Autoload is kinda convenient. Especially since in PEAR we have such a clear mapping of class name to file name. I do not really understand why some libraries and projects have these insane class name to file name mappings like ez Components and I think symfony also has them. Anyways while it seems kind of nice one must be weary of the idea that they speed up your code. If you are running a byte code cache (which anyone who cares about performance of course does) you will get quite the opposite. Here is Rasmus's answer on this taken straight from #php.pecl:

So I am working on two quick prototypes that are heavy on the Ajax/Javascript front. Been a while since I had to code stuff like that myself actually since in the past years I have always managed to push that kinda of work to other people on the team. Anyways in the prototypes I am doing a bit of drag and drop, I have sliders, moveable popup panels, forms that get submitted via Ajax and that update parts of the interface and web 2.0 stuff like that.

.. I am almost a bit embarrased to say it .. but my diploma thesis is now published on Lulu for print on demand. This is version 1.0.1 which contains a few typos fixes spotted by people in the community. Many thanks for that feedback!

MySQL AB invited me to Frankfurt to talk about their PHP native driver project. While I am not much of a C-hacker, I do know database API's and users preferences in them quite well from my work on MDB2. As such my job was to present a wish list for features. Do note that this is a wish list and any features I mention in this blog post are not yet in any definitive plan. Aside from that, I also took it upon me to make sure that whatever we do, we try to keep the API as clear and simple as possible.

So yesterday I pushed out the x.3.z releases of MDB2. Meaning I released MDB2 2.3.0 along with 1.3.0 versions of the driver packages. There were a lot of minor fixes for edge uses. The most important changes where in the sequence/autoincrement handling. I cleaned up a lot of little details that should lastInsertID() and currID() more consistent. The lastInsertID() method tries to return the last value generated on the given connection, whereas currID() returns the current value of a sequence. I also expanded the placeholder parser in prepare() to ignore things that look like placeholders inside quoted identifier and SQL comments. Previously only quoted strings where ignored. Aside from that I moved to the package.xml version 2 format, which means you now need PEAR 1.4 or higher to install MDB2, which according to Greg's stats should not hurt all too many users, but allows using some of the fancy features of the new format eventually.