Changeset 25994 for main/trunk/greenstone2/perllib

Timestamp:

2012-07-19T18:16:46+12:00 (12 years ago)

Author:

ak19

Message:

Moving the updated create_itemfile() subroutine from pdfpstoimg.pl to util so that PDFBoxConverter can easily reuse this method too, as the PDFBoxConverter is currently being modified to convert a PDF to images when the -pagedimage_IMGTYPE flag is specified.

File:

• main/trunk/greenstone2/perllib/util.pm (modified) (1 diff)

main/trunk/greenstone2/perllib/util.pm

@@ -1849 +1849 @@
 }
 
+# Used by pdfpstoimg.pl and PDFBoxConverter to create a .item file from
+# a directory containing sequentially numbered images.
+sub create_itemfile
+{
+    my ($output_dir, $convert_basename, $convert_to) = @_;
+    opendir(DIR, $output_dir) || die "can't opendir $output_dir: $!";  
+
+    my $page_num = "";
+    my @dir_files = grep {-f "$output_dir/$_"} readdir(DIR);
+
+    # Sort files in the directory by page_num    
+    sub page_number {
+    my ($dir) = @_;
+    my ($pagenum) =($dir =~ m/^.*[-\.]?(\d+)(\.(jpg|gif|png))?$/i);
+
+    $pagenum = 1 unless defined $pagenum;
+    return $pagenum;
+    }
+
+    # sort the files in the directory in the order of page_num rather than lexically. 
+    @dir_files = sort { page_number($a) <=> page_number($b) } @dir_files;
+
+    # work out if the numbering of the now sorted image files starts at 0 or not
+    # by checking the number of the first _image_ file (skipping item files)
+    my $starts_at_0 = 0;
+    my $firstfile = ($dir_files[0] !~ /\.item$/i) ? $dir_files[0] : $dir_files[1];
+    if(page_number($firstfile) == 0) { # 00 will evaluate to 0 too in this condition
+    $starts_at_0 = 1;
+    }
+
+    my $item_file = &util::filename_cat($output_dir, $convert_basename.".item");
+    open(FILE,">$item_file");    
+    print FILE "<PagedDocument>\n";
+
+    foreach my $file (@dir_files){
+    if ($file !~ /\.item/i){
+        $page_num = page_number($file);
+        $page_num++ if $starts_at_0; # image numbers start at 0, so add 1
+        print FILE "   <Page pagenum=\"$page_num\" imgfile=\"$file\" txtfile=\"\"/>\n";
+    } 
+    }
+
+    print FILE "</PagedDocument>\n";
+    closedir DIR;
+    return $item_file;
+}
+
 1;