Changeset 22597 for main/trunk/greenstone2/perllib/plugins/WordPlugin.pm

Timestamp:

2010-08-10T14:31:53+12:00 (14 years ago)

Author:

kjdon

Message:

code tidy up. rearranged how convertbinaryfile plugins set up their secondary plugins - now only set up the options for the one they are using. all subclass specific code moved out of convertbinaryfile.new into the appropriate plugin file.

File:

• main/trunk/greenstone2/perllib/plugins/WordPlugin.pm (modified) (3 diffs)

main/trunk/greenstone2/perllib/plugins/WordPlugin.pm

@@ -41 +41 @@
     eval("require OpenOfficeConverter");
     if ($@) {
-    # Useful debugging statement if there is a syntax error in OpenOfficeConverter  
+    # Useful debugging statement if there is a syntax error in OpenOfficeConverter:
     #print STDERR "$@\n";
     @WordPlugin::ISA = ('ConvertBinaryFile');
@@ -174 +174 @@
     }
 
-    my $outhandle = $self->{'outhandle'};
     $self->{'filename_extension'} = "doc";
     $self->{'file_type'} = "Word";
+
+    my $outhandle = $self->{'outhandle'};
 
     if ($self->{'windows_scripting'}) {
@@ -193 +194 @@
     }
 
-    # we always save as utf-8
-#    if ($self->{'input_encoding'} eq "auto") {
-#   $self->{'input_encoding'} = "utf8";
-#    }
-
+    # check convert_to
+    if ($self->{'convert_to'} eq "auto") {
+    $self->{'convert_to'} = "html";
+    }
+    # windows or open office scripting, outputs structuredHTML
+    if (defined $self->{'office_scripting'}) {
+    $self->{'convert_to'} = "structuredhtml";
+    } 
+
+    # set convert_to_plugin and convert_to_ext
+    $self->ConvertBinaryFile::set_standard_convert_settings();
+ 
+    my $secondary_plugin_name = $self->{'convert_to_plugin'};
     my $secondary_plugin_options = $self->{'secondary_plugin_options'};
-    if (defined $self->{'office_scripting'}) { 
-    if (!defined $secondary_plugin_options->{'StructuredHTMLPlugin'}){
-        $secondary_plugin_options->{'StructuredHTMLPlugin'} = [];
-        my $structhtml_options = $secondary_plugin_options->{'StructuredHTMLPlugin'};
-        
-        # Instruct HTMLPlugin (when eventually accessed through read_into_doc_obj)
-        # to extract these metadata fields from the HEAD META fields
-        push (@$structhtml_options, "-metadata_fields","Title,GENERATOR,date,author<Creator>");
-        push (@$structhtml_options, "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
-        push (@$structhtml_options, "-description_tags") if $self->{'office_scripting'}; 
-        push (@$structhtml_options, "-extract_language") if $self->{'extract_language'};
-        push (@$structhtml_options, "-delete_toc") if $self->{'delete_toc'};
-        push (@$structhtml_options, "-toc_header", $self->{'toc_header'}) if $self->{'toc_header'};
-        push (@$structhtml_options, "-title_header", $self->{'title_header'}) if $self->{'title_header'};
-        push (@$structhtml_options, "-level1_header", $self->{'level1_header'}) if $self->{'level1_header'};
-        push (@$structhtml_options, "-level2_header", $self->{'level2_header'})if $self->{'level2_header'};
-        push (@$structhtml_options, "-level3_header", $self->{'level3_header'}) if $self->{'level3_header'};
-        push (@$structhtml_options, "-metadata_fields", $self->{'metadata_fields'}) if $self->{'metadata_fields'};
-        push (@$structhtml_options, "-metadata_field_separator", $self->{'metadata_field_separator'}) if $self->{'metadata_field_separator'};
-    }
-    }
-    if (!defined $secondary_plugin_options->{'HTMLPlugin'}) {
-    $secondary_plugin_options->{'HTMLPlugin'} = [];
-    }
-    if (!defined $secondary_plugin_options->{'TextPlugin'}) {
-    $secondary_plugin_options->{'TextPlugin'} = [];
-    }
-
-    my $html_options = $secondary_plugin_options->{'HTMLPlugin'};
-    my $text_options = $secondary_plugin_options->{'TextPlugin'};
-    my $structhtml_options = $secondary_plugin_options->{'StructuredHTMLPlugin'};    
-    # tell the secondary plugins that they are processing tmp files
-    push(@$html_options, "-processing_tmp_files");
-    push(@$structhtml_options, "-processing_tmp_files");
-    
-    # wvWare will always produce html files encoded as utf-8, so make sure the secondary HTMLPlugin knows this
-    push(@$html_options,"-input_encoding", "utf8");
-    push(@$html_options,"-extract_language") if $self->{'extract_language'};
-    push(@$html_options, "-description_tags") if $self->{'description_tags'};
-
-    # Instruct HTMLPlugin (when eventually accessed through read_into_doc_obj)
-    # to extract these metadata fields from the HEAD META fields
-    push(@$html_options,"-metadata_fields","Title,GENERATOR,date,author<Creator>");
-    push(@$html_options , "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
+
+    if (!defined $secondary_plugin_options->{$secondary_plugin_name}) {
+    $secondary_plugin_options->{$secondary_plugin_name} = [];
+    }
+    my $specific_options = $secondary_plugin_options->{$secondary_plugin_name};
+
+    # following title_sub removes "Page 1" and a leading
+    # "1", which is often the page number at the top of the page. Bad Luck
+    # if your document title actually starts with "1 " - is there a better way?
+    push(@$specific_options , "-title_sub", '^(Page\s+\d+)?(\s*1\s+)?');
 
     my $associate_tail_re = $self->{'associate_tail_re'};
     if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
-    push(@$html_options, "-associate_tail_re", $associate_tail_re);
-    push(@$text_options, "-associate_tail_re", $associate_tail_re);
-    push(@$structhtml_options, "-associate_tail_re", $associate_tail_re) if defined $structhtml_options;
-    }
-
-    push(@$html_options, "-file_rename_method", "none");
-    push(@$text_options, "-file_rename_method", "none");
-    push(@$structhtml_options, "-file_rename_method", "none") if defined $structhtml_options;
+    push(@$specific_options, "-associate_tail_re", $associate_tail_re);
+    }
+    push(@$specific_options, "-file_rename_method", "none");
+
+    if ($secondary_plugin_name eq "StructuredHTMLPlugin") {
+    # Instruct HTMLPlugin (when eventually accessed through read_into_doc_obj)
+    # to extract these metadata fields from the HEAD META fields
+    push (@$specific_options, "-metadata_fields","Title,GENERATOR,date,author<Creator>");
+    push (@$specific_options, "-description_tags") if $self->{'office_scripting'}; 
+    push (@$specific_options, "-extract_language") if $self->{'extract_language'};
+    push (@$specific_options, "-delete_toc") if $self->{'delete_toc'};
+    push (@$specific_options, "-toc_header", $self->{'toc_header'}) if $self->{'toc_header'};
+    push (@$specific_options, "-title_header", $self->{'title_header'}) if $self->{'title_header'};
+    push (@$specific_options, "-level1_header", $self->{'level1_header'}) if $self->{'level1_header'};
+    push (@$specific_options, "-level2_header", $self->{'level2_header'})if $self->{'level2_header'};
+    push (@$specific_options, "-level3_header", $self->{'level3_header'}) if $self->{'level3_header'};
+    push (@$specific_options, "-metadata_fields", $self->{'metadata_fields'}) if $self->{'metadata_fields'};
+    push (@$specific_options, "-metadata_field_separator", $self->{'metadata_field_separator'}) if $self->{'metadata_field_separator'};
+    push(@$specific_options, "-processing_tmp_files");
+    
+    }
+    
+    elsif ($secondary_plugin_name eq "HTMLPlugin") {
+    push(@$specific_options, "-processing_tmp_files");
+    push(@$specific_options,"-input_encoding", "utf8");
+    push(@$specific_options,"-extract_language") if $self->{'extract_language'};
+    push(@$specific_options, "-description_tags") if $self->{'description_tags'};
+    # Instruct HTMLPlugin (when eventually accessed through read_into_doc_obj)
+    # to extract these metadata fields from the HEAD META fields
+    push(@$specific_options,"-metadata_fields","Title,GENERATOR,date,author<Creator>");
+    }
 
     $self = bless $self, $class;