Changeset 3350 for trunk/gsdl/bin/script/gsConvert.pl

Timestamp:

2002-08-13T12:28:15+12:00 (22 years ago)

Author:

sjboddie

Message:

Added -use_strings option to ConvertToPlug. The default behaviour for
plugins derived from ConvertToPlug (WordPlug, PDFPlug etc) is now to
exclude documents that can't be converted correctly. They won't use the
perl strings stuff to extract text unless the -use_strings option is
specified.

File:

• trunk/gsdl/bin/script/gsConvert.pl (modified) (5 diffs)

trunk/gsdl/bin/script/gsConvert.pl

@@ -57 +57 @@
 my $is_winnt_2000=eval {require Win32; return (Win32::IsWinNT()); return 0;};
 if (!defined($is_winnt_2000)) {$is_winnt_2000=0;}
+
+my $use_strings;
 
 sub print_usage
@@ -68 +70 @@
     print STDERR "\t-output\thtml|text\n";
     print STDERR "\t-timeout\t<max cpu seconds>\t(ulimit on unix systems)\n";
+    print STDERR "\t-use_strings\t(use strnigs to extract text if conversion fails)\n";
     exit(1);
 }
@@ -85 +88 @@
              'output/(html|text)/', \$output_type,
              'timeout/\d+/0',\$timeout,
-             'verbose/\d+/0',   \$verbose))
+             'verbose/\d+/0',   \$verbose,
+             'use_strings', \$use_strings))
     {
     print_usage();
@@ -469 +473 @@
         close FAILLOG if ($write_to_fail_log);
     }
-    print STDERR "Continuing...\n";
     return 0; # we can try any_to_text
     }
@@ -954 +957 @@
     ($input_filename, $output_filestem) = @_;
 
+    if (!$use_strings) {
+      return 0;
+    }
+
     open(IN, "<$input_filename") || return 0;
     binmode(IN);