Changeset 37051 for main

Timestamp:

2022-12-23T10:37:02+13:00 (16 months ago)

Author:

davidb

Message:

A variety of changes: added in call to post_process_doc_obj() which is an existing base-class method that should have really been there (for PagedImagePlugin not particulary relevant, but needed in the GoogleVisionPagedImagePlugin inherited version for Open Annotation JSON files to be produced; code also changed over the the add_dummy_text_if_empty(); some refactoring of methods to use supporting subroutines; Adjustment of print statements that say 'processing by <plugin>' to use reflect to determine the class name, allowing for Inherited Plugins names to be used

File:

• main/trunk/greenstone2/perllib/plugins/PagedImagePlugin.pm (modified) (9 diffs)

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

@@ -324 +324 @@
     my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
 
-    print $outhandle "PagedImagePlugin processing \"$filename_full_path\"\n"
+    my $toplevel_plugin_classname = ref($self);
+    print $outhandle "$toplevel_plugin_classname processing \"$filename_full_path\"\n"
     if $verbosity > 1;
-    print STDERR "<Processing n='$file' p='PagedImagePlugin'>\n" if ($gli);
+    print STDERR "<Processing n='$file' p='$toplevel_plugin_classname'>\n" if ($gli);
     
     $self->{'MaxImageWidth'} = 0;
@@ -366 +367 @@
 
     $self->add_OID($doc_obj);
+
+    $self->post_process_doc_obj($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli);
     return (1,$doc_obj);
 }
@@ -509 +512 @@
         $self->process_text ($self->{'xml_file_dir'}.$txtfile, $txtfile, $doc_obj, $self->{'current_section'});
     } else {
-        $self->add_dummy_text($doc_obj, $self->{'current_section'});
+        # A plugin inheriting from this might be able to derive text from the image
+        # (e.g., through GoogleVisionAPI), and so don't just assume there is no
+        # text for the image -- check its text length, and only set the dummy
+        # text if it is zero
+        $self->add_dummy_text_if_empty($doc_obj, $self->{'current_section'});
     }
     } elsif ($element eq "Metadata") {
@@ -564 +571 @@
     
     # create a new document
-    $self->{'doc_obj'} = new doc ($self->{'filename'}, "indexed_doc", $self->{'file_rename_method'});
-    # TODO is file filenmae_no_path??
-    $self->set_initial_doc_fields($self->{'doc_obj'}, $self->{'filename'}, $self->{'processor'}, $self->{'metadata'});
-
+    #$self->{'doc_obj'} = new doc ($self->{'filename'}, "indexed_doc", $self->{'file_rename_method'});
+    ## TODO is file filename_no_path??
+    #$self->set_initial_doc_fields($self->{'doc_obj'}, $self->{'filename'}, $self->{'processor'}, $self->{'metadata'});
+
+    # create a new document
+    my $doc_obj = $self->init_new_doc_item($self->{'filename'}, $self->{'processor'}, $self->{'metadata'});
+    
     my ($dir, $file) = $self->{'filename'} =~ /^(.*?)([^\/\\]*)$/;
     $self->{'xml_file_dir'} = $dir;
@@ -614 +624 @@
     $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
     $self->{'MaxImageWidth'} = undef;
-    $self->{'MaxImageHeight'} = undef;
-    
+    $self->{'MaxImageHeight'} = undef;    
 }
 
@@ -687 +696 @@
 }
 
-sub process_item {
-    my $self = shift (@_);
-    my ($filename_full_path, $dir, $filename_no_path, $processor, $metadata) = @_;
-
-    my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
+sub init_new_doc_item
+{    
+    my $self = shift (@_);
+    my ($filename_full_path, $processor, $metadata) = @_;
+
+    my $doc_obj = new doc($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
     $self->set_initial_doc_fields($doc_obj, $filename_full_path, $processor, $metadata);
+
+    return $doc_obj;
+}
+
+sub read_and_process_itemtxt
+{
+    my $self = shift (@_);
+    my ($filename_full_path, $dir, $filename_no_path, $processor, $metadata, $doc_obj) = @_;
+
     my $topsection = $doc_obj->get_top_section();
-    # simple item files are always paged unless user specified
-    if ($self->{'documenttype'} eq "auto") {
-    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "paged");
-    } else {
-    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", $self->{'documenttype'});
-    }
+    
     open (ITEMFILE, "<:encoding(UTF-8)", $filename_full_path) || die "couldn't open $filename_full_path\n";
     my $line = "";
@@ -742 +756 @@
         if (!defined $result2) {
             print "PagedImagePlugin: couldn't process text file \"$dir.$txtname\" for item \"$filename_full_path\"\n";
-            $self->add_dummy_text($doc_obj, $cursection);
+            $self->add_dummy_text_if_empty($doc_obj, $cursection);
         }
         } else {
         # otherwise add in some dummy text 
-        $self->add_dummy_text($doc_obj, $cursection);
+        $self->add_dummy_text_if_empty($doc_obj, $cursection);
         }
     }
@@ -753 +767 @@
     close ITEMFILE;
 
+    return $num;
+}
+
+    
+sub process_item {
+    my $self = shift (@_);
+    my ($filename_full_path, $dir, $filename_no_path, $processor, $metadata) = @_;
+
+    # create a new document
+    #my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
+    #$self->set_initial_doc_fields($doc_obj, $filename_full_path, $processor, $metadata);
+
+    # create a new document
+    my $doc_obj = $self->init_new_doc_item($filename_full_path, $processor, $metadata);
+        
+    my $num_pages = $self->read_and_process_itemtxt($filename_full_path, $dir, $filename_no_path, $processor, $metadata, $doc_obj);
+    
+    my $topsection = $doc_obj->get_top_section();
+    
+    # simple item files are always paged unless user specified
+    if ($self->{'documenttype'} eq "auto") {
+    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "paged");
+    } else {
+    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", $self->{'documenttype'});
+    }
+    
     # add numpages metadata
-    $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', "$num");
+    $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', "$num_pages");
 
     $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
@@ -760 +800 @@
     $self->{'MaxImageWidth'} = undef;
     $self->{'MaxImageHeight'} = undef;
-
 
     return $doc_obj;