Fixed copy and paste of prettified markup on IE.

If the output is a <pre> containing <br>s, IE 6 will lose newlines when converting the clipboard's content to plain text.

Added a step to rewrite <br> to CRLF on IE 6 only.

Author: mikesamuel

src/prettify.js

@@ -110,11 +110,17 @@ var PR_ATTRIB_VALUE = 'atv';
 /** the number of characters between tab columns */
 var PR_TAB_WIDTH = 8;
 
-// some string utilities
 function PR_isWordChar(ch) {
   return (ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z');
 }
 
+function PR_isIE6() {
+  var isIE6 = navigator && navigator.userAgent
+      && /\bMSIE 6\./.test(navigator.userAgent);
+  PR_isIE6 = function () { return isIE6; }
+  return isIE6;
+}
+
 /** Splice one array into another.
   * Like the python <code>
   * container[containerPosition:containerPosition + countReplaced] = inserted
@@ -827,7 +833,7 @@ function PR_recombineTagsAndDecorations(
       var htmlChunk = PR_textToHtml(
           tabExpander(sourceText.substring(outputIdx, sourceIdx)))
           .replace(/(\r\n?|\n| ) /g, '$1&nbsp;')
-          .replace(/\r\n?|\n/g, '<br>');
+          .replace(/\r\n?|\n/g, '<br />');
       html.push(htmlChunk);
       outputIdx = sourceIdx;
     }
@@ -928,6 +934,8 @@ var PR_SHOULD_USE_CONTINUATION = true;
   *     has been finished.
   */
 function prettyPrint(opt_whenDone) {
+  var isIE6 = PR_isIE6();
+  
   // fetch a list of nodes to rewrite
   var codeSegments = [
       document.getElementsByTagName('pre'),
@@ -988,8 +996,24 @@ function prettyPrint(opt_whenDone) {
               }
             }
             pre.innerHTML = newContent;
+
             // remove the old
             cs.parentNode.replaceChild(pre, cs);
+            pre = cs;
+          }
+
+          // Replace <br>s with line-feeds so that copying and pasting works on
+          // IE 6.
+          // Doing this on other browsers breaks lots of stuff since \r\n is
+          // treated as two newlines on Firefox, and doing this also slows down
+          // rendering.
+          if (isIE6 && cs.tagName === 'PRE') {
+            var lineBreaks = cs.getElementsByTagName('br');
+            for (var i = lineBreaks.length; --i >= 0;) {
+              var lineBreak = lineBreaks[i];
+              lineBreak.parentNode.replaceChild(
+                  document.createTextNode('\r\n'), lineBreak);
+            }
           }
         }
       }

tests/prettify_test.html

@@ -4,9 +4,10 @@
 <title>Code Prettifier</title>
 <script src="../src/prettify.js" type="text/javascript"
  onerror="alert('Error: failed to load ' + this.src)"></script>
-<script type="Text/javascript">
+<script type="text/javascript">
 // get accurate timing
 PR_SHOULD_USE_CONTINUATION = false;
+function PR_isIE6() { return false; }  // Ensure consistent output.
 </script>
 <link rel="stylesheet" type="text/css" href="../src/prettify.css" />
 <style type="text/css">
@@ -458,6 +459,21 @@ <h1>Bug 14b - comments not ignored</h1>
 &lt;p&gt;And another one inside the end tag&lt;/p<!-- GOTCHA!! -->&gt;
 </pre>
 
+<h1>Bug 20 - missing blank lines</h1>
+<pre class="prettyprint" id="issue20"
+>&lt;html&gt;
+
+&lt;head&gt;</pre>
+
+<h1>Bug 21 - code doesn't copy and paste well in IE</h1>
+<pre class="prettyprint" id="issue21"
+>&lt;html&gt;
+  &lt;head&gt;
+    &lt;title&gt;Test&lt;/title&gt;
+  &lt;/head&gt;
+&lt;/html&gt;</pre>
+<p>To test this bug, disable overriding of PR_isIE6 above, and copy and paste
+the above into Notepad.</p>
 
 </body>
 
@@ -1049,9 +1065,21 @@ <h1>Bug 14b - comments not ignored</h1>
       '`END`PUN{}`END'),
   issue14b: (
       '`COM&lt;!-- There\'s an `END<acronym title="tag soup">`COMHTML`END' +
-           '</acronym>`COM comment in my comment --&gt;`END`PLN<br>' +
-      '`END`PUN&lt;`END`TAGp`END`PUN&gt;`END`PLNAnd another one inside the end tag`END' +
-      '`PUN&lt;/`END`TAGp`END`PUN&gt;`END')
+          '</acronym>`COM comment in my comment --&gt;`END`PLN<br>' +
+      '`END`PUN&lt;`END`TAGp`END`PUN&gt;`END' +
+          '`PLNAnd another one inside the end tag`END' +
+      '`PUN&lt;/`END`TAGp`END`PUN&gt;`END'),
+  issue20: (
+      '`PUN&lt;`END`TAGhtml`END`PUN&gt;`END`PLN<br>' +
+      '<br>' +
+      '`END`PUN&lt;`END`TAGhead`END`PUN&gt;`END'),
+  issue21: (
+      '`PUN&lt;`END`TAGhtml`END`PUN&gt;`END`PLN<br>' +
+      '&nbsp; `END`PUN&lt;`END`TAGhead`END`PUN&gt;`END`PLN<br>' +
+      '&nbsp; &nbsp; `END`PUN&lt;`END`TAGtitle`END`PUN&gt;`END`PLNTest`END' +
+          '`PUN&lt;/`END`TAGtitle`END`PUN&gt;`END`PLN<br>' +
+      '&nbsp; `END`PUN&lt;/`END`TAGhead`END`PUN&gt;`END`PLN<br>' +
+      '`END`PUN&lt;/`END`TAGhtml`END`PUN&gt;`END')
 };