Add new option --force-single to support single-file edge case

This also
- adds a check for cases where the --file-list is empty
- adds a synonynomous option --force-no-index to --no-index

Resolves #2100

Author: pd3

NEWS

@@ -12,6 +12,10 @@ Changes affecting specific commands:
 
     - Add new option `-l, --file-list` to read the list of file names from a file
 
+* bcftools merge
+
+    - Add new option `--force-single` to support single-file edge case (#2100)
+
 * bcftools norm
 
     - Change the order of atomization and multiallelic splitting (when both -a,-m are given)

doc/bcftools.txt

@@ -1970,13 +1970,18 @@ also given explicitly using the *--print-header* and *--use-header* options.
 Note that only records from different files can be merged, never from the same file.
 For "vertical" merge take a look at *<<concat,bcftools concat>>* or *<<norm,bcftools norm>> -m* instead.
 
+*--force-no-index*::
+    synonymous to *--no-index*
 
 *--force-samples*::
     if the merged files contain duplicate samples names, proceed anyway.
     Duplicate sample names will be resolved by prepending the index of the file
     as it appeared on the command line to the conflicting sample name (see
     '2:S3' in the above example).
 
+*--force-single*::
+    run even if only one file is given on input
+
 *--print-header*::
     print only merged header and exit
 

test/merge.LPL.0.out

@@ -0,0 +1,15 @@
+##fileformat=VCFv4.3
+##FILTER=<ID=PASS,Description="All filters passed">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=PL,Number=G,Type=Integer,Description="Genotype Likelihoods">
+##FORMAT=<ID=GL,Number=G,Type=Float,Description="Genotype Likelihoods">
+##FORMAT=<ID=AD,Number=R,Type=Integer,Description="Allelic Depths">
+##FORMAT=<ID=DF,Number=R,Type=Float,Description="Dummy">
+##FORMAT=<ID=DD,Number=A,Type=Integer,Description="Dummy">
+##contig=<ID=1,assembly=b37,length=249250621>
+##reference=ref.fa
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	A
+1	30000	.	C	T	.	.	.	GT:PL:GL:AD:DF:DD	0/1:20,0,20:20,0,20:1,2:0.1,0.2:1
+1	30001	.	C	CA	.	.	.	GT:PL:GL:AD:DF:DD	0/1:10,0,10:10,0,10:1,2:0.1,0.2:1
+1	30002	.	C	CA,CAA	.	.	.	GT:PL:GL:AD:DF:DD	1/2:20,20,20,10,0,10:20,20,20,10,0,10:0,1,2:0,0.1,0.2:1,2
+1	30003	.	C	CA,CAA,CAAA	.	.	.	GT:PL:GL:AD:DF:DD	1/2:20,20,20,10,0,10,20,20,10,20:20,20,20,10,0,10,20,20,10,20:0,1,2,1:0,0.1,0.2,0.1:1,2,3

test/test.pl

@@ -60,6 +60,7 @@
 run_test(\&test_vcf_isec,$opts,in=>['isec-miss.2.1','isec-miss.2.2','isec-miss.2.3'],out=>'isec-miss.2.1.out',args=>'-n +1 -r 20:100,20:140,12:55,20:140,20:100');
 run_test(\&test_vcf_isec,$opts,in=>['isec-miss.2.1','isec-miss.2.2','isec-miss.2.3'],out=>'isec-miss.2.1.out',args=>'-R {PATH}/isec-miss.1.regs.txt -n +1');
 run_test(\&test_vcf_merge,$opts,in=>['merge.join.a','merge.join.b'],out=>'merge.join.1.out',args=>'-i AF:join');
+run_test(\&test_vcf_merge,$opts,in=>['merge.LPL.a'],out=>'merge.LPL.0.out',args=>'--force-single');
 run_test(\&test_vcf_merge,$opts,in=>['merge.LPL.a','merge.LPL.b','merge.LPL.c'],out=>'merge.LPL.1.out',args=>'--force-samples');
 run_test(\&test_vcf_merge,$opts,in=>['merge.LPL.a','merge.LPL.b','merge.LPL.c'],out=>'merge.LPL.2.out',args=>'--force-samples -L 1');
 run_test(\&test_vcf_merge,$opts,in=>['merge.LPL.a','merge.LPL.b','merge.LPL.c'],out=>'merge.LPL.3.out',args=>'--force-samples -L 2');

vcfmerge.c

@@ -1,6 +1,6 @@
 /*  vcfmerge.c -- Merge multiple VCF/BCF files to create one multi-sample file.
 
-    Copyright (C) 2012-2023 Genome Research Ltd.
+    Copyright (C) 2012-2024 Genome Research Ltd.
 
     Author: Petr Danecek <[email protected]>
 
@@ -174,7 +174,7 @@ typedef struct
     maux_t *maux;
     regidx_t *regs;    // apply regions only after the blocks are expanded
     regitr_t *regs_itr;
-    int header_only, collapse, output_type, force_samples, merge_by_id, do_gvcf, filter_logic, missing_to_ref, no_index;
+    int header_only, collapse, output_type, force_samples, force_single, merge_by_id, do_gvcf, filter_logic, missing_to_ref, no_index;
     char *header_fname, *output_fname, *regions_list, *info_rules, *file_list;
     faidx_t *gvcf_fai;
     info_rule_t *rules;
@@ -3465,7 +3465,9 @@ static void usage(void)
     fprintf(stderr, "Usage:   bcftools merge [options] <A.vcf.gz> <B.vcf.gz> [...]\n");
     fprintf(stderr, "\n");
     fprintf(stderr, "Options:\n");
+    fprintf(stderr, "        --force-no-index              Merge unindexed files, synonymous to --no-index\n");
     fprintf(stderr, "        --force-samples               Resolve duplicate sample names\n");
+    fprintf(stderr, "        --force-single                Run even if there is only one file on input\n");
     fprintf(stderr, "        --print-header                Print only the merged header and exit\n");
     fprintf(stderr, "        --use-header FILE             Use the provided header\n");
     fprintf(stderr, "    -0  --missing-to-ref              Assume genotypes at missing sites are 0/0\n");
@@ -3527,6 +3529,8 @@ int main_vcfmerge(int argc, char *argv[])
         {"missing-rules",required_argument,NULL,'M'},
         {"no-version",no_argument,NULL,8},
         {"no-index",no_argument,NULL,10},
+        {"force-no-index",no_argument,NULL,10},
+        {"force-single",no_argument,NULL,12},
         {"filter-logic",required_argument,NULL,'F'},
         {"write-index",no_argument,NULL,11},
         {NULL,0,NULL,0}
@@ -3608,14 +3612,13 @@ int main_vcfmerge(int argc, char *argv[])
             case  8 : args->record_cmd_line = 0; break;
             case 10 : args->no_index = 1; break;
             case 11 : args->write_index = 1; break;
+            case 12 : args->force_single = 1; break;
             case 'h':
             case '?': usage(); break;
             default: error("Unknown argument: %s\n", optarg);
         }
     }
     if ( argc==optind && !args->file_list ) usage();
-    if ( argc-optind<2 && !args->file_list ) usage();
-
     if ( args->no_index )
     {
         if ( args->regions_list ) error("Error: cannot combine --no-index with -r/-R\n");
@@ -3656,6 +3659,9 @@ int main_vcfmerge(int argc, char *argv[])
         for (i=0; i<nfiles; i++) free(files[i]);
         free(files);
     }
+    if ( !args->files->nreaders ) usage();
+    if ( args->files->nreaders==1 && !args->force_single ) error("Expected two or more files to merge, got only one. Use --force-single to proceed anyway\n");
+
     merge_vcf(args);
     bcf_sr_destroy(args->files);
     if ( args->regs ) regidx_destroy(args->regs);