1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # PDFBoxConverter - helper plugin that does pdf document conversion with PDFBox
|
---|
4 | #
|
---|
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) 2010 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 | package PDFBoxConverter;
|
---|
27 |
|
---|
28 | use BaseMediaConverter;
|
---|
29 |
|
---|
30 | use strict;
|
---|
31 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
32 |
|
---|
33 | use gsprintf 'gsprintf';
|
---|
34 |
|
---|
35 | # these two variables mustn't be initialised here or they will get stuck
|
---|
36 | # at those values.
|
---|
37 | our $pdfbox_conversion_available;
|
---|
38 | our $no_pdfbox_conversion_reason;
|
---|
39 |
|
---|
40 | BEGIN {
|
---|
41 | @PDFBoxConverter::ISA = ('BaseMediaConverter');
|
---|
42 |
|
---|
43 | # Check that PDFBox is installed and available on the path
|
---|
44 | $pdfbox_conversion_available = 1;
|
---|
45 | $no_pdfbox_conversion_reason = "";
|
---|
46 |
|
---|
47 | if (!defined $ENV{'GEXT_PDFBOX'}) {
|
---|
48 | $pdfbox_conversion_available = 0;
|
---|
49 | $no_pdfbox_conversion_reason = "gextpdfboxnotinstalled";
|
---|
50 | }
|
---|
51 | else {
|
---|
52 | my $gextpb_home = $ENV{'GEXT_PDFBOX'};
|
---|
53 | my $pbajar = &util::filename_cat($gextpb_home,"lib","java","pdfbox-app.jar");
|
---|
54 |
|
---|
55 | if (!-e $pbajar) {
|
---|
56 | print STDERR "Failed to find $pbajar\n";
|
---|
57 | $pdfbox_conversion_available = 0;
|
---|
58 | $no_pdfbox_conversion_reason = "gextpdfboxjarnotinstalled";
|
---|
59 | }
|
---|
60 | else {
|
---|
61 | # test to see if java is in path
|
---|
62 | my $cmd = "java";
|
---|
63 | if ($ENV{'GSDLOS'} =~ /^windows/i) {
|
---|
64 | $cmd .= " >nul 2>&1"; # java 2>&1 >null or java >null 2>&1 both work (%ERRORLEVEL% is 0)
|
---|
65 | }
|
---|
66 | else {
|
---|
67 | # On Ubuntu, java >/dev/null 2>&1 works,
|
---|
68 | # but java 2>&1 >/dev/null doesn't work: output goes to screen anyway
|
---|
69 | $cmd .= " >/dev/null 2>&1 &"; # Need the & at the end for Linux Centos (Ubuntu is fine without it), else parsing plugins list fails
|
---|
70 | }
|
---|
71 |
|
---|
72 | my $status = system($cmd);
|
---|
73 |
|
---|
74 | if ($status != 0) {
|
---|
75 | print STDERR "Testing for java\n";
|
---|
76 | print STDERR "Failed to run: $cmd\n";
|
---|
77 | print STDERR "$!\n";
|
---|
78 | $pdfbox_conversion_available = 0;
|
---|
79 | $no_pdfbox_conversion_reason = "couldnotrunjava";
|
---|
80 | }
|
---|
81 | }
|
---|
82 | }
|
---|
83 |
|
---|
84 | }
|
---|
85 |
|
---|
86 | my $arguments = [ ];
|
---|
87 |
|
---|
88 | my $options = { 'name' => "PDFBoxConverter",
|
---|
89 | 'desc' => "{PDFBoxConverter.desc}",
|
---|
90 | 'abstract' => "yes",
|
---|
91 | 'inherits' => "yes",
|
---|
92 | 'args' => $arguments };
|
---|
93 |
|
---|
94 | sub new {
|
---|
95 | my ($class) = shift (@_);
|
---|
96 | my ($pluginlist,$inputargs,$hashArgOptLists,$auxilary) = @_;
|
---|
97 | push(@$pluginlist, $class);
|
---|
98 |
|
---|
99 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
100 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
101 |
|
---|
102 |
|
---|
103 | my $self = new BaseMediaConverter($pluginlist, $inputargs,
|
---|
104 | $hashArgOptLists, $auxilary);
|
---|
105 |
|
---|
106 | if ($self->{'info_only'}) {
|
---|
107 | # don't worry about any options etc
|
---|
108 | return bless $self, $class;
|
---|
109 | }
|
---|
110 | if ($pdfbox_conversion_available) {
|
---|
111 | my $gextpb_home = $ENV{'GEXT_PDFBOX'};
|
---|
112 | my $pbajar = &util::filename_cat($gextpb_home,"lib","java","pdfbox-app.jar");
|
---|
113 | my $launch_cmd = "java -cp \"$pbajar\" -Dline.separator=\"<br />\" org.apache.pdfbox.ExtractText";
|
---|
114 |
|
---|
115 | $self->{'pdfbox_launch_cmd'} = $launch_cmd;
|
---|
116 | }
|
---|
117 | else {
|
---|
118 | $self->{'no_pdfbox_conversion_reason'} = $no_pdfbox_conversion_reason;
|
---|
119 |
|
---|
120 | my $outhandle = $self->{'outhandle'};
|
---|
121 | &gsprintf($outhandle, "PDFBoxConverter: {PDFBoxConverter.noconversionavailable} ({PDFBoxConverter.$no_pdfbox_conversion_reason})\n");
|
---|
122 | }
|
---|
123 |
|
---|
124 | $self->{'pdfbox_conversion_available'} = $pdfbox_conversion_available;
|
---|
125 |
|
---|
126 | return bless $self, $class;
|
---|
127 |
|
---|
128 | }
|
---|
129 |
|
---|
130 | sub init {
|
---|
131 | my $self = shift(@_);
|
---|
132 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
133 |
|
---|
134 | $self->{'pbtmp_file_paths'} = ();
|
---|
135 | }
|
---|
136 |
|
---|
137 | sub deinit {
|
---|
138 | my $self = shift(@_);
|
---|
139 |
|
---|
140 | $self->clean_up_temporary_files();
|
---|
141 | }
|
---|
142 |
|
---|
143 |
|
---|
144 | sub convert {
|
---|
145 | my $self = shift(@_);
|
---|
146 | my ($source_file_full_path, $target_file_type) = @_;
|
---|
147 |
|
---|
148 | return 0 unless $pdfbox_conversion_available;
|
---|
149 | # check the filename
|
---|
150 | return 0 if ( !-f $source_file_full_path);
|
---|
151 |
|
---|
152 | # the following line is necessary to avoid 'uninitialised variable' error
|
---|
153 | # messages concerning the converted_to member variable when PDFPlugin's
|
---|
154 | # use_sections option is checked.
|
---|
155 | # PDFBox plugin now processes use_sections option, when working with v1.5.0
|
---|
156 | # of the PDFBox jar file (which embeds each page in special <div> tags).
|
---|
157 | if ($target_file_type eq "html") {
|
---|
158 | $self->{'converted_to'} = "HTML";
|
---|
159 | } else {
|
---|
160 | $self->{'converted_to'} = "text";
|
---|
161 | }
|
---|
162 |
|
---|
163 | my $outhandle = $self->{'outhandle'};
|
---|
164 | my $verbosity = $self->{'verbosity'};
|
---|
165 |
|
---|
166 | my $source_file_no_path = &File::Basename::basename($source_file_full_path);
|
---|
167 | # Determine the full name and path of the output file
|
---|
168 | my $target_file_path;
|
---|
169 | if ($self->{'enable_cache'}) {
|
---|
170 | $self->init_cache_for_file($source_file_full_path);
|
---|
171 | my $cache_dir = $self->{'cached_dir'};
|
---|
172 | my $file_root = $self->{'cached_file_root'};
|
---|
173 | #$file_root .= "_$convert_id" if ($convert_id ne "");
|
---|
174 | my $target_file = "$file_root.$target_file_type";
|
---|
175 | $target_file_path = &util::filename_cat($cache_dir,$target_file);
|
---|
176 | }
|
---|
177 | else {
|
---|
178 | # this is in gsdl/tmp. get a tmp filename in collection instead???
|
---|
179 | $target_file_path = &util::get_tmp_filename($target_file_type);
|
---|
180 | push(@{$self->{'pbtmp_file_paths'}}, $target_file_path);
|
---|
181 | }
|
---|
182 |
|
---|
183 | # Generate and run the convert command
|
---|
184 | my $convert_cmd = $self->{'pdfbox_launch_cmd'};
|
---|
185 | $convert_cmd .= " -html" if ($target_file_type eq "html");
|
---|
186 | $convert_cmd .= " \"$source_file_full_path\" \"$target_file_path\"";
|
---|
187 |
|
---|
188 | if ($verbosity>2) {
|
---|
189 | print $outhandle "Convert command: $convert_cmd\n";
|
---|
190 | }
|
---|
191 |
|
---|
192 | my $print_info = { 'message_prefix' => "PDFBox Conversion",
|
---|
193 | 'message' => "Converting $source_file_no_path to: $target_file_type" };
|
---|
194 | # $print_info->{'cache_mode'} = $cache_mode if ($cache_mode ne "");
|
---|
195 |
|
---|
196 | my ($regenerated,$result,$had_error)
|
---|
197 | = $self->autorun_general_cmd($convert_cmd,$source_file_full_path, $target_file_path,$print_info);
|
---|
198 | if ($had_error) {
|
---|
199 | return (0, $result,$target_file_path);
|
---|
200 | }
|
---|
201 | return (1, $result,$target_file_path);
|
---|
202 | }
|
---|
203 |
|
---|
204 | sub convert_without_result {
|
---|
205 | my $self = shift(@_);
|
---|
206 |
|
---|
207 | my $source_file_path = shift(@_);
|
---|
208 | my $target_file_type = shift(@_);
|
---|
209 | my $convert_options = shift(@_) || "";
|
---|
210 | my $convert_id = shift(@_) || "";
|
---|
211 |
|
---|
212 | return $self->convert($source_file_path,$target_file_type,
|
---|
213 | $convert_options,$convert_id,"without_result");
|
---|
214 | }
|
---|
215 |
|
---|
216 | sub clean_up_temporary_files {
|
---|
217 | my $self = shift(@_);
|
---|
218 |
|
---|
219 | foreach my $pbtmp_file_path (@{$self->{'pbtmp_file_paths'}}) {
|
---|
220 | if (-e $pbtmp_file_path) {
|
---|
221 | &util::rm($pbtmp_file_path);
|
---|
222 | }
|
---|
223 | }
|
---|
224 |
|
---|
225 | $self->{'pbtmp_file_paths'} = ();
|
---|
226 | }
|
---|
227 |
|
---|
228 |
|
---|
229 |
|
---|
230 | 1;
|
---|