Re: How to "synchronize" file write accesses
- From: Rob <ratkinson@xxxxxxxxxxxxx>
- Date: Wed, 14 Jan 2009 04:02:44 -0800 (PST)
On Jan 13, 11:23 pm, "Netx" <user72cut-it-...@xxxxxx> wrote:
My .php script checks if the "_copper.txt" file is not older than
1h.
If it is older than the content of "_copper.txt" is
imported/updated from the other web page.
Then the content of "_copper.txt" file is displayed.
So in need of "update" I have the following fragment :
$myFile = '_copper.txt';
$fh = fopen($myFile, 'w') or die('can\'t open _copper.txt');
//--CODE TO IMPORT COPPER PRICE FROM WWW
//(...)
fwrite($fh,$copper1);
fclose($fh);
However I am afraid what will happen if two persons request the
page and two "threads" fire fopen() for write? I can get over
problems
with one badly processed request, but I am afraid that
"_copper.txt" file may be damaged what will lead to all
subsequent requests badly displaying copper price.
How would you suggest to correct it and if possible without
database?
Mark
function getCopper()
{
$filename = '_copper.txt';
$filemtime = filemtime($filename);
$currenttime = mktime();
$diff = $currenttime - $filemtime;
//seconds to refresh:
if ($diff > 3600) {
getCopperReally();
}
$fh = fopen($filename,'r') or die ('can\'t open
_copper.txt');
$theData = fread($fh, filesize($filename));
fclose($fh);
print ($theData);
}
function getCopperReally()
{
$myFile = '_copper.txt';
$fh = fopen($myFile, 'w') or die('can\'t open _copper.txt');
$result=get_web_page('www-address-goes-here');
//process the page to get copper price into $copper1
//(it works)
fwrite($fh,$copper1);
fclose($fh);
}
.
- References:
- How to "synchronize" file write accesses
- From: Netx
- How to "synchronize" file write accesses
- Prev by Date: nginx+fastcgi+php gets "No input file specified."
- Next by Date: Re: How to "synchronize" file write accesses
- Previous by thread: Re: How to "synchronize" file write accesses
- Next by thread: Re: How to "synchronize" file write accesses
- Index(es):
Relevant Pages
|
