Changeset 38728

Timestamp:

2024-02-08T18:53:34+13:00 (3 months ago)

Author:

davidb

Message:

Updated to dynamically add in JPEG2000 jars, if present of the filesystem

File:

• gs2-extensions/pdf-box/trunk/java/perllib/plugins/PDFBoxConverter.pm (modified) (5 diffs)

gs2-extensions/pdf-box/trunk/java/perllib/plugins/PDFBoxConverter.pm

@@ -126 +126 @@
     my $gextpb_home = $ENV{'GEXT_PDFBOX'};
     my $pbajar = &FileUtils::filenameConcatenate($gextpb_home,"lib","java","pdfbox-app.jar");
-    my $pbjbigjar = &FileUtils::filenameConcatenate($gextpb_home,"lib","java","jbig2-imageio-3.0.1.jar");
+    my $pb_jbig2_jar = &FileUtils::filenameConcatenate($gextpb_home,"lib","java","jbig2-imageio-3.0.1.jar");
+
+    my $pb_jaicore_jar = &FileUtils::filenameConcatenate($gextpb_home,"lib","java","jai-imageio-core-1.4.0.jar");
+    my $pb_jaijpeg2000_jar = &FileUtils::filenameConcatenate($gextpb_home,"lib","java","jai-imageio-jpeg2000-1.4.0.jar");
+        
     # Not including the following JPEG2000 jar, as it is under commercial license:
     # https://github.com/jai-imageio/jai-imageio-jpeg2000 leading to https://bintray.com/jai-imageio/maven/jai-imageio-jpeg2000# (Files tab)
     # my $pbjp2jar = &FileUtils::filenameConcatenate($gextpb_home,"lib","java","jai-imageio-jpeg2000-1.3.0.jar"); # jpeg2000
+    
     my $java = &util::get_java_command();
     $self->{'pdfbox_txt_launch_cmd'} = "$java -cp \"$pbajar\" org.apache.pdfbox.tools.ExtractText";
@@ -141 +146 @@
     # put the pdfbox jar, the jbig2-imageio library (Apache Software License 2.0)
     # and our build folder containing our custom PDFBox class on the classpath
-    my $classpath = &util::pathname_cat($pbajar, $pbjbigjar, $pdfbox_build);
+    my $classpath = &util::pathname_cat($pbajar, $pb_jbig2_jar);
+
+    if(!&FileUtils::filenameExists($pb_jaicore_jar)) {
+        $classpath = &util::pathname_cat($classpath, $pb_jaicore_jar);
+    }
+    if(!&FileUtils::filenameExists($pb_jaijpeg2000_jar)) {
+        $classpath = &util::pathname_cat($classpath, $pb_jaijpeg2000_jar);
+    }
+    $classpath = &util::pathname_cat($classpath, $pdfbox_build);
+    
 #   $self->{'pdfbox_img_launch_cmd'} = "java -cp \"$classpath\" org.apache.pdfbox.tools.PDFToImage"; # pdfbox 2.09 cmd for converting each PDF page to an image (jpg, png)  
     $self->{'pdfbox_imgtxt_launch_cmd'} = "java -cp \"$classpath\" org.greenstone.pdfbox.PDFBoxToImagesAndText";
@@ -212 +226 @@
     # Determine the full name and path of the output file
     my $target_file_path;
+    
     if ($self->{'enable_cache'}) {
     $self->init_cache_for_file($source_file_full_path);
@@ -236 +251 @@
     # this is in gsdl/tmp. get a tmp filename in collection instead???
     $target_file_path = &util::get_tmp_filename($target_file_type);
-
+    
     # for image files, remove the suffix, since we can have many output image files
     # per input PDF (one img for each page of the PDF, for example)
     if($img_output_mode || $paged_txt_output_mode) {
         $target_file_path =~ s/\.[^.]*$//g;
+        
         if(!&FileUtils::directoryExists($target_file_path)) {       
         mkdir($target_file_path);
@@ -274 +290 @@
         $convert_cmd .= " -textOnly";       
     }
+    
     $convert_cmd .= " -outputPrefix \"$output_prefix\"";
     $convert_cmd .= " \"$source_file_full_path\"";