooks/express/apicalls.js b/src/node/hooks/express/apicalls.js
index e57e1d350..0971a877b 100644
--- a/src/node/hooks/express/apicalls.js
+++ b/src/node/hooks/express/apicalls.js
@@ -57,4 +57,9 @@ exports.expressCreateServer = function (hook_name, args, cb) {
       res.end("OK");
     });
   });
+  
+  //Provide a possibility to query the latest available API version
+  args.app.get('/api', function (req, res) {
+     res.json({"currentVersion" : apiHandler.latestApiVersion});
+  });
 }

From 8c9ad6ee504bc1aa8d057118354a4eaacb077a50 Mon Sep 17 00:00:00 2001
From: Manuel Knitza <manuel.knitza@uni.kn>
Date: Tue, 12 Feb 2013 21:50:14 +0100
Subject: [PATCH 11/39] Update src/node/handler/APIHandler.js

---
 src/node/handler/APIHandler.js | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/node/handler/APIHandler.js b/src/node/handler/APIHandler.js
index 9f86277a0..8be5b5fe3 100644
--- a/src/node/handler/APIHandler.js
+++ b/src/node/handler/APIHandler.js
@@ -216,6 +216,9 @@ var version =
   }
 };
 
+// set the latest available API version here
+exports.latestApiVersion = '1.2.7';
+
 /**
  * Handles a HTTP API call
  * @param functionName the name of the called function

From fde10052ca5d9c80bc0837e00bdd87a2eeff4c72 Mon Sep 17 00:00:00 2001
From: Manuel Knitza <manuel.knitza@uni.kn>
Date: Tue, 12 Feb 2013 21:57:01 +0100
Subject: [PATCH 12/39] Update doc/api/http_api.md

Updated
---
 doc/api/http_api.md | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/doc/api/http_api.md b/doc/api/http_api.md
index 0543ef71a..7e05ff86d 100644
--- a/doc/api/http_api.md
+++ b/doc/api/http_api.md
@@ -61,7 +61,9 @@ Portal submits content into new blog post
 ## Usage
 
 ### API version
-The latest version is `1.2`
+The latest version is `1.2.7`
+
+The current version can be queried via /api.
 
 ### Request Format
 

From a4e5adab3e42d6436d89a1155ecf303dabc2dd9d Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 13 Feb 2013 02:47:24 +0000
Subject: [PATCH 13/39] Update Ueber DB to fix MySQL Bug!

---
 src/package.json | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/package.json b/src/package.json
index 03df87f9b..5b21334d8 100644
--- a/src/package.json
+++ b/src/package.json
@@ -16,7 +16,7 @@
                       "require-kernel"          : "1.0.5",
                       "resolve"                 : "0.2.x",
                       "socket.io"               : "0.9.x",
-                      "ueberDB"                 : "0.1.91",
+                      "ueberDB"                 : "0.1.92",
                       "async"                   : "0.1.x",
                       "express"                 : "3.x",
                       "connect"                 : "2.4.x",

From 500f9b8b48cb774c7ce73bf15a73c862fe0c4f11 Mon Sep 17 00:00:00 2001
From: Mark Fisher <eldiddio@gmail.com>
Date: Wed, 13 Feb 2013 15:25:27 +0000
Subject: [PATCH 14/39] Fixed typo thats made server hang

---
 src/node/handler/PadMessageHandler.js | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/node/handler/PadMessageHandler.js b/src/node/handler/PadMessageHandler.js
index 7cd945af3..eaf24b7a2 100644
--- a/src/node/handler/PadMessageHandler.js
+++ b/src/node/handler/PadMessageHandler.js
@@ -1429,7 +1429,7 @@ function composePadChangesets(padId, startNum, endNum, callback)
  */
 exports.padUsersCount = function (padID, callback) {
   callback(null, {
-    padUsersCount: socketio.sockets.clients(padId).length
+    padUsersCount: socketio.sockets.clients(padID).length
   });
 }
 

From d3f730e2ba060b0d2b26d03b55088f266fefb1b9 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 13 Feb 2013 18:01:15 +0000
Subject: [PATCH 15/39] fix various issues dont stop random *'s appearing

---
 src/node/utils/ExportHelper.js | 52 --------------------------------
 src/node/utils/ExportHtml.js   | 55 +++++++++++++++++++++++++++++++++-
 src/node/utils/ExportTxt.js    | 19 +++++-------
 3 files changed, 62 insertions(+), 64 deletions(-)

diff --git a/src/node/utils/ExportHelper.js b/src/node/utils/ExportHelper.js
index 030b0dc75..a939a8b6e 100644
--- a/src/node/utils/ExportHelper.js
+++ b/src/node/utils/ExportHelper.js
@@ -47,58 +47,6 @@ exports.getPadPlainText = function(pad, revNum){
   return pieces.join('');
 }
 
-// copied from ACE
-exports._processSpaces = function(s){
-  var doesWrap = true;
-  if (s.indexOf("<") < 0 && !doesWrap){
-    // short-cut
-    return s.replace(/ /g, '&nbsp;');
-  }
-  var parts = [];
-  s.replace(/<[^>]*>?| |[^ <]+/g, function (m){
-    parts.push(m);
-  });
-  if (doesWrap){
-    var endOfLine = true;
-    var beforeSpace = false;
-    // last space in a run is normal, others are nbsp,
-    // end of line is nbsp
-    for (var i = parts.length - 1; i >= 0; i--){
-      var p = parts[i];
-      if (p == " "){
-        if (endOfLine || beforeSpace) parts[i] = '&nbsp;';
-        endOfLine = false;
-        beforeSpace = true;
-      }
-      else if (p.charAt(0) != "<"){
-        endOfLine = false;
-        beforeSpace = false;
-      }
-    }
-    // beginning of line is nbsp
-    for (var i = 0; i < parts.length; i++){
-      var p = parts[i];
-      if (p == " "){
-        parts[i] = '&nbsp;';
-        break;
-      }
-      else if (p.charAt(0) != "<"){
-        break;
-      }
-    }
-  }
-  else
-  {
-    for (var i = 0; i < parts.length; i++){
-      var p = parts[i];
-      if (p == " "){
-        parts[i] = '&nbsp;';
-      }
-    }
-  }
-  return parts.join('');
-}
-
 
 exports._analyzeLine = function(text, aline, apool){
   var line = {};
diff --git a/src/node/utils/ExportHtml.js b/src/node/utils/ExportHtml.js
index a54f566c2..585694d4b 100644
--- a/src/node/utils/ExportHtml.js
+++ b/src/node/utils/ExportHtml.js
@@ -22,7 +22,6 @@ var ERR = require("async-stacktrace");
 var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
 var getPadPlainText = require('./ExportHelper').getPadPlainText
-var _processSpaces = require('./ExportHelper')._processSpaces;
 var _analyzeLine = require('./ExportHelper')._analyzeLine;
 var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
@@ -543,3 +542,57 @@ function _findURLs(text)
 
   return urls;
 }
+
+
+// copied from ACE
+function _processSpaces(s){
+  var doesWrap = true;
+  if (s.indexOf("<") < 0 && !doesWrap){
+    // short-cut
+    return s.replace(/ /g, '&nbsp;');
+  }
+  var parts = [];
+  s.replace(/<[^>]*>?| |[^ <]+/g, function (m){
+    parts.push(m);
+  });
+  if (doesWrap){
+    var endOfLine = true;
+    var beforeSpace = false;
+    // last space in a run is normal, others are nbsp,
+    // end of line is nbsp
+    for (var i = parts.length - 1; i >= 0; i--){
+      var p = parts[i];
+      if (p == " "){
+        if (endOfLine || beforeSpace) parts[i] = '&nbsp;';
+        endOfLine = false;
+        beforeSpace = true;
+      }
+      else if (p.charAt(0) != "<"){
+        endOfLine = false;
+        beforeSpace = false;
+      }
+    }
+    // beginning of line is nbsp
+    for (var i = 0; i < parts.length; i++){
+      var p = parts[i];
+      if (p == " "){
+        parts[i] = '&nbsp;';
+        break;
+      }
+      else if (p.charAt(0) != "<"){
+        break;
+      }
+    }
+  }
+  else
+  {
+    for (var i = 0; i < parts.length; i++){
+      var p = parts[i];
+      if (p == " "){
+        parts[i] = '&nbsp;';
+      }
+    }
+  }
+  return parts.join('');
+}
+
diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 30673a981..05847f162 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -25,7 +25,6 @@ var ERR = require("async-stacktrace");
 var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
 var getPadPlainText = require('./ExportHelper').getPadPlainText;
-var _processSpaces = require('./ExportHelper')._processSpaces;
 var _analyzeLine = require('./ExportHelper')._analyzeLine;
 var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
@@ -204,12 +203,12 @@ function getTXTFromAtext(pad, atext, authorColors)
           {
             if (propVals[i] === ENTER || propVals[i] === STAY)
             {
-              emitOpenTag(i);
               propVals[i] = true;
             }
           }
           // propVals is now all {true,false} again
         } // end if (propChanged)
+
         var chars = o.chars;
         if (o.lines)
         {
@@ -217,16 +216,15 @@ function getTXTFromAtext(pad, atext, authorColors)
         }
         
         var s = taker.take(chars);
-        
-        //removes the characters with the code 12. Don't know where they come 
-        //from but they break the abiword parser and are completly useless
-        s = s.replace(String.fromCharCode(12), "");
+
+        // removes the characters with the code 12. Don't know where they come 
+        // from but they break the abiword parser and are completly useless
+        // s = s.replace(String.fromCharCode(12), "");
 
         // remove * from s, it's just not needed on a blank line..  This stops
         // plugins from being able to display * at the beginning of a line
-        s = s.replace("*", "");
-        
-        // assem.append(_encodeWhitespace(Security.escapeHTML(s)));
+        // s = s.replace("*", ""); // Then remove it
+
         assem.append(_encodeWhitespace(s));
       } // end iteration over spans in line
       
@@ -242,8 +240,7 @@ function getTXTFromAtext(pad, atext, authorColors)
       
     } // end processNextChars
     processNextChars(text.length - idx);
-
-    return _processSpaces(assem.toString());
+    return(assem.toString());
   } // end getLineHTML
   var pieces = [css];
 

From be56272e66f4921f68fe12cd05cbaad8eb54894b Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 13 Feb 2013 18:30:55 +0000
Subject: [PATCH 16/39] allow non ascii chars in txt export

---
 src/node/utils/ExportHelper.js |   6 +-
 src/node/utils/ExportHtml.js   |   7 +-
 src/node/utils/ExportTxt.js    |  10 +-
 src/node/utils/padDiffHTML.js  | 554 +++++++++++++++++++++++++++++++++
 4 files changed, 562 insertions(+), 15 deletions(-)
 create mode 100644 src/node/utils/padDiffHTML.js

diff --git a/src/node/utils/ExportHelper.js b/src/node/utils/ExportHelper.js
index a939a8b6e..41d440f30 100644
--- a/src/node/utils/ExportHelper.js
+++ b/src/node/utils/ExportHelper.js
@@ -80,8 +80,4 @@ exports._analyzeLine = function(text, aline, apool){
 }
 
 
-exports._encodeWhitespace = function(s){
-  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c){
-    return "&#" +c.charCodeAt(0) + ";"
-  });
-}
+
diff --git a/src/node/utils/ExportHtml.js b/src/node/utils/ExportHtml.js
index 585694d4b..51a4b2c3d 100644
--- a/src/node/utils/ExportHtml.js
+++ b/src/node/utils/ExportHtml.js
@@ -23,7 +23,6 @@ var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
 var getPadPlainText = require('./ExportHelper').getPadPlainText
 var _analyzeLine = require('./ExportHelper')._analyzeLine;
-var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
 function getPadHTML(pad, revNum, callback)
 {
@@ -596,3 +595,9 @@ function _processSpaces(s){
   return parts.join('');
 }
 
+function _encodeWhitespace(s){
+  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c){
+    return "&#" +c.charCodeAt(0) + ";"
+  });
+}
+
diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 05847f162..4a3e458b4 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -26,7 +26,6 @@ var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
 var getPadPlainText = require('./ExportHelper').getPadPlainText;
 var _analyzeLine = require('./ExportHelper')._analyzeLine;
-var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
 // This is slightly different than the HTML method as it passes the output to getTXTFromAText
 function getPadTXT(pad, revNum, callback)
@@ -112,7 +111,6 @@ function getTXTFromAtext(pad, atext, authorColors)
     var taker = Changeset.stringIterator(text);
     var assem = Changeset.stringAssembler();
     var openTags = [];
-
     var idx = 0;
 
     function processNextChars(numChars)
@@ -225,7 +223,7 @@ function getTXTFromAtext(pad, atext, authorColors)
         // plugins from being able to display * at the beginning of a line
         // s = s.replace("*", ""); // Then remove it
 
-        assem.append(_encodeWhitespace(s));
+        assem.append(s);
       } // end iteration over spans in line
       
       var tags2close = [];
@@ -292,9 +290,3 @@ exports.getPadTXTDocument = function (padId, revNum, noDocType, callback)
   });
 }
 
-function _encodeWhitespace(s) {
-  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c)
-  {
-    return "&#" +c.charCodeAt(0) + ";"
-  });
-}
diff --git a/src/node/utils/padDiffHTML.js b/src/node/utils/padDiffHTML.js
new file mode 100644
index 000000000..00e967287
--- /dev/null
+++ b/src/node/utils/padDiffHTML.js
@@ -0,0 +1,554 @@
+var Changeset = require("../../static/js/Changeset");
+var async = require("async");
+var exportHtml = require('./ExportHtml');
+ 
+function padDiffHTML (pad, fromRev, toRev){      
+  //check parameters
+  if(!pad || !pad.id || !pad.atext || !pad.pool)
+  {
+    throw new Error('Invalid pad');
+  }
+ 
+  var range = pad.getValidRevisionRange(fromRev, toRev);
+  if(!range) { throw new Error('Invalid revision range.' + 
+      ' startRev: ' + fromRev +
+      ' endRev: ' + toRev); }
+  
+  this._pad = pad;
+  this._fromRev = range.startRev;
+  this._toRev = range.endRev;
+  this._html = null;
+  this._authors = [];
+}
+ 
+padDiffHTML.prototype._isClearAuthorship = function(changeset){
+  //unpack
+  var unpacked = Changeset.unpack(changeset);
+  
+  //check if there is nothing in the charBank
+  if(unpacked.charBank !== "")
+    return false;
+  
+  //check if oldLength == newLength
+  if(unpacked.oldLen !== unpacked.newLen)
+    return false;
+ 
+  //lets iterator over the operators
+  var iterator = Changeset.opIterator(unpacked.ops);
+  
+  //get the first operator, this should be a clear operator
+  var clearOperator = iterator.next();
+  
+  //check if there is only one operator
+  if(iterator.hasNext() === true)
+    return false;
+    
+  //check if this operator doesn't change text
+  if(clearOperator.opcode !== "=")
+    return false;
+    
+  //check that this operator applys to the complete text
+  //if the text ends with a new line, its exactly one character less, else it has the same length
+  if(clearOperator.chars !== unpacked.oldLen-1 && clearOperator.chars !== unpacked.oldLen)
+    return false;
+  
+  var attributes = [];
+  Changeset.eachAttribNumber(changeset, function(attrNum){
+    attributes.push(attrNum);
+  });
+    
+  //check that this changeset uses only one attribute
+  if(attributes.length !== 1)
+    return false;
+    
+  var appliedAttribute = this._pad.pool.getAttrib(attributes[0]);
+  
+  //check if the applied attribute is an anonymous author attribute
+  if(appliedAttribute[0] !== "author" || appliedAttribute[1] !== "")
+    return false;
+    
+  return true;
+}
+ 
+padDiffHTML.prototype._createClearAuthorship = function(rev, callback){
+  var self = this;
+  this._pad.getInternalRevisionAText(rev, function(err, atext){
+   if(err){
+     return callback(err);
+   }
+   
+   //build clearAuthorship changeset
+   var builder = Changeset.builder(atext.text.length);
+   builder.keepText(atext.text, [['author','']], self._pad.pool);
+   var changeset = builder.toString();
+   
+   callback(null, changeset);
+  });
+}
+ 
+padDiffHTML.prototype._createClearStartAtext = function(rev, callback){
+  var self = this;
+  
+  //get the atext of this revision
+  this._pad.getInternalRevisionAText(rev, function(err, atext){
+   if(err){
+     return callback(err);
+   }
+   
+   //create the clearAuthorship changeset
+   self._createClearAuthorship(rev, function(err, changeset){
+     if(err){
+       return callback(err);
+     }
+   
+     //apply the clearAuthorship changeset
+     var newAText = Changeset.applyToAText(changeset, atext, self._pad.pool);
+     
+     callback(null, newAText);
+   });
+  });
+}
+ 
+padDiffHTML.prototype._getChangesetsInBulk = function(startRev, count, callback) {
+  var self = this;
+  
+  //find out which revisions we need
+  var revisions = [];
+  for(var i=startRev;i<(startRev+count) && i<=this._pad.head;i++){
+    revisions.push(i);
+  }
+  
+  var changesets = [], authors = [];
+  
+  //get all needed revisions
+  async.forEach(revisions, function(rev, callback){
+    self._pad.getRevision(rev, function(err, revision){
+      if(err){
+        return callback(err)
+      }
+      
+      var arrayNum = rev-startRev;
+      
+      changesets[arrayNum] = revision.changeset;
+      authors[arrayNum] = revision.meta.author;
+  
+      callback();
+    });
+  }, function(err){
+    callback(err, changesets, authors);
+  });
+}
+ 
+padDiffHTML.prototype._addAuthors = function(authors) {
+  var self = this;
+  //add to array if not in the array
+  authors.forEach(function(author){
+    if(self._authors.indexOf(author) == -1){
+      self._authors.push(author);
+    }
+  });
+}
+ 
+padDiffHTML.prototype._createDiffAtext = function(callback) {
+  var self = this;
+  var bulkSize = 100;
+  
+  //get the cleaned startAText
+  self._createClearStartAtext(self._fromRev, function(err, atext){
+    if(err) { return callback(err); }
+    
+    var superChangeset = null;
+    
+    var rev = self._fromRev + 1;
+ 
+    //async while loop
+    async.whilst(
+      //loop condition
+      function () { return rev <= self._toRev; },
+      
+      //loop body
+      function (callback) {
+        //get the bulk
+        self._getChangesetsInBulk(rev,bulkSize,function(err, changesets, authors){
+          var addedAuthors = [];
+          
+          //run trough all changesets
+          for(var i=0;i<changesets.length && (rev+i)<=self._toRev;i++){
+            var changeset = changesets[i];
+            
+            //skip clearAuthorship Changesets
+            if(self._isClearAuthorship(changeset)){
+              continue;
+            }
+            
+            changeset = self._extendChangesetWithAuthor(changeset, authors[i], self._pad.pool);
+            
+            //add this author to the authorarray
+            addedAuthors.push(authors[i]);
+            
+            //compose it with the superChangset            
+            if(superChangeset === null){
+              superChangeset = changeset;
+            } else {           
+              superChangeset = Changeset.composeWithDeletions(superChangeset, changeset, self._pad.pool);
+            }
+          }
+          
+          //add the authors to the padDiffHTML authorArray
+          self._addAuthors(addedAuthors);
+        
+          //lets continue with the next bulk 
+          rev += bulkSize;
+          callback();
+        });
+      },
+      
+      //after the loop has ended
+      function (err) {
+        //if there are only clearAuthorship changesets, we don't get a superChangeset, so we can skip this step
+        if(superChangeset){
+          var deletionChangeset = self._createDeletionChangeset(superChangeset,atext,self._pad.pool);
+        
+          //apply the superChangeset, which includes all addings
+          atext = Changeset.applyToAText(superChangeset,atext,self._pad.pool);
+          //apply the deletionChangeset, which adds a deletions
+          atext = Changeset.applyToAText(deletionChangeset,atext,self._pad.pool);
+        }      
+ 
+        callback(err, atext);
+      }
+    );
+  });
+}
+ 
+padDiffHTML.prototype.getHtml = function(callback){
+  //cache the html
+  if(this._html != null){
+    return callback(null, this._html);
+  }
+  
+  var self = this;
+  var atext, html, authorColors;
+  
+  async.series([
+    //get the diff atext
+    function(callback){
+      self._createDiffAtext(function(err, _atext){
+        if(err){
+          return callback(err);
+        }
+        
+        atext = _atext;
+        callback();
+      });
+    },
+    //get the authorColor table
+    function(callback){
+      self._pad.getAllAuthorColors(function(err, _authorColors){
+        if(err){
+          return callback(err);
+        }
+        
+        authorColors = _authorColors;
+        callback();
+      });
+    },
+    //convert the atext to html
+    function(callback){
+      html = exportHtml.getHTMLFromAtext(self._pad, atext, authorColors);
+      self._html = html;
+      callback();
+    }
+  ], function(err){
+    callback(err, html);
+  });
+};
+ 
+padDiffHTML.prototype.getAuthors = function(callback){
+  var self = this;
+  
+  //check if html was already produced, if not produce it, this generates the author array at the same time
+  if(self._html == null){
+    self.getHtml(function(err){
+      if(err){
+        return callback(err);
+      }
+      
+      callback(null, self._authors);
+    });
+  } else {
+    callback(null, self._authors);
+  }
+}
+ 
+padDiffHTML.prototype._extendChangesetWithAuthor = function(changeset, author, apool) {  
+  //unpack
+  var unpacked = Changeset.unpack(changeset);  
+  
+  var iterator = Changeset.opIterator(unpacked.ops);
+  var assem = Changeset.opAssembler();
+  
+  //create deleted attribs
+  var authorAttrib = apool.putAttrib(["author", author || ""]);
+  var deletedAttrib = apool.putAttrib(["removed", true]);
+  var attribs = "*" + Changeset.numToString(authorAttrib) + "*" + Changeset.numToString(deletedAttrib);
+  
+  //iteratore over the operators of the changeset
+  while(iterator.hasNext()){
+    var operator = iterator.next();
+    
+    //this is a delete operator, extend it with the author
+    if(operator.opcode === "-"){
+      operator.attribs = attribs;
+    } 
+    //this is operator changes only attributes, let's mark which author did that
+    else if(operator.opcode === "=" && operator.attribs){
+      operator.attribs+="*"+Changeset.numToString(authorAttrib);
+    }
+    
+    //append the new operator to our assembler
+    assem.append(operator);
+  }
+  
+  //return the modified changeset
+  return Changeset.pack(unpacked.oldLen, unpacked.newLen, assem.toString(), unpacked.charBank);
+}
+ 
+//this method is 80% like Changeset.inverse. I just changed so instead of reverting, it adds deletions and attribute changes to to the atext.
+padDiffHTML.prototype._createDeletionChangeset = function(cs, startAText, apool) {
+  var lines = Changeset.splitTextLines(startAText.text);
+  var alines = Changeset.splitAttributionLines(startAText.attribs, startAText.text);
+  
+  // lines and alines are what the exports is meant to apply to.
+  // They may be arrays or objects with .get(i) and .length methods.
+  // They include final newlines on lines.
+ 
+  function lines_get(idx) {
+    if (lines.get) {
+      return lines.get(idx);
+    } else {
+      return lines[idx];
+    }
+  }
+ 
+  function lines_length() {
+    if ((typeof lines.length) == "number") {
+      return lines.length;
+    } else {
+      return lines.length();
+    }
+  }
+ 
+  function alines_get(idx) {
+    if (alines.get) {
+      return alines.get(idx);
+    } else {
+      return alines[idx];
+    }
+  }
+ 
+  function alines_length() {
+    if ((typeof alines.length) == "number") {
+      return alines.length;
+    } else {
+      return alines.length();
+    }
+  }
+ 
+  var curLine = 0;
+  var curChar = 0;
+  var curLineOpIter = null;
+  var curLineOpIterLine;
+  var curLineNextOp = Changeset.newOp('+');
+ 
+  var unpacked = Changeset.unpack(cs);
+  var csIter = Changeset.opIterator(unpacked.ops);
+  var builder = Changeset.builder(unpacked.newLen);
+ 
+  function consumeAttribRuns(numChars, func /*(len, attribs, endsLine)*/ ) {
+ 
+    if ((!curLineOpIter) || (curLineOpIterLine != curLine)) {
+      // create curLineOpIter and advance it to curChar
+      curLineOpIter = Changeset.opIterator(alines_get(curLine));
+      curLineOpIterLine = curLine;
+      var indexIntoLine = 0;
+      var done = false;
+      while (!done) {
+        curLineOpIter.next(curLineNextOp);
+        if (indexIntoLine + curLineNextOp.chars >= curChar) {
+          curLineNextOp.chars -= (curChar - indexIntoLine);
+          done = true;
+        } else {
+          indexIntoLine += curLineNextOp.chars;
+        }
+      }
+    }
+ 
+    while (numChars > 0) {
+      if ((!curLineNextOp.chars) && (!curLineOpIter.hasNext())) {
+        curLine++;
+        curChar = 0;
+        curLineOpIterLine = curLine;
+        curLineNextOp.chars = 0;
+        curLineOpIter = Changeset.opIterator(alines_get(curLine));
+      }
+      if (!curLineNextOp.chars) {
+        curLineOpIter.next(curLineNextOp);
+      }
+      var charsToUse = Math.min(numChars, curLineNextOp.chars);
+      func(charsToUse, curLineNextOp.attribs, charsToUse == curLineNextOp.chars && curLineNextOp.lines > 0);
+      numChars -= charsToUse;
+      curLineNextOp.chars -= charsToUse;
+      curChar += charsToUse;
+    }
+ 
+    if ((!curLineNextOp.chars) && (!curLineOpIter.hasNext())) {
+      curLine++;
+      curChar = 0;
+    }
+  }
+ 
+  function skip(N, L) {
+    if (L) {
+      curLine += L;
+      curChar = 0;
+    } else {
+      if (curLineOpIter && curLineOpIterLine == curLine) {
+        consumeAttribRuns(N, function () {});
+      } else {
+        curChar += N;
+      }
+    }
+  }
+ 
+  function nextText(numChars) {
+    var len = 0;
+    var assem = Changeset.stringAssembler();
+    var firstString = lines_get(curLine).substring(curChar);
+    len += firstString.length;
+    assem.append(firstString);
+ 
+    var lineNum = curLine + 1;
+    while (len < numChars) {
+      var nextString = lines_get(lineNum);
+      len += nextString.length;
+      assem.append(nextString);
+      lineNum++;
+    }
+ 
+    return assem.toString().substring(0, numChars);
+  }
+ 
+  function cachedStrFunc(func) {
+    var cache = {};
+    return function (s) {
+      if (!cache[s]) {
+        cache[s] = func(s);
+      }
+      return cache[s];
+    };
+  }
+ 
+  var attribKeys = [];
+  var attribValues = [];
+ 
+  //iterate over all operators of this changeset
+  while (csIter.hasNext()) {
+    var csOp = csIter.next();
+ 
+    if (csOp.opcode == '=') {      
+      var textBank = nextText(csOp.chars);
+ 
+      // decide if this equal operator is an attribution change or not. We can see this by checkinf if attribs is set.
+      // If the text this operator applies to is only a star, than this is a false positive and should be ignored
+      if (csOp.attribs && textBank != "*") {
+        var deletedAttrib = apool.putAttrib(["removed", true]);
+        var authorAttrib = apool.putAttrib(["author", ""]);;
+ 
+        attribKeys.length = 0;
+        attribValues.length = 0;
+        Changeset.eachAttribNumber(csOp.attribs, function (n) {
+          attribKeys.push(apool.getAttribKey(n));
+          attribValues.push(apool.getAttribValue(n));
+ 
+          if(apool.getAttribKey(n) === "author"){
+            authorAttrib = n;
+          };
+        });
+ 
+        var undoBackToAttribs = cachedStrFunc(function (attribs) {
+          var backAttribs = [];
+          for (var i = 0; i < attribKeys.length; i++) {
+            var appliedKey = attribKeys[i];
+            var appliedValue = attribValues[i];
+            var oldValue = Changeset.attribsAttributeValue(attribs, appliedKey, apool);
+            if (appliedValue != oldValue) {
+              backAttribs.push([appliedKey, oldValue]);
+            }
+          }
+          return Changeset.makeAttribsString('=', backAttribs, apool);
+        });
+ 
+        var oldAttribsAddition = "*" + Changeset.numToString(deletedAttrib) + "*" + Changeset.numToString(authorAttrib);
+ 
+        var textLeftToProcess = textBank;
+ 
+        while(textLeftToProcess.length > 0){
+          //process till the next line break or process only one line break
+          var lengthToProcess = textLeftToProcess.indexOf("\n");
+          var lineBreak = false;
+          switch(lengthToProcess){
+            case -1: 
+              lengthToProcess=textLeftToProcess.length;
+              break;
+            case 0:
+              lineBreak = true;
+              lengthToProcess=1;
+              break;
+          }
+ 
+          //get the text we want to procceed in this step
+          var processText = textLeftToProcess.substr(0, lengthToProcess);
+          textLeftToProcess = textLeftToProcess.substr(lengthToProcess);
+ 
+          if(lineBreak){
+            builder.keep(1, 1); //just skip linebreaks, don't do a insert + keep for a linebreak
+ 
+            //consume the attributes of this linebreak
+            consumeAttribRuns(1, function(){});
+          } else {
+            //add the old text via an insert, but add a deletion attribute + the author attribute of the author who deleted it
+            var textBankIndex = 0;
+            consumeAttribRuns(lengthToProcess, function (len, attribs, endsLine) {
+              //get the old attributes back
+              var attribs = (undoBackToAttribs(attribs) || "") + oldAttribsAddition;
+ 
+              builder.insert(processText.substr(textBankIndex, len), attribs);
+              textBankIndex += len;
+            });
+ 
+            builder.keep(lengthToProcess, 0);
+          }
+        }
+      } else {
+        skip(csOp.chars, csOp.lines);
+        builder.keep(csOp.chars, csOp.lines);
+      }
+    } else if (csOp.opcode == '+') {
+      builder.keep(csOp.chars, csOp.lines);
+    } else if (csOp.opcode == '-') {
+      var textBank = nextText(csOp.chars);
+      var textBankIndex = 0;
+      
+      consumeAttribRuns(csOp.chars, function (len, attribs, endsLine) {
+        builder.insert(textBank.substr(textBankIndex, len), attribs + csOp.attribs);
+        textBankIndex += len;
+      });
+    }
+  }
+ 
+  return Changeset.checkRep(builder.toString());
+};
+ 
+//export the constructor
+module.exports = padDiffHTML;

From dea892213e080b7637955413222384959b3b267f Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 13 Feb 2013 18:41:04 +0000
Subject: [PATCH 17/39] Revert "allow non ascii chars in txt export"

This reverts commit be56272e66f4921f68fe12cd05cbaad8eb54894b.
---
 src/node/utils/ExportHelper.js |   6 +-
 src/node/utils/ExportHtml.js   |   7 +-
 src/node/utils/ExportTxt.js    |  10 +-
 src/node/utils/padDiffHTML.js  | 554 ---------------------------------
 4 files changed, 15 insertions(+), 562 deletions(-)
 delete mode 100644 src/node/utils/padDiffHTML.js

diff --git a/src/node/utils/ExportHelper.js b/src/node/utils/ExportHelper.js
index 41d440f30..a939a8b6e 100644
--- a/src/node/utils/ExportHelper.js
+++ b/src/node/utils/ExportHelper.js
@@ -80,4 +80,8 @@ exports._analyzeLine = function(text, aline, apool){
 }
 
 
-
+exports._encodeWhitespace = function(s){
+  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c){
+    return "&#" +c.charCodeAt(0) + ";"
+  });
+}
diff --git a/src/node/utils/ExportHtml.js b/src/node/utils/ExportHtml.js
index 51a4b2c3d..585694d4b 100644
--- a/src/node/utils/ExportHtml.js
+++ b/src/node/utils/ExportHtml.js
@@ -23,6 +23,7 @@ var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
 var getPadPlainText = require('./ExportHelper').getPadPlainText
 var _analyzeLine = require('./ExportHelper')._analyzeLine;
+var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
 function getPadHTML(pad, revNum, callback)
 {
@@ -595,9 +596,3 @@ function _processSpaces(s){
   return parts.join('');
 }
 
-function _encodeWhitespace(s){
-  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c){
-    return "&#" +c.charCodeAt(0) + ";"
-  });
-}
-
diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 4a3e458b4..05847f162 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -26,6 +26,7 @@ var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
 var getPadPlainText = require('./ExportHelper').getPadPlainText;
 var _analyzeLine = require('./ExportHelper')._analyzeLine;
+var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
 // This is slightly different than the HTML method as it passes the output to getTXTFromAText
 function getPadTXT(pad, revNum, callback)
@@ -111,6 +112,7 @@ function getTXTFromAtext(pad, atext, authorColors)
     var taker = Changeset.stringIterator(text);
     var assem = Changeset.stringAssembler();
     var openTags = [];
+
     var idx = 0;
 
     function processNextChars(numChars)
@@ -223,7 +225,7 @@ function getTXTFromAtext(pad, atext, authorColors)
         // plugins from being able to display * at the beginning of a line
         // s = s.replace("*", ""); // Then remove it
 
-        assem.append(s);
+        assem.append(_encodeWhitespace(s));
       } // end iteration over spans in line
       
       var tags2close = [];
@@ -290,3 +292,9 @@ exports.getPadTXTDocument = function (padId, revNum, noDocType, callback)
   });
 }
 
+function _encodeWhitespace(s) {
+  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c)
+  {
+    return "&#" +c.charCodeAt(0) + ";"
+  });
+}
diff --git a/src/node/utils/padDiffHTML.js b/src/node/utils/padDiffHTML.js
deleted file mode 100644
index 00e967287..000000000
--- a/src/node/utils/padDiffHTML.js
+++ /dev/null
@@ -1,554 +0,0 @@
-var Changeset = require("../../static/js/Changeset");
-var async = require("async");
-var exportHtml = require('./ExportHtml');
- 
-function padDiffHTML (pad, fromRev, toRev){      
-  //check parameters
-  if(!pad || !pad.id || !pad.atext || !pad.pool)
-  {
-    throw new Error('Invalid pad');
-  }
- 
-  var range = pad.getValidRevisionRange(fromRev, toRev);
-  if(!range) { throw new Error('Invalid revision range.' + 
-      ' startRev: ' + fromRev +
-      ' endRev: ' + toRev); }
-  
-  this._pad = pad;
-  this._fromRev = range.startRev;
-  this._toRev = range.endRev;
-  this._html = null;
-  this._authors = [];
-}
- 
-padDiffHTML.prototype._isClearAuthorship = function(changeset){
-  //unpack
-  var unpacked = Changeset.unpack(changeset);
-  
-  //check if there is nothing in the charBank
-  if(unpacked.charBank !== "")
-    return false;
-  
-  //check if oldLength == newLength
-  if(unpacked.oldLen !== unpacked.newLen)
-    return false;
- 
-  //lets iterator over the operators
-  var iterator = Changeset.opIterator(unpacked.ops);
-  
-  //get the first operator, this should be a clear operator
-  var clearOperator = iterator.next();
-  
-  //check if there is only one operator
-  if(iterator.hasNext() === true)
-    return false;
-    
-  //check if this operator doesn't change text
-  if(clearOperator.opcode !== "=")
-    return false;
-    
-  //check that this operator applys to the complete text
-  //if the text ends with a new line, its exactly one character less, else it has the same length
-  if(clearOperator.chars !== unpacked.oldLen-1 && clearOperator.chars !== unpacked.oldLen)
-    return false;
-  
-  var attributes = [];
-  Changeset.eachAttribNumber(changeset, function(attrNum){
-    attributes.push(attrNum);
-  });
-    
-  //check that this changeset uses only one attribute
-  if(attributes.length !== 1)
-    return false;
-    
-  var appliedAttribute = this._pad.pool.getAttrib(attributes[0]);
-  
-  //check if the applied attribute is an anonymous author attribute
-  if(appliedAttribute[0] !== "author" || appliedAttribute[1] !== "")
-    return false;
-    
-  return true;
-}
- 
-padDiffHTML.prototype._createClearAuthorship = function(rev, callback){
-  var self = this;
-  this._pad.getInternalRevisionAText(rev, function(err, atext){
-   if(err){
-     return callback(err);
-   }
-   
-   //build clearAuthorship changeset
-   var builder = Changeset.builder(atext.text.length);
-   builder.keepText(atext.text, [['author','']], self._pad.pool);
-   var changeset = builder.toString();
-   
-   callback(null, changeset);
-  });
-}
- 
-padDiffHTML.prototype._createClearStartAtext = function(rev, callback){
-  var self = this;
-  
-  //get the atext of this revision
-  this._pad.getInternalRevisionAText(rev, function(err, atext){
-   if(err){
-     return callback(err);
-   }
-   
-   //create the clearAuthorship changeset
-   self._createClearAuthorship(rev, function(err, changeset){
-     if(err){
-       return callback(err);
-     }
-   
-     //apply the clearAuthorship changeset
-     var newAText = Changeset.applyToAText(changeset, atext, self._pad.pool);
-     
-     callback(null, newAText);
-   });
-  });
-}
- 
-padDiffHTML.prototype._getChangesetsInBulk = function(startRev, count, callback) {
-  var self = this;
-  
-  //find out which revisions we need
-  var revisions = [];
-  for(var i=startRev;i<(startRev+count) && i<=this._pad.head;i++){
-    revisions.push(i);
-  }
-  
-  var changesets = [], authors = [];
-  
-  //get all needed revisions
-  async.forEach(revisions, function(rev, callback){
-    self._pad.getRevision(rev, function(err, revision){
-      if(err){
-        return callback(err)
-      }
-      
-      var arrayNum = rev-startRev;
-      
-      changesets[arrayNum] = revision.changeset;
-      authors[arrayNum] = revision.meta.author;
-  
-      callback();
-    });
-  }, function(err){
-    callback(err, changesets, authors);
-  });
-}
- 
-padDiffHTML.prototype._addAuthors = function(authors) {
-  var self = this;
-  //add to array if not in the array
-  authors.forEach(function(author){
-    if(self._authors.indexOf(author) == -1){
-      self._authors.push(author);
-    }
-  });
-}
- 
-padDiffHTML.prototype._createDiffAtext = function(callback) {
-  var self = this;
-  var bulkSize = 100;
-  
-  //get the cleaned startAText
-  self._createClearStartAtext(self._fromRev, function(err, atext){
-    if(err) { return callback(err); }
-    
-    var superChangeset = null;
-    
-    var rev = self._fromRev + 1;
- 
-    //async while loop
-    async.whilst(
-      //loop condition
-      function () { return rev <= self._toRev; },
-      
-      //loop body
-      function (callback) {
-        //get the bulk
-        self._getChangesetsInBulk(rev,bulkSize,function(err, changesets, authors){
-          var addedAuthors = [];
-          
-          //run trough all changesets
-          for(var i=0;i<changesets.length && (rev+i)<=self._toRev;i++){
-            var changeset = changesets[i];
-            
-            //skip clearAuthorship Changesets
-            if(self._isClearAuthorship(changeset)){
-              continue;
-            }
-            
-            changeset = self._extendChangesetWithAuthor(changeset, authors[i], self._pad.pool);
-            
-            //add this author to the authorarray
-            addedAuthors.push(authors[i]);
-            
-            //compose it with the superChangset            
-            if(superChangeset === null){
-              superChangeset = changeset;
-            } else {           
-              superChangeset = Changeset.composeWithDeletions(superChangeset, changeset, self._pad.pool);
-            }
-          }
-          
-          //add the authors to the padDiffHTML authorArray
-          self._addAuthors(addedAuthors);
-        
-          //lets continue with the next bulk 
-          rev += bulkSize;
-          callback();
-        });
-      },
-      
-      //after the loop has ended
-      function (err) {
-        //if there are only clearAuthorship changesets, we don't get a superChangeset, so we can skip this step
-        if(superChangeset){
-          var deletionChangeset = self._createDeletionChangeset(superChangeset,atext,self._pad.pool);
-        
-          //apply the superChangeset, which includes all addings
-          atext = Changeset.applyToAText(superChangeset,atext,self._pad.pool);
-          //apply the deletionChangeset, which adds a deletions
-          atext = Changeset.applyToAText(deletionChangeset,atext,self._pad.pool);
-        }      
- 
-        callback(err, atext);
-      }
-    );
-  });
-}
- 
-padDiffHTML.prototype.getHtml = function(callback){
-  //cache the html
-  if(this._html != null){
-    return callback(null, this._html);
-  }
-  
-  var self = this;
-  var atext, html, authorColors;
-  
-  async.series([
-    //get the diff atext
-    function(callback){
-      self._createDiffAtext(function(err, _atext){
-        if(err){
-          return callback(err);
-        }
-        
-        atext = _atext;
-        callback();
-      });
-    },
-    //get the authorColor table
-    function(callback){
-      self._pad.getAllAuthorColors(function(err, _authorColors){
-        if(err){
-          return callback(err);
-        }
-        
-        authorColors = _authorColors;
-        callback();
-      });
-    },
-    //convert the atext to html
-    function(callback){
-      html = exportHtml.getHTMLFromAtext(self._pad, atext, authorColors);
-      self._html = html;
-      callback();
-    }
-  ], function(err){
-    callback(err, html);
-  });
-};
- 
-padDiffHTML.prototype.getAuthors = function(callback){
-  var self = this;
-  
-  //check if html was already produced, if not produce it, this generates the author array at the same time
-  if(self._html == null){
-    self.getHtml(function(err){
-      if(err){
-        return callback(err);
-      }
-      
-      callback(null, self._authors);
-    });
-  } else {
-    callback(null, self._authors);
-  }
-}
- 
-padDiffHTML.prototype._extendChangesetWithAuthor = function(changeset, author, apool) {  
-  //unpack
-  var unpacked = Changeset.unpack(changeset);  
-  
-  var iterator = Changeset.opIterator(unpacked.ops);
-  var assem = Changeset.opAssembler();
-  
-  //create deleted attribs
-  var authorAttrib = apool.putAttrib(["author", author || ""]);
-  var deletedAttrib = apool.putAttrib(["removed", true]);
-  var attribs = "*" + Changeset.numToString(authorAttrib) + "*" + Changeset.numToString(deletedAttrib);
-  
-  //iteratore over the operators of the changeset
-  while(iterator.hasNext()){
-    var operator = iterator.next();
-    
-    //this is a delete operator, extend it with the author
-    if(operator.opcode === "-"){
-      operator.attribs = attribs;
-    } 
-    //this is operator changes only attributes, let's mark which author did that
-    else if(operator.opcode === "=" && operator.attribs){
-      operator.attribs+="*"+Changeset.numToString(authorAttrib);
-    }
-    
-    //append the new operator to our assembler
-    assem.append(operator);
-  }
-  
-  //return the modified changeset
-  return Changeset.pack(unpacked.oldLen, unpacked.newLen, assem.toString(), unpacked.charBank);
-}
- 
-//this method is 80% like Changeset.inverse. I just changed so instead of reverting, it adds deletions and attribute changes to to the atext.
-padDiffHTML.prototype._createDeletionChangeset = function(cs, startAText, apool) {
-  var lines = Changeset.splitTextLines(startAText.text);
-  var alines = Changeset.splitAttributionLines(startAText.attribs, startAText.text);
-  
-  // lines and alines are what the exports is meant to apply to.
-  // They may be arrays or objects with .get(i) and .length methods.
-  // They include final newlines on lines.
- 
-  function lines_get(idx) {
-    if (lines.get) {
-      return lines.get(idx);
-    } else {
-      return lines[idx];
-    }
-  }
- 
-  function lines_length() {
-    if ((typeof lines.length) == "number") {
-      return lines.length;
-    } else {
-      return lines.length();
-    }
-  }
- 
-  function alines_get(idx) {
-    if (alines.get) {
-      return alines.get(idx);
-    } else {
-      return alines[idx];
-    }
-  }
- 
-  function alines_length() {
-    if ((typeof alines.length) == "number") {
-      return alines.length;
-    } else {
-      return alines.length();
-    }
-  }
- 
-  var curLine = 0;
-  var curChar = 0;
-  var curLineOpIter = null;
-  var curLineOpIterLine;
-  var curLineNextOp = Changeset.newOp('+');
- 
-  var unpacked = Changeset.unpack(cs);
-  var csIter = Changeset.opIterator(unpacked.ops);
-  var builder = Changeset.builder(unpacked.newLen);
- 
-  function consumeAttribRuns(numChars, func /*(len, attribs, endsLine)*/ ) {
- 
-    if ((!curLineOpIter) || (curLineOpIterLine != curLine)) {
-      // create curLineOpIter and advance it to curChar
-      curLineOpIter = Changeset.opIterator(alines_get(curLine));
-      curLineOpIterLine = curLine;
-      var indexIntoLine = 0;
-      var done = false;
-      while (!done) {
-        curLineOpIter.next(curLineNextOp);
-        if (indexIntoLine + curLineNextOp.chars >= curChar) {
-          curLineNextOp.chars -= (curChar - indexIntoLine);
-          done = true;
-        } else {
-          indexIntoLine += curLineNextOp.chars;
-        }
-      }
-    }
- 
-    while (numChars > 0) {
-      if ((!curLineNextOp.chars) && (!curLineOpIter.hasNext())) {
-        curLine++;
-        curChar = 0;
-        curLineOpIterLine = curLine;
-        curLineNextOp.chars = 0;
-        curLineOpIter = Changeset.opIterator(alines_get(curLine));
-      }
-      if (!curLineNextOp.chars) {
-        curLineOpIter.next(curLineNextOp);
-      }
-      var charsToUse = Math.min(numChars, curLineNextOp.chars);
-      func(charsToUse, curLineNextOp.attribs, charsToUse == curLineNextOp.chars && curLineNextOp.lines > 0);
-      numChars -= charsToUse;
-      curLineNextOp.chars -= charsToUse;
-      curChar += charsToUse;
-    }
- 
-    if ((!curLineNextOp.chars) && (!curLineOpIter.hasNext())) {
-      curLine++;
-      curChar = 0;
-    }
-  }
- 
-  function skip(N, L) {
-    if (L) {
-      curLine += L;
-      curChar = 0;
-    } else {
-      if (curLineOpIter && curLineOpIterLine == curLine) {
-        consumeAttribRuns(N, function () {});
-      } else {
-        curChar += N;
-      }
-    }
-  }
- 
-  function nextText(numChars) {
-    var len = 0;
-    var assem = Changeset.stringAssembler();
-    var firstString = lines_get(curLine).substring(curChar);
-    len += firstString.length;
-    assem.append(firstString);
- 
-    var lineNum = curLine + 1;
-    while (len < numChars) {
-      var nextString = lines_get(lineNum);
-      len += nextString.length;
-      assem.append(nextString);
-      lineNum++;
-    }
- 
-    return assem.toString().substring(0, numChars);
-  }
- 
-  function cachedStrFunc(func) {
-    var cache = {};
-    return function (s) {
-      if (!cache[s]) {
-        cache[s] = func(s);
-      }
-      return cache[s];
-    };
-  }
- 
-  var attribKeys = [];
-  var attribValues = [];
- 
-  //iterate over all operators of this changeset
-  while (csIter.hasNext()) {
-    var csOp = csIter.next();
- 
-    if (csOp.opcode == '=') {      
-      var textBank = nextText(csOp.chars);
- 
-      // decide if this equal operator is an attribution change or not. We can see this by checkinf if attribs is set.
-      // If the text this operator applies to is only a star, than this is a false positive and should be ignored
-      if (csOp.attribs && textBank != "*") {
-        var deletedAttrib = apool.putAttrib(["removed", true]);
-        var authorAttrib = apool.putAttrib(["author", ""]);;
- 
-        attribKeys.length = 0;
-        attribValues.length = 0;
-        Changeset.eachAttribNumber(csOp.attribs, function (n) {
-          attribKeys.push(apool.getAttribKey(n));
-          attribValues.push(apool.getAttribValue(n));
- 
-          if(apool.getAttribKey(n) === "author"){
-            authorAttrib = n;
-          };
-        });
- 
-        var undoBackToAttribs = cachedStrFunc(function (attribs) {
-          var backAttribs = [];
-          for (var i = 0; i < attribKeys.length; i++) {
-            var appliedKey = attribKeys[i];
-            var appliedValue = attribValues[i];
-            var oldValue = Changeset.attribsAttributeValue(attribs, appliedKey, apool);
-            if (appliedValue != oldValue) {
-              backAttribs.push([appliedKey, oldValue]);
-            }
-          }
-          return Changeset.makeAttribsString('=', backAttribs, apool);
-        });
- 
-        var oldAttribsAddition = "*" + Changeset.numToString(deletedAttrib) + "*" + Changeset.numToString(authorAttrib);
- 
-        var textLeftToProcess = textBank;
- 
-        while(textLeftToProcess.length > 0){
-          //process till the next line break or process only one line break
-          var lengthToProcess = textLeftToProcess.indexOf("\n");
-          var lineBreak = false;
-          switch(lengthToProcess){
-            case -1: 
-              lengthToProcess=textLeftToProcess.length;
-              break;
-            case 0:
-              lineBreak = true;
-              lengthToProcess=1;
-              break;
-          }
- 
-          //get the text we want to procceed in this step
-          var processText = textLeftToProcess.substr(0, lengthToProcess);
-          textLeftToProcess = textLeftToProcess.substr(lengthToProcess);
- 
-          if(lineBreak){
-            builder.keep(1, 1); //just skip linebreaks, don't do a insert + keep for a linebreak
- 
-            //consume the attributes of this linebreak
-            consumeAttribRuns(1, function(){});
-          } else {
-            //add the old text via an insert, but add a deletion attribute + the author attribute of the author who deleted it
-            var textBankIndex = 0;
-            consumeAttribRuns(lengthToProcess, function (len, attribs, endsLine) {
-              //get the old attributes back
-              var attribs = (undoBackToAttribs(attribs) || "") + oldAttribsAddition;
- 
-              builder.insert(processText.substr(textBankIndex, len), attribs);
-              textBankIndex += len;
-            });
- 
-            builder.keep(lengthToProcess, 0);
-          }
-        }
-      } else {
-        skip(csOp.chars, csOp.lines);
-        builder.keep(csOp.chars, csOp.lines);
-      }
-    } else if (csOp.opcode == '+') {
-      builder.keep(csOp.chars, csOp.lines);
-    } else if (csOp.opcode == '-') {
-      var textBank = nextText(csOp.chars);
-      var textBankIndex = 0;
-      
-      consumeAttribRuns(csOp.chars, function (len, attribs, endsLine) {
-        builder.insert(textBank.substr(textBankIndex, len), attribs + csOp.attribs);
-        textBankIndex += len;
-      });
-    }
-  }
- 
-  return Changeset.checkRep(builder.toString());
-};
- 
-//export the constructor
-module.exports = padDiffHTML;

From aefd8d8d0dd33a93a4a930816555332f66156c47 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 13 Feb 2013 18:45:45 +0000
Subject: [PATCH 18/39] nice chars n no cruft

---
 src/node/utils/ExportTxt.js | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/src/node/utils/ExportTxt.js b/src/node/utils/ExportTxt.js
index 05847f162..c57424f1d 100644
--- a/src/node/utils/ExportTxt.js
+++ b/src/node/utils/ExportTxt.js
@@ -26,7 +26,6 @@ var Security = require('ep_etherpad-lite/static/js/security');
 var hooks = require('ep_etherpad-lite/static/js/pluginfw/hooks');
 var getPadPlainText = require('./ExportHelper').getPadPlainText;
 var _analyzeLine = require('./ExportHelper')._analyzeLine;
-var _encodeWhitespace = require('./ExportHelper')._encodeWhitespace;
 
 // This is slightly different than the HTML method as it passes the output to getTXTFromAText
 function getPadTXT(pad, revNum, callback)
@@ -225,7 +224,7 @@ function getTXTFromAtext(pad, atext, authorColors)
         // plugins from being able to display * at the beginning of a line
         // s = s.replace("*", ""); // Then remove it
 
-        assem.append(_encodeWhitespace(s));
+        assem.append(s);
       } // end iteration over spans in line
       
       var tags2close = [];
@@ -292,9 +291,3 @@ exports.getPadTXTDocument = function (padId, revNum, noDocType, callback)
   });
 }
 
-function _encodeWhitespace(s) {
-  return s.replace(/[^\x21-\x7E\s\t\n\r]/g, function(c)
-  {
-    return "&#" +c.charCodeAt(0) + ";"
-  });
-}

From 1607af87531eb8bbaf76e82ff1e79b2cdff0e6a3 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 13 Feb 2013 20:12:46 +0000
Subject: [PATCH 19/39] Bitcoin yo.

---
 README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/README.md b/README.md
index e254b81d3..732a13d12 100644
--- a/README.md
+++ b/README.md
@@ -122,6 +122,7 @@ Join the [mailinglist](http://groups.google.com/group/etherpad-lite-dev) and mak
 # Donate!
 * [Flattr] (http://flattr.com/thing/71378/Etherpad-Foundation)
 * Paypal - Press the donate button on [etherpad.org](http://etherpad.org)
+* Bitcoin - (https://coinbase.com/checkouts/1e572bf8a82e4663499f7f1f66c2d15a)
 
 # License
 [Apache License v2](http://www.apache.org/licenses/LICENSE-2.0.html)

From c3972bcb87ccd9ac5fe0129a6a0dfe724f2be9ae Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 13 Feb 2013 20:14:15 +0000
Subject: [PATCH 20/39] Update README.md

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 732a13d12..fa0da3639 100644
--- a/README.md
+++ b/README.md
@@ -122,7 +122,7 @@ Join the [mailinglist](http://groups.google.com/group/etherpad-lite-dev) and mak
 # Donate!
 * [Flattr] (http://flattr.com/thing/71378/Etherpad-Foundation)
 * Paypal - Press the donate button on [etherpad.org](http://etherpad.org)
-* Bitcoin - (https://coinbase.com/checkouts/1e572bf8a82e4663499f7f1f66c2d15a)
+* [Bitcoin] (https://coinbase.com/checkouts/1e572bf8a82e4663499f7f1f66c2d15a)
 
 # License
 [Apache License v2](http://www.apache.org/licenses/LICENSE-2.0.html)

From 025a3aa7246be46c310f2d75e0c704ce1319c036 Mon Sep 17 00:00:00 2001
From: Siebrand Mazeland <s.mazeland@xs4all.nl>
Date: Sat, 16 Feb 2013 05:10:29 +0000
Subject: [PATCH 21/39] Localisation updates from http://translatewiki.net.

---
 src/locales/br.json      | 56 ++++++++++++++++++++++++++++++++++++++++
 src/locales/ja.json      |  4 +--
 src/locales/ksh.json     | 18 ++++++++++++-
 src/locales/pt-br.json   |  4 ++-
 src/locales/zh-hans.json |  2 +-
 src/locales/zh-hant.json |  7 ++---
 6 files changed, 83 insertions(+), 8 deletions(-)
 create mode 100644 src/locales/br.json

diff --git a/src/locales/br.json b/src/locales/br.json
new file mode 100644
index 000000000..d53ce591b
--- /dev/null
+++ b/src/locales/br.json
@@ -0,0 +1,56 @@
+{
+	"index.newPad": "Pad nevez",
+	"index.createOpenPad": "pe kroui\u00f1\/digeri\u00f1 ur pad gant an anv :",
+	"pad.toolbar.bold.title": "Tev (Ctrl-B)",
+	"pad.toolbar.italic.title": "Italek (Ctrl-I)",
+	"pad.toolbar.underline.title": "Islinenna\u00f1 (Ctrl-U)",
+	"pad.toolbar.indent.title": "Endanta\u00f1",
+	"pad.toolbar.unindent.title": "Diendanta\u00f1",
+	"pad.toolbar.settings.title": "Arventenno\u00f9",
+	"pad.colorpicker.save": "Enrolla\u00f1",
+	"pad.colorpicker.cancel": "Nulla\u00f1",
+	"pad.loading": "O karga\u00f1...",
+	"pad.settings.myView": "Ma diskwel",
+	"pad.settings.fontType.normal": "Reizh",
+	"pad.settings.language": "Yezh :",
+	"pad.importExport.import_export": "Enporzhia\u00f1\/Ezporzhia\u00f1",
+	"pad.importExport.import": "Enkarga\u00f1 un destenn pe ur restr",
+	"pad.importExport.exporthtml": "HTML",
+	"pad.importExport.exportplain": "Testenn blaen",
+	"pad.importExport.exportword": "Microsoft Word",
+	"pad.importExport.exportpdf": "PDF",
+	"pad.importExport.exportopen": "ODF (Open Document Format)",
+	"pad.importExport.exportdokuwiki": "DokuWiki",
+	"pad.modals.connected": "Kevreet.",
+	"pad.modals.slowcommit": "Digevreet.",
+	"pad.share.readonly": "Lenn hepken",
+	"pad.share.link": "Liamm",
+	"pad.chat": "Flap",
+	"timeslider.toolbar.authors": "Aozerien :",
+	"timeslider.toolbar.authorsList": "Aozer ebet",
+	"timeslider.toolbar.exportlink.title": "Ezporzhia\u00f1",
+	"timeslider.month.january": "Genver",
+	"timeslider.month.february": "C'hwevrer",
+	"timeslider.month.march": "Meurzh",
+	"timeslider.month.april": "Ebrel",
+	"timeslider.month.may": "Mae",
+	"timeslider.month.june": "Mezheven",
+	"timeslider.month.july": "Gouere",
+	"timeslider.month.august": "Eost",
+	"timeslider.month.september": "Gwengolo",
+	"timeslider.month.october": "Here",
+	"timeslider.month.november": "Du",
+	"timeslider.month.december": "Kerzu",
+	"pad.userlist.guest": "Den pedet",
+	"pad.userlist.deny": "Nac'h",
+	"pad.userlist.approve": "Aproui\u00f1",
+	"pad.impexp.importbutton": "Enporzhia\u00f1 brema\u00f1",
+	"pad.impexp.importing": "Oc'h enporzhia\u00f1...",
+	"pad.impexp.importfailed": "C'hwitet eo an enporzhiadenn",
+	"@metadata": {
+		"authors": [
+			"Fulup",
+			"Y-M D"
+		]
+	}
+}
\ No newline at end of file
diff --git a/src/locales/ja.json b/src/locales/ja.json
index c2f6da27b..f7173dd4f 100644
--- a/src/locales/ja.json
+++ b/src/locales/ja.json
@@ -19,7 +19,7 @@
 	"pad.toolbar.clearAuthorship.title": "\u4f5c\u8005\u306e\u8272\u5206\u3051\u3092\u6d88\u53bb",
 	"pad.toolbar.import_export.title": "\u4ed6\u306e\u5f62\u5f0f\u306e\u30d5\u30a1\u30a4\u30eb\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\/\u30a8\u30af\u30b9\u30dd\u30fc\u30c8",
 	"pad.toolbar.timeslider.title": "\u30bf\u30a4\u30e0\u30b9\u30e9\u30a4\u30c0\u30fc",
-	"pad.toolbar.savedRevision.title": "\u4fdd\u5b58\u6e08\u307f\u306e\u7248",
+	"pad.toolbar.savedRevision.title": "\u7248\u3092\u4fdd\u5b58",
 	"pad.toolbar.settings.title": "\u8a2d\u5b9a",
 	"pad.toolbar.embed.title": "\u3053\u306e\u30d1\u30c3\u30c9\u3092\u57cb\u3081\u8fbc\u3080",
 	"pad.toolbar.showusers.title": "\u3053\u306e\u30d1\u30c3\u30c9\u306e\u30e6\u30fc\u30b6\u30fc\u3092\u8868\u793a",
@@ -87,7 +87,7 @@
 	"timeslider.exportCurrent": "\u73fe\u5728\u306e\u7248\u3092\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u3059\u308b\u5f62\u5f0f:",
 	"timeslider.version": "\u30d0\u30fc\u30b8\u30e7\u30f3 {{version}}",
 	"timeslider.saved": "| {{year}}\u5e74{{month}}{{day}}\u65e5\u306b\u4fdd\u5b58",
-	"timeslider.dateformat": "{{year}}\u5e74{{month}}{{day}}\u65e5 {{hours}}:{{minutes}}:{{seconds}}",
+	"timeslider.dateformat": "{{year}}\u5e74{{month}}\u6708{{day}}\u65e5 {{hours}}:{{minutes}}:{{seconds}}",
 	"timeslider.month.january": "1\u6708",
 	"timeslider.month.february": "2\u6708",
 	"timeslider.month.march": "3\u6708",
diff --git a/src/locales/ksh.json b/src/locales/ksh.json
index e4557e855..648f95060 100644
--- a/src/locales/ksh.json
+++ b/src/locales/ksh.json
@@ -19,13 +19,16 @@
 	"pad.toolbar.clearAuthorship.title": "d\u00e4 Schriiver ier F\u00e4rve fottn\u00e4mme",
 	"pad.toolbar.import_export.title": "Vun ongerscheidlijje Dattei_Fommaate empotteere udder \u00e4xpotteere",
 	"pad.toolbar.timeslider.title": "Verjangeheid afschpelle",
-	"pad.toolbar.savedRevision.title": "Fa\u00dfjehallde Versione",
+	"pad.toolbar.savedRevision.title": "de Versjohn fa\u00dfhallde",
 	"pad.toolbar.settings.title": "Enscht\u00e4llonge",
 	"pad.toolbar.embed.title": "Donn dat Padd enbenge",
 	"pad.toolbar.showusers.title": "Verbonge Metschriiver aanzeije",
 	"pad.colorpicker.save": "Fa\u00dfhallde",
 	"pad.colorpicker.cancel": "Oph\u00fc\u00fcre",
 	"pad.loading": "Aam Laade&nbsp;&hellip;",
+	"pad.passwordRequired": "Do bruchs e Pa\u00dfwoot f\u00f6r heh dat P\u00e4dd.",
+	"pad.permissionDenied": "Do h\u00e4s nit dat R\u00e4\u00e4sch, op heh dat P\u00e4dd zohzejriife.",
+	"pad.wrongPassword": "Ding Pa\u00dfwoot wohr verkeht.",
 	"pad.settings.padSettings": "Dam P\u00e4dd sin Enscht\u00e4llonge",
 	"pad.settings.myView": "Anseesch",
 	"pad.settings.stickychat": "Donn der Klaaf emmer aanzeije",
@@ -38,6 +41,7 @@
 	"pad.settings.language": "Schprooch:",
 	"pad.importExport.import_export": "Empoot\/\u00c4xpoot",
 	"pad.importExport.import": "Donn jeede T\u00e4x udder jeede Zoot Dokem\u00e4nt huhlaade",
+	"pad.importExport.importSuccessful": "Jeschaff!",
 	"pad.importExport.export": "Don dat P\u00e4dd \u00e4xpoteere al\u00df:",
 	"pad.importExport.exporthtml": "HTML",
 	"pad.importExport.exportplain": "Eijfach T\u00e4x",
@@ -45,9 +49,11 @@
 	"pad.importExport.exportpdf": "PDF (Poteerbaa Dokem\u00e4nte Fommaat)",
 	"pad.importExport.exportopen": "ODF (Offe Dokem\u00e4nte-Fommaat)",
 	"pad.importExport.exportdokuwiki": "DokuWiki",
+	"pad.importExport.abiword.innerHTML": "Mer k\u00fcnne blo\u00df eijfaache T\u00e4xte udder HTML_Fommaate empoteere. Opw\u00e4ndejere M\u00fcjjeleschkeite f\u00f6 der Empoot jon och, dof\u00f6r bruch mer en <a href=\"https:\/\/github.com\/ether\/etherpad-lite\/wiki\/How-to-enable-importing-and-exporting-different-file-formats-in-Ubuntu-or-OpenSuse-or-SLES-with-AbiWord\">Enschtallazjuhn met <i lang=\"en\">Abiword<\/i><\/a>.",
 	"pad.modals.connected": "Verbonge.",
 	"pad.modals.reconnecting": "Ben wider aam Verbenge&nbsp;&hellip;",
 	"pad.modals.forcereconnect": "Wider verbenge",
+	"pad.modals.userdup": "En enem andere Finster en \u00c4rbeid",
 	"pad.modals.userdup.explanation": "Heh dat Padd schingk en mieh wi einem Finster vun enem Brauser op heh d\u00e4m R\u00e4\u00e4schner op ze sin.",
 	"pad.modals.userdup.advice": "En heh d\u00e4m Finster wider verbenge.",
 	"pad.modals.unauth": "Nit ber\u00e4\u00e4schtesch",
@@ -72,10 +78,12 @@
 	"pad.share.emebdcode": "URL enboue",
 	"pad.chat": "Klaaf",
 	"pad.chat.title": "Maach d\u00e4 Klaaf f\u00f6r heh dat P\u00e4dd op",
+	"pad.chat.loadmessages": "Mieh Nohresschte laade...",
 	"timeslider.pageTitle": "{{appTitle}} - Verjangeheid affschpelle",
 	"timeslider.toolbar.returnbutton": "Jangk retuur nohm P\u00e4dd",
 	"timeslider.toolbar.authors": "Schriiver:",
 	"timeslider.toolbar.authorsList": "Kein Schriivere",
+	"timeslider.toolbar.exportlink.title": "\u00c4xpoot",
 	"timeslider.exportCurrent": "Donn de meu\u00dfte V\u00e4sjohn \u00e4xpotteere al\u00df:",
 	"timeslider.version": "V\u00e4sjon {{version}}",
 	"timeslider.saved": "Fa\u00dfjehallde aam {{day}}. {{month}} {{year}}",
@@ -92,11 +100,19 @@
 	"timeslider.month.october": "Oktoober",
 	"timeslider.month.november": "Nov\u00e4mber",
 	"timeslider.month.december": "Dez\u00e4mber",
+	"timeslider.unnamedauthor": "{{num}} naameloose Schriever",
+	"timeslider.unnamedauthors": "{{num}} naameloose Schriever",
+	"pad.savedrevs.marked": "Heh di V\u00e4sjohn es j\u00e4z fa\u00dfjehallde.",
 	"pad.userlist.entername": "Jif Dinge Naame en",
 	"pad.userlist.unnamed": "naamelo\u00df\u00df",
 	"pad.userlist.guest": "Ja\u00df\u00df",
 	"pad.userlist.deny": "Aflehne",
+	"pad.userlist.approve": "Joodhei\u00dfe",
+	"pad.editbar.clearcolors": "Sulle mer de F\u00e4rve f\u00f6r de Schriiver uss_em janze T\u00e4x fott maache?",
+	"pad.impexp.importbutton": "J\u00e4z empoteere",
 	"pad.impexp.importing": "Ben aam Empotteere&nbsp;&hellip;",
+	"pad.impexp.confirmimport": "En Dattei ze empotteere m\u00e4\u00e4t der janze T\u00e4x em P\u00e4dd fott. Wess De dat verfaftesch hann?",
+	"pad.impexp.convertFailed": "Mer kunnte di Dattei nit empoteere. Nemm en ander Dattei-Fommaat udder donn d\u00e4 T\u00e4x vun Hand kopeere un ennf\u00f6\u00f6je.",
 	"pad.impexp.uploadFailed": "Et Huhlaade es don\u00e4vve jejange, bes esu jood un probeer et norr_ens",
 	"pad.impexp.importfailed": "Et Empoteere es don\u00e4vve jejange",
 	"pad.impexp.copypaste": "Bes esu jood un donn et koppeere un enf\u00f6\u00f6je",
diff --git a/src/locales/pt-br.json b/src/locales/pt-br.json
index 6562681a6..e029165db 100644
--- a/src/locales/pt-br.json
+++ b/src/locales/pt-br.json
@@ -19,7 +19,7 @@
 	"pad.toolbar.clearAuthorship.title": "Limpar as cores de identifica\u00e7\u00e3o de autoria",
 	"pad.toolbar.import_export.title": "Importar\/Exportar de\/para diferentes formatos de arquivo",
 	"pad.toolbar.timeslider.title": "Linha do tempo",
-	"pad.toolbar.savedRevision.title": "Revis\u00f5es Salvas",
+	"pad.toolbar.savedRevision.title": "Salvar revis\u00e3o",
 	"pad.toolbar.settings.title": "Configura\u00e7\u00f5es",
 	"pad.toolbar.embed.title": "Incorporar esta Nota",
 	"pad.toolbar.showusers.title": "Mostrar os usuarios nesta Nota",
@@ -100,6 +100,8 @@
 	"timeslider.month.october": "Outubro",
 	"timeslider.month.november": "Novembro",
 	"timeslider.month.december": "Dezembro",
+	"timeslider.unnamedauthor": "{{num}} autor desconhecido",
+	"timeslider.unnamedauthors": "{{num}} autores desconhecidos",
 	"pad.savedrevs.marked": "Esta revis\u00e3o foi marcada como salva",
 	"pad.userlist.entername": "Insira o seu nome",
 	"pad.userlist.unnamed": "Sem t\u00edtulo",
diff --git a/src/locales/zh-hans.json b/src/locales/zh-hans.json
index 7e29a3aef..a17358266 100644
--- a/src/locales/zh-hans.json
+++ b/src/locales/zh-hans.json
@@ -20,7 +20,7 @@
 	"pad.toolbar.redo.title": "\u91cd\u505a (Ctrl-Y)",
 	"pad.toolbar.clearAuthorship.title": "\u6e05\u9664\u4f5c\u540d\u989c\u8272",
 	"pad.toolbar.timeslider.title": "\u65f6\u95f4\u8f74",
-	"pad.toolbar.savedRevision.title": "\u5df2\u4fdd\u5b58\u7684\u4fee\u8ba2",
+	"pad.toolbar.savedRevision.title": "\u4fdd\u5b58\u4fee\u8ba2",
 	"pad.toolbar.settings.title": "\u8bbe\u7f6e",
 	"pad.toolbar.embed.title": "\u5d4c\u5165\u6b64\u8bb0\u4e8b\u672c",
 	"pad.toolbar.showusers.title": "\u663e\u793a\u6b64\u8bb0\u4e8b\u672c\u7684\u7528\u6237",
diff --git a/src/locales/zh-hant.json b/src/locales/zh-hant.json
index a5027a521..7c01860e0 100644
--- a/src/locales/zh-hant.json
+++ b/src/locales/zh-hant.json
@@ -1,7 +1,8 @@
 {
 	"@metadata": {
 		"authors": {
-			"1": "Simon Shek"
+			"0": "Shirayuki",
+			"2": "Simon Shek"
 		}
 	},
 	"index.newPad": "\u65b0Pad",
@@ -86,8 +87,8 @@
 	"timeslider.toolbar.exportlink.title": "\u532f\u51fa",
 	"timeslider.exportCurrent": "\u532f\u51fa\u7576\u524d\u7248\u672c\u70ba\uff1a",
 	"timeslider.version": "\u7248\u672c{{version}}",
-	"timeslider.saved": "{{year}}{{month}}{{day}}\u4fdd\u5b58",
-	"timeslider.dateformat": "{{year}}{{month}}{{day}} {{hours}}:{{minutes}}:{{seconds}}",
+	"timeslider.saved": "{{year}}\u5e74{{month}}{{day}}\u65e5\u4fdd\u5b58",
+	"timeslider.dateformat": "{{year}}\u5e74{{month}}\u6708{{day}}\u65e5 {{hours}}:{{minutes}}:{{seconds}}",
 	"timeslider.month.january": "1\u6708",
 	"timeslider.month.february": "2\u6708",
 	"timeslider.month.march": "3\u6708",

From 93d58b93be53fdefa62e4e3dbe4316799bbab6ba Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 17 Feb 2013 17:13:02 +0000
Subject: [PATCH 22/39] Update src/static/css/iframe_editor.css

---
 src/static/css/iframe_editor.css | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/static/css/iframe_editor.css b/src/static/css/iframe_editor.css
index dd7452b25..3e19cbbea 100644
--- a/src/static/css/iframe_editor.css
+++ b/src/static/css/iframe_editor.css
@@ -177,6 +177,10 @@ p {
 
 #overlaysdiv { position: absolute; left: -1000px; top: -1000px; }
 
+/* Stops super long lines without being spaces such as aaaaaaaaaaaaaa*100 breaking the editor
+   Commented out because it stops IE from being able to render the document, crazy IE bug is crazy. */
+/*
 .ace-line{
-  overflow:hidden; /* Stops super long lines without being spaces such as aaaaaaaaaaaaaa*100 breaking the editor */
+  overflow:hidden; 
 }
+*/

From 6b47fb69d077069521998470b6b8d692181a0555 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 17 Feb 2013 18:01:25 +0000
Subject: [PATCH 23/39] seems to be working pretty well

---
 src/static/js/ace2_inner.js | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/src/static/js/ace2_inner.js b/src/static/js/ace2_inner.js
index 8209c9bf8..70b5443bd 100644
--- a/src/static/js/ace2_inner.js
+++ b/src/static/js/ace2_inner.js
@@ -3731,6 +3731,33 @@ function Ace2Inner(){
             updateBrowserSelectionFromRep();
           }, 200);
         }
+
+        /* Attempt to apply some sanity to cursor handling in Chrome after a copy / paste event
+           We have to do this the way we do because rep. doesn't hold the value for keyheld events IE if the user
+           presses and holds the arrow key */
+        if((evt.which == 37 || evt.which == 38 || evt.which == 39 || evt.which == 40) && $.browser.chrome){
+          var isLeftArrow = evt.which === 37;
+          var isUpArrow = evt.which === 38;
+          var isRightArrow = evt.which === 39;
+          var isDownArrow = evt.which === 40;
+
+          var newVisibleLineRange = getVisibleLineRange(); // get the current visible range -- This works great.
+          var lineHeight = textLineHeight(); // what Is the height of each line?
+          var myselection = document.getSelection(); // get the current caret selection, can't use rep. here because that only gives us the start position not the current
+          var caretOffsetTop = myselection.focusNode.parentNode.offsetTop; // get the carets selection offset in px IE 214
+
+          if((isUpArrow || isLeftArrow || isRightArrow || isDownArrow) && caretOffsetTop){ // was it an up arrow or left arrow?
+            var lineNum = Math.round(caretOffsetTop / lineHeight) ; // Get the current Line Number IE 84
+            var caretIsVisible = (lineNum > newVisibleLineRange[0] && lineNum < newVisibleLineRange[1]); // Is the cursor in the visible Range IE ie 84 > 14 and 84 < 90?
+            if(!caretIsVisible){ // is the cursor no longer visible to the user?
+              // Oh boy the caret is out of the visible area, I need to scroll the browser window to lineNum.
+              // Get the new Y by getting the line number and multiplying by the height of each line.
+              var newY = lineHeight * (lineNum -1); // -1 to go to the line above
+              setScrollY(newY); // set the scroll height of the browser
+            }
+          }
+        }
+
       }
 
       if (type == "keydown")
@@ -3836,7 +3863,6 @@ function Ace2Inner(){
     selection.endPoint = getPointForLineAndChar(se);
 
     selection.focusAtStart = !! rep.selFocusAtStart;
-
     setSelection(selection);
   }
 

From 9ca22754328c96b98b7ae8d98e5f1ab0f6b9e291 Mon Sep 17 00:00:00 2001
From: Marcel Klehr <mklehr@gmx.net>
Date: Sun, 17 Feb 2013 21:35:46 +0100
Subject: [PATCH 24/39] Fixes #1498: Two-part locale specs in lang cookie
 wouldn't be read correctly

---
 src/static/js/l10n.js | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/static/js/l10n.js b/src/static/js/l10n.js
index a67a7c1a4..c79ea706d 100644
--- a/src/static/js/l10n.js
+++ b/src/static/js/l10n.js
@@ -1,8 +1,8 @@
 (function(document) {
   // Set language for l10n
-  var language = document.cookie.match(/language=((\w{2,3})(-w+)?)/);
+  var language = document.cookie.match(/language=((\w{2,3})(-\w+)?)/);
   if(language) language = language[1];
-  
+
   html10n.bind('indexed', function() {
     html10n.localize([language, navigator.language, navigator.userLanguage, 'en'])
   })

From 7e023ce8e1093977717c930237693d0bb5b874c2 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 17 Feb 2013 21:03:19 +0000
Subject: [PATCH 25/39] make the focus jump back to the left if it's required

---
 src/static/js/ace2_inner.js | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/src/static/js/ace2_inner.js b/src/static/js/ace2_inner.js
index 8209c9bf8..a4d3580a0 100644
--- a/src/static/js/ace2_inner.js
+++ b/src/static/js/ace2_inner.js
@@ -1622,10 +1622,19 @@ function Ace2Inner(){
         lines = ccData.lines;
         var lineAttribs = ccData.lineAttribs;
         var linesWrapped = ccData.linesWrapped;
+        var scrollToTheLeftNeeded = false;
 
         if (linesWrapped > 0)
         {
+          if(browser.chrome){
+            // chrome decides in it's infinite wisdom that its okay to put the browsers visisble window in the middle of the span
+            // an outcome of this is that the first chars of the string are no longer visible to the user..  Yay chrome..
+            // Move the browsers visible area to the left hand side of the span
+            var scrollToTheLeftNeeded = true;
+          }
           // console.log("Editor warning: " + linesWrapped + " long line" + (linesWrapped == 1 ? " was" : "s were") + " hard-wrapped into " + ccData.numLinesAfter + " lines.");
+        }else{
+          var scrollToTheLeftNeeded = false;
         }
 
         if (ss[0] >= 0) selStart = [ss[0] + a + netNumLinesChangeSoFar, ss[1]];
@@ -1692,6 +1701,10 @@ function Ace2Inner(){
       //console.log("removed: "+id);
     });
 
+    if(scrollToTheLeftNeeded){ // needed to stop chrome from breaking the ui when long strings without spaces are pasted
+      $("#innerdocbody").scrollLeft(0);
+    }
+
     p.mark("findsel");
     // if the nodes that define the selection weren't encountered during
     // content collection, figure out where those nodes are now.

From 51a9ecf1f0ff9a4d3238b90f49a9e7de0809c412 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Sun, 17 Feb 2013 21:19:15 +0000
Subject: [PATCH 26/39] better support for other browsers

---
 src/static/js/ace2_inner.js | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/src/static/js/ace2_inner.js b/src/static/js/ace2_inner.js
index a4d3580a0..182e40be5 100644
--- a/src/static/js/ace2_inner.js
+++ b/src/static/js/ace2_inner.js
@@ -1626,15 +1626,14 @@ function Ace2Inner(){
 
         if (linesWrapped > 0)
         {
-          if(browser.chrome){
+          if(!browser.ie){
             // chrome decides in it's infinite wisdom that its okay to put the browsers visisble window in the middle of the span
             // an outcome of this is that the first chars of the string are no longer visible to the user..  Yay chrome..
             // Move the browsers visible area to the left hand side of the span
+            // Firefox isn't quite so bad, but it's still pretty quirky.
             var scrollToTheLeftNeeded = true;
           }
           // console.log("Editor warning: " + linesWrapped + " long line" + (linesWrapped == 1 ? " was" : "s were") + " hard-wrapped into " + ccData.numLinesAfter + " lines.");
-        }else{
-          var scrollToTheLeftNeeded = false;
         }
 
         if (ss[0] >= 0) selStart = [ss[0] + a + netNumLinesChangeSoFar, ss[1]];

From 48ffbde7316a2aa21d76e300d0a272651b98b389 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 01:10:54 +0000
Subject: [PATCH 27/39] allow colon to indent line

---
 src/static/js/ace2_inner.js | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/static/js/ace2_inner.js b/src/static/js/ace2_inner.js
index 70b5443bd..5c0af0f15 100644
--- a/src/static/js/ace2_inner.js
+++ b/src/static/js/ace2_inner.js
@@ -1897,7 +1897,7 @@ function Ace2Inner(){
       var prevLine = rep.lines.prev(thisLine);
       var prevLineText = prevLine.text;
       var theIndent = /^ *(?:)/.exec(prevLineText)[0];
-      if (/[\[\(\{]\s*$/.exec(prevLineText)) theIndent += THE_TAB;
+      if (/[\[\(\:\{]\s*$/.exec(prevLineText)) theIndent += THE_TAB;
       var cs = Changeset.builder(rep.lines.totalWidth()).keep(
       rep.lines.offsetOfIndex(lineNum), lineNum).insert(
       theIndent, [

From 77d03d34731f1ec12d77f82fe4c260c4bb9f559a Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 01:40:34 +0000
Subject: [PATCH 28/39] Try to add some sanity to indentation

---
 src/static/js/ace2_inner.js  | 9 +++++----
 src/static/js/pad_editbar.js | 5 +----
 2 files changed, 6 insertions(+), 8 deletions(-)

diff --git a/src/static/js/ace2_inner.js b/src/static/js/ace2_inner.js
index 5c0af0f15..151a18068 100644
--- a/src/static/js/ace2_inner.js
+++ b/src/static/js/ace2_inner.js
@@ -3287,7 +3287,7 @@ function Ace2Inner(){
       listType = /([a-z]+)([12345678])/.exec(listType);
       var type  = listType[1];
       var level = Number(listType[2]);
-      
+
       //detect empty list item; exclude indentation
       if(text === '*' && type !== "indent")
       {
@@ -3317,8 +3317,10 @@ function Ace2Inner(){
 
   function doIndentOutdent(isOut)
   {
-    if (!(rep.selStart && rep.selEnd) ||
-        ((rep.selStart[0] == rep.selEnd[0]) && (rep.selStart[1] == rep.selEnd[1]) &&  rep.selEnd[1] > 1))
+    if (!((rep.selStart && rep.selEnd) ||
+        ((rep.selStart[0] == rep.selEnd[0]) && (rep.selStart[1] == rep.selEnd[1]) &&  rep.selEnd[1] > 1)) &&
+        (isOut != true)
+       )
     {
       return false;
     }
@@ -3326,7 +3328,6 @@ function Ace2Inner(){
     var firstLine, lastLine;
     firstLine = rep.selStart[0];
     lastLine = Math.max(firstLine, rep.selEnd[0] - ((rep.selEnd[1] === 0) ? 1 : 0));
-
     var mods = [];
     for (var n = firstLine; n <= lastLine; n++)
     {
diff --git a/src/static/js/pad_editbar.js b/src/static/js/pad_editbar.js
index 91a07bf96..cc9f8758c 100644
--- a/src/static/js/pad_editbar.js
+++ b/src/static/js/pad_editbar.js
@@ -156,10 +156,7 @@ var padeditbar = (function()
             else if (cmd == 'insertorderedlist') ace.ace_doInsertOrderedList();
             else if (cmd == 'indent')
             {
-              if (!ace.ace_doIndentOutdent(false))
-              {
-                ace.ace_doInsertUnorderedList();
-              }
+              ace.ace_doIndentOutdent(false);
             }
             else if (cmd == 'outdent')
             {

From b30e1de3e50e718769627b68ee3d70c9988b2f89 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 15:12:11 +0000
Subject: [PATCH 29/39] Update README.md

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index fa0da3639..aae935d66 100644
--- a/README.md
+++ b/README.md
@@ -28,7 +28,7 @@ documented codebase makes it easier for developers to improve the code and contr
 
 
 Etherpad Lite is designed to be easily embeddable and provides a [HTTP API](https://github.com/ether/etherpad-lite/wiki/HTTP-API) 
-that allows your web application to manage pads, users and groups. It is recommended to use the [available client implementations](https://github.com/ether/etherpad-lite/wiki/HTTP-API-client-libraries) in order to interact with this API. There is also a [jQuery plugin](https://github.com/johnyma22/etherpad-lite-jquery-plugin) that helps you to embed Pads into your website.
+that allows your web application to manage pads, users and groups. It is recommended to use the [available client implementations](https://github.com/ether/etherpad-lite/wiki/HTTP-API-client-libraries) in order to interact with this API. There is also a [jQuery plugin](https://github.com/etherpad/etherpad-lite-jquery-plugin) that helps you to embed Pads into your website.
 There's also a full-featured plugin framework, allowing you to easily add your own features.
 Finally, Etherpad Lite comes with translations into tons of different languages!
 

From abadfb7b6a2cb92f3fc736231ed2344e0344f32b Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 15:12:37 +0000
Subject: [PATCH 30/39] Update README.md

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index aae935d66..331353a3b 100644
--- a/README.md
+++ b/README.md
@@ -28,7 +28,7 @@ documented codebase makes it easier for developers to improve the code and contr
 
 
 Etherpad Lite is designed to be easily embeddable and provides a [HTTP API](https://github.com/ether/etherpad-lite/wiki/HTTP-API) 
-that allows your web application to manage pads, users and groups. It is recommended to use the [available client implementations](https://github.com/ether/etherpad-lite/wiki/HTTP-API-client-libraries) in order to interact with this API. There is also a [jQuery plugin](https://github.com/etherpad/etherpad-lite-jquery-plugin) that helps you to embed Pads into your website.
+that allows your web application to manage pads, users and groups. It is recommended to use the [available client implementations](https://github.com/ether/etherpad-lite/wiki/HTTP-API-client-libraries) in order to interact with this API. There is also a [jQuery plugin](https://github.com/ether/etherpad-lite-jquery-plugin) that helps you to embed Pads into your website.
 There's also a full-featured plugin framework, allowing you to easily add your own features.
 Finally, Etherpad Lite comes with translations into tons of different languages!
 

From 4cfac2f62452dfa80707d22d1a4575e6f8659c1d Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 08:29:25 -0800
Subject: [PATCH 31/39] fix extract and checkPad

---
 bin/extractPadData.js           |  4 +++-
 src/static/js/pluginfw/hooks.js | 10 ++++++----
 2 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/bin/extractPadData.js b/bin/extractPadData.js
index ea00f953a..1dbab3aa7 100644
--- a/bin/extractPadData.js
+++ b/bin/extractPadData.js
@@ -36,16 +36,18 @@ async.series([
     settings = require('../src/node/utils/Settings');
     db = require('../src/node/db/DB');
     dirty = require("../src/node_modules/ueberDB/node_modules/dirty")(padId + ".db");
+    callback();
   },
   //intallize the database
   function (callback)
   {
+console.log("dbe init");
     db.init(callback);
   },
   //get the pad 
   function (callback)
   {
-    padManager = require('../node/db/PadManager');
+    padManager = require('../src/node/db/PadManager');
     
     padManager.getPad(padId, function(err, _pad)  
     {
diff --git a/src/static/js/pluginfw/hooks.js b/src/static/js/pluginfw/hooks.js
index d9a14d857..4d5057949 100644
--- a/src/static/js/pluginfw/hooks.js
+++ b/src/static/js/pluginfw/hooks.js
@@ -70,10 +70,12 @@ exports.flatten = function (lst) {
 
 exports.callAll = function (hook_name, args) {
   if (!args) args = {};
-  if (exports.plugins.hooks[hook_name] === undefined) return [];
-  return _.flatten(_.map(exports.plugins.hooks[hook_name], function (hook) {
-    return hookCallWrapper(hook, hook_name, args);
-  }), true);
+  if (exports.plugins){
+    if (exports.plugins.hooks[hook_name] === undefined) return [];
+    return _.flatten(_.map(exports.plugins.hooks[hook_name], function (hook) {
+      return hookCallWrapper(hook, hook_name, args);
+    }), true);
+  }
 }
 
 exports.aCallAll = function (hook_name, args, cb) {

From f76cfad42b89cf58feb7bb91203d6544803009e6 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 08:30:42 -0800
Subject: [PATCH 32/39] remove cruft

---
 bin/extractPadData.js | 1 -
 1 file changed, 1 deletion(-)

diff --git a/bin/extractPadData.js b/bin/extractPadData.js
index 1dbab3aa7..4210e18d8 100644
--- a/bin/extractPadData.js
+++ b/bin/extractPadData.js
@@ -41,7 +41,6 @@ async.series([
   //intallize the database
   function (callback)
   {
-console.log("dbe init");
     db.init(callback);
   },
   //get the pad 

From 3f87a32a1289738ddbfebec44865c1532e3ec6e3 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 18:07:01 +0000
Subject: [PATCH 33/39] timestlider script block

---
 src/templates/timeslider.html | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/templates/timeslider.html b/src/templates/timeslider.html
index 4a8543c56..902c71918 100644
--- a/src/templates/timeslider.html
+++ b/src/templates/timeslider.html
@@ -40,8 +40,10 @@
   <% e.end_block(); %>
 
   <link rel="localizations" type="application/l10n+json" href="../../locales.json" />
+  <% e.begin_block("timesliderScripts"); %>
   <script type="text/javascript" src="../../static/js/html10n.js"></script>
   <script type="text/javascript" src="../../static/js/l10n.js"></script>
+  <% e.end_block(); %>
 </head>
 
 <% e.begin_block("timesliderBody"); %>

From fb97920163b740097f67c00f19ec06ec8a891679 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 19:32:07 +0000
Subject: [PATCH 34/39] update changelog

---
 CHANGELOG.md | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 5e3a5d77f..107bd2587 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -3,7 +3,13 @@
  * NEW: Plugins can now provide their own frontend tests
  * NEW: Improved server-side logging
  * NEW: Admin dashboard mobile device support and new hooks for Admin dashboard
+ * NEW: Get current API version from API
+ * Fix: Text Export indentation now supports multiple indentations
  * Fix: Bugfix getChatHistory API method
+ * Fix: Make colons on end of line create 4 spaces on indent
+ * Fix: Make indent when on middle of the line stop creating list
+ * Fix: Stop long strings breaking the UX by moving focus away from beginning of line
+ * Fix: padUsersCount no longer hangs server
  * Fix: Make plugin search case insensitive
  * Fix: Indentation and bullets on text export
  * Fix: Resolve various warnings on dependencies during install

From c37875e09ada4b1c70a6d520bf04b1d79db02e9b Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 19:33:31 +0000
Subject: [PATCH 35/39] update changelog

---
 CHANGELOG.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 107bd2587..c1b38380a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,10 +6,12 @@
  * NEW: Get current API version from API
  * Fix: Text Export indentation now supports multiple indentations
  * Fix: Bugfix getChatHistory API method
+ * Fix: Stop Chrome losing caret after paste is texted
  * Fix: Make colons on end of line create 4 spaces on indent
  * Fix: Make indent when on middle of the line stop creating list
  * Fix: Stop long strings breaking the UX by moving focus away from beginning of line
  * Fix: padUsersCount no longer hangs server
+ * Fix: Issue with two part locale specs not working
  * Fix: Make plugin search case insensitive
  * Fix: Indentation and bullets on text export
  * Fix: Resolve various warnings on dependencies during install

From 19964498f3b66a1c98da29c92f3f3f7532fd4d9e Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 20:38:32 +0000
Subject: [PATCH 36/39] no need to parse already parsed json

---
 bin/extractPadData.js | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/bin/extractPadData.js b/bin/extractPadData.js
index 4210e18d8..c18af5b0e 100644
--- a/bin/extractPadData.js
+++ b/bin/extractPadData.js
@@ -83,7 +83,10 @@ async.series([
       db.db.db.wrappedDB.get(dbkey, function(err, dbvalue)
       {
         if(err) { callback(err); return}
-        dbvalue=JSON.parse(dbvalue);
+
+        if(typeof dbvalue != 'object'){
+          dbvalue=JSON.parse(dbvalue); // if its not json then parse it as json
+        }
         
         dirty.set(dbkey, dbvalue, callback);
       });

From 0a195895097bff67e2c06652938d22393b91bbf6 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Mon, 18 Feb 2013 20:40:34 +0000
Subject: [PATCH 37/39] fix path for windows

---
 bin/extractPadData.js | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/bin/extractPadData.js b/bin/extractPadData.js
index c18af5b0e..0d1acaaee 100644
--- a/bin/extractPadData.js
+++ b/bin/extractPadData.js
@@ -35,7 +35,7 @@ async.series([
   function(callback) {
     settings = require('../src/node/utils/Settings');
     db = require('../src/node/db/DB');
-    dirty = require("../src/node_modules/ueberDB/node_modules/dirty")(padId + ".db");
+    dirty = require("../node_modules/ep_etherpad-lite/node_modules/ueberDB/node_modules/dirty")(padId + ".db");
     callback();
   },
   //intallize the database

From fb3e4a6232a9e58de22bcb37aa23bb568ac1fcc3 Mon Sep 17 00:00:00 2001
From: John McLear <john@mclear.co.uk>
Date: Wed, 20 Feb 2013 16:10:27 +0000
Subject: [PATCH 38/39] only show clients on this pad resolves issue #1544

---
 src/node/handler/PadMessageHandler.js | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/node/handler/PadMessageHandler.js b/src/node/handler/PadMessageHandler.js
index eaf24b7a2..7ffeb6e5a 100644
--- a/src/node/handler/PadMessageHandler.js
+++ b/src/node/handler/PadMessageHandler.js
@@ -1035,7 +1035,7 @@ function handleClientReady(client, message)
       }
 
       // notify all existing users about new user
-      client.broadcast.to(padIds.padIds).json.send(messageToTheOtherUsers);
+      client.broadcast.to(padIds.padId).json.send(messageToTheOtherUsers);
       
       //Run trough all sessions of this pad
       async.forEach(socketio.sockets.clients(padIds.padId), function(roomClient, callback)

From 813583c9cebbf96381358c607fc57874d3321c4b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Uli=20K=C3=B6hler?= <ulikoehler@online.de>
Date: Wed, 20 Feb 2013 21:38:24 +0100
Subject: [PATCH 39/39] Update link to dev guidelines

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 331353a3b..91410b873 100644
--- a/README.md
+++ b/README.md
@@ -108,7 +108,7 @@ You know all this and just want to know how you can help?
 
 Look at the [TODO list](https://github.com/ether/etherpad-lite/wiki/TODO) and our [Issue tracker](https://github.com/ether/etherpad-lite/issues). (Please consider using [jshint](http://www.jshint.com/about/), if you plan to contribute code.)
 
-Also, and most importantly, read our [**Developer Guidelines**](https://github.com/ether/etherpad-lite/wiki/Developer-Guidelines), really!
+Also, and most importantly, read our [**Developer Guidelines**](https://github.com/ether/etherpad-lite/blob/master/CONTRIBUTING.md), really!
 
 # Get in touch
 Join the [mailinglist](http://groups.google.com/group/etherpad-lite-dev) and make some noise on our freenode irc channel [#etherpad-lite-dev](http://webchat.freenode.net?channels=#etherpad-lite-dev)!