used one more defer

guess it helps to prevent memory leak when the connection is interrupted?

Author: lh3

bgt-server.go

@@ -329,6 +329,7 @@ func bgs_query(w http.ResponseWriter, r *http.Request) {
 
 	// read through
 	b := C.bcf_init1();
+	defer C.bcf_destroy1(b);
 	n_read := 0;
 	for {
 		if n_read > max_read || uint64(bm.n_gt_read) > bgt_max_gt {
@@ -349,7 +350,6 @@ func bgs_query(w http.ResponseWriter, r *http.Request) {
 		}
 		n_read += 1;
 	}
-	C.bcf_destroy1(b);
 
 	// print hapcnt and/or sample list
 	if !vcf_out && int(bm.n_aal) > 0 {

tex/bgt.tex

@@ -203,7 +203,7 @@ \section{Results}
 The command line above takes 12 CPU seconds with most of time spent on reading
 through the variant annotation file to find matching alleles. The BGT server
 reads the entire file into memory to alleviate the overhead, but a better
-solution would be to set up a proper database for variant annotations.
+solution would be to use a proper database for variant annotations.
 
 We have also set up a public BGT server at http://bgtdemo.herokuapp.com. It
 hosts 1000 Genomes data for chromosome 11 and 20.