Question

Code in the post has been modified.

I was thinking that replacing \n with
with javascript was quite a simple task, but it seems not to be so. Posts in Ask Ben or StackOverflow suggest that something as simple as:

var myRe = new RegExp(/\r?\n/g); // to avoid the re literal caching problem
lDes = $("div.descr").html(); 
lDes = lDes.replace (myRe, "<br/>"); 
lDes = lDes.replace (/(http:\/\/\S+)/g, "<a target=\"blank\" href=\"$1\">$1</a>"); 
$("div.descr").html(lDes);

will get the job done. Indeed, this work in FF and Safari but not in IE.

Or, using postie (great hint!)

Text has been created in a textarea and then stored in a database, then retrieved without further processing. It works using FF on windows and Safari on Mac. IE on windows, nada. Is it a major bug in my head? Is it a JQuery issue?

Have some idea about how to solve this? And possible reason for?

Many thanks

Answer 1

Instead of putting the text in question into a <div> on the page, perhaps you can put it directly into a block of Javascript. Now that will require that you have some server-side code to “protect” the Javascript string constant syntax, which is a strangely rare facility but easy to create. You just have to make sure that quote characters, as well as control characters and characters outside the 7-bit range, are appropriately “escaped” the way Javascript expects string constants to look.

Here’s what it’d look like in one of my applications. This is a Java/JSP example, so probably not what you’re using, and of course the “escape” function I call is my own invention, but just so you see what I mean:

var theText = '${pointy:escapeJS(data.theText)}';
$('#target').html(theText.replace(/\n/g, '<br>'));

Of course if you’re receiving the text as an AJAX response, things are simpler.

Now that example leaves out something important: you have to make sure the text is HTML-escaped before dropping it into the document. Perhaps that’s been done server-side, or if not you can do that in Javascript too:

$('#target').html(theText
  .replace(/\r/g, '') // get rid of carriage returns
  .replace(/&/g, '&')
  .replace(/</g, '&lt;')
  .replace(/>/g, '&gt;')
  .replace(/\n\n/g, '\n&nbsp;\n')
  .replace(/\n/g, '<br>')
);

(There are better/faster ways to do HTML escaping of course; that’s just an example.) Note that I stick explicit blank padding between successive newlines – that’s something I cribbed from some code I have and I think I did that because successive <br> elements without intervening “stuff” might not give multiple blank lines in the result; not sure however.