Switch or If ?

Tekime

How do you decide between using a switch or an if statement?

Say I have a file admin.php that has four administration modes: user, group, type and definition. A few common requests may look like:

admin.php?mode=user&action=modify&user_id=143

admin.php?mode=type&action=create

In admin.php I have an if...else control structure for the modes:

if($gForm['mode'] == 'user') {

  do stuff....

} elseif($gForm['mode'] == 'group') {

  do stuff....

} elseif($gForm['mode'] == 'type') {

Would there be any benefit from using a switch($gForm['mode']) statement instead of the if..else type of structure? I can only iamgine there might be a difference in speed, aside from that I'm not sure what kind of benefit you would have either way.

diego25

The answer to your question is here:
http://www.php.net/manual/en/control-structures.switch.php

Diego

Tekime

Thanks diego, I have read through the switch docs but I guess I was still unclear of the speed benefit. Do you think it is always going to be faster using a switch when you are only evaluating a single variable, then?

Since I am evaluating a single variable $gForm['mode'], would it always be smarter to use a switch statement?

diego25

Well, it may be a bit faster than doing elseif($gForm['mode'] == 'type') , but it won't be much faster. Especially if you have only 4 comparisons, the speed increase won't be substantial...

I guess it's mostly a matter of taste...I use switch statements coz it's less typing 🙂 But if there's a lot of code in between cases, I use if statements (but one can argue that I can put those lines in a function so that there's just one function call in between cases).

Diego

Tekime

Okay, thanks again diego. I guess I like if..elseif for what I'm doing, since there is quite a bit of code between the control statements. I can't justify putting all of that code into functions, since most of it already uses class functions for all the real work, it's just a lot of if..else statements and mangling.

I guess I should probably run some scripts with microtime and see if there is a noticeable difference between the two; even 1/1000 of a second difference can add up in the end.