[14661] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
[15872] | 3 | # PagedImagePlugin.pm -- plugin for sets of images and OCR text that
|
---|
[14661] | 4 | # make up a document
|
---|
| 5 | # A component of the Greenstone digital library software
|
---|
| 6 | # from the New Zealand Digital Library Project at the
|
---|
| 7 | # University of Waikato, New Zealand.
|
---|
| 8 | #
|
---|
| 9 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
| 10 | #
|
---|
| 11 | # This program is free software; you can redistribute it and/or modify
|
---|
| 12 | # it under the terms of the GNU General Public License as published by
|
---|
| 13 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 14 | # (at your option) any later version.
|
---|
| 15 | #
|
---|
| 16 | # This program is distributed in the hope that it will be useful,
|
---|
| 17 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 18 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 19 | # GNU General Public License for more details.
|
---|
| 20 | #
|
---|
| 21 | # You should have received a copy of the GNU General Public License
|
---|
| 22 | # along with this program; if not, write to the Free Software
|
---|
| 23 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 24 | #
|
---|
| 25 | ###########################################################################
|
---|
| 26 |
|
---|
[15872] | 27 | # PagedImagePlugin
|
---|
[14661] | 28 | # processes sequences of images, with optional OCR text
|
---|
| 29 | #
|
---|
| 30 | # This plugin takes *.item files, which contain metadata and lists of image
|
---|
| 31 | # files, and produces a document containing sections, one for each page.
|
---|
| 32 | # The files should be named something.item, then you can have more than one
|
---|
| 33 | # book in a directory. You will need to create these files, one for each
|
---|
| 34 | # document/book.
|
---|
| 35 | #
|
---|
| 36 | #There are two formats for the item files: a plain text format, and an xml
|
---|
| 37 | #format. You can use either format, and can have both formats in the same
|
---|
| 38 | #collection if you like. If you use the plain format, you must not start the
|
---|
| 39 | #file off with <PagedDocument>
|
---|
| 40 |
|
---|
| 41 | #### PLAIN FORMAT
|
---|
| 42 | # The format of the xxx.item file is as follows:
|
---|
| 43 | # The first lines contain any metadata for the whole document
|
---|
| 44 | # <metadata-name>metadata-value
|
---|
| 45 | # eg.
|
---|
| 46 | # <Title>Snail farming
|
---|
| 47 | # <Date>19230102
|
---|
| 48 | # Then comes a list of pages, one page per line, each line has the format
|
---|
| 49 | #
|
---|
| 50 | # pagenum:imagefile:textfile:r
|
---|
| 51 | #
|
---|
| 52 | # page num and imagefile are required. pagenum is used for the Title
|
---|
| 53 | # of the section, and in the display is shown as page <pagenum>.
|
---|
| 54 | # imagefile is the image for the page. textfile is an optional text
|
---|
| 55 | # file containing the OCR (or any) text for the page - this gets added
|
---|
| 56 | # as the text for the section. r is optional, and signals that the image
|
---|
| 57 | # should be rotated 180deg. Eg use this if the image has been made upside down.
|
---|
| 58 | # So an example item file looks like:
|
---|
| 59 | # <Title>Snail farming
|
---|
| 60 | # <Date>19960403
|
---|
| 61 | # 1:p1.gif:p1.txt:
|
---|
| 62 | # 2:p2.gif::
|
---|
| 63 | # 3:p3.gif:p3.txt:
|
---|
| 64 | # 3b:p3b.gif:p3b.txt:r
|
---|
| 65 | # The second page has no text, the fourth page is a back page, and
|
---|
| 66 | # should be rotated.
|
---|
| 67 | #
|
---|
| 68 |
|
---|
| 69 | #### XML FORMAT
|
---|
| 70 | # The xml format looks like the following
|
---|
| 71 | #<PagedDocument>
|
---|
| 72 | #<Metadata name="Title">The Title of the entire document</Metadata>
|
---|
[15018] | 73 | #<Page pagenum="1" imgfile="xxx.jpg" txtfile="yyy.txt">
|
---|
[14661] | 74 | #<Metadata name="Title">The Title of this page</Metadata>
|
---|
| 75 | #</Page>
|
---|
| 76 | #... more pages
|
---|
| 77 | #</PagedDocument>
|
---|
| 78 | #PagedDocument contains a list of Pages, Metadata and PageGroups. Any metadata
|
---|
| 79 | #that is not inside another tag will belong to the document.
|
---|
| 80 | #Each Page has a pagenum (not used at the moment), an imgfile and/or a txtfile.
|
---|
| 81 | #These are both optional - if neither is used, the section will have no content.
|
---|
| 82 | #Pages can also have metadata associated with them.
|
---|
| 83 | #PageGroups can be introduced at any point - they can contain Metadata and Pages and other PageGroups. They are used to introduce hierarchical structure into the document.
|
---|
| 84 | #For example
|
---|
| 85 | #<PagedDocument>
|
---|
| 86 | #<PageGroup>
|
---|
| 87 | #<Page>
|
---|
| 88 | #<Page>
|
---|
| 89 | #</PageGroup>
|
---|
| 90 | #<Page>
|
---|
| 91 | #</PagedDocument>
|
---|
| 92 | #would generate a structure like
|
---|
| 93 | #X
|
---|
| 94 | #--X
|
---|
| 95 | # --X
|
---|
| 96 | # --X
|
---|
| 97 | #--X
|
---|
| 98 | #PageGroup tags can also have imgfile/textfile metadata if you like - this way they get some content themselves.
|
---|
| 99 |
|
---|
| 100 | #Currently the XML structure doesn't work very well with the paged document type, unless you use numerical Titles for each section.
|
---|
| 101 | #There is still a bit of work to do on this format:
|
---|
| 102 | #* enable other text file types, eg html, pdf etc
|
---|
| 103 | #* make the document paging work properly
|
---|
| 104 | #* add pagenum as Title unless a Title is present?
|
---|
| 105 |
|
---|
| 106 | # All the supplemetary image amd text files should be in the same folder as
|
---|
| 107 | # the .item file.
|
---|
| 108 | #
|
---|
| 109 | # To display the images instead of the document text, you can use [srcicon]
|
---|
| 110 | # in the DocumentText format statement.
|
---|
| 111 | # For example,
|
---|
| 112 | #
|
---|
| 113 | # format DocumentText "<center><table width=_pagewidth_><tr><td>[srcicon]</td></tr></table></center>"
|
---|
| 114 | #
|
---|
[15905] | 115 | # To have it create thumbnail size images, use the '-create_thumbnail' option.
|
---|
| 116 | # To have it create medium size images for display, use the '-create_screenview'
|
---|
[14661] | 117 | # option. As usual, running
|
---|
[15872] | 118 | # 'perl -S pluginfo.pl PagedImagePlugin' will list all the options.
|
---|
[14661] | 119 |
|
---|
| 120 | # If you want the resulting documents to be presented with a table of
|
---|
| 121 | # contents, use '-documenttype hierarchy', otherwise they will have
|
---|
| 122 | # next and previous arrows, and a goto page X box.
|
---|
| 123 |
|
---|
[15905] | 124 | # If you have used -create_screenview, you can also use [screenicon] in the format
|
---|
[14661] | 125 | # statement to display the smaller image. Here is an example that switches
|
---|
| 126 | # between the two:
|
---|
| 127 | #
|
---|
| 128 | # format DocumentText "<center><table width=_pagewidth_><tr><td>{If}{_cgiargp_ eq full,<a href='_httpdocument_&d=_cgiargd_&p=small'>Switch to small version.</a>,<a href='_httpdocument_&d=_cgiargd_&p=full'>Switch to fullsize version</a>}</td></tr><tr><td>{If}{_cgiargp_ eq full,<a href='_httpdocument_&d=_cgiargd_&p=small' title='Switch to small version'>[srcicon]</a>,<a href='_httpdocument_&d=_cgiargd_&p=full' title='Switch to fullsize version'>[screenicon]</a>}</td></tr></table></center>"
|
---|
| 129 | #
|
---|
| 130 | # Additional metadata can be added into the .item files, alternatively you can
|
---|
| 131 | # use normal metadata.xml files, with the name of the xxx.item file as the
|
---|
| 132 | # FileName (only for document level metadata).
|
---|
| 133 |
|
---|
[15872] | 134 | package PagedImagePlugin;
|
---|
[14661] | 135 |
|
---|
[15872] | 136 | use ReadXMLFile;
|
---|
| 137 | use ReadTextFile;
|
---|
| 138 | use ImageConverter;
|
---|
| 139 |
|
---|
[14661] | 140 | use strict;
|
---|
| 141 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
| 142 |
|
---|
| 143 | sub BEGIN {
|
---|
[15872] | 144 | @PagedImagePlugin::ISA = ('ReadXMLFile', 'ReadTextFile', 'ImageConverter');
|
---|
[14661] | 145 | }
|
---|
| 146 |
|
---|
| 147 | my $type_list =
|
---|
| 148 | [ { 'name' => "paged",
|
---|
[15872] | 149 | 'desc' => "{PagedImagePlugin.documenttype.paged}" },
|
---|
[14661] | 150 | { 'name' => "hierarchy",
|
---|
[15872] | 151 | 'desc' => "{PagedImagePlugin.documenttype.hierarchy}" } ];
|
---|
[14661] | 152 |
|
---|
| 153 | my $arguments =
|
---|
| 154 | [ { 'name' => "process_exp",
|
---|
[16009] | 155 | 'desc' => "{BasePlugin.process_exp}",
|
---|
[14661] | 156 | 'type' => "string",
|
---|
| 157 | 'deft' => &get_default_process_exp(),
|
---|
| 158 | 'reqd' => "no" },
|
---|
| 159 | { 'name' => "title_sub",
|
---|
[16009] | 160 | 'desc' => "{HTMLPlugin.title_sub}",
|
---|
[14661] | 161 | 'type' => "string",
|
---|
| 162 | 'deft' => "" },
|
---|
| 163 | { 'name' => "headerpage",
|
---|
[15872] | 164 | 'desc' => "{PagedImagePlugin.headerpage}",
|
---|
[14661] | 165 | 'type' => "flag",
|
---|
| 166 | 'reqd' => "no" },
|
---|
| 167 | { 'name' => "documenttype",
|
---|
[15872] | 168 | 'desc' => "{PagedImagePlugin.documenttype}",
|
---|
[14661] | 169 | 'type' => "enum",
|
---|
| 170 | 'list' => $type_list,
|
---|
| 171 | 'deft' => "paged",
|
---|
[20791] | 172 | 'reqd' => "no" },
|
---|
| 173 | {'name' => "processing_tmp_files",
|
---|
| 174 | 'desc' => "{BasePlugin.processing_tmp_files}",
|
---|
| 175 | 'type' => "flag",
|
---|
| 176 | 'hiddengli' => "yes"}
|
---|
| 177 | ];
|
---|
[14661] | 178 |
|
---|
| 179 |
|
---|
[15872] | 180 | my $options = { 'name' => "PagedImagePlugin",
|
---|
| 181 | 'desc' => "{PagedImagePlugin.desc}",
|
---|
[14661] | 182 | 'abstract' => "no",
|
---|
| 183 | 'inherits' => "yes",
|
---|
| 184 | 'args' => $arguments };
|
---|
| 185 |
|
---|
| 186 | sub new {
|
---|
| 187 | my ($class) = shift (@_);
|
---|
| 188 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
| 189 | push(@$pluginlist, $class);
|
---|
| 190 |
|
---|
[15872] | 191 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
| 192 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
[14661] | 193 |
|
---|
[16827] | 194 | my $imc_self = new ImageConverter($pluginlist, $inputargs, $hashArgOptLists);
|
---|
| 195 | my $rtf_self = new ReadTextFile($pluginlist, $inputargs, $hashArgOptLists, 1);
|
---|
| 196 | my $rxf_self = new ReadXMLFile($pluginlist, $inputargs, $hashArgOptLists);
|
---|
| 197 |
|
---|
| 198 | my $self = BasePlugin::merge_inheritance($imc_self,$rtf_self,$rxf_self);
|
---|
| 199 |
|
---|
| 200 | # Update $self used by XML::Parser so it finds callback functions
|
---|
| 201 | # such as start_document here and not in ReadXMLFile (which is what
|
---|
| 202 | # $self was when new XML::Parser was done)
|
---|
| 203 | #
|
---|
| 204 | # If the $self returned by this constructor is the same as the one
|
---|
[17741] | 205 | # used in ReadXMLFile (e.g. in the GreenstoneXMLPlugin) then this step isn't necessary
|
---|
[16827] | 206 | #
|
---|
| 207 | # Consider embedding this type of assignment into merge_inheritance
|
---|
| 208 | # to help catch all cases?
|
---|
| 209 |
|
---|
| 210 | $rxf_self->{'parser'}->{'PluginObj'} = $self;
|
---|
| 211 |
|
---|
[14661] | 212 | return bless $self, $class;
|
---|
| 213 | }
|
---|
| 214 |
|
---|
[15872] | 215 |
|
---|
| 216 | sub init {
|
---|
| 217 | my $self = shift (@_);
|
---|
| 218 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
| 219 |
|
---|
| 220 | $self->SUPER::init(@_);
|
---|
| 221 | $self->ImageConverter::init();
|
---|
| 222 | }
|
---|
| 223 |
|
---|
[16847] | 224 | sub begin {
|
---|
| 225 | my $self = shift (@_);
|
---|
| 226 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
| 227 |
|
---|
[16849] | 228 | $self->SUPER::begin(@_);
|
---|
[18468] | 229 | $self->ImageConverter::begin(@_);
|
---|
[16847] | 230 | }
|
---|
| 231 |
|
---|
[14661] | 232 | sub get_default_process_exp {
|
---|
| 233 | my $self = shift (@_);
|
---|
| 234 |
|
---|
| 235 | return q^\.item$^;
|
---|
| 236 | }
|
---|
| 237 |
|
---|
| 238 | sub get_doctype {
|
---|
| 239 | my $self = shift(@_);
|
---|
| 240 |
|
---|
| 241 | return "PagedDocument";
|
---|
| 242 | }
|
---|
| 243 |
|
---|
| 244 |
|
---|
[17822] | 245 | # want to use BasePlugin's version of this, not ReadXMLFile's
|
---|
| 246 | sub can_process_this_file {
|
---|
| 247 | my $self = shift(@_);
|
---|
| 248 | return $self->BasePlugin::can_process_this_file(@_);
|
---|
| 249 | }
|
---|
| 250 |
|
---|
[22565] | 251 | # instead of a block exp, now we scan the file and record all text and img files mentioned there for blocking.
|
---|
| 252 | sub store_block_files
|
---|
| 253 | {
|
---|
| 254 | my $self = shift (@_);
|
---|
| 255 | my ($filename_full_path, $block_hash) = @_;
|
---|
| 256 |
|
---|
| 257 | my $xml_version = $self->is_xml_item_file($filename_full_path);
|
---|
| 258 |
|
---|
[22814] | 259 | # do we need to do this?
|
---|
| 260 | # does BOM interfere just with XML parsing? In that case don't need it here
|
---|
| 261 | # if we do it here, we are modifying the file before we have worked out if
|
---|
| 262 | # its new or not, so it will always be reimported.
|
---|
| 263 | #$self->tidy_item_file($filename_full_path);
|
---|
[22565] | 264 |
|
---|
| 265 | my ($dir, $file) = $filename_full_path =~ /^(.*?)([^\/\\]*)$/;
|
---|
| 266 | if ($xml_version) {
|
---|
| 267 |
|
---|
| 268 | # do something
|
---|
| 269 | $self->scan_xml_for_files_to_block($filename_full_path, $dir, $block_hash);
|
---|
| 270 | } else {
|
---|
| 271 |
|
---|
| 272 | $self->scan_item_for_files_to_block($filename_full_path, $dir, $block_hash);
|
---|
| 273 | }
|
---|
| 274 |
|
---|
| 275 | }
|
---|
| 276 |
|
---|
[15914] | 277 | # we want to use BasePlugin's read, not ReadXMLFile's
|
---|
| 278 | sub read
|
---|
| 279 | {
|
---|
[14661] | 280 | my $self = shift (@_);
|
---|
[16827] | 281 |
|
---|
[15914] | 282 | $self->BasePlugin::read(@_);
|
---|
[15872] | 283 | }
|
---|
[14661] | 284 |
|
---|
| 285 |
|
---|
[18468] | 286 |
|
---|
[14661] | 287 | sub read_into_doc_obj {
|
---|
| 288 | my $self = shift (@_);
|
---|
[16384] | 289 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
[15872] | 290 | my $outhandle = $self->{'outhandle'};
|
---|
[16827] | 291 | my $verbosity = $self->{'verbosity'};
|
---|
[14661] | 292 |
|
---|
[16384] | 293 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[15872] | 294 |
|
---|
| 295 | print $outhandle "PagedImagePlugin processing \"$filename_full_path\"\n"
|
---|
[16827] | 296 | if $verbosity > 1;
|
---|
[15872] | 297 | print STDERR "<Processing n='$file' p='PagedImagePlugin'>\n" if ($gli);
|
---|
[14661] | 298 |
|
---|
[19054] | 299 | $self->{'MaxImageWidth'} = 0;
|
---|
| 300 | $self->{'MaxImageHeight'} = 0;
|
---|
[15872] | 301 |
|
---|
[14661] | 302 | # here we need to decide if we have an old text .item file, or a new xml
|
---|
[15872] | 303 | # .item file
|
---|
| 304 | my $xml_version = $self->is_xml_item_file($filename_full_path);
|
---|
| 305 |
|
---|
[22814] | 306 | $self->tidy_item_file($filename_full_path);
|
---|
[15872] | 307 |
|
---|
| 308 | my $doc_obj;
|
---|
| 309 | if ($xml_version) {
|
---|
| 310 | # careful checking needed here!! are we using local xml handlers or super ones
|
---|
[16384] | 311 | $self->ReadXMLFile::read($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli);
|
---|
[15872] | 312 | $doc_obj = $self->{'doc_obj'};
|
---|
| 313 | } else {
|
---|
| 314 | my ($dir);
|
---|
| 315 | ($dir, $file) = $filename_full_path =~ /^(.*?)([^\/\\]*)$/;
|
---|
| 316 |
|
---|
| 317 | #process the .item file
|
---|
| 318 | $doc_obj = $self->process_item($filename_full_path, $dir, $file, $processor);
|
---|
| 319 |
|
---|
| 320 | }
|
---|
| 321 |
|
---|
| 322 | my $section = $doc_obj->get_top_section();
|
---|
| 323 |
|
---|
| 324 | $doc_obj->add_utf8_metadata($section, "Plugin", "$self->{'plugin_type'}");
|
---|
| 325 | $doc_obj->add_metadata($section, "FileFormat", "PagedImage");
|
---|
| 326 |
|
---|
| 327 | # include any metadata passed in from previous plugins
|
---|
| 328 | # note that this metadata is associated with the top level section
|
---|
| 329 | $self->add_associated_files($doc_obj, $filename_full_path);
|
---|
| 330 | $self->extra_metadata ($doc_obj, $section, $metadata);
|
---|
| 331 | $self->auto_extract_metadata ($doc_obj);
|
---|
| 332 |
|
---|
| 333 | # if we haven't found any Title so far, assign one
|
---|
| 334 | $self->title_fallback($doc_obj,$section,$filename_no_path);
|
---|
| 335 |
|
---|
| 336 | $self->add_OID($doc_obj);
|
---|
| 337 | return (1,$doc_obj);
|
---|
| 338 | }
|
---|
| 339 |
|
---|
| 340 | # for now, the test is if the first non-empty line is <PagedDocument>, then its xml
|
---|
| 341 | sub is_xml_item_file {
|
---|
| 342 | my $self = shift(@_);
|
---|
| 343 | my ($filename) = @_;
|
---|
| 344 |
|
---|
[14661] | 345 | my $xml_version = 0;
|
---|
| 346 | open (ITEMFILE, $filename) || die "couldn't open $filename\n";
|
---|
[19999] | 347 |
|
---|
[14661] | 348 | my $line = "";
|
---|
| 349 | my $num = 0;
|
---|
[19999] | 350 |
|
---|
[14661] | 351 | $line = <ITEMFILE>;
|
---|
[19999] | 352 | while (defined ($line) && ($line !~ /\w/)) {
|
---|
[14661] | 353 | $line = <ITEMFILE>;
|
---|
| 354 | }
|
---|
[19999] | 355 |
|
---|
| 356 | if (defined $line) {
|
---|
| 357 | chomp $line;
|
---|
| 358 | if ($line =~ /<PagedDocument/) {
|
---|
| 359 | $xml_version = 1;
|
---|
| 360 | }
|
---|
| 361 | }
|
---|
| 362 |
|
---|
[15872] | 363 | close ITEMFILE;
|
---|
| 364 | return $xml_version;
|
---|
| 365 | }
|
---|
| 366 |
|
---|
| 367 | sub tidy_item_file {
|
---|
| 368 | my $self = shift(@_);
|
---|
| 369 | my ($filename) = @_;
|
---|
| 370 |
|
---|
[14661] | 371 | open (ITEMFILE, $filename) || die "couldn't open $filename\n";
|
---|
[15872] | 372 | my $backup_filename = "backup.item";
|
---|
| 373 | open (BACKUP,">$backup_filename")|| die "couldn't write to $backup_filename\n";
|
---|
| 374 | my $line = "";
|
---|
[14661] | 375 | $line = <ITEMFILE>;
|
---|
| 376 | $line =~ s/^\xEF\xBB\xBF//; # strip BOM
|
---|
| 377 | $line =~ s/\x0B+//ig;
|
---|
| 378 | $line =~ s/&/&/g;
|
---|
| 379 | print BACKUP ($line);
|
---|
| 380 | #Tidy up the item file some metadata title contains \vt-vertical tab
|
---|
| 381 | while ($line = <ITEMFILE>) {
|
---|
| 382 | $line =~ s/\x0B+//ig;
|
---|
| 383 | $line =~ s/&/&/g;
|
---|
| 384 | print BACKUP ($line);
|
---|
| 385 | }
|
---|
| 386 | close ITEMFILE;
|
---|
| 387 | close BACKUP;
|
---|
| 388 | &File::Copy::copy ($backup_filename, $filename);
|
---|
| 389 | &util::rm($backup_filename);
|
---|
| 390 |
|
---|
| 391 | }
|
---|
[15914] | 392 |
|
---|
| 393 | sub rotate_image {
|
---|
[14661] | 394 | my $self = shift (@_);
|
---|
[15914] | 395 | my ($filename_full_path) = @_;
|
---|
| 396 |
|
---|
| 397 | my ($this_filetype) = $filename_full_path =~ /\.([^\.]*)$/;
|
---|
| 398 | my $result = $self->convert($filename_full_path, $this_filetype, "-rotate 180", "ROTATE");
|
---|
| 399 | my ($new_filename) = ($result =~ /=>(.*\.$this_filetype)/);
|
---|
| 400 | if (-e "$new_filename") {
|
---|
| 401 | return $new_filename;
|
---|
| 402 | }
|
---|
| 403 | # somethings gone wrong
|
---|
| 404 | return $filename_full_path;
|
---|
| 405 |
|
---|
[14661] | 406 | }
|
---|
| 407 |
|
---|
[15914] | 408 | sub process_image {
|
---|
| 409 | my $self = shift(@_);
|
---|
| 410 | my ($filename_full_path, $filename_no_path, $doc_obj, $section, $rotate) = @_;
|
---|
[20778] | 411 | # check the filenames
|
---|
| 412 | return 0 if ($filename_no_path eq "" || !-f $filename_full_path);
|
---|
| 413 |
|
---|
[20791] | 414 | # remember that this image file was one of our source files, but only
|
---|
| 415 | # if we are not processing a tmp file
|
---|
| 416 | if (!$self->{'processing_tmp_files'} ) {
|
---|
| 417 | $doc_obj->associate_source_file($filename_full_path);
|
---|
| 418 | }
|
---|
[15914] | 419 | # do rotation
|
---|
[16827] | 420 | if ((defined $rotate) && ($rotate eq "r")) {
|
---|
[15914] | 421 | # we get a new temporary file which is rotated
|
---|
| 422 | $filename_full_path = $self->rotate_image($filename_full_path);
|
---|
| 423 | }
|
---|
| 424 |
|
---|
| 425 | # do generate images
|
---|
[16009] | 426 | my $result = 0;
|
---|
| 427 | if ($self->{'image_conversion_available'} == 1) {
|
---|
[16384] | 428 | # do we need to convert $filename_no_path to utf8? We are already reading in from a file, what encoding is it in???
|
---|
[16009] | 429 | $result = $self->generate_images($filename_full_path, $filename_no_path, $doc_obj, $section);
|
---|
| 430 | }
|
---|
[15914] | 431 | #overwrite one set in ImageConverter
|
---|
| 432 | $doc_obj->set_metadata_element ($section, "FileFormat", "PagedImage");
|
---|
| 433 | return $result;
|
---|
| 434 | }
|
---|
| 435 |
|
---|
| 436 |
|
---|
[14661] | 437 | sub xml_start_tag {
|
---|
| 438 | my $self = shift(@_);
|
---|
| 439 | my ($expat, $element) = @_;
|
---|
| 440 | $self->{'element'} = $element;
|
---|
| 441 |
|
---|
| 442 | my $doc_obj = $self->{'doc_obj'};
|
---|
| 443 | if ($element eq "PagedDocument") {
|
---|
| 444 | $self->{'current_section'} = $doc_obj->get_top_section();
|
---|
| 445 | } elsif ($element eq "PageGroup" || $element eq "Page") {
|
---|
| 446 | # create a new section as a child
|
---|
| 447 | $self->{'current_section'} = $doc_obj->insert_section($doc_obj->get_end_child($self->{'current_section'}));
|
---|
| 448 | $self->{'num_pages'}++;
|
---|
| 449 | # assign pagenum as what??
|
---|
| 450 | my $pagenum = $_{'pagenum'}; #TODO!!
|
---|
| 451 | if (defined $pagenum) {
|
---|
| 452 | $doc_obj->set_utf8_metadata_element($self->{'current_section'}, 'PageNum', $pagenum);
|
---|
| 453 | }
|
---|
| 454 | my ($imgfile) = $_{'imgfile'};
|
---|
| 455 | if (defined $imgfile) {
|
---|
[16827] | 456 | # *****
|
---|
| 457 | # What about support for rotate image (e.g. old ':r' notation)?
|
---|
| 458 | $self->process_image($self->{'xml_file_dir'}.$imgfile, $imgfile, $doc_obj, $self->{'current_section'});
|
---|
[14661] | 459 | }
|
---|
| 460 | my ($txtfile) = $_{'txtfile'};
|
---|
| 461 | if (defined($txtfile)&& $txtfile ne "") {
|
---|
[16827] | 462 | $self->process_text ($self->{'xml_file_dir'}.$txtfile, $txtfile, $doc_obj, $self->{'current_section'});
|
---|
[15872] | 463 | } else {
|
---|
| 464 | $self->add_dummy_text($doc_obj, $self->{'current_section'});
|
---|
[14661] | 465 | }
|
---|
| 466 | } elsif ($element eq "Metadata") {
|
---|
| 467 | $self->{'metadata_name'} = $_{'name'};
|
---|
| 468 | }
|
---|
| 469 | }
|
---|
| 470 |
|
---|
| 471 | sub xml_end_tag {
|
---|
| 472 | my $self = shift(@_);
|
---|
| 473 | my ($expat, $element) = @_;
|
---|
| 474 |
|
---|
| 475 | my $doc_obj = $self->{'doc_obj'};
|
---|
| 476 | if ($element eq "Page" || $element eq "PageGroup") {
|
---|
| 477 | # if Title hasn't been assigned, set PageNum as Title
|
---|
| 478 | if (!defined $doc_obj->get_metadata_element ($self->{'current_section'}, "Title") && defined $doc_obj->get_metadata_element ($self->{'current_section'}, "PageNum" )) {
|
---|
| 479 | $doc_obj->add_utf8_metadata ($self->{'current_section'}, "Title", $doc_obj->get_metadata_element ($self->{'current_section'}, "PageNum" ));
|
---|
| 480 | }
|
---|
| 481 | # move the current section back to the parent
|
---|
| 482 | $self->{'current_section'} = $doc_obj->get_parent_section($self->{'current_section'});
|
---|
| 483 | } elsif ($element eq "Metadata") {
|
---|
[22349] | 484 | my $meta_name = $self->{'metadata_name'};
|
---|
| 485 | if ($meta_name =~ /\./) {
|
---|
| 486 | $meta_name = "ex.$meta_name";
|
---|
| 487 | }
|
---|
| 488 | $doc_obj->add_utf8_metadata ($self->{'current_section'}, $meta_name, $self->{'metadata_value'});
|
---|
[14661] | 489 | $self->{'metadata_name'} = "";
|
---|
| 490 | $self->{'metadata_value'} = "";
|
---|
| 491 |
|
---|
| 492 | }
|
---|
| 493 | # otherwise we ignore the end tag
|
---|
| 494 | }
|
---|
| 495 |
|
---|
| 496 |
|
---|
| 497 | sub xml_text {
|
---|
| 498 | my $self = shift(@_);
|
---|
| 499 | my ($expat) = @_;
|
---|
| 500 |
|
---|
| 501 | if ($self->{'element'} eq "Metadata" && $self->{'metadata_name'}) {
|
---|
| 502 | $self->{'metadata_value'} .= $_;
|
---|
| 503 | }
|
---|
| 504 | }
|
---|
| 505 |
|
---|
| 506 | sub xml_doctype {
|
---|
| 507 | }
|
---|
| 508 |
|
---|
| 509 | sub open_document {
|
---|
| 510 | my $self = shift(@_);
|
---|
| 511 |
|
---|
| 512 | # create a new document
|
---|
[18327] | 513 | $self->{'doc_obj'} = new doc ($self->{'filename'}, "indexed_doc", $self->{'file_rename_method'});
|
---|
[15872] | 514 | # TODO is file filenmae_no_path??
|
---|
| 515 | $self->set_initial_doc_fields($self->{'doc_obj'}, $self->{'file'}, $self->{'processor'});
|
---|
| 516 |
|
---|
[14661] | 517 | my ($dir, $file) = $self->{'filename'} =~ /^(.*?)([^\/\\]*)$/;
|
---|
[16827] | 518 | $self->{'xml_file_dir'} = $dir;
|
---|
[14661] | 519 | $self->{'num_pages'} = 0;
|
---|
| 520 |
|
---|
| 521 | }
|
---|
| 522 |
|
---|
| 523 | sub close_document {
|
---|
| 524 | my $self = shift(@_);
|
---|
| 525 | my $doc_obj = $self->{'doc_obj'};
|
---|
| 526 |
|
---|
| 527 | # add numpages metadata
|
---|
[19054] | 528 | my $topsection = $doc_obj->get_top_section();
|
---|
[14661] | 529 |
|
---|
[19054] | 530 | $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', $self->{'num_pages'});
|
---|
| 531 |
|
---|
| 532 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
|
---|
| 533 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
|
---|
| 534 | $self->{'MaxImageWidth'} = undef;
|
---|
| 535 | $self->{'MaxImageHeight'} = undef;
|
---|
[14661] | 536 |
|
---|
| 537 | }
|
---|
| 538 |
|
---|
| 539 |
|
---|
[15872] | 540 | sub set_initial_doc_fields {
|
---|
| 541 | my $self = shift(@_);
|
---|
| 542 | my ($doc_obj, $filename_no_path, $processor) = @_;
|
---|
| 543 |
|
---|
[14661] | 544 | my $topsection = $doc_obj->get_top_section();
|
---|
| 545 |
|
---|
| 546 | if ($self->{'documenttype'} eq 'paged') {
|
---|
| 547 | # set the gsdlthistype metadata to Paged - this ensures this document will
|
---|
| 548 | # be treated as a Paged doc, even if Titles are not numeric
|
---|
| 549 | $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Paged");
|
---|
| 550 | } else {
|
---|
| 551 | $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Hierarchy");
|
---|
| 552 | }
|
---|
| 553 |
|
---|
[15872] | 554 | $self->set_Source_metadata($doc_obj, $filename_no_path);
|
---|
| 555 |
|
---|
| 556 | # if we want a header page, we need to add some text into the top section, otherwise this section will become invisible
|
---|
| 557 | if ($self->{'headerpage'}) {
|
---|
[15911] | 558 | $self->add_dummy_text($doc_obj, $topsection);
|
---|
[15872] | 559 | }
|
---|
[14661] | 560 |
|
---|
[15872] | 561 |
|
---|
| 562 | }
|
---|
| 563 |
|
---|
[22565] | 564 | sub scan_xml_for_files_to_block
|
---|
| 565 | {
|
---|
| 566 | my $self = shift (@_);
|
---|
| 567 | my ($filename_full_path, $dir, $block_hash) = @_;
|
---|
[15872] | 568 |
|
---|
[22565] | 569 | open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path to work out which files to block\n";
|
---|
| 570 | my $line = "";
|
---|
| 571 | while (defined ($line = <ITEMFILE>)) {
|
---|
| 572 | next unless $line =~ /\w/;
|
---|
| 573 |
|
---|
| 574 | if ($line =~ /imgfile=\"([^\"]+)\"/) {
|
---|
| 575 | $block_hash->{'file_blocks'}->{$dir.$1} = 1;
|
---|
| 576 | }
|
---|
| 577 | if ($line =~ /txtfile=\"([^\"]+)\"/) {
|
---|
| 578 | $block_hash->{'file_blocks'}->{$dir.$1} = 1;
|
---|
| 579 | }
|
---|
| 580 | }
|
---|
| 581 | close ITEMFILE;
|
---|
| 582 |
|
---|
| 583 | }
|
---|
| 584 |
|
---|
| 585 | sub scan_item_for_files_to_block
|
---|
| 586 | {
|
---|
| 587 | my $self = shift (@_);
|
---|
| 588 | my ($filename_full_path, $dir, $block_hash) = @_;
|
---|
| 589 |
|
---|
| 590 | open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path to work out which files to block\n";
|
---|
| 591 | my $line = "";
|
---|
| 592 | while (defined ($line = <ITEMFILE>)) {
|
---|
| 593 | next unless $line =~ /\w/;
|
---|
| 594 | chomp $line;
|
---|
| 595 | next if $line =~ /^#/; # ignore comment lines
|
---|
| 596 | next if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/); # ignore metadata lines
|
---|
| 597 | # line should be like page:imagefilename:textfilename:r
|
---|
| 598 | $line =~ s/^\s+//; #remove space at the front
|
---|
| 599 | $line =~ s/\s+$//; #remove space at the end
|
---|
| 600 | my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;
|
---|
| 601 |
|
---|
| 602 | # find the image file if there is one
|
---|
| 603 | if (defined $imgname && $imgname ne "") {
|
---|
| 604 | $block_hash->{'file_blocks'}->{$dir.$imgname}=1;
|
---|
| 605 | }
|
---|
| 606 | # find the text file if there is one
|
---|
| 607 | if (defined $txtname && $txtname ne "") {
|
---|
| 608 | $block_hash->{'file_blocks'}->{$dir.$txtname} = 1;
|
---|
| 609 | }
|
---|
| 610 | }
|
---|
| 611 | close ITEMFILE;
|
---|
| 612 |
|
---|
| 613 | }
|
---|
| 614 |
|
---|
[15872] | 615 | sub process_item {
|
---|
| 616 | my $self = shift (@_);
|
---|
| 617 | my ($filename_full_path, $dir, $filename_no_path, $processor) = @_;
|
---|
| 618 |
|
---|
[18327] | 619 | my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
|
---|
[15872] | 620 | $self->set_initial_doc_fields($doc_obj, $filename_no_path, $processor);
|
---|
| 621 | my $topsection = $doc_obj->get_top_section();
|
---|
| 622 | open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path\n";
|
---|
[14661] | 623 | my $line = "";
|
---|
| 624 | my $num = 0;
|
---|
| 625 | while (defined ($line = <ITEMFILE>)) {
|
---|
| 626 | next unless $line =~ /\w/;
|
---|
| 627 | chomp $line;
|
---|
| 628 | next if $line =~ /^#/; # ignore comment lines
|
---|
| 629 | if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/) {
|
---|
[22349] | 630 | my $meta_name = $1;
|
---|
| 631 | my $meta_value = $2;
|
---|
| 632 | if ($meta_name =~ /\./) {
|
---|
| 633 | $meta_name = "ex.$meta_name";
|
---|
| 634 | }
|
---|
| 635 | $doc_obj->set_utf8_metadata_element ($topsection, $meta_name, $meta_value);
|
---|
[14661] | 636 | #$meta->{$1} = $2;
|
---|
| 637 | } else {
|
---|
| 638 | $num++;
|
---|
| 639 | # line should be like page:imagefilename:textfilename:r - the r is optional -> means rotate the image 180 deg
|
---|
| 640 | $line =~ s/^\s+//; #remove space at the front
|
---|
| 641 | $line =~ s/\s+$//; #remove space at the end
|
---|
| 642 | my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;
|
---|
| 643 |
|
---|
| 644 | # create a new section for each image file
|
---|
| 645 | my $cursection = $doc_obj->insert_section($doc_obj->get_end_child($topsection));
|
---|
| 646 | # the page number becomes the Title
|
---|
| 647 | $doc_obj->set_utf8_metadata_element($cursection, 'Title', $pagenum);
|
---|
| 648 |
|
---|
| 649 | # process the image for this page if there is one
|
---|
| 650 | if (defined $imgname && $imgname ne "") {
|
---|
| 651 | my $result1 = $self->process_image($dir.$imgname, $imgname, $doc_obj, $cursection, $rotate);
|
---|
| 652 | if (!defined $result1)
|
---|
| 653 | {
|
---|
[21721] | 654 | print "PagedImagePlugin: couldn't process image \"$dir$imgname\" for item \"$filename_full_path\"\n";
|
---|
[14661] | 655 | }
|
---|
| 656 | }
|
---|
| 657 | # process the text file if one is there
|
---|
| 658 | if (defined $txtname && $txtname ne "") {
|
---|
| 659 | my $result2 = $self->process_text ($dir.$txtname, $txtname, $doc_obj, $cursection);
|
---|
| 660 |
|
---|
| 661 | if (!defined $result2) {
|
---|
[15872] | 662 | print "PagedImagePlugin: couldn't process text file \"$dir.$txtname\" for item \"$filename_full_path\"\n";
|
---|
| 663 | $self->add_dummy_text($doc_obj, $cursection);
|
---|
[14661] | 664 | }
|
---|
| 665 | } else {
|
---|
| 666 | # otherwise add in some dummy text
|
---|
[15872] | 667 | $self->add_dummy_text($doc_obj, $cursection);
|
---|
| 668 | }
|
---|
[14661] | 669 | }
|
---|
| 670 | }
|
---|
| 671 |
|
---|
| 672 | close ITEMFILE;
|
---|
| 673 |
|
---|
| 674 | # add numpages metadata
|
---|
| 675 | $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', "$num");
|
---|
[19054] | 676 |
|
---|
| 677 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
|
---|
| 678 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
|
---|
| 679 | $self->{'MaxImageWidth'} = undef;
|
---|
| 680 | $self->{'MaxImageHeight'} = undef;
|
---|
| 681 |
|
---|
| 682 |
|
---|
[14661] | 683 | return $doc_obj;
|
---|
| 684 | }
|
---|
| 685 |
|
---|
| 686 | sub process_text {
|
---|
| 687 | my $self = shift (@_);
|
---|
[15872] | 688 | my ($filename_full_path, $file, $doc_obj, $cursection) = @_;
|
---|
[14661] | 689 |
|
---|
| 690 | # check that the text file exists!!
|
---|
[15872] | 691 | if (!-f $filename_full_path) {
|
---|
| 692 | print "PagedImagePlugin: ERROR: File $filename_full_path does not exist, skipping\n";
|
---|
[14661] | 693 | return 0;
|
---|
| 694 | }
|
---|
| 695 |
|
---|
[20791] | 696 | # remember that this text file was one of our source files, but only
|
---|
| 697 | # if we are not processing a tmp file
|
---|
| 698 | if (!$self->{'processing_tmp_files'} ) {
|
---|
| 699 | $doc_obj->associate_source_file($filename_full_path);
|
---|
| 700 | }
|
---|
[14661] | 701 | # Do encoding stuff
|
---|
[15872] | 702 | my ($language, $encoding) = $self->textcat_get_language_encoding ($filename_full_path);
|
---|
[14661] | 703 |
|
---|
| 704 | my $text="";
|
---|
[15872] | 705 | &ReadTextFile::read_file($self, $filename_full_path, $encoding, $language, \$text);
|
---|
[14661] | 706 | if (!length ($text)) {
|
---|
| 707 | # It's a bit unusual but not out of the question to have no text, so just give a warning
|
---|
[15872] | 708 | print "PagedImagePlugin: WARNING: $filename_full_path contains no text\n";
|
---|
[14661] | 709 | }
|
---|
| 710 |
|
---|
| 711 | # we need to escape the escape character, or else mg will convert into
|
---|
| 712 | # eg literal newlines, instead of leaving the text as '\n'
|
---|
| 713 | $text =~ s/\\/\\\\/g; # macro language
|
---|
| 714 | $text =~ s/_/\\_/g; # macro language
|
---|
[15018] | 715 |
|
---|
| 716 |
|
---|
[16827] | 717 | if ($text =~ m/<html.*?>\s*<head.*?>.*<\/head>\s*<body.*?>(.*)<\/body>\s*<\/html>\s*$/is) {
|
---|
[15018] | 718 | # looks like HTML input
|
---|
| 719 | # no need to escape < and > or put in <pre> tags
|
---|
| 720 |
|
---|
| 721 | $text = $1;
|
---|
| 722 |
|
---|
[22565] | 723 | # add text to document object
|
---|
[15018] | 724 | $doc_obj->add_utf8_text($cursection, "$text");
|
---|
| 725 | }
|
---|
| 726 | else {
|
---|
| 727 | $text =~ s/</</g;
|
---|
| 728 | $text =~ s/>/>/g;
|
---|
| 729 |
|
---|
| 730 | # insert preformat tags and add text to document object
|
---|
| 731 | $doc_obj->add_utf8_text($cursection, "<pre>\n$text\n</pre>");
|
---|
| 732 | }
|
---|
| 733 |
|
---|
[14661] | 734 |
|
---|
| 735 | return 1;
|
---|
| 736 | }
|
---|
| 737 |
|
---|
| 738 |
|
---|
[15872] | 739 | sub clean_up_after_doc_obj_processing {
|
---|
| 740 | my $self = shift(@_);
|
---|
| 741 |
|
---|
| 742 | $self->ImageConverter::clean_up_temporary_files();
|
---|
| 743 | }
|
---|
| 744 |
|
---|
[14661] | 745 | 1;
|
---|