From 8be5ca74606889751ddc9784e78430bb09ffbfe5 Mon Sep 17 00:00:00 2001
From: Rob Davies <rmd+git@sanger.ac.uk>
Date: Tue, 5 Aug 2025 16:50:41 +0100
Subject: [PATCH] Ignore first phasing bit in process_gt_to_hap /
 process_gt_to_hap2

HTSlib version 1.22 introduced prefixed phasing support for VCF
files with version 4.4 or later.  Prior to this the first GT phase
(when reading VCF) was always zero; after it is set either
explicitly or when all other alleles are phased.  This updates
process_gt_to_hap() and process_gt_to_hap2() to ignore the first
phase bit, removing an assumption it is always zero.

Fixes incorrect reporting of phase when using HTSlib 1.22 to read
VCF files with version 4.4 or 4.5.
---
 convert.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/convert.c b/convert.c
index 9faaded6e..6a36bc708 100644
--- a/convert.c
+++ b/convert.c
@@ -878,7 +878,7 @@ static void process_gt_to_hap(convert_t *convert, bcf1_t *line, fmt_t *fmt, int
         ptr += fmt_gt->n;
         if ( fmt_gt->n==1 ) // haploid genotypes
         {
-            if ( ptr[0]==2 ) /* 0 */
+            if ( ( ptr[0] >> 1 )==1 ) /* 0 */
             {
                 str->s[str->l++] = '0'; str->s[str->l++] = ' '; str->s[str->l++] = '-'; str->s[str->l++] = ' ';
             }
@@ -886,7 +886,7 @@ static void process_gt_to_hap(convert_t *convert, bcf1_t *line, fmt_t *fmt, int
             {
                 str->s[str->l++] = '?'; str->s[str->l++] = ' '; str->s[str->l++] = '?'; str->s[str->l++] = ' ';
             }
-            else if ( ptr[0]==4 ) /* 1 */
+            else if ( ( ptr[0] >> 1 )==2 ) /* 1 */
             {
                 str->s[str->l++] = '1'; str->s[str->l++] = ' '; str->s[str->l++] = '-'; str->s[str->l++] = ' ';
             }
@@ -895,7 +895,7 @@ static void process_gt_to_hap(convert_t *convert, bcf1_t *line, fmt_t *fmt, int
                 kputw(bcf_gt_allele(ptr[0]),str); str->s[str->l++] = ' '; str->s[str->l++] = '-'; str->s[str->l++] = ' ';
             }
         }
-        else if ( ptr[0]==2 )
+        else if ( ( ptr[0] >> 1 )==1 )
         {
             if ( ptr[1]==3 ) /* 0|0 */
             {
@@ -934,7 +934,7 @@ static void process_gt_to_hap(convert_t *convert, bcf1_t *line, fmt_t *fmt, int
                 str->s[str->l++] = '*'; str->s[str->l++] = ' ';
             }
         }
-        else if ( ptr[0]==4 )
+        else if ( ( ptr[0] >> 1 )==2 )
         {
             if ( ptr[1]==3 ) /* 1|0 */
             {
@@ -1028,7 +1028,7 @@ static void process_gt_to_hap2(convert_t *convert, bcf1_t *line, fmt_t *fmt, int
     for (i=0; i<convert->nsamples; i++)
     {
         ptr += fmt_gt->n;
-        if ( ptr[0]==2 )
+        if ( ( ptr[0] >> 1 )==1 )
         {
             if ( ptr[1]==3 ) /* 0|0 */
             {
@@ -1067,7 +1067,7 @@ static void process_gt_to_hap2(convert_t *convert, bcf1_t *line, fmt_t *fmt, int
                 str->s[str->l++] = '*'; str->s[str->l++] = ' ';
             }
         }
-        else if ( ptr[0]==4 )
+        else if ( ( ptr[0] >> 1 )==2 )
         {
             if ( ptr[1]==3 ) /* 1|0 */
             {