major restructure of the training code

1 files changed, 18 insertions, 0 deletions

diff --git a/training/pro/mr_pro_generate_mapper_input.pl b/training/pro/mr_pro_generate_mapper_input.pl
new file mode 100755
index 00000000..b30fc4fd
--- /dev/null
+++ b/training/pro/mr_pro_generate_mapper_input.pl
@@ -0,0 +1,18 @@
+#!/usr/bin/perl -w
+use strict;
+
+die "Usage: $0 HG_DIR\n" unless scalar @ARGV == 1;
+my $d = shift @ARGV;
+die "Can't find directory $d" unless -d $d;
+
+opendir(DIR, $d) or die "Can't read $d: $!";
+my @hgs = grep { /\.gz$/ } readdir(DIR);
+closedir DIR;
+
+for my $hg (@hgs) {
+  my $file = $hg;
+  my $id = $hg;
+  $id =~ s/(\.json)?\.gz//;
+  print "$d/$file $id\n";
+}
+