1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # PagedImagePlugin.pm -- plugin for sets of images and OCR text that
|
---|
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 |
|
---|
27 | # PagedImagePlugin
|
---|
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>
|
---|
73 | #<Page pagenum="1" imgfile="xxx.jpg" txtfile="yyy.txt">
|
---|
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 | #
|
---|
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'
|
---|
117 | # option. As usual, running
|
---|
118 | # 'perl -S pluginfo.pl PagedImagePlugin' will list all the options.
|
---|
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 |
|
---|
124 | # If you have used -create_screenview, you can also use [screenicon] in the format
|
---|
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 |
|
---|
134 | package PagedImagePlugin;
|
---|
135 |
|
---|
136 | use Encode;
|
---|
137 | use ReadXMLFile;
|
---|
138 | use ReadTextFile;
|
---|
139 | use ImageConverter;
|
---|
140 | use MetadataRead;
|
---|
141 |
|
---|
142 | use strict;
|
---|
143 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
144 |
|
---|
145 | sub BEGIN {
|
---|
146 | @PagedImagePlugin::ISA = ('MetadataRead', 'ReadXMLFile', 'ReadTextFile', 'ImageConverter');
|
---|
147 | }
|
---|
148 |
|
---|
149 | my $gs2_type_list =
|
---|
150 | [ { 'name' => "auto",
|
---|
151 | 'desc' => "{PagedImagePlugin.documenttype.auto2}" },
|
---|
152 | { 'name' => "paged",
|
---|
153 | 'desc' => "{PagedImagePlugin.documenttype.paged2}" },
|
---|
154 | { 'name' => "hierarchy",
|
---|
155 | 'desc' => "{PagedImagePlugin.documenttype.hierarchy}" }
|
---|
156 | ];
|
---|
157 |
|
---|
158 | my $gs3_type_list =
|
---|
159 | [ { 'name' => "auto",
|
---|
160 | 'desc' => "{PagedImagePlugin.documenttype.auto3}" },
|
---|
161 | { 'name' => "paged",
|
---|
162 | 'desc' => "{PagedImagePlugin.documenttype.paged3}" },
|
---|
163 | { 'name' => "hierarchy",
|
---|
164 | 'desc' => "{PagedImagePlugin.documenttype.hierarchy}" },
|
---|
165 | { 'name' => "pagedhierarchy",
|
---|
166 | 'desc' => "{PagedImagePlugin.documenttype.pagedhierarchy}" }
|
---|
167 | ];
|
---|
168 |
|
---|
169 | my $arguments =
|
---|
170 | [ { 'name' => "process_exp",
|
---|
171 | 'desc' => "{BaseImporter.process_exp}",
|
---|
172 | 'type' => "string",
|
---|
173 | 'deft' => &get_default_process_exp(),
|
---|
174 | 'reqd' => "no" },
|
---|
175 | { 'name' => "title_sub",
|
---|
176 | 'desc' => "{HTMLPlugin.title_sub}",
|
---|
177 | 'type' => "string",
|
---|
178 | 'deft' => "" },
|
---|
179 | { 'name' => "headerpage",
|
---|
180 | 'desc' => "{PagedImagePlugin.headerpage}",
|
---|
181 | 'type' => "flag",
|
---|
182 | 'reqd' => "no" },
|
---|
183 | # { 'name' => "documenttype",
|
---|
184 | # 'desc' => "{PagedImagePlugin.documenttype}",
|
---|
185 | # 'type' => "enum",
|
---|
186 | # 'list' => $type_list,
|
---|
187 | # 'deft' => "auto",
|
---|
188 | # 'reqd' => "no" },
|
---|
189 | {'name' => "processing_tmp_files",
|
---|
190 | 'desc' => "{BaseImporter.processing_tmp_files}",
|
---|
191 | 'type' => "flag",
|
---|
192 | 'hiddengli' => "yes"}
|
---|
193 | ];
|
---|
194 |
|
---|
195 | my $doc_type_opt = { 'name' => "documenttype",
|
---|
196 | 'desc' => "{PagedImagePlugin.documenttype}",
|
---|
197 | 'type' => "enum",
|
---|
198 | 'deft' => "auto",
|
---|
199 | 'reqd' => "no" };
|
---|
200 |
|
---|
201 | my $options = { 'name' => "PagedImagePlugin",
|
---|
202 | 'desc' => "{PagedImagePlugin.desc}",
|
---|
203 | 'abstract' => "no",
|
---|
204 | 'inherits' => "yes",
|
---|
205 | 'args' => $arguments };
|
---|
206 |
|
---|
207 | sub new {
|
---|
208 | my ($class) = shift (@_);
|
---|
209 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
210 | push(@$pluginlist, $class);
|
---|
211 |
|
---|
212 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
213 |
|
---|
214 | my $imc_self = new ImageConverter($pluginlist, $inputargs, $hashArgOptLists);
|
---|
215 | # we can use this plugin to check gs3 version
|
---|
216 | if ($imc_self->{'gs_version'} eq "3") {
|
---|
217 | $doc_type_opt->{'list'} = $gs3_type_list;
|
---|
218 | }
|
---|
219 | else {
|
---|
220 | $doc_type_opt->{'list'} = $gs2_type_list;
|
---|
221 | }
|
---|
222 | push(@$arguments,$doc_type_opt);
|
---|
223 | # now we add the args to the list for parsing
|
---|
224 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
225 |
|
---|
226 | my $rtf_self = new ReadTextFile($pluginlist, $inputargs, $hashArgOptLists, 1);
|
---|
227 | my $rxf_self = new ReadXMLFile($pluginlist, $inputargs, $hashArgOptLists);
|
---|
228 |
|
---|
229 | my $self = BaseImporter::merge_inheritance($rxf_self, $imc_self,$rtf_self);
|
---|
230 |
|
---|
231 | # Update $self used by XML::Parser so it finds callback functions
|
---|
232 | # such as start_document here and not in ReadXMLFile (which is what
|
---|
233 | # $self was when new XML::Parser was done)
|
---|
234 | #
|
---|
235 | # If the $self returned by this constructor is the same as the one
|
---|
236 | # used in ReadXMLFile (e.g. in the GreenstoneXMLPlugin) then this step isn't necessary
|
---|
237 | #
|
---|
238 | # Consider embedding this type of assignment into merge_inheritance
|
---|
239 | # to help catch all cases?
|
---|
240 |
|
---|
241 | $rxf_self->{'parser'}->{'PluginObj'} = $self;
|
---|
242 |
|
---|
243 | return bless $self, $class;
|
---|
244 | }
|
---|
245 |
|
---|
246 |
|
---|
247 | sub init {
|
---|
248 | my $self = shift (@_);
|
---|
249 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
250 |
|
---|
251 | $self->SUPER::init(@_);
|
---|
252 | $self->ImageConverter::init();
|
---|
253 | }
|
---|
254 |
|
---|
255 | sub begin {
|
---|
256 | my $self = shift (@_);
|
---|
257 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
258 |
|
---|
259 | $self->SUPER::begin(@_);
|
---|
260 | $self->ImageConverter::begin(@_);
|
---|
261 | }
|
---|
262 |
|
---|
263 | sub get_default_process_exp {
|
---|
264 | my $self = shift (@_);
|
---|
265 |
|
---|
266 | return q^\.item$^;
|
---|
267 | }
|
---|
268 |
|
---|
269 | sub get_doctype {
|
---|
270 | my $self = shift(@_);
|
---|
271 |
|
---|
272 | return "PagedDocument";
|
---|
273 | }
|
---|
274 |
|
---|
275 |
|
---|
276 | # want to use BaseImporter's version of this, not ReadXMLFile's
|
---|
277 | sub can_process_this_file {
|
---|
278 | my $self = shift(@_);
|
---|
279 | return $self->BaseImporter::can_process_this_file(@_);
|
---|
280 | }
|
---|
281 |
|
---|
282 | # instead of a block exp, now we scan the file and record all text and img files mentioned there for blocking.
|
---|
283 | sub store_block_files
|
---|
284 | {
|
---|
285 | my $self = shift (@_);
|
---|
286 | my ($filename_full_path, $block_hash) = @_;
|
---|
287 |
|
---|
288 | my $xml_version = $self->is_xml_item_file($filename_full_path);
|
---|
289 |
|
---|
290 | # do we need to do this?
|
---|
291 | # does BOM interfere just with XML parsing? In that case don't need it here
|
---|
292 | # if we do it here, we are modifying the file before we have worked out if
|
---|
293 | # its new or not, so it will always be reimported.
|
---|
294 | #$self->tidy_item_file($filename_full_path);
|
---|
295 |
|
---|
296 | my ($dir, $file) = $filename_full_path =~ /^(.*?)([^\/\\]*)$/;
|
---|
297 | if ($xml_version) {
|
---|
298 |
|
---|
299 | # do something
|
---|
300 | $self->scan_xml_for_files_to_block($filename_full_path, $dir, $block_hash);
|
---|
301 | } else {
|
---|
302 |
|
---|
303 | $self->scan_item_for_files_to_block($filename_full_path, $dir, $block_hash);
|
---|
304 | }
|
---|
305 |
|
---|
306 | }
|
---|
307 |
|
---|
308 | # we want to use BaseImporter's read, not ReadXMLFile's
|
---|
309 | sub read
|
---|
310 | {
|
---|
311 | my $self = shift (@_);
|
---|
312 |
|
---|
313 | $self->BaseImporter::read(@_);
|
---|
314 | }
|
---|
315 |
|
---|
316 |
|
---|
317 |
|
---|
318 | sub read_into_doc_obj {
|
---|
319 | my $self = shift (@_);
|
---|
320 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
321 | my $outhandle = $self->{'outhandle'};
|
---|
322 | my $verbosity = $self->{'verbosity'};
|
---|
323 |
|
---|
324 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
325 |
|
---|
326 | print $outhandle "PagedImagePlugin processing \"$filename_full_path\"\n"
|
---|
327 | if $verbosity > 1;
|
---|
328 | print STDERR "<Processing n='$file' p='PagedImagePlugin'>\n" if ($gli);
|
---|
329 |
|
---|
330 | $self->{'MaxImageWidth'} = 0;
|
---|
331 | $self->{'MaxImageHeight'} = 0;
|
---|
332 |
|
---|
333 | # here we need to decide if we have an old text .item file, or a new xml
|
---|
334 | # .item file
|
---|
335 | my $xml_version = $self->is_xml_item_file($filename_full_path);
|
---|
336 |
|
---|
337 | $self->tidy_item_file($filename_full_path);
|
---|
338 |
|
---|
339 | my $doc_obj;
|
---|
340 | if ($xml_version) {
|
---|
341 | # careful checking needed here!! are we using local xml handlers or super ones
|
---|
342 | $self->ReadXMLFile::read($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli);
|
---|
343 | $doc_obj = $self->{'doc_obj'};
|
---|
344 | } else {
|
---|
345 | my ($dir, $item_file);
|
---|
346 | ($dir, $item_file) = $filename_full_path =~ /^(.*?)([^\/\\]*)$/;
|
---|
347 |
|
---|
348 | #process the .item file
|
---|
349 | $doc_obj = $self->process_item($filename_full_path, $dir, $item_file, $processor, $metadata);
|
---|
350 |
|
---|
351 | }
|
---|
352 |
|
---|
353 | my $section = $doc_obj->get_top_section();
|
---|
354 |
|
---|
355 | $doc_obj->add_utf8_metadata($section, "Plugin", "$self->{'plugin_type'}");
|
---|
356 | $doc_obj->add_metadata($section, "FileFormat", "PagedImage");
|
---|
357 |
|
---|
358 | # include any metadata passed in from previous plugins
|
---|
359 | # note that this metadata is associated with the top level section
|
---|
360 | $self->add_associated_files($doc_obj, $filename_full_path);
|
---|
361 | $self->extra_metadata ($doc_obj, $section, $metadata);
|
---|
362 | $self->auto_extract_metadata ($doc_obj);
|
---|
363 | $self->plugin_specific_process($base_dir, $file, $doc_obj, $gli);
|
---|
364 | # if we haven't found any Title so far, assign one
|
---|
365 | $self->title_fallback($doc_obj,$section,$filename_no_path);
|
---|
366 |
|
---|
367 | $self->add_OID($doc_obj);
|
---|
368 | return (1,$doc_obj);
|
---|
369 | }
|
---|
370 | # override this for an inheriting plugin to add extra metadata etc
|
---|
371 | sub plugin_specific_process {
|
---|
372 | my $self = shift(@_);
|
---|
373 | my ($base_dir, $file, $doc_obj, $gli) = @_;
|
---|
374 |
|
---|
375 | }
|
---|
376 |
|
---|
377 | # for now, the test is if the first non-empty line is <PagedDocument>, then its xml
|
---|
378 | sub is_xml_item_file {
|
---|
379 | my $self = shift(@_);
|
---|
380 | my ($filename) = @_;
|
---|
381 |
|
---|
382 | my $xml_version = 0;
|
---|
383 | open (ITEMFILE, $filename) || die "couldn't open $filename\n";
|
---|
384 |
|
---|
385 | my $line = "";
|
---|
386 | my $num = 0;
|
---|
387 |
|
---|
388 | $line = <ITEMFILE>;
|
---|
389 | while (defined ($line) && ($line !~ /\w/)) {
|
---|
390 | $line = <ITEMFILE>;
|
---|
391 | }
|
---|
392 |
|
---|
393 | if (defined $line) {
|
---|
394 | chomp $line;
|
---|
395 | if ($line =~ /<PagedDocument/) {
|
---|
396 | $xml_version = 1;
|
---|
397 | }
|
---|
398 | }
|
---|
399 |
|
---|
400 | close ITEMFILE;
|
---|
401 | return $xml_version;
|
---|
402 | }
|
---|
403 |
|
---|
404 | sub tidy_item_file {
|
---|
405 | my $self = shift(@_);
|
---|
406 | my ($filename) = @_;
|
---|
407 |
|
---|
408 | open (ITEMFILE, "<:encoding(UTF-8)", $filename) || die "couldn't open $filename\n";
|
---|
409 | my $backup_filename = "backup.item";
|
---|
410 | open (BACKUP,">$backup_filename")|| die "couldn't write to $backup_filename\n";
|
---|
411 | binmode(BACKUP, ":utf8");
|
---|
412 | my $line = "";
|
---|
413 | $line = <ITEMFILE>;
|
---|
414 | #$line =~ s/^\xEF\xBB\xBF//; # strip BOM in text file read in as a sequence of bytes (not unicode aware strings)
|
---|
415 | $line =~ s/^\x{FEFF}//; # strip BOM in file opened *as UTF-8*. Strings in the file just read in are now unicode-aware,
|
---|
416 | # this means the BOM is now a unicode codepoint instead of a byte sequence
|
---|
417 | # See http://en.wikipedia.org/wiki/Byte_order_mark and http://perldoc.perl.org/5.14.0/perlunicode.html
|
---|
418 | $line =~ s/\x{0B}+//ig; # removing \vt-vertical tabs using the unicode codepoint for \vt
|
---|
419 | $line =~ s/&/&/g;
|
---|
420 | print BACKUP ($line);
|
---|
421 | #Tidy up the item file some metadata title contains \vt-vertical tab
|
---|
422 | while ($line = <ITEMFILE>) {
|
---|
423 | $line =~ s/\x{0B}+//ig; # removing \vt-vertical tabs using the unicode codepoint for \vt
|
---|
424 | $line =~ s/&/&/g;
|
---|
425 | print BACKUP ($line);
|
---|
426 | }
|
---|
427 | close ITEMFILE;
|
---|
428 | close BACKUP;
|
---|
429 | &File::Copy::copy ($backup_filename, $filename);
|
---|
430 | &FileUtils::removeFiles($backup_filename);
|
---|
431 |
|
---|
432 | }
|
---|
433 |
|
---|
434 | sub rotate_image {
|
---|
435 | my $self = shift (@_);
|
---|
436 | my ($filename_full_path) = @_;
|
---|
437 |
|
---|
438 | my ($this_filetype) = $filename_full_path =~ /\.([^\.]*)$/;
|
---|
439 | my ($result, $result_str, $new_filename) = $self->convert($filename_full_path, $this_filetype, "-rotate 180", "ROTATE");
|
---|
440 |
|
---|
441 | if (-e "$new_filename") {
|
---|
442 | return $new_filename;
|
---|
443 | }
|
---|
444 | # somethings gone wrong
|
---|
445 | return $filename_full_path;
|
---|
446 |
|
---|
447 | }
|
---|
448 |
|
---|
449 | sub process_image {
|
---|
450 | my $self = shift(@_);
|
---|
451 | my ($filename_full_path, $filename_no_path, $doc_obj, $section, $rotate) = @_;
|
---|
452 | # check the filenames
|
---|
453 | return 0 if ($filename_no_path eq "" || !-f $filename_full_path);
|
---|
454 |
|
---|
455 | # remember that this image file was one of our source files, but only
|
---|
456 | # if we are not processing a tmp file
|
---|
457 | if (!$self->{'processing_tmp_files'} ) {
|
---|
458 | $doc_obj->associate_source_file($filename_full_path);
|
---|
459 | }
|
---|
460 | # do rotation
|
---|
461 | if ((defined $rotate) && ($rotate eq "r")) {
|
---|
462 | # we get a new temporary file which is rotated
|
---|
463 | $filename_full_path = $self->rotate_image($filename_full_path);
|
---|
464 | }
|
---|
465 |
|
---|
466 | # do generate images
|
---|
467 | my $result = 0;
|
---|
468 | if ($self->{'image_conversion_available'} == 1) {
|
---|
469 | # do we need to convert $filename_no_path to utf8/url encoded?
|
---|
470 | # We are already reading in from a file, what encoding is it in???
|
---|
471 | my $url_encoded_full_filename
|
---|
472 | = &unicode::raw_filename_to_url_encoded($filename_full_path);
|
---|
473 | $result = $self->generate_images($filename_full_path, $url_encoded_full_filename, $doc_obj, $section);
|
---|
474 | }
|
---|
475 | #overwrite one set in ImageConverter
|
---|
476 | $doc_obj->set_metadata_element ($section, "FileFormat", "PagedImage");
|
---|
477 | return $result;
|
---|
478 | }
|
---|
479 |
|
---|
480 |
|
---|
481 | sub xml_start_tag {
|
---|
482 | my $self = shift(@_);
|
---|
483 | my ($expat, $element) = @_;
|
---|
484 | $self->{'element'} = $element;
|
---|
485 |
|
---|
486 | my $doc_obj = $self->{'doc_obj'};
|
---|
487 | if ($element eq "PagedDocument") {
|
---|
488 | $self->{'current_section'} = $doc_obj->get_top_section();
|
---|
489 | } elsif ($element eq "PageGroup" || $element eq "Page") {
|
---|
490 | if ($element eq "PageGroup") {
|
---|
491 | $self->{'has_internal_structure'} = 1;
|
---|
492 | }
|
---|
493 | # create a new section as a child
|
---|
494 | $self->{'current_section'} = $doc_obj->insert_section($doc_obj->get_end_child($self->{'current_section'}));
|
---|
495 | $self->{'num_pages'}++;
|
---|
496 | # assign pagenum as what??
|
---|
497 | my $pagenum = $_{'pagenum'}; #TODO!!
|
---|
498 | if (defined $pagenum) {
|
---|
499 | $doc_obj->set_utf8_metadata_element($self->{'current_section'}, 'PageNum', $pagenum);
|
---|
500 | }
|
---|
501 | my ($imgfile) = $_{'imgfile'};
|
---|
502 | if (defined $imgfile) {
|
---|
503 | # *****
|
---|
504 | # What about support for rotate image (e.g. old ':r' notation)?
|
---|
505 | $self->process_image($self->{'xml_file_dir'}.$imgfile, $imgfile, $doc_obj, $self->{'current_section'});
|
---|
506 | }
|
---|
507 | my ($txtfile) = $_{'txtfile'};
|
---|
508 | if (defined($txtfile)&& $txtfile ne "") {
|
---|
509 | $self->process_text ($self->{'xml_file_dir'}.$txtfile, $txtfile, $doc_obj, $self->{'current_section'});
|
---|
510 | } else {
|
---|
511 | $self->add_dummy_text($doc_obj, $self->{'current_section'});
|
---|
512 | }
|
---|
513 | } elsif ($element eq "Metadata") {
|
---|
514 | $self->{'metadata_name'} = $_{'name'};
|
---|
515 | }
|
---|
516 | }
|
---|
517 |
|
---|
518 | sub xml_end_tag {
|
---|
519 | my $self = shift(@_);
|
---|
520 | my ($expat, $element) = @_;
|
---|
521 |
|
---|
522 | my $doc_obj = $self->{'doc_obj'};
|
---|
523 | if ($element eq "Page" || $element eq "PageGroup") {
|
---|
524 | # if Title hasn't been assigned, set PageNum as Title
|
---|
525 | if (!defined $doc_obj->get_metadata_element ($self->{'current_section'}, "Title") && defined $doc_obj->get_metadata_element ($self->{'current_section'}, "PageNum" )) {
|
---|
526 | $doc_obj->add_utf8_metadata ($self->{'current_section'}, "Title", $doc_obj->get_metadata_element ($self->{'current_section'}, "PageNum" ));
|
---|
527 | }
|
---|
528 | # move the current section back to the parent
|
---|
529 | $self->{'current_section'} = $doc_obj->get_parent_section($self->{'current_section'});
|
---|
530 | } elsif ($element eq "Metadata") {
|
---|
531 |
|
---|
532 | # text read in by XML::Parser is in Perl's binary byte value
|
---|
533 | # form ... need to explicitly make it UTF-8
|
---|
534 | my $meta_name = decode("utf-8",$self->{'metadata_name'});
|
---|
535 | my $metadata_value = decode("utf-8",$self->{'metadata_value'});
|
---|
536 |
|
---|
537 | if ($meta_name =~ /\./) {
|
---|
538 | $meta_name = "ex.$meta_name";
|
---|
539 | }
|
---|
540 |
|
---|
541 | $doc_obj->add_utf8_metadata ($self->{'current_section'}, $meta_name, $metadata_value);
|
---|
542 | $self->{'metadata_name'} = "";
|
---|
543 | $self->{'metadata_value'} = "";
|
---|
544 |
|
---|
545 | }
|
---|
546 | # otherwise we ignore the end tag
|
---|
547 | }
|
---|
548 |
|
---|
549 |
|
---|
550 | sub xml_text {
|
---|
551 | my $self = shift(@_);
|
---|
552 | my ($expat) = @_;
|
---|
553 |
|
---|
554 | if ($self->{'element'} eq "Metadata" && $self->{'metadata_name'}) {
|
---|
555 | $self->{'metadata_value'} .= $_;
|
---|
556 | }
|
---|
557 | }
|
---|
558 |
|
---|
559 | sub xml_doctype {
|
---|
560 | }
|
---|
561 |
|
---|
562 | sub open_document {
|
---|
563 | my $self = shift(@_);
|
---|
564 |
|
---|
565 | # create a new document
|
---|
566 | $self->{'doc_obj'} = new doc ($self->{'filename'}, "indexed_doc", $self->{'file_rename_method'});
|
---|
567 | # TODO is file filenmae_no_path??
|
---|
568 | $self->set_initial_doc_fields($self->{'doc_obj'}, $self->{'filename'}, $self->{'processor'}, $self->{'metadata'});
|
---|
569 |
|
---|
570 | my ($dir, $file) = $self->{'filename'} =~ /^(.*?)([^\/\\]*)$/;
|
---|
571 | $self->{'xml_file_dir'} = $dir;
|
---|
572 | $self->{'num_pages'} = 0;
|
---|
573 | $self->{'has_internal_structure'} = 0;
|
---|
574 |
|
---|
575 | }
|
---|
576 |
|
---|
577 | sub close_document {
|
---|
578 | my $self = shift(@_);
|
---|
579 | my $doc_obj = $self->{'doc_obj'};
|
---|
580 |
|
---|
581 | my $topsection = $doc_obj->get_top_section();
|
---|
582 |
|
---|
583 | # add numpages metadata
|
---|
584 | $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', $self->{'num_pages'});
|
---|
585 |
|
---|
586 | # set the document type
|
---|
587 | my $final_doc_type = "";
|
---|
588 | if ($self->{'documenttype'} eq "auto") {
|
---|
589 | if ($self->{'has_internal_structure'}) {
|
---|
590 | if ($self->{'gs_version'} eq "3") {
|
---|
591 | $final_doc_type = "pagedhierarchy";
|
---|
592 | }
|
---|
593 | else {
|
---|
594 | $final_doc_type = "hierarchy";
|
---|
595 | }
|
---|
596 | } else {
|
---|
597 | $final_doc_type = "paged";
|
---|
598 | }
|
---|
599 | } else {
|
---|
600 | # set to what doc type option was set to
|
---|
601 | $final_doc_type = $self->{'documenttype'};
|
---|
602 | }
|
---|
603 | $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", $final_doc_type);
|
---|
604 | ### capiatalisation????
|
---|
605 | # if ($self->{'documenttype'} eq 'paged') {
|
---|
606 | # set the gsdlthistype metadata to Paged - this ensures this document will
|
---|
607 | # be treated as a Paged doc, even if Titles are not numeric
|
---|
608 | # $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Paged");
|
---|
609 | # } else {
|
---|
610 | # $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Hierarchy");
|
---|
611 | # }
|
---|
612 |
|
---|
613 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
|
---|
614 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
|
---|
615 | $self->{'MaxImageWidth'} = undef;
|
---|
616 | $self->{'MaxImageHeight'} = undef;
|
---|
617 |
|
---|
618 | }
|
---|
619 |
|
---|
620 |
|
---|
621 | sub set_initial_doc_fields {
|
---|
622 | my $self = shift(@_);
|
---|
623 | my ($doc_obj, $filename_full_path, $processor, $metadata) = @_;
|
---|
624 |
|
---|
625 | my $topsection = $doc_obj->get_top_section();
|
---|
626 |
|
---|
627 | my $plugin_filename_encoding = $self->{'filename_encoding'};
|
---|
628 | my $filename_encoding = $self->deduce_filename_encoding($filename_full_path,$metadata,$plugin_filename_encoding);
|
---|
629 | $self->set_Source_metadata($doc_obj, $filename_full_path, $filename_encoding);
|
---|
630 |
|
---|
631 | # if we want a header page, we need to add some text into the top section, otherwise this section will become invisible
|
---|
632 | if ($self->{'headerpage'}) {
|
---|
633 | $self->add_dummy_text($doc_obj, $topsection);
|
---|
634 | }
|
---|
635 | }
|
---|
636 |
|
---|
637 | sub scan_xml_for_files_to_block
|
---|
638 | {
|
---|
639 | my $self = shift (@_);
|
---|
640 | my ($filename_full_path, $dir, $block_hash) = @_;
|
---|
641 |
|
---|
642 | open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path to work out which files to block\n";
|
---|
643 | my $line = "";
|
---|
644 | while (defined ($line = <ITEMFILE>)) {
|
---|
645 | next unless $line =~ /\w/;
|
---|
646 |
|
---|
647 | if ($line =~ /imgfile=\"([^\"]+)\"/) {
|
---|
648 | $self->block_raw_filename($block_hash,&FileUtils::filenameConcatenate($dir,$1));
|
---|
649 | }
|
---|
650 | if ($line =~ /txtfile=\"([^\"]+)\"/) {
|
---|
651 | $self->block_raw_filename($block_hash,&FileUtils::filenameConcatenate($dir,$1));
|
---|
652 | }
|
---|
653 | }
|
---|
654 | close ITEMFILE;
|
---|
655 |
|
---|
656 | }
|
---|
657 |
|
---|
658 | sub scan_item_for_files_to_block
|
---|
659 | {
|
---|
660 | my $self = shift (@_);
|
---|
661 | my ($filename_full_path, $dir, $block_hash) = @_;
|
---|
662 |
|
---|
663 |
|
---|
664 | open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path to work out which files to block\n";
|
---|
665 | my $line = "";
|
---|
666 | while (defined ($line = <ITEMFILE>)) {
|
---|
667 | next unless $line =~ /\w/;
|
---|
668 | chomp $line;
|
---|
669 | next if $line =~ /^#/; # ignore comment lines
|
---|
670 | next if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/); # ignore metadata lines
|
---|
671 | # line should be like page:imagefilename:textfilename:r
|
---|
672 | $line =~ s/^\s+//; #remove space at the front
|
---|
673 | $line =~ s/\s+$//; #remove space at the end
|
---|
674 | my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;
|
---|
675 |
|
---|
676 | # find the image file if there is one
|
---|
677 | if (defined $imgname && $imgname ne "") {
|
---|
678 | $self->block_raw_filename($block_hash, &FileUtils::filenameConcatenate( $dir,$imgname));
|
---|
679 | }
|
---|
680 | # find the text file if there is one
|
---|
681 | if (defined $txtname && $txtname ne "") {
|
---|
682 | $self->block_raw_filename($block_hash, &FileUtils::filenameConcatenate($dir,$txtname));
|
---|
683 | }
|
---|
684 | }
|
---|
685 | close ITEMFILE;
|
---|
686 |
|
---|
687 | }
|
---|
688 |
|
---|
689 | sub process_item {
|
---|
690 | my $self = shift (@_);
|
---|
691 | my ($filename_full_path, $dir, $filename_no_path, $processor, $metadata) = @_;
|
---|
692 |
|
---|
693 | my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
|
---|
694 | $self->set_initial_doc_fields($doc_obj, $filename_full_path, $processor, $metadata);
|
---|
695 | my $topsection = $doc_obj->get_top_section();
|
---|
696 | # simple item files are always paged unless user specified
|
---|
697 | if ($self->{'documenttype'} eq "auto") {
|
---|
698 | $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "paged");
|
---|
699 | } else {
|
---|
700 | $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", $self->{'documenttype'});
|
---|
701 | }
|
---|
702 | open (ITEMFILE, "<:encoding(UTF-8)", $filename_full_path) || die "couldn't open $filename_full_path\n";
|
---|
703 | my $line = "";
|
---|
704 | my $num = 0;
|
---|
705 | while (defined ($line = <ITEMFILE>)) {
|
---|
706 |
|
---|
707 | next unless $line =~ /\w/;
|
---|
708 | chomp $line;
|
---|
709 | next if $line =~ /^#/; # ignore comment lines
|
---|
710 | if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/) {
|
---|
711 | my $meta_name = $1;
|
---|
712 | my $meta_value = $2;
|
---|
713 | if ($meta_name =~ /\./) {
|
---|
714 | $meta_name = "ex.$meta_name";
|
---|
715 | }
|
---|
716 | $doc_obj->set_utf8_metadata_element ($topsection, $meta_name, $meta_value);
|
---|
717 | #$meta->{$1} = $2;
|
---|
718 | } else {
|
---|
719 | $num++;
|
---|
720 | # line should be like page:imagefilename:textfilename:r - the r is optional -> means rotate the image 180 deg
|
---|
721 | $line =~ s/^\s+//; #remove space at the front
|
---|
722 | $line =~ s/\s+$//; #remove space at the end
|
---|
723 | my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;
|
---|
724 |
|
---|
725 | # create a new section for each image file
|
---|
726 | my $cursection = $doc_obj->insert_section($doc_obj->get_end_child($topsection));
|
---|
727 | # the page number becomes the Title
|
---|
728 | $doc_obj->set_utf8_metadata_element($cursection, 'Title', $pagenum);
|
---|
729 |
|
---|
730 | # process the image for this page if there is one
|
---|
731 | if (defined $imgname && $imgname ne "") {
|
---|
732 | my $result1 = $self->process_image($dir.$imgname, $imgname, $doc_obj, $cursection, $rotate);
|
---|
733 | if (!defined $result1)
|
---|
734 | {
|
---|
735 | print "PagedImagePlugin: couldn't process image \"$dir$imgname\" for item \"$filename_full_path\"\n";
|
---|
736 | }
|
---|
737 | }
|
---|
738 | # process the text file if one is there
|
---|
739 | if (defined $txtname && $txtname ne "") {
|
---|
740 | my $result2 = $self->process_text ($dir.$txtname, $txtname, $doc_obj, $cursection);
|
---|
741 |
|
---|
742 | if (!defined $result2) {
|
---|
743 | print "PagedImagePlugin: couldn't process text file \"$dir.$txtname\" for item \"$filename_full_path\"\n";
|
---|
744 | $self->add_dummy_text($doc_obj, $cursection);
|
---|
745 | }
|
---|
746 | } else {
|
---|
747 | # otherwise add in some dummy text
|
---|
748 | $self->add_dummy_text($doc_obj, $cursection);
|
---|
749 | }
|
---|
750 | }
|
---|
751 | }
|
---|
752 |
|
---|
753 | close ITEMFILE;
|
---|
754 |
|
---|
755 | # add numpages metadata
|
---|
756 | $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', "$num");
|
---|
757 |
|
---|
758 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
|
---|
759 | $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
|
---|
760 | $self->{'MaxImageWidth'} = undef;
|
---|
761 | $self->{'MaxImageHeight'} = undef;
|
---|
762 |
|
---|
763 |
|
---|
764 | return $doc_obj;
|
---|
765 | }
|
---|
766 |
|
---|
767 | sub process_text {
|
---|
768 | my $self = shift (@_);
|
---|
769 | my ($filename_full_path, $file, $doc_obj, $cursection) = @_;
|
---|
770 |
|
---|
771 | # check that the text file exists!!
|
---|
772 | if (!-f $filename_full_path) {
|
---|
773 | print "PagedImagePlugin: ERROR: File $filename_full_path does not exist, skipping\n";
|
---|
774 | return 0;
|
---|
775 | }
|
---|
776 |
|
---|
777 | # remember that this text file was one of our source files, but only
|
---|
778 | # if we are not processing a tmp file
|
---|
779 | if (!$self->{'processing_tmp_files'} ) {
|
---|
780 | $doc_obj->associate_source_file($filename_full_path);
|
---|
781 | }
|
---|
782 | # Do encoding stuff
|
---|
783 | my ($language, $encoding) = $self->textcat_get_language_encoding ($filename_full_path);
|
---|
784 |
|
---|
785 | my $text="";
|
---|
786 | &ReadTextFile::read_file($self, $filename_full_path, $encoding, $language, \$text); # already decoded as utf8
|
---|
787 | if (!length ($text)) {
|
---|
788 | # It's a bit unusual but not out of the question to have no text, so just give a warning
|
---|
789 | print "PagedImagePlugin: WARNING: $filename_full_path contains no text\n";
|
---|
790 | }
|
---|
791 |
|
---|
792 | # we need to escape the escape character, or else mg will convert into
|
---|
793 | # eg literal newlines, instead of leaving the text as '\n'
|
---|
794 | $text =~ s/\\/\\\\/g; # macro language
|
---|
795 | $text =~ s/_/\\_/g; # macro language
|
---|
796 |
|
---|
797 |
|
---|
798 | if ($text =~ m/<html.*?>\s*<head.*?>.*<\/head>\s*<body.*?>(.*)<\/body>\s*<\/html>\s*$/is) {
|
---|
799 | # looks like HTML input
|
---|
800 | # no need to escape < and > or put in <pre> tags
|
---|
801 |
|
---|
802 | $text = $1;
|
---|
803 |
|
---|
804 | # add text to document object
|
---|
805 | $doc_obj->add_utf8_text($cursection, "$text");
|
---|
806 | }
|
---|
807 | else {
|
---|
808 | $text =~ s/</</g;
|
---|
809 | $text =~ s/>/>/g;
|
---|
810 |
|
---|
811 | # insert preformat tags and add text to document object
|
---|
812 | $doc_obj->add_utf8_text($cursection, "<pre>\n$text\n</pre>");
|
---|
813 | }
|
---|
814 |
|
---|
815 |
|
---|
816 | return 1;
|
---|
817 | }
|
---|
818 |
|
---|
819 |
|
---|
820 | sub clean_up_after_doc_obj_processing {
|
---|
821 | my $self = shift(@_);
|
---|
822 |
|
---|
823 | $self->ImageConverter::clean_up_temporary_files();
|
---|
824 | }
|
---|
825 |
|
---|
826 | 1;
|
---|