perlcraft
holygrail
#!/usr/bin/perl -w
##############################################################################
#
# XML2PDF
#
# Author : HolyGrail
# Version: 0.7
#
# Usage: xml2pdf {<xmlfile>}
#
# This conversionprogram translates XML into PDF. You can specify more than one
# xmlfile at the commandline and they will all be processed. The PDF files will
# replace ".xml" with ".pdf" or add ".pdf" if the sourcefile does not end with
# ".xml". It does no XML validation at all. The XML file should conform to the
# following DTD:
#
# <!ELEMENT document (section1)+>
# <!ATTLIST document
# title CDATA #REQUIRED
# >
#
# <!ELEMENT section1 (section2 | p | img)+>
# <!ATTLIST section1
# title CDATA #REQUIRED
# >
#
# <!ELEMENT section2 (p | img)+>
# <!ATTLIST section2
# title CDATA #REQUIRED
# >
#
# <!ELEMENT p (img | #PCDATA)>
#
# <!ELEMENT img>
# <!ATTLIST img
# src CDATA #REQUIRED
# >
#
# Known issues:
# * 'img'-elements are processed after a complete 'p' element is processed,
# so the images that are meant to come 'in' the text, come 'after' the
# text. The workaround for this should be to split the text up in multiple
# p-elements and put the img-element between two p-elements.
# * Images that are wider than the page fall off the page on the right side
#
# Bugs:
# * Lots of them, no doubt
#
# Changes:
#
# 0.7: 'img' tags can now also be url's and no longer just local files
# 0.6: It now also accepts images (JPG, GIF and PNG), changed the DTD
# accordingly.
# 0.5: Changed the program call, so you can process multiple files in one call
# 0.4: It now can handle sections that don't fit on one page. It starts new
# pages if needed.
# 0.3: Adapted it to conform to OeufMayo's DTD
# 0.2: Minor change (0.1 re-declared the global vars as 'my', thanks to davorg
# 0.1: First release
#
##############################################################################
use strict;
use PDFLib;
use XML::XPath;
use LWP::Simple;
use vars qw ( $TOPMARGIN
$BOTTOMMARGIN
$LEFTMARGIN
$RIGHTMARGIN
$PAGEHEIGHTPOINTS
$PAGEWIDTHPOINTS
$MAXYPOS
%FONTDEFINITION
%IMAGES
);
# Define the fonts for each element
%FONTDEFINITION = ( section1 => [face => "Helvetica-Bold", size => "16.0"],
section2 => [face => "Helvetica-Bold", size => "14.0"],
p => [face => "Helvetica", size => "12.0"]
);
# Pagesetup
my $pagename = "a4";
$PAGEHEIGHTPOINTS = 842; # See PDFLib's documentation for
$PAGEWIDTHPOINTS = 595; # the size of your page
($TOPMARGIN, $BOTTOMMARGIN, $LEFTMARGIN, $RIGHTMARGIN) = (10, 10, 15, 15);
foreach (@ARGV)
{
# Documentsetup
my $xmlfile = $_;
my $xml = XML::XPath->new(filename => $xmlfile);
s/\.xml$/\.pdf/i if /\.xml$/i;
my $pdffile = /\.pdf$/i ? $_ : "$_.pdf";
my $pdf = PDFLib->new( filename => $pdffile,
papersize=> $pagename,
creator => "XML2PDF",
title => (($xml->findnodes('document'))[0]->getAttribute('name'))
);
print "Converting $xmlfile to $pdffile \n";
# pre-process the images:
foreach my $img ($xml->findnodes('//img'))
{
my $filetype = 'jpeg' if $img->getAttribute('src')=~ /[jpg|jpeg]$/i;
$filetype = 'gif' if $img->getAttribute('src')=~ /gif$/i;
$filetype = 'png' if $img->getAttribute('src')=~ /png$/i;
my $filename = $img->getAttribute('src');
if ($filename =~ /^http:\/\//i)
{
$filename =~ /.*\/(.*\..*)/;
print $1."\n";
getstore($img->getAttribute('src'), $1);
$filename=$1;
}
$IMAGES{$img->getAttribute('src')} =
$pdf->load_image(filetype => $filetype || 'gif',
filename => $filename );
}
# Process the document
foreach my $section ($xml->findnodes('document/section1'))
{
#start every section on a new page
$pdf->start_page();
# Calculate the starting Y-axis value
$MAXYPOS = $PAGEHEIGHTPOINTS - $TOPMARGIN - $BOTTOMMARGIN;
# Print the header
printpdf($pdf, $FONTDEFINITION{section1},
"\n".$section->getAttribute('title')."\n\n", $MAXYPOS);
processsection($pdf, $section);
}
}
sub printpdf
{
my ($pdf, $font, $what, $ypos) = @_;
$pdf->set_font( @{$font} );
my $charsnotprinted = length($what);
while ($charsnotprinted)
{
$charsnotprinted = $pdf->print_boxed($what,
mode => "left",
x => $LEFTMARGIN,
y => $BOTTOMMARGIN,
w => $PAGEWIDTHPOINTS - $RIGHTMARGIN -
$LEFTMARGIN,
h => $ypos || $pdf->get_value("texty")
);
$what = substr($what, (length($what) - $charsnotprinted), length($what));
$pdf->start_page() if $charsnotprinted;
$ypos = $PAGEHEIGHTPOINTS - $TOPMARGIN - $BOTTOMMARGIN;
}
}
sub processsection
{
my ($pdf, $section) = @_;
foreach my $node ($section->findnodes('*'))
{
my $text;
if ($node->getName() eq "section2")
{
printpdf($pdf, $FONTDEFINITION{$node->getName()},
"\n".$node->getAttribute('title')."\n");
processsection($pdf, $node); #recurse!
}
if ($node->getName() eq "p")
{
printpdf($pdf, $FONTDEFINITION{$node->getName()},
"\n".$node->string_value()."\n");
processsection($pdf, $node); #recurse!
}
if ($node->getName() eq "img")
{
if ($pdf->get_value('texty')-$IMAGES{$node->getAttribute('src')}->height() < 0)
{
$pdf->start_page();
$pdf->set_text_pos($pdf->get_value('textx'), $PAGEHEIGHTPOINTS - $TOPMARGIN);
}
$pdf->add_image(img => $IMAGES{$node->getAttribute('src')},
x => $LEFTMARGIN,
y => $pdf->get_value('texty')-$IMAGES{$node->getAttribute('src')}->height());
$pdf->set_text_pos($pdf->get_value('textx'),
$pdf->get_value('texty')-$IMAGES{$node->getAttribute('src')}->height());
}
}
}