source: trunk/gsdl/bin/script/pdftohtml.pl@ 2599

#!/usr/bin/perl -w


###########################################################################
#
# pdftohtml.pl -- convert documents to HTML or TEXT format
#
# A component of the Greenstone digital library software
# from the New Zealand Digital Library Project at the 
# University of Waikato, New Zealand.
#
# Copyright (C) 1999 New Zealand Digital Library Project
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
###########################################################################

# pdftohtml.pl is a wrapper for running pdftohtml utility which converts
# PDF documents to HTML, and converts images to PNG format for display in
# the HTML pages generated

BEGIN {
    die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
    unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
}

use parsargv;
use util;
use Cwd;
use File::Basename;

sub print_usage {
# note - we don't actually ever use most of these options...
print STDERR  
    ("pdftohtml version 0.22 - modified for NZDL use\n",
     "Usage: pdftohtml [options] <PDF-file> <html-file>\n",
     "  -f <int>      : first page to convert\n",
     "  -l <int>      : last page to convert\n",
     "  -d <dir>      : target directory (default: basename of pdf-file)\n",
     "  -o <file>     : name of output file; - means stdout (default index.html)\n",
     "  -q            : don't print any messages or errors\n",
     "  -h            : print this usage information\n",
     "  -p            : exchange .pdf links by .html\n",
# these options now have no effect in gs-custom pdftohtml
#     "  -c            : generate complex HTML document\n",
#     "  -F            : don't use frames in HTML document\n",
     "  -i            : ignore images\n",
     "  -e <string>   : set extension for images (in the Html-file) (default png)\n"
     );
exit (1);
}

sub main {
    my (@ARGV) = @_;
    my ($first,$last,$target_dir,$out_file,$img_ext,
    $optq,$opth,$optp,$optF,$opti);
    
    # read command-line arguments so that
    # you can change the command in this script
    if (!parsargv::parse(\@ARGV,
             'f/\d+/1', \$first,
             'l/\d+/1', \$last,
             'd/[\S]*/', \$target_dir,
             'o/[\S]*/', \$out_file,
             'e/[\S]*/', \$img_ext,
             'q', \$optq,
             'h', \$opth,
             'p', \$optp,
#            'c', \$optc,
             'F', \$optF,
             'i', \$opti
             ))
    {
    print_usage();
    }

    # Make sure the input file exists and can be opened for reading
    if (scalar(@ARGV!=2)) {
    print_usage();
    }

    my $input_filename = $ARGV[0];
    my $output_filestem = $ARGV[1];
    $output_filestem =~ s/\.html$//; # pdftohtml adds this suffix

    my @dir = split (/(\/|\\)/, $input_filename);
    pop(@dir);
    my $dir = join ("", @dir);

    if (!-r $input_filename) {
    print STDERR "Error: unable to open $input_filename for reading\n";
    exit(1);
    }

    # Heuristical code removed due to pdftohtml being "fixed" to not
    # create bitmaps for each char in some pdfs. However, this means we
    # now create .html files even if we can't extract any text. We should
    # check for that now instead someday...


    # formulate the command
    my $cmd = &util::filename_cat($ENV{'GSDLHOME'}, "bin", $ENV{'GSDLOS'}, "pdftohtml");

    # don't include path on windows (to avoid having to play about
    # with quoting when GSDLHOME might contain spaces) but assume
    # that the PATH is set up correctly.
    $cmd = "pdftohtml" if ($ENV{'GSDLOS'} =~ /^windows$/);

    if ($timeout) {$cmd = "ulimit -t $timeout; $cmd";}
    $cmd .= " -noframes \"$input_filename\" \"$output_filestem.html\"";
    $cmd .= " > \"$output_filestem.out\"";

    # attempting to redirect STDERR on windows 95/98 is a bad idea
    $cmd .= " 2> \"$output_filestem.err\"" 
    if $ENV{'GSDLOS'} !~ /^windows$/i;

# system() returns -1 if it can't run, otherwise it's $cmds ret val.
    if (system($cmd)!=0) {
    print STDERR "Error executing $cmd: $!\n";
    &util::rm("$output_filestem.text") if (-e "$output_filestem.text");
    &util::rm("$output_filestem.err") if (-e "$output_filestem.err");
    return 0;
    }

# post-process to remove </b><b> and </i><i>, as these break up
# words, screwing up indexing and searching.
    &util::mv("$output_filestem.html","$output_filestem.html.tmp");
    open INFILE, "$output_filestem.html.tmp" ||
    die "Couldn't open file: $!";
    open OUTFILE, ">$output_filestem.html" ||
    die "Couldn't open file for writing: $!";
    my $line;
    while ($line=<INFILE>) {
    $line =~ s#</b><b>##g;
    $line =~ s#</i><i>##g;
    print OUTFILE $line;
    }
    close INFILE;
    close OUTFILE;
    &util::rm("$output_filestem.html.tmp");


    # Need to convert images from PPM format to PNG format
    my @images;


    open (IMAGES, "images.log") || 
    open (IMAGES, "image.log") ||
          print STDERR "Error opening image log:$!\n";
    while (<IMAGES>) {
    push (@images, $_);
    }
    close IMAGES;

    for $image (@images) {
    chomp($image);
    my $cmd = "";
    if ($ENV{'GSDLOS'} =~ /^windows/i) {
        $cmd = "pnmtopng $image";
        if (system($cmd)!=0) {
        print STDERR "Error executing $cmd\n";
        #return 0; # not sure about whether to leave this one in or take it out
        next;
        }
    } else {
        my @nameparts = split(/\./, $image);
        my $image_base = shift(@nameparts);

        $cmd = "pnmtopng $image > $image_base.png 2>/dev/null";
        if (system($cmd)!=0) {
        $cmd = "convert $image $image_base.png 2>/dev/null";
        if (system($cmd)!=0) {
            print STDERR "Cannot convert $image into PNG format (tried `pnmtopng' and `convert')...\n";
            #return 0; # not sure about whether to leave this one in or take it out
            next;
        }
        }
    }
    &util::rm($image);
    }

    return 1;
}

# indicate our error status
if (&main(@ARGV)) {exit 0;}
exit 1;