Re: [PHP] Benchmarking check to see if array key is set
Ford, Mike wrote: > On 06 November 2007 12:57, Christoph Boget wrote: > >> Consider the following test code: > > [...snip...] > >> Running that I found that >> >> if( isset( $myArray[$key] )) >> >> is faster than >> >> if( array key exists( $key, $myArray )) >> >> is faster than >> >> if( $myArray[$key] ) >> >> To be honest, I was surprised. I would have thought that if( >> $myArray[$key] ) would have been the fastest way to check. I'm >> wondering if those in the know could explain why it's the slowest and >> also where the discrepancies come from w/r/t the difference in the >> amount of time it takes >> to make each >> check. > > OK, I'll take a stab. > > Firstly, isset() is a language construct not an actual function, so the only > thing your isset() does is check whether the array element exists. > > Next, array_key_exists() is the only one that involves a real function call, > so it's going to bear that cost. However, the function then does a pretty > simple test which again only involves checking whether the key exists, so the > function itself is going to be pretty quick. > > Finally, the if() is the only one that actually gets the value in > $myArray[$key], and then casts it to Boolean to boot, and the cost of this > must therefore outweigh the cost of a function call plus a check for a key's > existence. > > I'm certainly not surprised that isset() is the quickest, because it does the > simplest test, but to be honest I'm not sure if I'd have predicted the > relative positions of the other two.. > > Cheers! > > Mike > For convenience let's label the cases: 1: if( isset( $myArray[$key] )) 2: if( array key exists( $key, $myArray )) 3: if( $myArray[$key] ) Now, surely 3 is not comparable at all to 1 and 2. Cases 1 and 2 check that the key exists in the array. Case 3 does a boolean test on the value referenced by the key. That has several consequences: Firstly, if there is no element with key "$key" in the arrray, the interpreter will throw an error (undefined index ...); this error could be suppressed and ignored, I suppose. Secondly, if the value referenced by the key is equivalent to FALSE, then the test will return FALSE; for example, if you have set $myArray[$key]='' or $myArray[$key]=0. This could be very confusing. So if you want to test that the key exists, case 3 is NOT the way to do it. In fact, you will often want to check that the key exists(with case 1, ideally) BEFORE using a test of the form of case 3. Cheers Pete Ford (no relation, I think, although I do have a cousin called Michael...) -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
RE: [PHP] Benchmarking check to see if array key is set
On 06 November 2007 12:57, Christoph Boget wrote: > Consider the following test code: [...snip...] > Running that I found that > > if( isset( $myArray[$key] )) > > is faster than > > if( array key exists( $key, $myArray )) > > is faster than > > if( $myArray[$key] ) > > To be honest, I was surprised. I would have thought that if( > $myArray[$key] ) would have been the fastest way to check. I'm > wondering if those in the know could explain why it's the slowest and > also where the discrepancies come from w/r/t the difference in the > amount of time it takes > to make each > check. OK, I'll take a stab. Firstly, isset() is a language construct not an actual function, so the only thing your isset() does is check whether the array element exists. Next, array_key_exists() is the only one that involves a real function call, so it's going to bear that cost. However, the function then does a pretty simple test which again only involves checking whether the key exists, so the function itself is going to be pretty quick. Finally, the if() is the only one that actually gets the value in $myArray[$key], and then casts it to Boolean to boot, and the cost of this must therefore outweigh the cost of a function call plus a check for a key's existence. I'm certainly not surprised that isset() is the quickest, because it does the simplest test, but to be honest I'm not sure if I'd have predicted the relative positions of the other two. Cheers! Mike - Mike Ford, Electronic Information Services Adviser, JG125, The Headingley Library, James Graham Building, Leeds Metropolitan University, Headingley Campus, LEEDS, LS6 3QS, United Kingdom Email: [EMAIL PROTECTED] Tel: +44 113 812 4730 Fax: +44 113 812 3211 To view the terms under which this email is distributed, please go to http://disclaimer.leedsmet.ac.uk/email.htm -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
[PHP] Benchmarking check to see if array key is set
Consider the following test code:
';
$startTime = microtime( TRUE );
$foundCount = 0;
for( $i = 0; $i <= 1; $i++ )
{
$date = microtime( TRUE );
$key = rand( $date, $date * rand( 1, 5000 ));
if( array key exists( $key, $myArray ))
{
$foundCount++;
}
}
$endTime = microtime( TRUE );
echo 'array key exists took [' . ( $endTime - $startTime ) . '] seconds;
found: [' . $foundCount . '] keys';
$startTime = microtime( TRUE );
$foundCount = 0;
for( $i = 0; $i <= 1; $i++ )
{
$date = microtime( TRUE );
$key = rand( $date, $date * rand( 1, 5000 ));
if( isset( $myArray[$key] ))
{
$foundCount++;
}
}
$endTime = microtime( TRUE );
echo 'isset took [' . ( $endTime - $startTime ) . '] seconds; found: [' .
$foundCount . '] keys';
$startTime = microtime( TRUE );
$foundCount = 0;
for( $i = 0; $i <= 1; $i++ )
{
$date = microtime( TRUE );
$key = rand( $date, $date * rand( 1, 5000 ));
if( $myArray[$key] )
{
$foundCount++;
}
}
$endTime = microtime( TRUE );
echo 'IF() took [' . ( $endTime - $startTime ) . '] seconds; found: [' .
$foundCount . '] keys';
?>
Running that I found that
if( isset( $myArray[$key] ))
is faster than
if( array key exists( $key, $myArray ))
is faster than
if( $myArray[$key] )
To be honest, I was surprised. I would have thought that if( $myArray[$key]
) would have been the fastest way to check. I'm wondering if those in the
know could explain why it's the slowest and also where the discrepancies
come from w/r/t the difference in the amount of time it takes to make each
check.
thnx,
Chris
