Added another minor hack for making sure we only get the first page in the PNG.

[webpdf] / createpdf.pl

diff --git a/createpdf.pl b/createpdf.pl
index bd39e9c386e4958de35aa3f891de92a4e1efe34c..e74ab0e0119a08a9e3297e5b4891d634d7d9aca1 100755 (executable)
--- a/createpdf.pl
+++ b/createpdf.pl
@@ -22,12 +22,14 @@ use CGI;
 use POSIX;
 use strict;
 use File::Temp;
+require './config.pm';
 
-$ENV{"HOME"} = "/home/cassarossa/itk/sesse/sesse-pdf/";
+$ENV{"HOME"} = $pdfweb::config::homedir;
 
 my $cgi = CGI->new;
 my $filename = $cgi->param('input');
 my $file = $cgi->upload('input');
+my $url = $cgi->param('url');
 
 # It kind of sucks that we just can't get the temporary file name from
 # CGI.pm, but OK, here goes :-)
@@ -53,7 +55,14 @@ if ($cgi->param('preset') eq 'screen') {
        $psopts = "";
 }
 
-if ($filename =~ /(.*)\.(?:e?ps|pdf)$/i) {
+if ($url =~ /^http/i) {
+       $outname = "web.pdf";
+
+       # Render through Gecko
+       $ENV{'DISPLAY'} = $pdfweb::config::xserver;
+       system("gnome-web-print", "-f", $url, "$pdfweb::config::outputdir/$pdf_filename.ps");
+       system("gs $pdfopts -dCompatbilityLevel=1.4 -dNOPAUSE -dPATCH -sDEVICE=pdfwrite -dSAFER -sOutputFile=output/$pdf_filename -c '.setpdfwrite $psopts' -f - < $pdfweb::config::outputdir/$pdf_filename.ps >&2");
+} elsif ($filename =~ /(.*)\.(?:e?ps|pdf)$/i) {
        $outname = "$1.pdf";
 
        # Yay, just a round through GhostScript
@@ -69,7 +78,7 @@ if ($filename =~ /(.*)\.(?:e?ps|pdf)$/i) {
        $outname = "$1.pdf";
 
        # Run through ImageMagick first of all, then gs
-       open PIPE, "| convert $2:- ps:- | gs $pdfopts -dCompatbilityLevel=1.4 -dNOPAUSE -dPATCH -sDEVICE=pdfwrite -dSAFER -sOutputFile=output/$pdf_filename -c '.setpdfwrite $psopts' -f - >&2"
+       open PIPE, "| convert $2:- pdf:- | gs $pdfopts -dCompatbilityLevel=1.4 -dNOPAUSE -dPATCH -sDEVICE=pdfwrite -dSAFER -sOutputFile=output/$pdf_filename -c '.setpdfwrite $psopts' -f - >&2"
                 or die "convert: $!";
 
        my ($buf, $ret);
@@ -103,9 +112,17 @@ if ($filename =~ /(.*)\.(?:e?ps|pdf)$/i) {
        close DOC;
 
        # Create PostScript from OOo :-)
-       system("/usr/lib/openoffice/program/soffice -display :25 -headless -pt pdf /home/cassarossa/itk/sesse/public_html/pdf/output/$pdf_filename.$ext");
+       system("/usr/lib/openoffice/program/soffice -display $pdfweb::config::xserver -headless -pt pdf $pdfweb::config::outputdir/$pdf_filename.$ext");
+
+       # This is quite hideous -- it looks like OO.o calls the file something slightly
+       # different depending on the time format, phase of the moon or something... So
+       # we try both.
+       my $inp_ps = "output/$pdf_filename.$ext.pdf";
+       if (! -r $inp_ps) {
+               $inp_ps = "output/$pdf_filename.pdf";
+       }
 
-       system("gs $pdfopts -dCompatbilityLevel=1.4 -dNOPAUSE -dPATCH -sDEVICE=pdfwrite -dSAFER -sOutputFile=output/$pdf_filename -c '.setpdfwrite $psopts' -f - < output/$pdf_filename.pdf >&2");
+       system("gs $pdfopts -dCompatbilityLevel=1.4 -dNOPAUSE -dPATCH -sDEVICE=pdfwrite -dSAFER -sOutputFile=output/$pdf_filename -c '.setpdfwrite $psopts' -f - < $inp_ps >&2");
 } elsif ($filename =~ /(.*)\.(c|cc|cpp|cs|h|py|rb|pl|diff|patch|js|php[1-5]?|hs|f|f90|java|css|sql|l|y|s?ml|sh|awk|m|v)$/i) {
        $outname = "$1.pdf";
        my $ext = $2;
@@ -121,7 +138,7 @@ if ($filename =~ /(.*)\.(?:e?ps|pdf)$/i) {
        (my $sanitized_filename = $filename) =~ tr/a-zA-Z0-9./_/c;
 
        # This is just perverse :-P 
-       system("vim -Z -R +\"set printheader=\%<$sanitized_filename\%h\%m\%=Page\\ \%N\" +\"ha >/home/cassarossa/itk/sesse/public_html/pdf/output/$pdf_filename.ps\" +:q output/$pdf_filename.$ext");
+       system("vim -Z -R +\"set printheader=\%<$sanitized_filename\%h\%m\%=Page\\ \%N\" +\"ha >$pdfweb::config::outputdir/$pdf_filename.ps\" +:q output/$pdf_filename.$ext");
 
        system("gs $pdfopts -dCompatbilityLevel=1.4 -dNOPAUSE -dPATCH -sDEVICE=pdfwrite -dSAFER -sOutputFile=output/$pdf_filename -c '.setpdfwrite $psopts' -f - < output/$pdf_filename.ps >&2");
 } else {
@@ -138,8 +155,28 @@ EOF
 
 my $size = -s "output/$pdf_filename";
 
+if (defined($size) && $size > 0) {
+       # Make a thumbnail from the finished PDF, for later reference. (Output to
+       # stdout is so we make sure we get only the first page; it's the simplest
+       # hack I can find offhand. :-) )
+       system("convert -resize 192x192 output/$pdf_filename png:- | convert png:- png:- > output/$pdf_filename.png");
+       open DESC, ">output/$pdf_filename.desc";
+
+       if ($url =~ /^http/i) {
+               $url =~ tr/\n//d;
+               print DESC "$url\n";
+       } else {
+               $filename =~ tr/\n//d;
+               print DESC "$filename\n";
+       }
+
+       close DESC;
+}
+       
+(my $sanitized_outname = $outname) =~ tr/a-zA-Z0-9. -/_/c;
+
 print "Content-type: application/pdf\n";
-print "Content-disposition: attachment; filename=\"$outname\"\n";  # FIXME: XSS problems?
+print "Content-disposition: attachment; filename=\"$sanitized_outname\"\n";
 print "Content-length: $size\n\n";
 
 system("cat output/$pdf_filename");  # yuck?