Fair enough. The only things I would disagree with is that 1) Just
because there are only 2500 in the source code doesn't mean that they
don't get looped over quite a bit, and 2) my Mac Pro Quad Xeon with 4GB
of memory behaves all that much different than the Quad Xeon 4GB
webservers that we and our clients have deployed :)
The original question came out of the coder module which recommended
these changes for better "performance", and I looked into it, and
several other projects recommend using lower case for these constants,
as well as a couple bugs on the
php.net bug list. I actually prefer
writing with upper case, so the only point was if there was any real
world gain.
Here are the links I had found:
http://bugs.php.net/bug.php?id=27286&edit=1
http://framework.zend.com/wiki/display/ZFDEV/PHP+Coding+Standard+(draft)?showComments=false#PHPCodingStandard%28draft%29-BooleansandtheNULLValue
http://drupal.org/node/121388
http://ilia.ws/archives/12-PHP-Optimization-Tricks.html
Other PHP software that use lower case true, false and null standard:
PEAR Standard -
http://pear.php.net/manual/en/standards.naming.php
Joomla -
http://help.joomla.org/content/view/826/125/
Mambo -
http://www.source.mambo-foundation.org/content/view/117/69/
EZ Publish -
http://ez.no/ezpublish/documentation/development/standards/php
Flyspray -
http://flyspray.org/internal:conventions#constant_names
OSCommerce -
http://svn.oscommerce.com/fisheye/browse/osCommerce/tags/oscommerce-2.2rc1/STANDARD?r=1662
phpMyAdmin -
http://phpmyadmin.sourceforge.net/documentation/
Zen Cart -
http://www.zen-cart.com/wiki/index.php/Developers_Standards#Naming_Conventions
phpBB -
http://area51.phpbb.com/docs/coding-guidelines.html#optimizing
And finally, thank you Peter for updating the docs to clarify.
Sorry to be a performance nut, but I can't help it. I think a lot of
us are a little OCD sometimes :D
Steve Rude
Konstantin Käfer wrote:
1326ms vs. 1382ms is 60ms, and when normal
page load times are somewhere around 200ms, it makes a substantial
difference.
Even if we disregard the fact that microtime tests have a big stddev,
your test is not very convincing: You are using these constants several
hundred thousand times whereas Drupal only has 2500+ in its *entire*
source code. And of that source code, only a fraction gets to
execution. I'd estimate the number of constants that are actually
executed on a Drupal page view to about 500, and even that is probably
too high. If you extrapolate your results (60ms for 100000), you'll get
0.3 msecs. Besides, a real world server behaves quite different from a
desktop machine, especially in such edge cases.
Konstantin