Re: try - catch is not so clear to me...

[Lamp Lists <lamp.lists@xxxxxxxxx>]

________________________________
From: Kyle Smith <kyle.smith@xxxxxxxxxxxxxx>
To: Lamp Lists <lamp.lists@xxxxxxxxx>
Cc: php-general@xxxxxxxxxxxxx
Sent: Monday, April 13, 2009 9:52:36 AM
Subject: Re:  try - catch is not so clear to me...

Lamp Lists wrote: 
hi to all!

actually, the statement in the Subject line is not 100% correct. I understand the purpose and how it works (at least I think I understand :-)) but to me it's so complicated way?

let's take a look in example from php.net(http://us3.php.net/try)


<?php
function inverse($x) {
    if (!$x) {
        throw new Exception('Division by zero.');
    }
    else return 1/$x;
}

try {
    echo inverse(5) . "\n";
    echo inverse(0) . "\n";
} catch (Exception $e) {
    echo 'Caught exception: ',  $e->getMessage(), "\n";
}

// Continue execution
echo 'Hello World';
?>  
I would do the same thing, I think, less complicated:

<?php
function inverse($x)
{
    if (!$x) {
        echo 'Division by zero';
    }
    else return 1/$x;

}

echo inverse(5);
echo inverse(0);

// Continue execution
echo 'Hello world';
?>

I know this is "too simple, maybe not the best example", but can somebody please explain "the purpose" of try/catch?

Thanks.

-LL




Your example kind of defeats the point.  The point of a try {} block is
that it will attempt to execute code and execute catch on a true
failure.  Your function already is protected against failure.

Consider this

$x = 0;

try {
    $y = 4 / $x;  // This will divide by zero, not good.
} catch (Exception $e) {
    echo "Error: $e"
}



More importantly, the try/catch should be in your function, not around
the invocations of your function:

function inverse($x) {
	try {
		return $x/0;
	} catch(Exception $e) {
		return false;
	}
}

Consider this also, simply echoing an error on divide by Zero might not
be great if your function is called, say, before headers.  Throwing
exceptions can be re-caught by executing code, which can easily avoid
pre-header output.

Does that clear up the purpose a bit?  I'm no expert, but that's my
understanding.

HTH,
Kyle




Yes and No...
Right now I was thinking to start using try/catch block on places where could be problem in my code and, if there is an error - send to myself email with error info.
E.g., I use right now something like this:

function send_conf_email($to, $subject, $content)
{    
  $headers =    "MIME-Versin: 1.0\n" .
                 "Content-type: text/plain; charset=ISO-8859-1; format=flowed\n" .
                 "Content-Transfer-Encoding: 8bit\n" .
                            "Reply-To: <ll@xxxxxxxxx>\n".
                 "From: LL\n" .
                 "X-Mailer: PHP" . phpversion();
        
    mail($to, $subject, $body, $headers);
}


function send_error_email_to_admin($email)
{    
    $to = 'LL@xxxxxxxxx';
    $subject = '[Error Report] '.$email['subject'];
    $content = $email['content'];
    
    send_conf_email($to, $subject, $content);

}



if (!send_conf_email($to, $subject, $content))
{
   send_error_email_to_admin($subject, $content);
}


how would/should be the code by using try/catch() block? something like this:

try
{
    if (!send_plain_email($this->submitted_email, $subject, $body_plain))
    {
        throw new Exception('Confirmation email is not sent');
    }
}
catch (Exception $e)
{
    send_error_email_to_admin($subject, $content . $e->getMessage());
}


there is something "fishy"
:-)

-LL