faster alternative to eregi?

Anon

I have a script that scans a html page for some special tags in order to execute specific functions such as displaying a form for example.

The tags have the syntax {function_name:argument1;arg2;arg3:function_name}

The code looks like this:
if(eregi("([^{]){\=function_name🙁[^{]):function_name\=}([^])",}} $this->html, $tmp3))
{
// execute a function with some parameters
}

Using eregi like this it is very slow (over 2 seconds), when I pass generic arguments instead of doing the eregi it takes under 0.5 seconds.

Has anybody an idea on how to speed this up?
Str_replace doesn't seem to handle such complex matchings, maybe preg_match would, but I'm pretty ignorant when it comes to perl style regex.

I hope you can help :-)

Anon

Ok, in fact eregi was not the problem. I just seem to have written a bad matching regex, because when I do it without the ([^]) at the beginning and the end of the string it works well :-)

Anon

Although you found your problem already, using the perl-compatible (preg_match, etc) regular expressions are MUCH MUCH faster (I can't quite explain how much faster it is!) than the POSIX-extended (ereg, etc) ones..

They use almost the exact same syntax, and can be converted easily to use the perl compatible ones.

For example, your eregi could be converted as so:

eregi("([^{]){\=function_name🙁[^{]):function_name\=}([^])",}} $this->html, $tmp3)

preg_match("~{\=function_name🙁[^{]):function_name\=}~i",$this->html,$tmp3)}

And you should get the exact same result (Not tested since I don't have an example input or output).

Hope this helps,
-Josh B

van__

Man, is everybody doing this these days?
I have my own object that allows setting of variables and function calls through templates as well.

Damn!

But, Josh is right, preg is soooooooo much faster that ereg. Incredibly so, in fact I don't even bother to use ereg at all these days, because it seems like a waste.

you really should check out preg_match_all()

It allows you to do a pattern match through an entire file and return an array of all occurences of the regex and it's sub componants.

That's what you want to be using here.

Anon

Thanks for your advices guys. Although it works right now, I surely will check out those nifty preg_match functions as they don't seem to be so much difficult :-)