Changeset 10276

Timestamp:

2005-07-25T13:56:23+12:00 (19 years ago)

Author:

chi

Message:

Add a read_into_doc_obj() for enabling secondary_pluging function. Also, a numer of minor modifications
to identify undefined variable or text file contains no text.

File:

• trunk/gsdl/perllib/plugins/PagedImgPlug.pm (modified) (13 diffs)

trunk/gsdl/perllib/plugins/PagedImgPlug.pm

@@ -228 +228 @@
     if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
     if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
-
+    
     my $self = (defined $hashArgOptLists)? new XMLPlug($pluginlist,$inputargs,$hashArgOptLists): new XMLPlug($pluginlist,$inputargs);
 
@@ -256 +256 @@
     my $section = shift (@_); #the current section
     my $rotate = shift (@_); # whether to rotate the image or not
-
+    $rotate = 0 unless defined $rotate;
+    
     my $top=0;
     if ($section eq $doc_obj->get_top_section()) {
@@ -279 +280 @@
     my $converted = 0;
     my $rotated=0;
+
     if ($converttotype ne "" && $filename !~ /$converttotype$/) {
     $converted=1;
@@ -528 +530 @@
 # include directories
 
-sub read {
+sub read_into_doc_obj {
     my $self = shift (@_);
     my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
     my $outhandle   = $self->{'outhandle'};
     my $smart_block = $self->{'smart_block'};
-
+    
     my $filename = &util::filename_cat($base_dir, $file);
-
+    
     if ($self->associate_with($file,$filename,$metadata)) {
     # a form of smart block
@@ -541 +543 @@
     return 0; # blocked
     }
-
+    
     if ($smart_block) {
     if (defined $self->{'file_blocks'}->{$filename} && $self->{'file_blocks'}->{$filename} == 1){
@@ -557 +559 @@
     
     print $outhandle "PagedImgPlug processing \"$filename\"\n"
-        if $self->{'verbosity'} > 1;
+    if $self->{'verbosity'} > 1;
     print STDERR "<Processing n='$file' p='PagedImgPlug'>\n" if ($gli);
-
+    
     # here we need to decide if we have an old text .item file, or a new xml 
     # .item file - for now the test is if the first non-empty line is 
@@ -572 +574 @@
     }
     chomp $line;
-    if ($line =~ /^<PagedDocument/) {
+    if ($line =~ /<PagedDocument/) {
     $xml_version = 1;
     }
     close ITEMFILE;
+    #print STDERR "xml version = $xml_version\n";
     my $doc_obj;
     if ($xml_version) {
-
     $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
     $self->{'file'} = $file;
@@ -644 +646 @@
     my $section = $doc_obj->get_top_section();
     $self->extra_metadata ($doc_obj, $section, $metadata);
-
-    # do plugin specific processing of doc_obj - don't need this unless
-    # something inherits from PagedImgPlug
+    #my $text="";
+    # do plugin specific processing of doc_obj
     #unless (defined ($self->process(\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj))) {
-#   print STDERR "<ProcessingError n='$file'>\n" if ($gli);
-#   return -1;
-#    }
-
+    #print STDERR "<ProcessingError n='$file'>\n" if ($gli);
+    #return -1;
+    #}
     # do any automatic metadata extraction
     $self->auto_extract_metadata ($doc_obj);
 
-    # process the document
-    $processor->process($doc_obj);
-
+    $self->{'num_processed'}++;
+    return (1,$doc_obj);
+}
+
+sub read
+{
+    my $self = shift (@_);
+    my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
+
+    my ($process_status,$doc_obj) = $self->read_into_doc_obj(@_);
+
+    if ((defined $process_status) && ($process_status == 1)) {
+    # process the document
+    $processor->process($doc_obj);
+
+    #if(defined($self->{'places_filename'})){
+    #    &util::rm($self->{'places_filename'});
+    #    $self->{'places_filename'} = undef;
+    #}
+    #$self->{'num_processed'} ++;
+    undef $doc_obj; 
+    }
+    
     # clean up temporary files - we do this here instead of in  
     # process_image becuase associated files aren't actually copied 
@@ -663 +683 @@
     if (defined $self->{'tmp_filename1'} && 
     -e $self->{'tmp_filename1'}) {
-      &util::rm($self->{'tmp_filename1'})
+    &util::rm($self->{'tmp_filename1'})
     }
     if (defined $self->{'tmp_filename2'} && 
     -e $self->{'tmp_filename2'}) {
-      &util::rm($self->{'tmp_filename2'})
+    &util::rm($self->{'tmp_filename2'})
     }
     if (defined $self->{'tmp_filename3'} && 
@@ -673 +693 @@
       &util::rm($self->{'tmp_filename3'})
     }
-
-    $self->{'num_processed'}++;
-
-    return 1;
+    # if process_status == 1, then the file has been processed.
+    return $process_status;
 }
 
@@ -699 +717 @@
     }
     my ($txtfile) = $_{'txtfile'};
-    if (defined($txtfile)) {
+    if (defined($txtfile)&& $txtfile ne "") {
         $self->process_text ($self->{'base_dir'}.$txtfile, $txtfile, $doc_obj, $self->{'current_section'});
     } else {
@@ -793 +811 @@
     # set the gsdlthistype metadata to Paged - this ensures this document will
     # be treated as a Paged doc, even if Titles are not numeric
-    
     $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Paged");
     } else {
@@ -862 +879 @@
     # Do encoding stuff
     my ($language, $encoding) = $self->textcat_get_language_encoding ($fullpath);
- 
+
     my $text="";
     &BasPlug::read_file($self, $fullpath, $encoding, $language, \$text);