--- "G.P.Gaudreault" <[EMAIL PROTECTED]> wrote:
> All,
>
> Which is faster, using the CGI.pm param function to get form input, or parsing out
>the hash
> within my script? I've got a script that currently uses a lot of system resources,
>and I'm
> looking for ways to optimize it.
[snip]
> Is it faster to run this, or use CGI.pm and call param()?
>
> -GPG
Your routine is faster. It's also chock full o'bugs. For a good discussion as to
why, check out
http://www.easystreet.com/~ovid/cgi_course/lesson_two/lesson_two.html.
Ironically, earlier today I asked someone to parse their form-handling code stating
that I would
find at least 5 bugs. The person replied that s?he hadn't really written a
form-handling routine
but a form *validating* routine. Silly me, not seeing the context. Now, you come
along with your
form-handling routine and I just happen to have my gun loaded, so please don't take
any of the
following personally :)
Your routine has a variety of issues, including some novel ones that I haven't
stumbled across
before. All comments will refer to the line of code following them. I am not trying
to be cruel,
this is just another example of why "rolling your own" is usually a bad idea.
######################################################################
sub getform
# Looks like we're not using strict! Need to declare variables with 'my'.
$buffer = "";
# Doesn't allow for GET requests (or HEAD or a variety of others, but we rarely
# encounter those in CGI scripts.
# There is no test to see if the read() is successful
read(STDIN, $buffer, $ENV{'CONTENT_LENGTH'});
# Doesn't allow for an alternate delimiter. The semicolon (;) is supposed to be
the
# new delimiter and systems are slowly moving towards that. Those will break with
this
# code.
@pairs=split(/&/,$buffer);
foreach $pair (@pairs) {
@a = split(/=/,$pair);
$name=$a[0];
$value=$a[1];
# why didn't we do this with $name?
$value =~ s/\+/ /g;
# why didn't we do this with $name?
$value =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C", hex($1))/eg;
# why are we throwing away data that we might need in the future?
# what if you need data from a textarea tag?
$value =~ s/[\r\n]//g;
# why not add them to the %form hash?
push (@data,$name);
push (@data,$value);
}
# Oops! We just lost all multiple values.
%form=@data;
# returning a reference to the hash instead of the hash is probably faster,
# especially if you're working with a large hash.
%form;
}
######################################################################
Let us know if you have any more questions.
Cheers,
Curtis Poe
=====
Senior Programmer
Onsite! Technology (http://www.onsitetech.com/)
"Ovid" on http://www.perlmonks.org/
__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail
http://personal.mail.yahoo.com/
