| 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 | #
|
|---|
| 8 | # Copyright (C) 1999-2001 New Zealand Digital Library Project
|
|---|
| 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 | ###########################################################################
|
|---|
| 25 | package PDFPlug;
|
|---|
| 26 |
|
|---|
| 27 | use ConvertToPlug;
|
|---|
| 28 | use unicode;
|
|---|
| 29 | use strict;
|
|---|
| 30 | no strict 'refs'; # so we can use a var for filehandles (eg STDERR)
|
|---|
| 31 |
|
|---|
| 32 | sub BEGIN {
|
|---|
| 33 | @PDFPlug::ISA = ('ConvertToPlug');
|
|---|
| 34 | }
|
|---|
| 35 |
|
|---|
| 36 | my $arguments =
|
|---|
| 37 | [ { 'name' => "process_exp",
|
|---|
| 38 | 'desc' => "{BasPlug.process_exp}",
|
|---|
| 39 | 'type' => "regexp",
|
|---|
| 40 | 'deft' => &get_default_process_exp(),
|
|---|
| 41 | 'reqd' => "no" },
|
|---|
| 42 | { 'name' => "block_exp",
|
|---|
| 43 | 'desc' => "{BasPlug.block_exp}",
|
|---|
| 44 | 'type' => "regexp",
|
|---|
| 45 | 'deft' => &get_default_block_exp() },
|
|---|
| 46 | { 'name' => "noimages",
|
|---|
| 47 | 'desc' => "{PDFPlug.noimages}",
|
|---|
| 48 | 'type' => "flag" },
|
|---|
| 49 | { 'name' => "complex",
|
|---|
| 50 | 'desc' => "{PDFPlug.complex}",
|
|---|
| 51 | 'type' => "flag" },
|
|---|
| 52 | { 'name' => "nohidden",
|
|---|
| 53 | 'desc' => "{PDFPlug.nohidden}",
|
|---|
| 54 | 'type' => "flag" },
|
|---|
| 55 | { 'name' => "zoom",
|
|---|
| 56 | 'desc' => "{PDFPlug.zoom}",
|
|---|
| 57 | 'deft' => "2",
|
|---|
| 58 | 'range' => "1,3", # actually the range is 0.5-3
|
|---|
| 59 | 'type' => "int" },
|
|---|
| 60 | { 'name' => "use_sections",
|
|---|
| 61 | 'desc' => "{PDFPlug.use_sections}",
|
|---|
| 62 | 'type' => "flag" } ];
|
|---|
| 63 |
|
|---|
| 64 | my $options = { 'name' => "PDFPlug",
|
|---|
| 65 | 'desc' => "{PDFPlug.desc}",
|
|---|
| 66 | 'abstract' => "no",
|
|---|
| 67 | 'inherits' => "yes",
|
|---|
| 68 | 'args' => $arguments };
|
|---|
| 69 |
|
|---|
| 70 | sub new {
|
|---|
| 71 | my ($class) = shift (@_);
|
|---|
| 72 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
|---|
| 73 | push(@$pluginlist, $class);
|
|---|
| 74 |
|
|---|
| 75 | push(@$inputargs,"-title_sub");
|
|---|
| 76 | push(@$inputargs,'^(Page\s+\d+)?(\s*1\s+)?');
|
|---|
| 77 |
|
|---|
| 78 | if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
|
|---|
| 79 | if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
|
|---|
| 80 |
|
|---|
| 81 | my @arg_array = @$inputargs;
|
|---|
| 82 | my $self = (defined $hashArgOptLists)? new ConvertToPlug($pluginlist,$inputargs,$hashArgOptLists): new ConvertToPlug($pluginlist,$inputargs);
|
|---|
| 83 |
|
|---|
| 84 | # these are passed through to gsConvert.pl by ConvertToPlug.pm
|
|---|
| 85 | my $zoom = $self->{"zoom"};
|
|---|
| 86 | $self->{'convert_options'} = "-pdf_zoom $zoom";
|
|---|
| 87 | $self->{'convert_options'} .= " -pdf_complex" if $self->{"complex"};
|
|---|
| 88 | $self->{'convert_options'} .= " -pdf_nohidden" if $self->{"nohidden"};
|
|---|
| 89 | $self->{'convert_options'} .= " -pdf_ignore_images" if $self->{"noimages"};
|
|---|
| 90 |
|
|---|
| 91 | my $secondary_plugin_options = $self->{'secondary_plugin_options'};
|
|---|
| 92 |
|
|---|
| 93 | if (!defined $secondary_plugin_options->{'HTMLPlug'}) {
|
|---|
| 94 | $secondary_plugin_options->{'HTMLPlug'} = [];
|
|---|
| 95 | }
|
|---|
| 96 | if (!defined $secondary_plugin_options->{'TEXTPlug'}) {
|
|---|
| 97 | $secondary_plugin_options->{'TEXTPlug'} = [];
|
|---|
| 98 | }
|
|---|
| 99 | if (defined $self->{'convert_to'} && $self->{'convert_to'} =~ /pagedimg.*/i) {
|
|---|
| 100 | if (!defined $secondary_plugin_options->{'PagedImgPlug'}){
|
|---|
| 101 | $secondary_plugin_options->{'PagedImgPlug'} = [];
|
|---|
| 102 | my $pagedimg_options = $secondary_plugin_options->{'PagedImgPlug'};
|
|---|
| 103 | push(@$pagedimg_options, "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
|
|---|
| 104 | }
|
|---|
| 105 | }
|
|---|
| 106 | my $html_options = $secondary_plugin_options->{'HTMLPlug'};
|
|---|
| 107 | my $text_options = $secondary_plugin_options->{'TEXTPlug'};
|
|---|
| 108 |
|
|---|
| 109 | if ($self->{'input_encoding'} eq "auto") {
|
|---|
| 110 | # pdftohtml will always produce html files encoded as utf-8
|
|---|
| 111 | # => restrict primary PDFPlug and secondary HTML plugin to use
|
|---|
| 112 | # utf8 and extract language.
|
|---|
| 113 | $self->{'input_encoding'} = "utf8";
|
|---|
| 114 | $self->{'extract_language'} = 1;
|
|---|
| 115 |
|
|---|
| 116 | push(@$html_options,"-input_encoding", "utf8");
|
|---|
| 117 | push(@$html_options,"-extract_language");
|
|---|
| 118 | }
|
|---|
| 119 |
|
|---|
| 120 | # Instruct HTMLPlug (when eventually accessed through read_into_doc_obj)
|
|---|
| 121 | # to extract these metadata fields from the HEAD META fields
|
|---|
| 122 | push(@$html_options,"-metadata_fields","Title,GENERATOR,date,author<Creator>");
|
|---|
| 123 |
|
|---|
| 124 | if ($self->{'use_sections'}) {
|
|---|
| 125 | $self->{'description_tags'} = 1;
|
|---|
| 126 | push(@$html_options,"-description_tags");
|
|---|
| 127 | }
|
|---|
| 128 |
|
|---|
| 129 | # following title_sub removes "Page 1" added by pdftohtml, and a leading
|
|---|
| 130 | # "1", which is often the page number at the top of the page. Bad Luck
|
|---|
| 131 | # if your document title actually starts with "1 " - is there a better way?
|
|---|
| 132 | push(@$html_options , "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
|
|---|
| 133 | push(@$text_options , "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
|
|---|
| 134 |
|
|---|
| 135 | $self = bless $self, $class;
|
|---|
| 136 | $self->load_secondary_plugins($class,$secondary_plugin_options,$hashArgOptLists);
|
|---|
| 137 | return $self;
|
|---|
| 138 | }
|
|---|
| 139 |
|
|---|
| 140 | sub get_default_process_exp {
|
|---|
| 141 | my $self = shift (@_);
|
|---|
| 142 |
|
|---|
| 143 | return q^(?i)\.pdf$^;
|
|---|
| 144 | }
|
|---|
| 145 |
|
|---|
| 146 | # so we don't inherit HTMLPlug's block exp...
|
|---|
| 147 | sub get_default_block_exp {
|
|---|
| 148 | return "";
|
|---|
| 149 | }
|
|---|
| 150 |
|
|---|
| 151 | sub convert_post_process
|
|---|
| 152 | {
|
|---|
| 153 | my $self = shift (@_);
|
|---|
| 154 | my ($conv_filename) = @_;
|
|---|
| 155 |
|
|---|
| 156 | my $outhandle=$self->{'outhandle'};
|
|---|
| 157 |
|
|---|
| 158 | my ($language, $encoding) = $self->textcat_get_language_encoding ($conv_filename);
|
|---|
| 159 |
|
|---|
| 160 | # read in file ($text will be in utf8)
|
|---|
| 161 | my $text = "";
|
|---|
| 162 | $self->read_file ($conv_filename, $encoding, $language, \$text);
|
|---|
| 163 |
|
|---|
| 164 | # Calculate number of pages based on <a ...> tags (we have a <a name=1> etc
|
|---|
| 165 | # for each page). Metadata based on this calculation not set until process()
|
|---|
| 166 | #
|
|---|
| 167 | # Note: this is done even if we are not breaking to document into pages as it might
|
|---|
| 168 | # be useful to give an indication of document length in browser through setting
|
|---|
| 169 | # num_pages as metadata.
|
|---|
| 170 | my @pages = ($text =~ /\<[Aa] name=\"?\w+\"?>/ig);
|
|---|
| 171 | my $num_pages = scalar(@pages);
|
|---|
| 172 | $self->{'num_pages'} = $num_pages;
|
|---|
| 173 |
|
|---|
| 174 | if ($self->{'use_sections'}
|
|---|
| 175 | && $self->{'converted_to'} eq "HTML") {
|
|---|
| 176 |
|
|---|
| 177 | print $outhandle "PDFPlug: Calculating sections...\n";
|
|---|
| 178 |
|
|---|
| 179 | # we have "<a name=1></a>" etc for each page
|
|---|
| 180 | # it may be <A name=
|
|---|
| 181 | my @sections = split('<[Aa] name=', $text);
|
|---|
| 182 |
|
|---|
| 183 | my $top_section = "";
|
|---|
| 184 |
|
|---|
| 185 | if (scalar (@sections) == 1) { #only one section - no split!
|
|---|
| 186 | print $outhandle "PDFPlug: warning - no sections found\n";
|
|---|
| 187 | } else {
|
|---|
| 188 | $top_section .= shift @sections; # keep HTML header etc as top_section
|
|---|
| 189 | }
|
|---|
| 190 |
|
|---|
| 191 | # handle first section specially for title? Or all use first 100...
|
|---|
| 192 |
|
|---|
| 193 | my $title = $sections[0];
|
|---|
| 194 | $title =~ s/^\"?\w+\"?>//; # specific for pdftohtml...
|
|---|
| 195 | $title =~ s/<\/([^>]+)><\1>//g; # (eg) </b><b> - no space
|
|---|
| 196 | $title =~ s/<[^>]*>/ /g;
|
|---|
| 197 | $title =~ s/(?: |\xc2\xa0)/ /g; # utf-8 for nbsp...
|
|---|
| 198 | $title =~ s/^\s+//s;
|
|---|
| 199 | $title =~ s/\s+$//;
|
|---|
| 200 | $title =~ s/\s+/ /gs;
|
|---|
| 201 | $title =~ s/^$self->{'title_sub'}// if ($self->{'title_sub'});
|
|---|
| 202 | $title =~ s/^\s+//s; # in case title_sub introduced any...
|
|---|
| 203 | $title = substr ($title, 0, 100);
|
|---|
| 204 | $title =~ s/\s\S*$/.../;
|
|---|
| 205 |
|
|---|
| 206 |
|
|---|
| 207 | if (scalar (@sections) == 1) { # no sections found
|
|---|
| 208 | $top_section .= $sections[0];
|
|---|
| 209 | @sections=();
|
|---|
| 210 | } else {
|
|---|
| 211 | $top_section .= "<!--<Section>\n<Metadata name=\"Title\">$title</Metadata>\n-->\n <!--</Section>-->\n";
|
|---|
| 212 | }
|
|---|
| 213 |
|
|---|
| 214 | # add metadata per section...
|
|---|
| 215 | foreach my $section (@sections) {
|
|---|
| 216 | # section names are not always just digits, may be like "outline"
|
|---|
| 217 | $section =~ s@^\"?(\w+)\"?></a>@@; # leftover from split expression...
|
|---|
| 218 |
|
|---|
| 219 | $title = $1; # Greenstone does magic if sections are titled digits
|
|---|
| 220 | if (! defined($title) ) {
|
|---|
| 221 | print STDERR "no title: $section\n";
|
|---|
| 222 | $title = " "; # get rid of the undefined warning in next line
|
|---|
| 223 | }
|
|---|
| 224 | my $newsection = "<!-- from PDFPlug -->\n<!-- <Section>\n";
|
|---|
| 225 | $newsection .= "<Metadata name=\"Title\">" . $title
|
|---|
| 226 | . "</Metadata>\n--><p>\n";
|
|---|
| 227 | $newsection .= $section;
|
|---|
| 228 | $newsection .= "<!--</Section>-->\n";
|
|---|
| 229 | $section = $newsection;
|
|---|
| 230 | }
|
|---|
| 231 |
|
|---|
| 232 | $text=join('', ($top_section, @sections));
|
|---|
| 233 | }
|
|---|
| 234 |
|
|---|
| 235 | # turn any high bytes that aren't valid utf-8 into utf-8.
|
|---|
| 236 | unicode::ensure_utf8(\$text);
|
|---|
| 237 |
|
|---|
| 238 | # Write it out again!
|
|---|
| 239 | $self->utf8_write_file (\$text, $conv_filename);
|
|---|
| 240 | }
|
|---|
| 241 |
|
|---|
| 242 |
|
|---|
| 243 | # do plugin specific processing of doc_obj for HTML type
|
|---|
| 244 | sub process {
|
|---|
| 245 | my $self = shift (@_);
|
|---|
| 246 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
|---|
| 247 |
|
|---|
| 248 | my $result = $self->process_type("pdf",$base_dir,$file,$doc_obj);
|
|---|
| 249 |
|
|---|
| 250 | # fix up the extracted date metadata to be in Greenstone date format,
|
|---|
| 251 | # and fix the capitalisation of 'date'
|
|---|
| 252 | my $cursection = $doc_obj->get_top_section();
|
|---|
| 253 |
|
|---|
| 254 | foreach my $datemeta (@{$doc_obj->get_metadata($cursection, "date")}) {
|
|---|
| 255 | $doc_obj->delete_metadata($cursection, "date", $datemeta);
|
|---|
| 256 |
|
|---|
| 257 | # We're just interested in the date bit, not the time
|
|---|
| 258 | # some pdf creators (eg "Acrobat 5.0 Scan Plug-in for Windows")
|
|---|
| 259 | # set a /CreationDate, and set /ModDate to 000000000. pdftohtml
|
|---|
| 260 | # extracts the ModDate, so it is 0...
|
|---|
| 261 | $datemeta =~ /(\d+)-(\d+)-(\d+)/;
|
|---|
| 262 | my ($year, $month, $day) = ($1,$2,$3);
|
|---|
| 263 | if (defined($year) && defined($month) && defined($day)) {
|
|---|
| 264 | if ($year == 0) {next}
|
|---|
| 265 | if ($year < 100) {$year += 1900} # just to be safe
|
|---|
| 266 | if ($month =~ /^\d$/) {$month="0$month"} # single digit
|
|---|
| 267 | if ($day =~ /^\d$/) {$day="0$day"} # single digit
|
|---|
| 268 | my $date="$year$month$day";
|
|---|
| 269 | $doc_obj->add_utf8_metadata($cursection, "Date", $date);
|
|---|
| 270 | }
|
|---|
| 271 | }
|
|---|
| 272 |
|
|---|
| 273 | $doc_obj->add_utf8_metadata($cursection, "NumPages", $self->{'num_pages'});
|
|---|
| 274 |
|
|---|
| 275 | if ($self->{'use_sections'} && $self->{'converted_to'} eq "HTML") {
|
|---|
| 276 | # we explicitly make it a paged document, cos greenstone won't get it
|
|---|
| 277 | # right if any section has an empty title, or one with letters in it
|
|---|
| 278 | $doc_obj->set_utf8_metadata_element ($cursection, "gsdlthistype", "Paged");
|
|---|
| 279 | }
|
|---|
| 280 |
|
|---|
| 281 | return $result;
|
|---|
| 282 | }
|
|---|
| 283 |
|
|---|
| 284 | 1;
|
|---|
