ignoring softclips for lengths

as reported in
#16

.gitignore

@@ -2,3 +2,4 @@
 .vscode
 test.feather
 Cargo.lock
+test.txt

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "cramino"
-version = "0.12.0"
+version = "0.13.0"
 edition = "2021"
 authors = ["Wouter De Coster [email protected]"]
 license = "MIT"

src/extract_from_bam.rs

@@ -58,7 +58,11 @@ pub fn extract(args: &crate::Cli) -> Data {
         .inspect(|_| all_counts += 1)
         .filter(|read| filter_closure(read))
     {
-        lengths.push(read.seq_len() as u64);
+        lengths.push(
+            read.seq_len() as u64
+                - read.cigar().leading_softclips() as u64
+                - read.cigar().trailing_softclips() as u64,
+        );
         if args.karyotype || args.phased {
             tids.push(read.tid());
         }

src/main.rs

@@ -154,7 +154,7 @@ fn generate_main_output(
         panic!();
     }
     let data_yield: u64 = lengths.iter().sum::<u64>();
-    println!("Number of reads\t{num_reads}");
+    println!("Number of alignments\t{num_reads}");
     println!(
         "% from total reads\t{:.2}",
         (num_reads as f64) / (all_reads as f64) * 100.0
@@ -213,7 +213,7 @@ fn extract() {
 #[test]
 fn extract_ubam() {
     let args = Cli {
-        input: "test-data/small-test-phased.bam".to_string(),
+        input: "test-data/small-test-ubam.bam".to_string(),
         threads: 8,
         reference: None,
         min_read_len: 0,