added script to remove identical matches from LAST alignment

gringer · May 1, 2016 · 195d11a · 195d11a
1 parent 3eb4d6a
commit 195d11a
Showing 1 changed file with 31 additions and 0 deletions.
diff --git a/last-rmident.pl b/last-rmident.pl
@@ -0,0 +1,31 @@
+#!/usr/bin/perl
+
+my @lineBuffer = ();
+my @ids = ();
+
+while(<>){
+  my $line = $_;
+  if(/^([a-z])\s(.*?)\s/){
+    my ($flag, $id) = ($1, $2);
+    if($flag eq "s"){
+        push(@ids, $id);
+    } elsif($flag eq "a"){ # new alignment
+      if((scalar(@ids) != 2) || ($ids[0] ne $ids[1])){
+        print(join("",@lineBuffer));
+      }
+      # if(@ids){
+      #   printf("a '%s' '%s' %d %d\n", $ids[0], $ids[1],
+      #          scalar(@lineBuffer), scalar(@ids));
+      # } else {
+      #   print("a\n");
+      # }
+      @lineBuffer = ();
+      @ids = ();
+    }
+  }
+  push(@lineBuffer, $line);
+}
+
+if((scalar(@ids) != 2) || ($ids[0] ne $ids[1])){
+  print(join("",@lineBuffer));
+}