Parsing URL using regular expression

[knsridhar]

Hi Monks

I have written a script which parses mbox and updates into the database. When i get a mail with 8859 character set or quoted printable format, i parse it by

     $message_body =~ s/=([ 0-9A-Fa-f ]{2})/chr(hex($1))/ge;
     $message_body =~ s/=\n//g;
     $message_body =~ s/=3D/=/g;
     $message_body =~ s/>/>\n/g;

Now the problem is while parsing mails whose body contains hyperlinks with key/value pair, like

http://www.foo.com/bar.pl?aid=3D2744&bid=3D&cn=3DOM=
Ddemo2&rid=3D0

the message body written to db doesnt contain any hyperlinks. Is there any way to solve this problem

Thanks
Sridhar

[Sec]

I suspect your mistake lies somewhere else. The above mentioned snipped works as expected and keeps the URL parameters:

#!perl -l
$message_body="http://www.foo.com/bar.pl?aid=3D2744&bid=3D&cn=3DOM=
Ddemo2&rid=3D0";
$message_body =~ s/=([0-9A-Fa-f]{2})/chr(hex($1))/ge;
$message_body =~ s/=\n//g;
$message_body =~ s/=3D/=/g;
$message_body =~ s/>/>\n/g;
print $message_body;

results in http://www.foo.com/bar.pl?aid=2744&bid=&cn=OMDdemo2&rid=0
While we're at it. the third substitute (the one with =3D) seems unnecessary, as the first substitute already transforms =3D into =.

[Jaap]

What do you mean with: "the message body written to db doesnt contain any hyperlinks" Do you mean the entire URL is not in the database?

[knsridhar]

Yes, the entire URL is not written to the database

[Anonymous Monk]

What is your question actually? How to add URLs to a database? How to parse URLs? How to extract URLs from a body of text? How to deal with quoted-printable?

Confused Monk

[knsridhar]

My requirement to parse a mail which contains hyperlink with key/value pair in its message body Sorry if i was not clean