[22227] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # OpenOfficeConverter - helper plugin that does office document conversion
|
---|
| 4 | # using jodconverter combined with OpenOffice
|
---|
| 5 | #
|
---|
| 6 | # A component of the Greenstone digital library software
|
---|
| 7 | # from the New Zealand Digital Library Project at the
|
---|
| 8 | # University of Waikato, New Zealand.
|
---|
| 9 | #
|
---|
| 10 | # Copyright (C) 2010 New Zealand Digital Library Project
|
---|
| 11 | #
|
---|
| 12 | # This program is free software; you can redistribute it and/or modify
|
---|
| 13 | # it under the terms of the GNU General Public License as published by
|
---|
| 14 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 15 | # (at your option) any later version.
|
---|
| 16 | #
|
---|
| 17 | # This program is distributed in the hope that it will be useful,
|
---|
| 18 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 19 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 20 | # GNU General Public License for more details.
|
---|
| 21 | #
|
---|
| 22 | # You should have received a copy of the GNU General Public License
|
---|
| 23 | # along with this program; if not, write to the Free Software
|
---|
| 24 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 25 | #
|
---|
| 26 | ###########################################################################
|
---|
| 27 | package OpenOfficeConverter;
|
---|
| 28 |
|
---|
[22494] | 29 | use ConvertBinaryFile;
|
---|
[22227] | 30 | use BaseMediaConverter;
|
---|
| 31 |
|
---|
| 32 | use strict;
|
---|
| 33 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
| 34 |
|
---|
| 35 | use gsprintf 'gsprintf';
|
---|
| 36 |
|
---|
[22508] | 37 | # these two variables mustn't be initialised here or they will get stuck
|
---|
| 38 | # at those values.
|
---|
| 39 | our $openoffice_conversion_available;
|
---|
| 40 | our $no_openoffice_conversion_reason;
|
---|
| 41 |
|
---|
[22227] | 42 | BEGIN {
|
---|
[22609] | 43 | @OpenOfficeConverter::ISA = ('BaseMediaConverter');
|
---|
[22227] | 44 |
|
---|
| 45 | # Check that OpenOffice and jodconverter are installed and available on
|
---|
| 46 | # the path
|
---|
[22508] | 47 | $openoffice_conversion_available = 1;
|
---|
| 48 | $no_openoffice_conversion_reason = "";
|
---|
[22227] | 49 |
|
---|
[22375] | 50 | if (! defined $ENV{'GEXT_OPENOFFICE'}) {
|
---|
[22227] | 51 | $openoffice_conversion_available = 0;
|
---|
| 52 | $no_openoffice_conversion_reason = "gextopenofficenotinstalled";
|
---|
| 53 | }
|
---|
| 54 | else {
|
---|
[22375] | 55 | my $gextoo_home = $ENV{'GEXT_OPENOFFICE'};
|
---|
[22609] | 56 | my $jodjar = &util::filename_cat($gextoo_home,"lib","java","jodconverterx.jar");
|
---|
[22227] | 57 |
|
---|
| 58 | if (!-e $jodjar) {
|
---|
| 59 | print STDERR "Failed to find $jodjar\n";
|
---|
| 60 | $openoffice_conversion_available = 0;
|
---|
| 61 | $no_openoffice_conversion_reason = "gextjodconverternotinstalled";
|
---|
| 62 | }
|
---|
| 63 | else {
|
---|
| 64 | # test to see if soffice in in path
|
---|
[22494] | 65 | if ($ENV{'GSDLOS'} =~ m/^windows$/) {
|
---|
| 66 | my $ooffice_dir_guess =
|
---|
| 67 | &util::filename_cat($ENV{'ProgramFiles'},"OpenOffice.org 3",
|
---|
| 68 | "program");
|
---|
| 69 | if (-d $ooffice_dir_guess) {
|
---|
| 70 | &util::envvar_append("PATH",$ooffice_dir_guess);
|
---|
| 71 | }
|
---|
| 72 | }
|
---|
[22502] | 73 | my $cmd = "soffice -headless 2>&1";
|
---|
[22494] | 74 |
|
---|
| 75 | my $status = system($cmd);
|
---|
| 76 | if ($status != 0) {
|
---|
[22227] | 77 | print STDERR "Failed to run: $cmd\n";
|
---|
| 78 | print STDERR "$!\n";
|
---|
| 79 | $openoffice_conversion_available = 0;
|
---|
| 80 | $no_openoffice_conversion_reason = "openofficenotinstalled";
|
---|
[22502] | 81 | }
|
---|
[22227] | 82 | }
|
---|
| 83 | }
|
---|
| 84 |
|
---|
[22508] | 85 | }
|
---|
[22227] | 86 |
|
---|
[22508] | 87 | my $arguments = [
|
---|
| 88 | { 'name' => "openoffice_port",
|
---|
| 89 | 'desc' => "{OpenOfficeConverter.openoffice_port}",
|
---|
| 90 | 'type' => "int",
|
---|
| 91 | 'deft' => "8100",
|
---|
| 92 | 'range' => "81,",
|
---|
| 93 | 'reqd' => "no" },
|
---|
| 94 | ];
|
---|
| 95 |
|
---|
| 96 | my $options = { 'name' => "OpenOfficeConverter",
|
---|
| 97 | 'desc' => "{OpenOfficeConverter.desc}",
|
---|
| 98 | 'abstract' => "yes",
|
---|
| 99 | 'inherits' => "yes",
|
---|
| 100 | 'args' => $arguments };
|
---|
| 101 |
|
---|
| 102 | sub new {
|
---|
| 103 | my ($class) = shift (@_);
|
---|
| 104 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
| 105 | push(@$pluginlist, $class);
|
---|
| 106 |
|
---|
[22227] | 107 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
| 108 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
| 109 |
|
---|
[22609] | 110 | my $self = new BaseMediaConverter($pluginlist, $inputargs, $hashArgOptLists);
|
---|
[22494] | 111 |
|
---|
[22508] | 112 | if ($self->{'info_only'}) {
|
---|
| 113 | # don't worry about any options etc
|
---|
| 114 | return bless $self, $class;
|
---|
| 115 | }
|
---|
[22227] | 116 | if ($openoffice_conversion_available) {
|
---|
[22506] | 117 | my $oo_port = $self->{'openoffice_port'};
|
---|
| 118 | my $launch_cmd = "soffice";
|
---|
| 119 | $launch_cmd .= " \"-accept=socket,host=localhost,port=$oo_port;urp;StarOffice.ServiceManager\"";
|
---|
| 120 | $launch_cmd .= " -headless";
|
---|
| 121 |
|
---|
| 122 | $self->{'openoffice_launch_cmd'} = $launch_cmd;
|
---|
[22227] | 123 | }
|
---|
| 124 | else {
|
---|
| 125 | $self->{'no_openoffice_conversion_reason'} = $no_openoffice_conversion_reason;
|
---|
| 126 |
|
---|
| 127 | my $outhandle = $self->{'outhandle'};
|
---|
| 128 | &gsprintf($outhandle, "OpenOfficeConverter: {OpenOfficeConverter.noconversionavailable} ({OpenOfficeConverter.$no_openoffice_conversion_reason})\n");
|
---|
| 129 | }
|
---|
[22494] | 130 |
|
---|
| 131 | $self->{'openoffice_conversion_available'} = $openoffice_conversion_available;
|
---|
[22227] | 132 |
|
---|
| 133 | return bless $self, $class;
|
---|
| 134 |
|
---|
| 135 | }
|
---|
| 136 |
|
---|
| 137 | sub init {
|
---|
| 138 | my $self = shift(@_);
|
---|
[22494] | 139 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
[22227] | 140 |
|
---|
| 141 | $self->{'ootmp_file_paths'} = ();
|
---|
| 142 | }
|
---|
| 143 |
|
---|
| 144 | sub deinit {
|
---|
| 145 | my $self = shift(@_);
|
---|
| 146 |
|
---|
| 147 | $self->clean_up_temporary_files();
|
---|
| 148 | }
|
---|
[22494] | 149 |
|
---|
| 150 |
|
---|
[22227] | 151 | sub convert {
|
---|
| 152 | my $self = shift(@_);
|
---|
[22609] | 153 | my ($source_file_full_path, $target_file_type) = @_;
|
---|
| 154 |
|
---|
| 155 | return 0 unless $openoffice_conversion_available;
|
---|
| 156 | # check the filename
|
---|
| 157 | return 0 if ( !-f $source_file_full_path);
|
---|
| 158 |
|
---|
| 159 | my $source_file_no_path = &File::Basename::basename($source_file_full_path);
|
---|
| 160 | # Determine the full name and path of the output file
|
---|
| 161 | my $target_file_path;
|
---|
| 162 | if ($self->{'enable_cache'}) {
|
---|
| 163 | $self->init_cache_for_file($source_file_full_path);
|
---|
| 164 | my $cache_dir = $self->{'cached_dir'};
|
---|
| 165 | my $file_root = $self->{'cached_file_root'};
|
---|
| 166 | #$file_root .= "_$convert_id" if ($convert_id ne "");
|
---|
| 167 | my $target_file = "$file_root.$target_file_type";
|
---|
| 168 | $target_file_path = &util::filename_cat($cache_dir,$target_file);
|
---|
| 169 | }
|
---|
| 170 | else {
|
---|
| 171 | # this is in gsdl/tmp. get a tmp filename in collection instead???
|
---|
| 172 | $target_file_path = &util::get_tmp_filename($target_file_type);
|
---|
| 173 | push(@{$self->{'ootmp_file_paths'}}, $target_file_path);
|
---|
| 174 | }
|
---|
| 175 |
|
---|
| 176 | # Generate and run the convert command
|
---|
| 177 | my $convert_command = "oo2html \"$source_file_full_path\" \"$target_file_path\"";
|
---|
| 178 | print STDERR "convert_command = $convert_command\n";
|
---|
| 179 | my $print_info = { 'message_prefix' => "OpenOffice Conversion",
|
---|
| 180 | 'message' => "Converting $source_file_no_path to: $target_file_type" };
|
---|
| 181 | #$print_info->{'cache_mode'} = $cache_mode if ($cache_mode ne "");
|
---|
| 182 |
|
---|
| 183 | my ($regenerated,$result,$had_error)
|
---|
| 184 | = $self->autorun_general_cmd($convert_command,$source_file_full_path, $target_file_path,$print_info);
|
---|
| 185 | if ($had_error) {
|
---|
| 186 | return (0, $result,$target_file_path);
|
---|
| 187 | }
|
---|
| 188 | return (1, $result,$target_file_path);
|
---|
| 189 | }
|
---|
| 190 |
|
---|
| 191 |
|
---|
| 192 |
|
---|
| 193 | sub convert_orig {
|
---|
| 194 | my $self = shift(@_);
|
---|
[22227] | 195 | my $source_file_path = shift(@_);
|
---|
| 196 | my $target_file_type = shift(@_);
|
---|
| 197 | my $convert_options = shift(@_) || "";
|
---|
| 198 | my $convert_id = shift(@_) || "";
|
---|
| 199 | my $cache_mode = shift(@_) || "";
|
---|
| 200 |
|
---|
| 201 | my $outhandle = $self->{'outhandle'};
|
---|
| 202 | my $verbosity = $self->{'verbosity'};
|
---|
| 203 |
|
---|
| 204 | my $source_file_no_path = &File::Basename::basename($source_file_path);
|
---|
| 205 |
|
---|
| 206 | # Determine the full name and path of the output file
|
---|
| 207 | my $target_file_path;
|
---|
| 208 | if ($self->{'enable_cache'}) {
|
---|
| 209 | my $cached_image_dir = $self->{'cached_dir'};
|
---|
| 210 | my $image_root = $self->{'cached_file_root'};
|
---|
| 211 | $image_root .= "_$convert_id" if ($convert_id ne "");
|
---|
| 212 | my $image_file = "$image_root.$target_file_type";
|
---|
| 213 | $target_file_path = &util::filename_cat($cached_image_dir,$image_file);
|
---|
| 214 | }
|
---|
| 215 | else {
|
---|
| 216 | $target_file_path = &util::get_tmp_filename($target_file_type);
|
---|
| 217 | push(@{$self->{'ootmp_file_paths'}}, $target_file_path);
|
---|
| 218 | }
|
---|
| 219 |
|
---|
| 220 | # Generate and run the convert command
|
---|
| 221 | my $convert_command = "oo2html $convert_options \"$source_file_path\" \"$target_file_path\"";
|
---|
| 222 |
|
---|
| 223 | my $print_info = { 'message_prefix' => $convert_id,
|
---|
| 224 | 'message' => "Converting $source_file_no_path to: $convert_id $target_file_type" };
|
---|
| 225 | $print_info->{'cache_mode'} = $cache_mode if ($cache_mode ne "");
|
---|
| 226 |
|
---|
| 227 | my ($regenerated,$result,$had_error)
|
---|
| 228 | = $self->autorun_general_cmd($convert_command,$target_file_path,$print_info);
|
---|
| 229 |
|
---|
| 230 | return ($result,$target_file_path);
|
---|
| 231 | }
|
---|
| 232 |
|
---|
| 233 | sub convert_without_result {
|
---|
| 234 | my $self = shift(@_);
|
---|
| 235 |
|
---|
| 236 | my $source_file_path = shift(@_);
|
---|
| 237 | my $target_file_type = shift(@_);
|
---|
| 238 | my $convert_options = shift(@_) || "";
|
---|
| 239 | my $convert_id = shift(@_) || "";
|
---|
| 240 |
|
---|
| 241 | return $self->convert($source_file_path,$target_file_type,
|
---|
| 242 | $convert_options,$convert_id,"without_result");
|
---|
| 243 | }
|
---|
| 244 |
|
---|
| 245 |
|
---|
[22494] | 246 | sub tmp_area_convert_fileXX {
|
---|
| 247 | my $self = shift (@_);
|
---|
| 248 | my ($output_ext, $input_filename, $textref) = @_;
|
---|
| 249 |
|
---|
| 250 | my $outhandle = $self->{'outhandle'};
|
---|
| 251 | my $convert_to = $self->{'convert_to'};
|
---|
| 252 | my $failhandle = $self->{'failhandle'};
|
---|
| 253 | my $convert_to_ext = $self->{'convert_to_ext'};
|
---|
| 254 |
|
---|
| 255 | # derive tmp filename from input filename
|
---|
| 256 | my ($tailname, $dirname, $suffix)
|
---|
| 257 | = &File::Basename::fileparse($input_filename, "\\.[^\\.]+\$");
|
---|
| 258 | }
|
---|
[22227] | 259 |
|
---|
[22494] | 260 |
|
---|
[22227] | 261 | sub clean_up_temporary_files {
|
---|
| 262 | my $self = shift(@_);
|
---|
| 263 |
|
---|
| 264 | foreach my $ootmp_file_path (@{$self->{'ootmp_file_paths'}}) {
|
---|
| 265 | if (-e $ootmp_file_path) {
|
---|
| 266 | &util::rm($ootmp_file_path);
|
---|
| 267 | }
|
---|
| 268 | }
|
---|
| 269 |
|
---|
| 270 | $self->{'ootmp_file_paths'} = ();
|
---|
| 271 | }
|
---|
| 272 |
|
---|
| 273 |
|
---|
| 274 |
|
---|
| 275 | 1;
|
---|