[1410] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # PDFPlug.pm -- reasonably with-it pdf plugin
|
---|
| 4 | # A component of the Greenstone digital library software
|
---|
| 5 | # from the New Zealand Digital Library Project at the
|
---|
| 6 | # University of Waikato, New Zealand.
|
---|
| 7 | #
|
---|
[2661] | 8 | # Copyright (C) 1999-2001 New Zealand Digital Library Project
|
---|
[1410] | 9 | #
|
---|
| 10 | # This program is free software; you can redistribute it and/or modify
|
---|
| 11 | # it under the terms of the GNU General Public License as published by
|
---|
| 12 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 13 | # (at your option) any later version.
|
---|
| 14 | #
|
---|
| 15 | # This program is distributed in the hope that it will be useful,
|
---|
| 16 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 17 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 18 | # GNU General Public License for more details.
|
---|
| 19 | #
|
---|
| 20 | # You should have received a copy of the GNU General Public License
|
---|
| 21 | # along with this program; if not, write to the Free Software
|
---|
| 22 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 23 | #
|
---|
| 24 | ###########################################################################
|
---|
[8227] | 25 | use strict;
|
---|
| 26 | no strict 'refs'; # so we can use a var for filehandles (eg STDERR)
|
---|
[1410] | 27 |
|
---|
| 28 | package PDFPlug;
|
---|
| 29 |
|
---|
| 30 | use ConvertToPlug;
|
---|
[8218] | 31 | use unicode;
|
---|
[1410] | 32 |
|
---|
[8716] | 33 | sub BEGIN {
|
---|
[8227] | 34 | @PDFPlug::ISA = ('ConvertToPlug');
|
---|
[8716] | 35 | }
|
---|
[1410] | 36 |
|
---|
[4744] | 37 | my $arguments =
|
---|
| 38 | [ { 'name' => "process_exp",
|
---|
[4873] | 39 | 'desc' => "{BasPlug.process_exp}",
|
---|
[6408] | 40 | 'type' => "regexp",
|
---|
[4744] | 41 | 'deft' => &get_default_process_exp(),
|
---|
| 42 | 'reqd' => "no" },
|
---|
| 43 | { 'name' => "block_exp",
|
---|
[4873] | 44 | 'desc' => "{BasPlug.block_exp}",
|
---|
[6408] | 45 | 'type' => "regexp",
|
---|
[4873] | 46 | 'deft' => &get_default_block_exp() },
|
---|
[4744] | 47 | { 'name' => "noimages",
|
---|
[4873] | 48 | 'desc' => "{PDFPlug.noimages}",
|
---|
[4744] | 49 | 'type' => "flag" },
|
---|
| 50 | { 'name' => "complex",
|
---|
[4873] | 51 | 'desc' => "{PDFPlug.complex}",
|
---|
[4744] | 52 | 'type' => "flag" },
|
---|
| 53 | { 'name' => "nohidden",
|
---|
[4873] | 54 | 'desc' => "{PDFPlug.nohidden}",
|
---|
[4744] | 55 | 'type' => "flag" },
|
---|
| 56 | { 'name' => "zoom",
|
---|
[4873] | 57 | 'desc' => "{PDFPlug.zoom}",
|
---|
[4744] | 58 | 'deft' => "2",
|
---|
[7107] | 59 | 'range' => "1,3", # actually the range is 0.5-3
|
---|
[4744] | 60 | 'type' => "int" },
|
---|
| 61 | { 'name' => "use_sections",
|
---|
[4873] | 62 | 'desc' => "{PDFPlug.use_sections}",
|
---|
[4744] | 63 | 'type' => "flag" } ];
|
---|
[3540] | 64 |
|
---|
| 65 | my $options = { 'name' => "PDFPlug",
|
---|
[5680] | 66 | 'desc' => "{PDFPlug.desc}",
|
---|
[6408] | 67 | 'abstract' => "no",
|
---|
[3540] | 68 | 'inherits' => "yes",
|
---|
| 69 | 'args' => $arguments };
|
---|
| 70 |
|
---|
[1410] | 71 | sub new {
|
---|
| 72 | my $class = shift (@_);
|
---|
[2452] | 73 |
|
---|
[4103] | 74 | my ($noimages, $complex, $zoom, $use_sections, $nohidden);
|
---|
[5616] | 75 |
|
---|
| 76 | my @args=@_;
|
---|
| 77 |
|
---|
[3720] | 78 | if (!parsargv::parse(\@_,
|
---|
| 79 | q^noimages^, \$noimages,
|
---|
| 80 | q^complex^, \$complex,
|
---|
| 81 | q^zoom/\d+/2^, \$zoom,
|
---|
[4103] | 82 | q^nohidden^, \$nohidden,
|
---|
[5139] | 83 | q^use_sections^, \$use_sections,
|
---|
[3720] | 84 | "allow_extra_options")) {
|
---|
| 85 |
|
---|
[8227] | 86 | my $self = new ConvertToPlug($class, @_);
|
---|
[7019] | 87 | my $outhandle=$self->{'outhandle'};
|
---|
| 88 | print $outhandle "\nIncorrect options passed to PDFPlug, check your collect.cfg configuration file\n";
|
---|
[4873] | 89 | $self->print_txt_usage(""); # Use default resource bundle
|
---|
[7019] | 90 | exit 1;
|
---|
[3720] | 91 | }
|
---|
| 92 |
|
---|
[3833] | 93 |
|
---|
| 94 | if ($use_sections) {
|
---|
| 95 | push (@args, "-description_tags");
|
---|
| 96 | }
|
---|
| 97 |
|
---|
[2452] | 98 | # following title_sub removes "Page 1" added by pdftohtml, and a leading
|
---|
| 99 | # "1", which is often the page number at the top of the page. Bad Luck
|
---|
| 100 | # if your document title actually starts with "1 " - is there a better way?
|
---|
| 101 |
|
---|
[3411] | 102 | my $self = new ConvertToPlug ($class, @args, "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
|
---|
[5924] | 103 | $self->{'plugin_type'} = "PDFPlug";
|
---|
[3411] | 104 | if ($use_sections) {
|
---|
| 105 | $self->{'use_sections'}=1;
|
---|
| 106 | }
|
---|
[3833] | 107 |
|
---|
[3540] | 108 | # 14-05-02 To allow for proper inheritance of arguments - John Thompson
|
---|
| 109 | my $option_list = $self->{'option_list'};
|
---|
| 110 | push( @{$option_list}, $options );
|
---|
[3720] | 111 |
|
---|
| 112 | # these are passed through to gsConvert.pl by ConvertToPlug.pm
|
---|
| 113 | $self->{'convert_options'} = "-pdf_zoom $zoom";
|
---|
| 114 | $self->{'convert_options'} .= " -pdf_complex" if $complex;
|
---|
[4103] | 115 | $self->{'convert_options'} .= " -pdf_nohidden" if $nohidden;
|
---|
[3720] | 116 | $self->{'convert_options'} .= " -pdf_ignore_images" if $noimages;
|
---|
| 117 |
|
---|
| 118 | # pdftohtml will always produce html files encoded as utf-8
|
---|
| 119 | if ($self->{'input_encoding'} eq "auto") {
|
---|
| 120 | $self->{'input_encoding'} = "utf8";
|
---|
| 121 | $self->{'extract_language'} = 1;
|
---|
| 122 | }
|
---|
[5616] | 123 |
|
---|
[1410] | 124 | return bless $self, $class;
|
---|
| 125 | }
|
---|
| 126 |
|
---|
| 127 | sub get_default_process_exp {
|
---|
| 128 | my $self = shift (@_);
|
---|
| 129 |
|
---|
| 130 | return q^(?i)\.pdf$^;
|
---|
| 131 | }
|
---|
[2661] | 132 |
|
---|
| 133 | # so we don't inherit HTMLPlug's block exp...
|
---|
| 134 | sub get_default_block_exp {
|
---|
| 135 | return "";
|
---|
| 136 | }
|
---|
[1410] | 137 |
|
---|
| 138 |
|
---|
| 139 | # do plugin specific processing of doc_obj for HTML type
|
---|
| 140 | sub process {
|
---|
| 141 | my $self = shift (@_);
|
---|
[7019] | 142 | my $outhandle=$self->{'outhandle'};
|
---|
| 143 |
|
---|
[8218] | 144 | my $textref=$_[0];
|
---|
| 145 |
|
---|
[3411] | 146 | if ($self->{'use_sections'}
|
---|
| 147 | && $self->{'converted_to'} eq "HTML") {
|
---|
| 148 |
|
---|
[7019] | 149 | print $outhandle "PDFPlug: Calculating sections...\n";
|
---|
[3411] | 150 |
|
---|
[3614] | 151 | # we have "<a name=1></a>" etc for each page
|
---|
[8795] | 152 | # it may be <A name=
|
---|
| 153 | my @sections = split('<[Aa] name=', $$textref);
|
---|
[3411] | 154 |
|
---|
[7019] | 155 | if (scalar (@sections) == 1) { #only one section - no split!
|
---|
| 156 | print $outhandle "PDFPlug: warning - no sections found\n";
|
---|
| 157 | } else {
|
---|
| 158 | shift @sections; # don't need HTML header, etc
|
---|
| 159 | }
|
---|
| 160 |
|
---|
[3411] | 161 | # handle first section specially for title? Or all use first 100...
|
---|
| 162 |
|
---|
| 163 | my $title = $sections[0];
|
---|
[8795] | 164 | $title =~ s/^\"?\w+\"?>//; # specific for pdftohtml...
|
---|
[3411] | 165 | $title =~ s/<\/([^>]+)><\1>//g; # (eg) </b><b> - no space
|
---|
| 166 | $title =~ s/<[^>]*>/ /g;
|
---|
| 167 | $title =~ s/(?: |\xc2\xa0)/ /g; # utf-8 for nbsp...
|
---|
| 168 | $title =~ s/^\s+//s;
|
---|
| 169 | $title =~ s/\s+$//;
|
---|
| 170 | $title =~ s/\s+/ /gs;
|
---|
| 171 | $title =~ s/^$self->{'title_sub'}// if ($self->{'title_sub'});
|
---|
| 172 | $title =~ s/^\s+//s; # in case title_sub introduced any...
|
---|
| 173 | $title = substr ($title, 0, 100);
|
---|
| 174 | $title =~ s/\s\S*$/.../;
|
---|
| 175 |
|
---|
[7019] | 176 | my $top_section;
|
---|
| 177 | if (scalar (@sections) == 1) { # no sections found
|
---|
| 178 | $top_section=$sections[0];
|
---|
| 179 | @sections=();
|
---|
| 180 | } else {
|
---|
| 181 | $top_section = "<!--<Section>\n<Metadata name=\"Title\">$title</Metadata>\n-->\n <!--</Section>-->\n";
|
---|
| 182 | }
|
---|
[3411] | 183 |
|
---|
| 184 | # add metadata per section...
|
---|
| 185 | foreach my $section (@sections) {
|
---|
[8795] | 186 | # section names are not always just digits, may be like "outline"
|
---|
| 187 | $section =~ s@^\"?(\w+)\"?></a>@@; # leftover from split expression...
|
---|
[3614] | 188 |
|
---|
[3411] | 189 | $title = $1; # Greenstone does magic if sections are titled digits
|
---|
| 190 | if (! defined($title) ) {
|
---|
| 191 | print STDERR "no title: $section\n";
|
---|
[8795] | 192 | $title = " "; # get rid of the undefined warning in next line
|
---|
[3411] | 193 | }
|
---|
| 194 | my $newsection = "<!-- from PDFPlug -->\n<!-- <Section>\n";
|
---|
| 195 | $newsection .= "<Metadata name=\"Title\">" . $title
|
---|
| 196 | . "</Metadata>\n--><p>\n";
|
---|
| 197 | $newsection .= $section;
|
---|
| 198 | $newsection .= "<!--</Section>-->\n";
|
---|
| 199 | $section = $newsection;
|
---|
| 200 | }
|
---|
| 201 |
|
---|
| 202 | $$textref=join('', ($top_section, @sections));
|
---|
| 203 | }
|
---|
| 204 |
|
---|
[8218] | 205 | # turn any high bytes that aren't valid utf-8 into utf-8.
|
---|
| 206 | unicode::ensure_utf8($textref);
|
---|
| 207 |
|
---|
[2979] | 208 | print $outhandle "PDFPlug: passing $_[3] on to $self->{'converted_to'}Plug\n"
|
---|
[1410] | 209 | if $self->{'verbosity'} > 1;
|
---|
[7287] | 210 |
|
---|
[8226] | 211 | # tell htmlplug to extract these metadata fields from the HEAD META fields
|
---|
| 212 | $self->{'metadata_fields'} .= ",date,author<Creator>";
|
---|
[7287] | 213 |
|
---|
| 214 | my $result = ConvertToPlug::process_type($self,"pdf",@_);
|
---|
| 215 |
|
---|
| 216 | my $doc_obj = pop(@_);
|
---|
[8226] | 217 |
|
---|
| 218 | # fix up the extracted date metadata to be in Greenstone date format,
|
---|
| 219 | # and fix the capitalisation of 'date'
|
---|
[8227] | 220 | my $cursection = $doc_obj->get_top_section();
|
---|
| 221 |
|
---|
| 222 | foreach my $datemeta (@{$doc_obj->get_metadata($cursection, "date")}) {
|
---|
[7287] | 223 | $doc_obj->delete_metadata($cursection, "date", $datemeta);
|
---|
| 224 |
|
---|
| 225 | # We're just interested in the date bit, not the time
|
---|
[8278] | 226 | # some pdf creators (eg "Acrobat 5.0 Scan Plug-in for Windows")
|
---|
| 227 | # set a /CreationDate, and set /ModDate to 000000000. pdftohtml
|
---|
| 228 | # extracts the ModDate, so it is 0...
|
---|
| 229 | $datemeta =~ /(\d+)-(\d+)-(\d+)/;
|
---|
| 230 | my ($year, $month, $day) = ($1,$2,$3);
|
---|
| 231 | if (defined($year) && defined($month) && defined($day)) {
|
---|
| 232 | if ($year == 0) {next}
|
---|
| 233 | if ($year < 100) {$year += 1900} # just to be safe
|
---|
| 234 | if ($month =~ /^\d$/) {$month="0$month"} # single digit
|
---|
| 235 | if ($day =~ /^\d$/) {$day="0$day"} # single digit
|
---|
| 236 | my $date="$year$month$day";
|
---|
| 237 | $doc_obj->add_utf8_metadata($cursection, "Date", $date);
|
---|
| 238 | }
|
---|
[7287] | 239 | }
|
---|
| 240 |
|
---|
[8139] | 241 | # Add NumPages metadata (we have "<a name=1>" etc for each page)
|
---|
[8795] | 242 | my @pages = ($$textref =~ /\<[Aa] name=\"?\w+\"?>/ig);
|
---|
[8139] | 243 | $doc_obj->add_utf8_metadata($cursection, "NumPages", scalar(@pages));
|
---|
[8795] | 244 |
|
---|
| 245 | if ($self->{'use_sections'} && $self->{'converted_to'} eq "HTML") {
|
---|
| 246 | # we explicitly make it a paged document, cos greenstone won't get it
|
---|
| 247 | # right if any section has an empty title, or one with letters in it
|
---|
| 248 | $doc_obj->set_utf8_metadata_element ($cursection, "gsdlthistype", "Paged");
|
---|
| 249 | }
|
---|
[7287] | 250 | return $result;
|
---|
[1410] | 251 | }
|
---|
| 252 |
|
---|
| 253 | 1;
|
---|