add filter-by-source-word-count.perl

1 files changed, 27 insertions, 0 deletions

diff --git a/contrib/other-builds/extract-mixed-syntax/filter-by-source-word-count.perl b/contrib/other-builds/extract-mixed-syntax/filter-by-source-word-count.perl
new file mode 100755
index 000000000..d0e482a02
--- /dev/null
+++ b/contrib/other-builds/extract-mixed-syntax/filter-by-source-word-count.perl
@@ -0,0 +1,27 @@
+#!/usr/bin/perl
+
+use strict;
+
+binmode(STDIN, ":utf8");
+binmode(STDOUT, ":utf8");
+binmode(STDERR, ":utf8");
+
+my $maxNumWords = $ARGV[0];
+
+while (my $line = <STDIN>) {
+  chomp($line);
+  my @toks = split(/ /,$line);
+  
+  my $numSourceWords = 0;
+  my $tok = $toks[$numSourceWords];
+  while ($tok ne "|||") {
+    ++$numSourceWords;
+    $tok = $toks[$numSourceWords];
+  }
+
+  if ($numSourceWords <= $maxNumWords) {
+    print "$line\n";
+  }
+}
+
+