Changeset 15872 for gsdl/trunk/perllib/plugins/HBPlugin.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/HBPlugin.pm (modified) (10 diffs)

gsdl/trunk/perllib/plugins/HBPlugin.pm

@@ -1 +1 @@
 ###########################################################################
 #
-# HBPlug.pm --
+# HBPlugin.pm --
 # A component of the Greenstone digital library software
 # from the New Zealand Digital Library Project at the 
@@ -38 +38 @@
 # Humanity Library collections
 
-package HBPlug;
+package HBPlugin;
 
 use ghtml;
-use BasPlug;
+use BasePlugin;
 use unicode;
 use util;
@@ -50 +50 @@
 
 sub BEGIN {
-    @HBPlug::ISA = ('BasPlug');
-}
-
+    @HBPlugin::ISA = ('BasePlugin');
+}
+my $encoding_list =     
+    [ { 'name' => "ascii",
+    'desc' => "{ReadTextFile.input_encoding.ascii}" },
+      { 'name' => "iso_8859_1",
+    'desc' => "Latin1 (western languages)" } ];
+ 
 my $arguments = 
     [ { 'name' => "process_exp",
-    'desc' => "{BasPlug.process_exp}",
+    'desc' => "{BasePlugin.process_exp}",
     'type' => "regexp",
     'reqd' => "no",
-    'deft' => &get_default_process_exp() }
+    'deft' => &get_default_process_exp() },
+      { 'name' => "input_encoding",
+    'desc' => "{ReadTextFile.input_encoding}",
+    'type' => "enum",
+    'deft' => "iso_8859_1",
+    'list' => $encoding_list,
+    'reqd' => "no" }
       ];
 
-my $options = { 'name'     => "HBPlug",
-        'desc'     => "{HBPlug.desc}",
+my $options = { 'name'     => "HBPlugin",
+        'desc'     => "{HBPlugin.desc}",
         'abstract' => "no",
         'inherits' => "yes",
@@ -72 +83 @@
     push(@$pluginlist, $class);
 
-    if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
-    if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
-
-    my $self = new BasPlug($pluginlist, $inputargs, $hashArgOptLists);
+    push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
+    push(@{$hashArgOptLists->{"OptList"}},$options);
+
+    my $self = new BasePlugin($pluginlist, $inputargs, $hashArgOptLists);
 
     return bless $self, $class;
 }
 
-sub init {
-    my $self = shift (@_);
-    my ($verbosity, $outhandle) = @_;
-
-    $self->BasPlug::init($verbosity, $outhandle);
-    $self->{'input_encoding'} = "iso_8859_1";
-
-    # this plugin only handles ascii encodings
-    if ($self->{'input_encoding'} !~ /^(iso_8859_1|ascii)$/) {
-    die "ERROR: HBPlug can handle only iso_8859_1 or ascii encodings.\n" .
-        $self->{'input_encoding'} . " is not an acceptable input_encoding value\n";
-    } 
-}
-
 # this is included only to prevent warnings being printed out 
-# from BasPlug::init. The process_exp is not used by this plugin
+# from BasePlugin::init. The process_exp is not used by this plugin
 sub get_default_process_exp {
     my $self = shift (@_);
@@ -148 +145 @@
     if ($line =~ /<font [^>]*?face\s*=\s*\"?(\w+)\"?/i) {
         my $font = $1;
-        print $outhandle "HBPlug::HB_gettext - warning removed font $font\n" 
+        print $outhandle "HBPlugin::HB_gettext - warning removed font $font\n" 
         if ($font !~ /^arial$/i);
     }
@@ -217 +214 @@
 }
 
-# if input_encoding is ascii we can call add_utf8_metadata
-# directly but if it's iso_8859_1 (the default) we need to call
-# add_metadata so that the ascii2utf8 conversion is done first
-# this should speed things up a little if processing an ascii only
-# document with input_encoding set to ascii
-sub HB_add_metadata {
-    my $self = shift (@_);
-    my ($doc_obj, $cursection, $field, $value) = @_;
-
-# All text should now be in utf-8
-#    if ($self->{'input_encoding'} eq "ascii") {
-    $doc_obj->add_utf8_metadata ($cursection, $field, $value);
-#    } else {
-#   $doc_obj->add_metadata ($cursection, $field, $value);
-#    }
-}
-
 # return number of files processed, undef if can't process
 # Note that $base_dir might be "" and that $file might 
@@ -251 +231 @@
     return undef unless -e $htmlfile;
 
-    print STDERR "<Processing n='$file' p='HBPlug'>\n" if ($gli);
-    print $outhandle "HBPlug: processing $file\n";
+    print STDERR "<Processing n='$file' p='HBPlugin'>\n" if ($gli);
+    print $outhandle "HBPlugin: processing $file\n";
 
     # read in the file and do basic html cleaning (removing header etc)
@@ -276 +256 @@
     # $metadata->{$field} may be an array reference
     if (ref ($metadata->{$field}) eq "ARRAY") {
-        map { 
-        $self->HB_add_metadata ($doc_obj, $cursection, $field, $_); 
+        map {
+        $doc_obj->add_utf8_metadata($cursection, $field, $_);
         } @{$metadata->{$field}};
     } else {
-        $self->HB_add_metadata ($doc_obj, $cursection, $field, $metadata->{$field}); 
+        $doc_obj->add_utf8_metadata($cursection, $field, $metadata->{$field}); 
     }
     }
@@ -321 +301 @@
 
         # add the metadata to this section
-        $self->HB_add_metadata ($doc_obj, $cursection, "Title", $title);
+        $doc_obj->add_utf8_metadata($cursection, "Title", $title);
 
         # clean up the section html
@@ -332 +312 @@
 
         # add the text for this section
-# All read text should now be in utf-8
-#       if ($self->{'input_encoding'} eq "ascii") {
         $doc_obj->add_utf8_text ($cursection, $sectiontext);
-#       } else {
-#       $doc_obj->add_text ($cursection, $sectiontext);
-#       }
     } else {
         print $outhandle "WARNING - leftover text\n" , $self->shorten($html),