Interesting socket_read problem

[utexas_pjm]

For the past couple of weeks I've been writing a webserver in PHP.  Everything has been going relativeley smooth, until recently when I noticed that sometimes the HTTP request is getting truncated:

 

POST /test.php HTTP/1.1
Host: www.patrickmizer.com:9191
User-Agent: Mozilla/5.0 (X11; you; Linux i686 (x86_64); en-US; rv:1.8.1.1) Gecko/20061208 Firefox/2.0.0.1
Accept: text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Referer: http://www.patrickmizer.com:9191/test.php
Content-Type: application/x-www-form-urlencoded
Content-Length: 28

height=0&width=0&submitted=1

 

POST /test.php HTTP/1.1
Host: www.patrickmizer.com:9191
User-Agent: Mozilla/5.0 (X11; you; Linux i686 (x86_64); en-US; rv:1.8.1.1) Gecko/20061208 Firefox/2.0.0.1
Accept: text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Referer: http://www.patrickmizer.com:9191/test.php

 

I've tracked the problem down to the socket_read function.  I'm calling it from within a socket wrapper class that I've written.  The code is berlow:

 

<?php
// ...
    public function readBinary($size = 4096)
    {
	if (false === ($inputl = socket_read($this->socket, $size, PHP_BINARY_READ))) {
		throw new Exception("Could not read data from socket.");
		return false;
	}
	return trim($input); 	
    }
// ...
?>

 

I think what is happening is that all of the data hasn't been received by the socket before it is read from the socket_read function.  I say this, because if I add a sleep(1) call as the first line in the readBinary method this problem goes away.  Obviously this isn't a viable solution, any suggestions?

 

Thanks,

 

Patrick

 

P.S.

 

The problem can be seen here (on the POST): http://www.patrickmizer.com:9191/test.php