A tiny fix to the jQuery hint plugin

Here’s a tiny fix to Remy Sharp‘s excellent jQuery Text box hints plug-in. Without this fix, jQuery‘s val function will return the hint text if the text box hasn’t been filled out by the user yet.

Here’s the patch:

@@ -20,7 +23,7 @@
       $win = $(window);

     function remove() {
-      if ($input.val() === title && $input.hasClass(blurClass)) {
+      if ($input.realval() === title && $input.hasClass(blurClass)) {
         $input.val('').removeClass(blurClass);
       }
     }
@@ -41,4 +44,17 @@
   });
 };

+
+$.fn.realval = $.fn.val;
+
+$.fn.val = function (value) {
+  var i = $(this);
+  if (value === undefined) {
+    return (i.realval() === i.attr('title')) ? '' : i.realval();
+  } else {
+    return i.realval(value);
+  }
+}
+
+
 })(jQuery);

And here’s the full plugin with the patch applied:

/**
* @author Remy Sharp
* @url http://remysharp.com/2007/01/25/jquery-tutorial-text-box-hints/
*
* better val() method added by Matt Chisholm, 2009/07/27
* http://glyphobet.net/blog/essay/878
*/

(function ($) {

$.fn.hint = function (blurClass) {
  if (!blurClass) {
    blurClass = 'blur';
  }

  return this.each(function () {
    // get jQuery version of 'this'
    var $input = $(this),

    // capture the rest of the variable to allow for reuse
      title = $input.attr('title'),
      $form = $(this.form),
      $win = $(window);

    function remove() {
      if ($input.realval() === title && $input.hasClass(blurClass)) {
        $input.val('').removeClass(blurClass);
      }
    }

    // only apply logic if the element has the attribute
    if (title) {
      // on blur, set value to title attr if text is blank
      $input.blur(function () {
        if (this.value === '') {
          $input.val(title).addClass(blurClass);
        }
      }).focus(remove).blur(); // now change all inputs to title

      // clear the pre-defined text when form is submitted
      $form.submit(remove);
      $win.unload(remove); // handles Firefox's autocomplete
    }
  });
};

$.fn.realval = $.fn.val;

$.fn.val = function (value) {
  var i = $(this);
  if (value === undefined) {
    return (i.realval() === i.attr('title')) ? '' : i.realval();
  } else {
    return i.realval(value);
  }
}

})(jQuery);

My identity is not for your profit

OkCupid is one of the best-designed websites out there. It’s addictive, captivating, easy to use, and pretty. But since it’s free and completely funded by advertising, the ultimate design goal of the site must be to get users to visit more pages, view more ads, and click on some. And this design goal can be taken to extremes that are at odds with the goal of providing quality matching. Consider this screenshot:

why-im-not-on-okcupid-anymore

This was from the unauthenticated view of my (now-disabled) OkCupid profile. The black text in the bottom half of this screen-shot was my somewhat snide response to OkCupid’s seventh-grade reading level getting-to-know-you question. And at the top is a plug for a different user and an ad for a book that has absolutely nothing to do with me or my tastes1.

The last thing that any OkCupid user should want is for readers to be distracted from their profile like this. And I really didn’t want some site co-opting my identity and interspersing advertisements and links to generate more page views into the middle of it. Even if it was to support a free site. So, OkCupid, you have joined the ranks of other online dating sites that just don’t cut it. Thanks, but no thanks.

  1. A Summer Affair‘s plot summary on Amazon makes it sound like an upscale romance novel with a helping of East Coast affluence-porn thrown in. [↩]

Bring on the patches

Zed Shaw’s recent Python Neglect article raises a few points that are unquestionably valid: easy_install and mx.DateTime suck big time, and it sounds like the email tools have some serious problems. But I would like to hear better explanations from him about:

  • What the datetime module (introduced to replace mx.DateTime and fix shortcomings with the time module) is missing? It’s always handled everything I’ve thrown at it.
  • What is wrong with optparse? Usage? Functionality? Implementation?

I also just plain don’t understand his complaint about del versus remove. I think mystuff.remove(mything) and del mystuff[4] do exactly the same thing, so what’s the problem? That del exists?

His promise to provide patches will be an interesting test of the Python community’s ability to judge patches on their objective merit, not by their contributor, since Shaw seems to have a tendency to ruffle feathers1.

And a “WSGI for templating,” and efficient state machines with generators? Sounds exciting.

  1. This is the pot calling the kettle black here. [↩]