[SOLVED] Most efficient way to perform case-irrelevant comparison

[jordanwb]

In my User class' Register function it calls a function that checks if an already existing user has a field whose value is the same as the one that the new user is trying to use. An example is Email address. Now I want to perform a comparison where case does not matter. Now I know how I would do that, but that's not the question:

 

[1]: strtolower ("string") == strlower ("STRING")

[2]: preg_match ("#^[string]$#i", "STRING")

 

But what I'm asking is which would be more efficient? Is there perhaps another algorithm that would be better?

[genericnumber1]

probably the fastest way of comparing two strings case insensitively is with strcasecmp(). Keep in mind that this function won't return what is immediately expected in that it returns 0 if the strings are equal. I'm only guessing that it's faster since it's one function call instead of two, and regex is inherently slow, so I may be mistaken in it being the fastest alternative . Benchmark if you really care about it being the perfect alternative.

 

<?php
$str1 = 'stuff';
$str2 = 'stUfF';

if(strcasecmp($str1, $str2) == 0)
{
   echo "$str1, $str2 ARE equal to eachother!";
}
?>

[JasonLewis]

I don't think there would be to much difference. I think I convert both to lowercase before comparison. I think.

[jordanwb]

Benchmark if you really care about it being the perfect alternative.

 

That's not a bad idea:

 

<?php

$string_1 = "Jordan";
$string_2 = "jORDAN";

$tests = 100000;
$start = microtime ();

for ($i = 0; $i < $tests; $i++)
{
strcasecmp ($string_1, $string_2);
}

print "strcasecmp took: " . (microtime () - $start) . " microseconds.<br />";

$start = microtime ();

for ($i = 0; $i < $tests; $i++)
{
strtolower ($string_1) == strtolower ($string_2);
}

print "strtolower took: " . (microtime () - $start) . " microseconds.<br />";

$start = microtime ();

for ($i = 0; $i < $tests; $i++)
{
preg_match ("#^[".$string_1."]$#i", $string_2);
}

print "preg_match took: " . (microtime () - $start) . " microseconds.<br />";

?>

 

Output:

 

strcasecmp took: 0.165722 microseconds.

strtolower took: -0.668401 microseconds.

preg_match took: 0.361707 microseconds.

 

I'm not sure what to think regarding strtolower. I re-ran it a bunch of times and the output that was negative kept changing.

 

strcasecmp took: 0.166892 microseconds.

strtolower took: 0.334652 microseconds.

preg_match took: 0.374544 microseconds.

 

in these results strcasecmp seems the quickest.

[DarkWater]

<?php

$string_1 = "Jordan";
$string_2 = "jORDAN";

$tests = 100000;
$start = microtime (true);

for ($i = 0; $i < $tests; $i++)
{



strcasecmp ($string_1, $string_2);
}

print "strcasecmp took: " . (microtime (true) - $start) . " microseconds.<br />";

$start = microtime (true);

for ($i = 0; $i < $tests; $i++)
{



strtolower ($string_1) == strtolower ($string_2);
}

print "strtolower took: " . (microtime (true) - $start) . " microseconds.<br />";

$start = microtime (true);

for ($i = 0; $i < $tests; $i++)
{



preg_match ("#^[".$string_1."]$#i", $string_2);
}

print "preg_match took: " . (microtime (true) - $start) . " microseconds.<br />";

?>

 

Try specifying TRUE as the first argument to microtime.  I get the following from doing that:

strcasecmp took: 0.287498950958 microseconds.<br />
strtolower took: 0.728889942169 microseconds.<br />
preg_match took: 0.907059192657 microseconds.<br />

 

EDIT: The line breaks are there because I ran this in the shell.

[jordanwb]

Aww the process no longer goes back in time.  :'(

[DarkWater]

Aww the process no longer goes back in time.  :'(

 

Lol. >_<  You get the proper results from the first parameter being true. =P