########################################################################### # # ConvertToPlug.pm -- plugin that inherits from HTML or TEXT Plug, depending # on plugin argument convert_to # # A component of the Greenstone digital library software # from the New Zealand Digital Library Project at the # University of Waikato, New Zealand. # # Copyright (C) 1999 New Zealand Digital Library Project # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. # ########################################################################### # The plugin is inherited by such plugins as WordPlug and PDFPlug. # It facilitates the conversion of these document types to either HTML # or TEXT by setting up variable that instruct ConvertToBasPlug # how to work. # It works by dynamically inheriting HTMLPlug or TEXTPlug based on # the plugin argument 'convert_to'. If the argument is not present, # the default is to inherit HTMLPlug. package ConvertToPlug; use BasPlug; use HTMLPlug; use TEXTPlug; sub BEGIN { @ISA = ('HTMLPlug'); # @ISA = ('HTMLPlug', 'TEXTPlug'); # @ISA = ('BasPlug'); #, 'HTMLPlug', 'TEXTPlug'); } # use strict; # this breaks 'print $outhandle ' error msgs. sub print_usage { my ($plugin_name) = @_; # for when this function is called directly by pluginfo.pl if (ref ($plugin_name)) { $plugin_name = ref ($plugin_name); } print STDERR "\n usage: plugin $plugin_name [options]\n\n"; print STDERR " options:\n"; print STDERR " -convert_to (html|text) plugin converts to TEXT or HTML\n"; print STDERR " (default html)\n"; } sub parse_args { my $class = shift (@_); my ($args) = @_; my $plugin_name = $class; $plugin_name =~ s/\.pm$//; my $generate_format; my $kea_arg; if (!parsargv::parse($args, q^extract_keyphrases^, \$kea_arg->{'kea'}, #with extra options q^extract_keyphrase_options/.*/^, \$kea_arg->{'kea_options'}, #no extra options q^convert_to/(html|text)/html^, \$generate_format, "allow_extra_options")) { print STDERR "\nIncorrect options passed to $plugin_name, "; print STDERR "check your collect.cfg configuration file\n"; &print_usage($plugin_name); die "\n"; } return ($plugin_name,$generate_format, $kea_arg); } sub new { my $class = shift (@_); my ($plugin_name,$generate_format, $kea_arg) = $class->parse_args(\@_); my $self; if ($generate_format eq "text") { $self = new TEXTPlug ($class, @_); $self->{'convert_to'} = "TEXT"; $self->{'convert_to_ext'} = "txt"; } else { $self = new HTMLPlug ($class, @_); $self->{'convert_to'} = "HTML"; $self->{'convert_to_ext'} = "html"; $self->{'rename_assoc_files'} = 1; $self->{'metadata_fields'} .= ",GENERATOR"; } #if kea data to be extracted... $self->{'kea'} = 1 if($kea_arg->{'kea'}); $self->{'kea_options'} = 1 if($kea_arg->{'kea_options'}); return bless $self, $class; } # Run conversion utility on the input file. # # The conversion takes place in a collection specific 'tmp' directory so # that we don't accidentally damage the input. # # The desired output type is indicated by $output_ext. This is usually # something like "html" or "word", but can be "best" (or the empty string) # to indicate that the conversion utility should do the best it can. sub tmp_area_convert_file { my $self = shift (@_); my ($output_ext,$input_filename, $textref) = @_; my $convert_to = $self->{'convert_to'}; # softlink to collection tmp dir my $colname = &util::use_collection(); my $tmp_dirname = &util::filename_cat($ENV{'GSDLHOME'},"collect",$colname,"tmp"); &util::mk_dir($tmp_dirname) if (!-e $tmp_dirname); # derive tmp filename from input filename my ($tailname,$dirname,$suffix) = File::Basename::fileparse($input_filename,'\.[^\.]+$'); my $tmp_filename = &util::filename_cat($tmp_dirname,"$tailname$suffix"); # Remove any white space from filename -- no risk of name collision, and # makes later conversion by utils simpler $tmp_filename =~ s/\s+//g; &util::soft_link($input_filename,$tmp_filename); my $verbosity = $self->{'verbosity'}; if ($verbosity>0) { print STDERR "Converting $tailname$suffix to $convert_to format\n"; } # Execute the conversion command and get the type of the result, # making sure the converter gives us the appropriate output type my $output_type = lc($convert_to); my $cmd = "gsConvert.pl -verbose $verbosity -output $output_type \"$tmp_filename\""; $output_type = `$cmd`; # Check STDERR here chomp $output_type; if ($output_type eq "fail") { print STDERR "Could not convert $tailname$suffix to $convert_to format\n"; return ""; ### exit 1; } # remove symbolic link to original file &util::rm($tmp_filename); # store the *actual* output type and return the output filename $self->{'convert_to_ext'} = $output_type; my $output_filename = $tmp_filename; $output_filename =~ s/$suffix$/.$output_type/; return $output_filename; } # Remove collection specific tmp directory and all its contents. sub cleanup_tmp_area { my $self = shift (@_); my $colname = &util::use_collection(); my $tmp_dirname = &util::filename_cat($ENV{'GSDLHOME'},"collect",$colname,"tmp"); &util::rm_r($tmp_dirname); &util::mk_dir($tmp_dirname); } # Override BasPlug read # We don't want to get language encoding stuff until after we've converted # our file to either TEXT or HTML. sub read { my $self = shift (@_); my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs) = @_; # if ($self->is_recursive()) { # die "BasPlug::read function must be implemented in sub-class for recursive plugins\n"; # } my $outhandle = $self->{'outhandle'}; my $filename = &util::filename_cat($base_dir, $file); return 0 if $self->{'block_exp'} ne "" && $filename =~ /$self->{'block_exp'}/; if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) { return undef; } my $plugin_name = ref ($self); $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up # read in file ($text will be in utf8) my $text = ""; my $output_ext = $self->{'convert_to_ext'}; my $conv_filename = $self->tmp_area_convert_file($output_ext,$filename); if ("$conv_filename" eq "") {return 0;} # allows continue on errors $self->{'conv_filename'} = $conv_filename; # Do encoding stuff my ($language, $encoding); if ($self->{'input_encoding'} eq "auto") { # use textcat to automatically work out the input encoding and language ($language, $encoding) = $self->get_language_encoding ($conv_filename); } elsif ($self->{'extract_language'}) { # use textcat to get language metadata my ($language, $extracted_encoding) = $self->get_language_encoding ($conv_filename); $encoding = $self->{'input_encoding'}; if ($extracted_encoding ne $encoding && $self->{'verbosity'}) { print $outhandle "$plugin_name: WARNING: $file was read using $encoding encoding but "; print $outhandle "appears to be encoded as $extracted_encoding.\n"; } } else { $language = $self->{'default_language'}; $encoding = $self->{'input_encoding'}; } BasPlug::read_file($self,$conv_filename, $encoding, \$text); if (!length ($text)) { print $outhandle "$plugin_name: ERROR: $file contains no text\n" if $self->{'verbosity'}; return 0; } # create a new document my $doc_obj = new doc ($conv_filename, "indexed_doc"); $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Language", $language); $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Encoding", $encoding); # include any metadata passed in from previous plugins # note that this metadata is associated with the top level section $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata); # do plugin specific processing of doc_obj return undef unless defined ($self->process (\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj)); # do any automatic metadata extraction $self->auto_extract_metadata ($doc_obj); # add an OID $doc_obj->set_OID(); # process the document $processor->process($doc_obj); $self->cleanup_tmp_area(); return 1; } # do plugin specific processing of doc_obj for HTML type sub process_type { my $self = shift (@_); my ($doc_ext, $textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_; my $conv_filename = $self->{'conv_filename'}; my $tmp_dirname = File::Basename::dirname($conv_filename); my $tmp_tailname = File::Basename::basename($conv_filename); my $convert_to = $self->{'convert_to'}; my $ret_val; if ($convert_to eq "TEXT") { $ret_val = TEXTPlug::process($self,$textref,$pluginfo, $tmp_dirname,$tmp_tailname, $metadata,$doc_obj); } else { $ret_val = HTMLPlug::process($self,$textref,$pluginfo, $tmp_dirname,$tmp_tailname, $metadata,$doc_obj); } # associate original file with doc object my $cursection = $doc_obj->get_top_section(); my $filename = &util::filename_cat($base_dir,$file); $doc_obj->associate_file($filename, "doc.$doc_ext", undef, $cursection); my $doclink = ""; $doc_obj->add_utf8_metadata ($cursection, "srclink", $doclink); $doc_obj->add_utf8_metadata ($cursection, "srcicon", "_icon".$doc_ext."_"); $doc_obj->add_utf8_metadata ($cursection, "/srclink", ""); return $ret_val; } 1;