[1410] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
[15872] | 3 | # PDFPlugin.pm -- reasonably with-it pdf plugin
|
---|
[1410] | 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 | ###########################################################################
|
---|
[15872] | 25 | package PDFPlugin;
|
---|
[1410] | 26 |
|
---|
[15872] | 27 | use ConvertBinaryFile;
|
---|
| 28 | use ReadTextFile;
|
---|
[8218] | 29 | use unicode;
|
---|
[10353] | 30 | use strict;
|
---|
| 31 | no strict 'refs'; # so we can use a var for filehandles (eg STDERR)
|
---|
[1410] | 32 |
|
---|
[8716] | 33 | sub BEGIN {
|
---|
[15872] | 34 | @PDFPlugin::ISA = ('ConvertBinaryFile', 'ReadTextFile');
|
---|
[8716] | 35 | }
|
---|
[1410] | 36 |
|
---|
[10452] | 37 | my $convert_to_list =
|
---|
| 38 | [ { 'name' => "auto",
|
---|
[15872] | 39 | 'desc' => "{ConvertBinaryFile.convert_to.auto}" },
|
---|
[10452] | 40 | { 'name' => "html",
|
---|
[15872] | 41 | 'desc' => "{ConvertBinaryFile.convert_to.html}" },
|
---|
[10452] | 42 | { 'name' => "text",
|
---|
[15872] | 43 | 'desc' => "{ConvertBinaryFile.convert_to.text}" },
|
---|
[10452] | 44 | { 'name' => "pagedimg_jpg",
|
---|
[15872] | 45 | 'desc' => "{ConvertBinaryFile.convert_to.pagedimg_jpg}"},
|
---|
[10452] | 46 | { 'name' => "pagedimg_gif",
|
---|
[15872] | 47 | 'desc' => "{ConvertBinaryFile.convert_to.pagedimg_gif}"},
|
---|
[10452] | 48 | { 'name' => "pagedimg_png",
|
---|
[15872] | 49 | 'desc' => "{ConvertBinaryFile.convert_to.pagedimg_png}"},
|
---|
[10452] | 50 | ];
|
---|
| 51 |
|
---|
| 52 |
|
---|
| 53 | my $arguments =
|
---|
[10889] | 54 | [
|
---|
| 55 | { 'name' => "convert_to",
|
---|
[15872] | 56 | 'desc' => "{ConvertBinaryFile.convert_to}",
|
---|
[10889] | 57 | 'type' => "enum",
|
---|
| 58 | 'reqd' => "yes",
|
---|
| 59 | 'list' => $convert_to_list,
|
---|
| 60 | 'deft' => "html" },
|
---|
| 61 | { 'name' => "process_exp",
|
---|
[15872] | 62 | 'desc' => "{BasePlugin.process_exp}",
|
---|
[10889] | 63 | 'type' => "regexp",
|
---|
| 64 | 'deft' => &get_default_process_exp(),
|
---|
| 65 | 'reqd' => "no" },
|
---|
| 66 | { 'name' => "block_exp",
|
---|
[15872] | 67 | 'desc' => "{BasePlugin.block_exp}",
|
---|
[10889] | 68 | 'type' => "regexp",
|
---|
| 69 | 'deft' => &get_default_block_exp() },
|
---|
| 70 | { 'name' => "metadata_fields",
|
---|
[15872] | 71 | 'desc' => "{HTMLPlugin.metadata_fields}",
|
---|
[10889] | 72 | 'type' => "string",
|
---|
| 73 | 'deft' => "" },
|
---|
| 74 | { 'name' => "noimages",
|
---|
[15872] | 75 | 'desc' => "{PDFPlugin.noimages}",
|
---|
[10889] | 76 | 'type' => "flag" },
|
---|
| 77 | { 'name' => "allowimagesonly",
|
---|
[15872] | 78 | 'desc' => "{PDFPlugin.allowimagesonly}",
|
---|
[10889] | 79 | 'type' => "flag" },
|
---|
| 80 | { 'name' => "complex",
|
---|
[15872] | 81 | 'desc' => "{PDFPlugin.complex}",
|
---|
[10889] | 82 | 'type' => "flag" },
|
---|
| 83 | { 'name' => "nohidden",
|
---|
[15872] | 84 | 'desc' => "{PDFPlugin.nohidden}",
|
---|
[10889] | 85 | 'type' => "flag" },
|
---|
| 86 | { 'name' => "zoom",
|
---|
[15872] | 87 | 'desc' => "{PDFPlugin.zoom}",
|
---|
[10889] | 88 | 'deft' => "2",
|
---|
| 89 | 'range' => "1,3", # actually the range is 0.5-3
|
---|
| 90 | 'type' => "int" },
|
---|
| 91 | { 'name' => "use_sections",
|
---|
[15872] | 92 | 'desc' => "{PDFPlugin.use_sections}",
|
---|
[10889] | 93 | 'type' => "flag" },
|
---|
| 94 | { 'name' => "description_tags",
|
---|
[15872] | 95 | 'desc' => "{HTMLPlugin.description_tags}",
|
---|
[10889] | 96 | 'type' => "flag" }
|
---|
| 97 | ];
|
---|
[3540] | 98 |
|
---|
[15872] | 99 | my $options = { 'name' => "PDFPlugin",
|
---|
| 100 | 'desc' => "{PDFPlugin.desc}",
|
---|
[6408] | 101 | 'abstract' => "no",
|
---|
[3540] | 102 | 'inherits' => "yes",
|
---|
[15114] | 103 | 'srcreplaceable' => "yes", # Source docs in PDF can be replaced with GS-generated html
|
---|
[3540] | 104 | 'args' => $arguments };
|
---|
| 105 |
|
---|
[1410] | 106 | sub new {
|
---|
[10218] | 107 | my ($class) = shift (@_);
|
---|
| 108 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
| 109 | push(@$pluginlist, $class);
|
---|
[2452] | 110 |
|
---|
[10218] | 111 | push(@$inputargs,"-title_sub");
|
---|
| 112 | push(@$inputargs,'^(Page\s+\d+)?(\s*1\s+)?');
|
---|
[5616] | 113 |
|
---|
[15872] | 114 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
| 115 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
[10429] | 116 |
|
---|
| 117 | my @arg_array = @$inputargs;
|
---|
[15872] | 118 | my $self = new ConvertBinaryFile($pluginlist, $inputargs, $hashArgOptLists);
|
---|
[10353] | 119 |
|
---|
[10580] | 120 | if ($self->{'info_only'}) {
|
---|
| 121 | # don't worry about any options etc
|
---|
| 122 | return bless $self, $class;
|
---|
| 123 | }
|
---|
| 124 |
|
---|
[15872] | 125 | $self->{'filename_extension'} = "pdf";
|
---|
| 126 | $self->{'file_type'} = "PDF";
|
---|
| 127 |
|
---|
| 128 | # these are passed through to gsConvert.pl by ConvertBinaryFile.pm
|
---|
[10218] | 129 | my $zoom = $self->{"zoom"};
|
---|
[3720] | 130 | $self->{'convert_options'} = "-pdf_zoom $zoom";
|
---|
[10218] | 131 | $self->{'convert_options'} .= " -pdf_complex" if $self->{"complex"};
|
---|
| 132 | $self->{'convert_options'} .= " -pdf_nohidden" if $self->{"nohidden"};
|
---|
| 133 | $self->{'convert_options'} .= " -pdf_ignore_images" if $self->{"noimages"};
|
---|
[10452] | 134 | $self->{'convert_options'} .= " -pdf_allow_images_only" if $self->{"allowimagesonly"};
|
---|
[3720] | 135 |
|
---|
[10273] | 136 | my $secondary_plugin_options = $self->{'secondary_plugin_options'};
|
---|
| 137 |
|
---|
[15872] | 138 | if (!defined $secondary_plugin_options->{'HTMLPlugin'}) {
|
---|
| 139 | $secondary_plugin_options->{'HTMLPlugin'} = [];
|
---|
[10273] | 140 | }
|
---|
[15872] | 141 | if (!defined $secondary_plugin_options->{'TextPlugin'}) {
|
---|
| 142 | $secondary_plugin_options->{'TextPlugin'} = [];
|
---|
[10273] | 143 | }
|
---|
[10353] | 144 | if (defined $self->{'convert_to'} && $self->{'convert_to'} =~ /pagedimg.*/i) {
|
---|
[15872] | 145 | if (!defined $secondary_plugin_options->{'PagedImagePlugin'}){
|
---|
| 146 | $secondary_plugin_options->{'PagedImagePlugin'} = [];
|
---|
| 147 | my $pagedimg_options = $secondary_plugin_options->{'PagedImagePlugin'};
|
---|
[10429] | 148 | push(@$pagedimg_options, "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
|
---|
[10353] | 149 | }
|
---|
| 150 | }
|
---|
[15872] | 151 | my $html_options = $secondary_plugin_options->{'HTMLPlugin'};
|
---|
| 152 | my $text_options = $secondary_plugin_options->{'TextPlugin'};
|
---|
| 153 | my $pagedimg_options = $secondary_plugin_options->{'PagedImagePlugin'};
|
---|
[10429] | 154 |
|
---|
[15904] | 155 | # if ($self->{'input_encoding'} eq "auto") {
|
---|
| 156 | # $self->{'input_encoding'} = "utf8";
|
---|
| 157 | # }
|
---|
[10273] | 158 |
|
---|
[10835] | 159 | # if pdftohtml is always producing utf8, then htmlplug always needs this option
|
---|
| 160 | push(@$html_options,"-input_encoding", "utf8");
|
---|
[12834] | 161 | push(@$html_options,"-extract_language") if $self->{'extract_language'};
|
---|
[10273] | 162 | # Instruct HTMLPlug (when eventually accessed through read_into_doc_obj)
|
---|
| 163 | # to extract these metadata fields from the HEAD META fields
|
---|
[10724] | 164 | my $required_metadata;
|
---|
| 165 | if (defined $self->{'metadata_fields'} && $self->{'metadata_fields'} =~ /\S/) {
|
---|
| 166 | push(@$html_options,"-metadata_fields",$self->{'metadata_fields'});
|
---|
| 167 | } else {
|
---|
| 168 | push(@$html_options,"-metadata_fields","Title,GENERATOR,date,author<Creator>");
|
---|
| 169 | }
|
---|
| 170 | #push(@$html_options,"-metadata_fields","Title,GENERATOR,date,author<Creator>");
|
---|
| 171 |
|
---|
[10514] | 172 | if ($self->{'use_sections'} || $self->{'description_tags'}) {
|
---|
[10273] | 173 | $self->{'description_tags'} = 1;
|
---|
| 174 | push(@$html_options,"-description_tags");
|
---|
| 175 | }
|
---|
[10429] | 176 |
|
---|
[10273] | 177 | # following title_sub removes "Page 1" added by pdftohtml, and a leading
|
---|
| 178 | # "1", which is often the page number at the top of the page. Bad Luck
|
---|
| 179 | # if your document title actually starts with "1 " - is there a better way?
|
---|
| 180 | push(@$html_options , "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
|
---|
| 181 | push(@$text_options , "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
|
---|
[11122] | 182 |
|
---|
| 183 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 184 | if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
|
---|
| 185 | push(@$html_options, "-associate_tail_re", $associate_tail_re);
|
---|
| 186 | push(@$text_options, "-associate_tail_re", $associate_tail_re);
|
---|
| 187 | push(@$pagedimg_options, "-associate_tail_re", $associate_tail_re);
|
---|
| 188 | }
|
---|
| 189 |
|
---|
[10429] | 190 |
|
---|
[10273] | 191 | $self = bless $self, $class;
|
---|
[10429] | 192 | $self->load_secondary_plugins($class,$secondary_plugin_options,$hashArgOptLists);
|
---|
[10273] | 193 | return $self;
|
---|
[1410] | 194 | }
|
---|
| 195 |
|
---|
| 196 | sub get_default_process_exp {
|
---|
| 197 | my $self = shift (@_);
|
---|
| 198 |
|
---|
| 199 | return q^(?i)\.pdf$^;
|
---|
| 200 | }
|
---|
[2661] | 201 |
|
---|
| 202 | # so we don't inherit HTMLPlug's block exp...
|
---|
| 203 | sub get_default_block_exp {
|
---|
| 204 | return "";
|
---|
| 205 | }
|
---|
[1410] | 206 |
|
---|
[10273] | 207 | sub convert_post_process
|
---|
| 208 | {
|
---|
[1410] | 209 | my $self = shift (@_);
|
---|
[10273] | 210 | my ($conv_filename) = @_;
|
---|
[9465] | 211 |
|
---|
[7019] | 212 | my $outhandle=$self->{'outhandle'};
|
---|
| 213 |
|
---|
[15963] | 214 | #$self->{'input_encoding'} = "utf8"; # The output is always in utf8 (is it?? it is for html, but what about other types?)
|
---|
| 215 | #my ($language, $encoding) = $self->textcat_get_language_encoding ($conv_filename);
|
---|
[8218] | 216 |
|
---|
[10273] | 217 | # read in file ($text will be in utf8)
|
---|
| 218 | my $text = "";
|
---|
[15963] | 219 | # encoding will be utf8 for html files - what about other types? will we do this step for them anyway?
|
---|
| 220 | $self->read_file ($conv_filename, "utf8", "", \$text);
|
---|
[10273] | 221 |
|
---|
| 222 | # Calculate number of pages based on <a ...> tags (we have a <a name=1> etc
|
---|
| 223 | # for each page). Metadata based on this calculation not set until process()
|
---|
| 224 | #
|
---|
| 225 | # Note: this is done even if we are not breaking to document into pages as it might
|
---|
| 226 | # be useful to give an indication of document length in browser through setting
|
---|
| 227 | # num_pages as metadata.
|
---|
| 228 | my @pages = ($text =~ /\<[Aa] name=\"?\w+\"?>/ig);
|
---|
| 229 | my $num_pages = scalar(@pages);
|
---|
| 230 | $self->{'num_pages'} = $num_pages;
|
---|
| 231 |
|
---|
[3411] | 232 | if ($self->{'use_sections'}
|
---|
| 233 | && $self->{'converted_to'} eq "HTML") {
|
---|
| 234 |
|
---|
[15872] | 235 | print $outhandle "PDFPlugin: Calculating sections...\n";
|
---|
[3411] | 236 |
|
---|
[3614] | 237 | # we have "<a name=1></a>" etc for each page
|
---|
[8795] | 238 | # it may be <A name=
|
---|
[10273] | 239 | my @sections = split('<[Aa] name=', $text);
|
---|
[3411] | 240 |
|
---|
[10273] | 241 | my $top_section = "";
|
---|
| 242 |
|
---|
[7019] | 243 | if (scalar (@sections) == 1) { #only one section - no split!
|
---|
[15872] | 244 | print $outhandle "PDFPlugin: warning - no sections found\n";
|
---|
[7019] | 245 | } else {
|
---|
[10273] | 246 | $top_section .= shift @sections; # keep HTML header etc as top_section
|
---|
[7019] | 247 | }
|
---|
| 248 |
|
---|
[3411] | 249 | # handle first section specially for title? Or all use first 100...
|
---|
| 250 |
|
---|
| 251 | my $title = $sections[0];
|
---|
[8795] | 252 | $title =~ s/^\"?\w+\"?>//; # specific for pdftohtml...
|
---|
[3411] | 253 | $title =~ s/<\/([^>]+)><\1>//g; # (eg) </b><b> - no space
|
---|
| 254 | $title =~ s/<[^>]*>/ /g;
|
---|
| 255 | $title =~ s/(?: |\xc2\xa0)/ /g; # utf-8 for nbsp...
|
---|
| 256 | $title =~ s/^\s+//s;
|
---|
| 257 | $title =~ s/\s+$//;
|
---|
| 258 | $title =~ s/\s+/ /gs;
|
---|
| 259 | $title =~ s/^$self->{'title_sub'}// if ($self->{'title_sub'});
|
---|
| 260 | $title =~ s/^\s+//s; # in case title_sub introduced any...
|
---|
| 261 | $title = substr ($title, 0, 100);
|
---|
| 262 | $title =~ s/\s\S*$/.../;
|
---|
| 263 |
|
---|
[10273] | 264 |
|
---|
[7019] | 265 | if (scalar (@sections) == 1) { # no sections found
|
---|
[10273] | 266 | $top_section .= $sections[0];
|
---|
[7019] | 267 | @sections=();
|
---|
| 268 | } else {
|
---|
[10273] | 269 | $top_section .= "<!--<Section>\n<Metadata name=\"Title\">$title</Metadata>\n-->\n <!--</Section>-->\n";
|
---|
[7019] | 270 | }
|
---|
[3411] | 271 |
|
---|
| 272 | # add metadata per section...
|
---|
| 273 | foreach my $section (@sections) {
|
---|
[8795] | 274 | # section names are not always just digits, may be like "outline"
|
---|
| 275 | $section =~ s@^\"?(\w+)\"?></a>@@; # leftover from split expression...
|
---|
[3614] | 276 |
|
---|
[3411] | 277 | $title = $1; # Greenstone does magic if sections are titled digits
|
---|
| 278 | if (! defined($title) ) {
|
---|
| 279 | print STDERR "no title: $section\n";
|
---|
[8795] | 280 | $title = " "; # get rid of the undefined warning in next line
|
---|
[3411] | 281 | }
|
---|
[15872] | 282 | my $newsection = "<!-- from PDFPlugin -->\n<!-- <Section>\n";
|
---|
[3411] | 283 | $newsection .= "<Metadata name=\"Title\">" . $title
|
---|
| 284 | . "</Metadata>\n--><p>\n";
|
---|
| 285 | $newsection .= $section;
|
---|
| 286 | $newsection .= "<!--</Section>-->\n";
|
---|
| 287 | $section = $newsection;
|
---|
| 288 | }
|
---|
| 289 |
|
---|
[10273] | 290 | $text=join('', ($top_section, @sections));
|
---|
[3411] | 291 | }
|
---|
| 292 |
|
---|
[8218] | 293 | # turn any high bytes that aren't valid utf-8 into utf-8.
|
---|
[10273] | 294 | unicode::ensure_utf8(\$text);
|
---|
[8218] | 295 |
|
---|
[10273] | 296 | # Write it out again!
|
---|
| 297 | $self->utf8_write_file (\$text, $conv_filename);
|
---|
| 298 | }
|
---|
[7287] | 299 |
|
---|
| 300 |
|
---|
[10273] | 301 | # do plugin specific processing of doc_obj for HTML type
|
---|
| 302 | sub process {
|
---|
| 303 | my $self = shift (@_);
|
---|
[15872] | 304 | my ($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
[7287] | 305 |
|
---|
[15963] | 306 | my $result = $self->process_type($base_dir,$file,$doc_obj);
|
---|
[10273] | 307 |
|
---|
[8226] | 308 | # fix up the extracted date metadata to be in Greenstone date format,
|
---|
| 309 | # and fix the capitalisation of 'date'
|
---|
[8227] | 310 | my $cursection = $doc_obj->get_top_section();
|
---|
| 311 | foreach my $datemeta (@{$doc_obj->get_metadata($cursection, "date")}) {
|
---|
[7287] | 312 | $doc_obj->delete_metadata($cursection, "date", $datemeta);
|
---|
| 313 |
|
---|
| 314 | # We're just interested in the date bit, not the time
|
---|
[8278] | 315 | # some pdf creators (eg "Acrobat 5.0 Scan Plug-in for Windows")
|
---|
| 316 | # set a /CreationDate, and set /ModDate to 000000000. pdftohtml
|
---|
| 317 | # extracts the ModDate, so it is 0...
|
---|
| 318 | $datemeta =~ /(\d+)-(\d+)-(\d+)/;
|
---|
| 319 | my ($year, $month, $day) = ($1,$2,$3);
|
---|
| 320 | if (defined($year) && defined($month) && defined($day)) {
|
---|
| 321 | if ($year == 0) {next}
|
---|
| 322 | if ($year < 100) {$year += 1900} # just to be safe
|
---|
| 323 | if ($month =~ /^\d$/) {$month="0$month"} # single digit
|
---|
| 324 | if ($day =~ /^\d$/) {$day="0$day"} # single digit
|
---|
| 325 | my $date="$year$month$day";
|
---|
| 326 | $doc_obj->add_utf8_metadata($cursection, "Date", $date);
|
---|
| 327 | }
|
---|
[7287] | 328 | }
|
---|
| 329 |
|
---|
[10273] | 330 | $doc_obj->add_utf8_metadata($cursection, "NumPages", $self->{'num_pages'});
|
---|
[8795] | 331 |
|
---|
| 332 | if ($self->{'use_sections'} && $self->{'converted_to'} eq "HTML") {
|
---|
| 333 | # we explicitly make it a paged document, cos greenstone won't get it
|
---|
| 334 | # right if any section has an empty title, or one with letters in it
|
---|
| 335 | $doc_obj->set_utf8_metadata_element ($cursection, "gsdlthistype", "Paged");
|
---|
| 336 | }
|
---|
[10273] | 337 |
|
---|
[7287] | 338 | return $result;
|
---|
[1410] | 339 | }
|
---|
| 340 |
|
---|
| 341 | 1;
|
---|