#!@PERL@ # # formfind.pl # # This script gets a HTML page from the specified URL and presents form # information you may need in order to machine-make a respond to the form. # # Written to use 'curl' for URL fetching. # # Author: Daniel Stenberg # Version: 0.1 Nov 12, 1998 # # HISTORY # # 0.1 - Created now! # # TODO # respect file:// URLs for local file fetches! $in=""; $usestdin = 0; if($ARGV[0] eq "" ) { $usestdin = 1; } else { $geturl = $ARGV[0]; } if(($geturl eq "") && !$usestdin) { print "Usage: $0 \n", " Use a traling slash for directory URLs!\n"; exit; } # If you need a proxy for web access, edit your .curlrc file to feature # -x # linkchecker, URL will be appended to the right of this command line # this is the one using HEAD: $linkcheck = "curl -s -m 20 -I"; # as a second attempt, this will be used. This is not using HEAD but will # get the whole frigging document! $linkcheckfull = "curl -s -m 20 -i"; # htmlget, URL will be appended to the right of this command line $htmlget = "curl -s"; # urlget, URL will be appended to the right of this command line # this stores the file with the remote file name in the current dir $urlget = "curl -O -s"; # Parse the input URL and split it into the relevant parts: sub SplitURL { my $inurl = $_[0]; if($inurl=~ /^([^:]+):\/\/([^\/]*)\/(.*)\/(.*)/ ) { $getprotocol = $1; $getserver = $2; $getpath = $3; $getdocument = $4; } elsif ($inurl=~ /^([^:]+):\/\/([^\/]*)\/(.*)/ ) { $getprotocol = $1; $getserver = $2; $getpath = $3; $getdocument = ""; if($getpath !~ /\//) { $getpath =""; $getdocument = $3; } } elsif ($inurl=~ /^([^:]+):\/\/(.*)/ ) { $getprotocol = $1; $getserver = $2; $getpath = ""; $getdocument = ""; } else { print "Couldn't parse the specified URL, retry please!\n"; exit; } } if(!$usestdin) { &SplitURL($geturl); #print "protocol = $getprotocol\n"; #print "server = $getserver\n"; #print "path = $getpath\n"; #print "document = $getdocument\n"; #exit; open(HEADGET, "$linkcheck $geturl|") || die "Couldn't get web page for some reason"; headget: while() { # print $_; if($_ =~ /HTTP\/.*3\d\d /) { $pagemoved=1; } elsif($pagemoved && ($_ =~ /^Location: (.*)/)) { $geturl = $1; &SplitURL($geturl); $pagemoved++; last headget; } } close(HEADGET); if($pagemoved == 1) { print "Page is moved but we don't know where. Did you forget the ", "traling slash?\n"; exit; } open(WEBGET, "$htmlget $geturl|") || die "Couldn't get web page for some reason"; while() { $line = $_; push @indoc, $line; $line=~ s/\n//g; $line=~ s/\r//g; # print $line."\n"; $in=$in.$line; } close(WEBGET); } else { while() { $line = $_; push @indoc, $line; $line=~ s/\n//g; $line=~ s/\r//g; $in=$in.$line; } } getlinkloop: while($in =~ /[^<]*(<[^>]+>)/g ) { # we have a tag in $1 $tag = $1; if($tag =~ /^