Changeset 15872 for gsdl/trunk/perllib/plugins/NulPlugin.pm

Timestamp:

2008-06-05T09:29:32+12:00 (16 years ago)

Author:

kjdon

Message:

plugin overhaul: plugins renamed to xxPlugin, and in some cases the names are made more sensible. They now use the new base plugins. Hopefully we have better code reuse. Some of the plugins still need work done as I didn't want to spend another month doing this before committing it. Alos, I haven't really tested anything yet...

File:

• gsdl/trunk/perllib/plugins/NulPlugin.pm (modified) (6 diffs)

gsdl/trunk/perllib/plugins/NulPlugin.pm

@@ -1 +1 @@
 ###########################################################################
 #
-# NULPlug.pm -- Plugin for dummy (.nul) files
+# NulPlugin.pm -- Plugin for dummy (.nul) files
 #
 # A component of the Greenstone digital library software from the New
@@ -26 +26 @@
 ###########################################################################
 
-# NULPlug - a plugin for dummy files
+# NulPlugin - a plugin for dummy files
 
 # This is a simple Plugin for importing dummy files, along with
@@ -35 +35 @@
 # databases
 
-package NULPlug;
+package NulPlugin;
 
-use BasPlug;
+use BasePlugin;
 
 use strict;
@@ -43 +43 @@
 
 sub BEGIN {
-    @NULPlug::ISA = ('BasPlug');
+    @NulPlugin::ISA = ('BasePlugin');
 }
 
 my $arguments = 
     [ { 'name' => "process_exp",
-    'desc' => "{BasPlug.process_exp}",
+    'desc' => "{BasePlugin.process_exp}",
     'type' => "regexp",
     'reqd' => "no",
     'deft' => &get_default_process_exp() },
       { 'name' => "assoc_field",
-    'desc' => "{NULPlug.assoc_field}",
+    'desc' => "{NulPlugin.assoc_field}",
     'type' => "string",
-    'deft' => "",
+    'deft' => "null_file",
     'reqd' => "no" },
       { 'name' => "add_metadata_as_text",
-    'desc' => "{NULPlug.add_metadata_as_text}",
+    'desc' => "{NulPlugin.add_metadata_as_text}",
     'type' => "flag" },
       { 'name' => "remove_namespace_for_text",
-    'desc' => "{NULPlug.remove_namespace_for_text}",
+    'desc' => "{NulPlugin.remove_namespace_for_text}",
     'type' => "flag" }
       ];
 
-my $options = { 'name'     => "NULPlug",
-        'desc'     => "{NULPlug.desc}",
+my $options = { 'name'     => "NulPlugin",
+        'desc'     => "{NulPlugin.desc}",
         'abstract' => "no",
         'inherits' => "yes",
@@ -77 +77 @@
     push(@$pluginlist, $class);
 
-    if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
-    if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
+    push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
+    push(@{$hashArgOptLists->{"OptList"}},$options);
 
-    my $self = new BasPlug($pluginlist, $inputargs, $hashArgOptLists);
+    my $self = new BasePlugin($pluginlist, $inputargs, $hashArgOptLists);
     
     return bless $self, $class;
@@ -89 +89 @@
 }
 
-# The NULPlug read() function. This function does all the right
-# things to make general options work for a given plugin.  NULPlug
-# overrides read() because there is no need to read the actual text of
-# the file in, because the contents of the file is not text...
-#
-#
-# Return number of files processed, undef if can't process 
-#
-# Note that $base_dir might be "" and that $file might include directories
+# NulPlugin specific processing of doc_obj. 
+sub process {
+    my $self = shift (@_);
+    my ($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
+    
+    my $topsection = $doc_obj->get_top_section();
+       
+    my $assoc_field = $self->{'assoc_field'}; # || "null_file"; TODO, check this
+    $doc_obj->add_metadata ($topsection, $assoc_field, $file);
 
-sub read {
-    my $self = shift (@_);
-    my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
-
-    my $outhandle = $self->{'outhandle'};
-
-    #check for associate_ext, blocking etc
-    my ($block_status,$filename) = $self->read_block(@_);    
-    return $block_status if ((!defined $block_status) || ($block_status==0));
-
-    print STDERR "<Processing n='$file' p='NULPlug'>\n" if ($gli);
-    print $outhandle "NULPlug processing \"$filename\"\n"
-        if $self->{'verbosity'} > 1;
-
-    #if there's a leading directory name, eat it...
-    $file =~ s/^.*[\/\\]//;
-    
-    # create a new document
-    my $doc_obj = new doc ($filename, "indexed_doc");
-    my $top_section = $doc_obj->get_top_section();
-
-    $doc_obj->set_OIDtype ($processor->{'OIDtype'}, $processor->{'OIDmetadata'});    
-    #$doc_obj->set_OIDtype ("incremental");
-    $doc_obj->add_utf8_metadata($top_section, "Plugin", "$self->{'plugin_type'}");
-    $doc_obj->add_metadata($top_section, "Source", $file); # set the filename as Source metadata to be consistent with other plugins
-
-    $doc_obj->add_utf8_metadata($top_section, "FileSize", (-s $filename));
-
-    # the metadata NoText is used to suppress the dummy text 'This document has no text.'
-    $doc_obj->add_metadata ($top_section, "NoText",    "1");
-
-    my $assoc_field = $self->{'assoc_field'} || "null_file";
-    $doc_obj->add_metadata ($top_section, $assoc_field, $file);
-    
-     if ($self->{'cover_image'}) {
-    $self->associate_cover_image($doc_obj, $filename);
-    }
-
-    # include any metadata passed in from previous plugins 
-    my $section = $doc_obj->get_top_section();
-    $self->extra_metadata ($doc_obj, $section, $metadata);
-    
     # format the metadata passed in (presumably from metadata.xml)
     my $text = "";
     if ($self->{'add_metadata_as_text'}) {
     $text = &metadatautil::format_metadata_as_table($metadata, $self->{'remove_namespace_for_text'});
+    $doc_obj->add_utf8_text($topsection, $text);
     } else {
-    #create an empty text string so we don't break downstream plugins 
-    $text = &gsprintf::lookup_string("{BasPlug.dummy_text}",1);
+    $self->add_dummy_text($doc_obj, $topsection);
     }
-    $self->title_fallback($doc_obj,$section,$file);
-    
-    # do plugin specific processing of doc_obj
-    unless (defined ($self->process(\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj))) {
-    print STDERR "<ProcessingError n='$file'>\n" if ($gli);
-    return -1;
-    }
-
-    # do any automatic metadata extraction
-    $self->auto_extract_metadata ($doc_obj);
-
-    # add an OID
-    $doc_obj->set_OID();
-    $doc_obj->add_utf8_text($section, $text);
-    
-    # process the document
-    $processor->process($doc_obj);
-
-    $self->{'num_processed'} ++;
-    return 1;
-}
-
-
-# NULPlug processing of doc_obj.  In practice we don't need to do
-# anything here because the read function takes care of everything.
-
-sub process {
-    my $self = shift (@_);
-    my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
-    my $outhandle = $self->{'outhandle'};
     
     return 1;