> Please forgive the off topic post but I don't subscribe to a
> perl list, and hate to for the occasional question.
Don't neglect the usenet newsgroups. There's always
comp.lang.perl.misc, or possibly comp.infosystems.www.misc
(or you might even try comp.infosystems.www.authoring.cgi,
which has more traffic, though this isn't technically a cgi
question).
> OK, I give up. I know this should be a simple task but I
> cannot get it to work. I am using the perl libwww request
> object to retrieve the results of a POST. I get the result
> back (html page source) in one big string. How do I parse
> the string variable into an array with each element
> containing one line of the result.
>
> I figured I sould do something like :
> $resstring = $ua->request($req)->as_string;
>
> @resarray= split /\n/, $resstring;
> for (@resarray) {
> if (/string/) {
> further processing
> }
> }
I'm not familiar with the libwww module, but there isn't
anything wrong with doing this:
@resarray= split /\n/, $resstring;
So the problem must be that the returned string simply
doesn't have any newlines in it. Remember, newlines aren't
required in HTML: human beings tend to stick them in for
readability, but machine generated code tends to skip them
for the sake of speed (why transmit a character that doesn't
do anything?).
So you're going to have to chunk this material in a
different way. For example, something like this *might*
work to find end tags, and use those as line terminators:
while ( $resstring =~ m|(.*?)</(.*?)>|g ) {
$chunk = "$1</$2>";
push @resarray, $chunk;
}
But if you're serious about this stuff, I'd recommend
looking at the O'Reilly book "Web Client Programming" by
Clinton Wong. He talks about the LWP and HTML perl modules,
and presents lots of example code. E.g. on page 90, there's
a 10 line script that can extract all the URLs used on a
given web page...
_______________________________________________
Redhat-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-list
