fprintf(stderr, " -i only show lines/consensus with indels\n");
fprintf(stderr, " -m INT filtering reads with bits in INT [%d]\n", d->mask);
fprintf(stderr, " -M INT cap mapping quality at INT [%d]\n", d->c->cap_mapQ);
- fprintf(stderr, " -t FILE list of reference sequences (assume the input is in SAM)\n");
+ fprintf(stderr, " -t FILE list of reference sequences (force -S)\n");
fprintf(stderr, " -l FILE list of sites at which pileup is output\n");
fprintf(stderr, " -f FILE reference sequence in the FASTA format\n\n");
fprintf(stderr, " -c output the maq consensus sequence\n");
}
if (d->fai == 0 && (d->format & (BAM_PLF_CNS|BAM_PLF_INDEL_ONLY)))
fprintf(stderr, "[bam_pileup] indels will not be called when -f is absent.\n");
+ if (fn_fa && is_SAM && fn_list == 0) fn_list = samfaipath(fn_fa);
+
{
samfile_t *fp;
fp = is_SAM? samopen(argv[optind], "r", fn_list) : samopen(argv[optind], "rb", 0);
#include <string.h>
+#include <unistd.h>
+#include "faidx.h"
#include "sam.h"
#define TYPE_BAM 1
bam_destroy1(b);
return 0;
}
+
+char *samfaipath(const char *fn_ref)
+{
+ char *fn_list = 0;
+ fn_list = calloc(strlen(fn_ref) + 5, 1);
+ strcat(strcpy(fn_list, fn_ref), ".fai");
+ if (access(fn_list, R_OK) == -1) { // fn_list is unreadable
+ if (access(fn_ref, R_OK) == -1) {
+ fprintf(stderr, "[samfaipath] fail to read file %s.\n", fn_ref);
+ } else {
+ fprintf(stderr, "[samfaipath] build FASTA index...\n");
+ if (fai_build(fn_ref) == -1) {
+ fprintf(stderr, "[samfaipath] fail to build FASTA index.\n");
+ free(fn_list); fn_list = 0;
+ }
+ }
+ }
+ return fn_list;
+}
@param aux auxiliary data; if mode[0]=='w', aux points to
bam_header_t; if strcmp(mode, "rb")!=0 and @SQ header lines in SAM
are absent, aux points the file name of the list of the reference;
- aux is not used otherwise.
+ aux is not used otherwise. If @SQ header lines are present in SAM,
+ aux is not used, either.
@return SAM/BAM file handler
*/
*/
int sampileup(samfile_t *fp, int mask, bam_pileup_f func, void *data);
+ char *samfaipath(const char *fn_ref);
+
#ifdef __cplusplus
}
#endif
#include <stdio.h>
#include <unistd.h>
#include "sam.h"
+#include "faidx.h"
static int g_min_mapQ = 0, g_flag_on = 0, g_flag_off = 0;
static char *g_library, *g_rg;
int c, is_header = 0, is_header_only = 0, is_bamin = 1, ret = 0, is_uncompressed = 0, is_bamout = 0;
int of_type = BAM_OFDEC, is_long_help = 0;
samfile_t *in = 0, *out = 0;
- char in_mode[5], out_mode[5], *fn_out = 0, *fn_list = 0;
+ char in_mode[5], out_mode[5], *fn_out = 0, *fn_list = 0, *fn_ref = 0;
/* parse command-line options */
strcpy(in_mode, "r"); strcpy(out_mode, "w");
- while ((c = getopt(argc, argv, "Sbt:hHo:q:f:F:ul:r:xX?")) >= 0) {
+ while ((c = getopt(argc, argv, "Sbt:hHo:q:f:F:ul:r:xX?T:")) >= 0) {
switch (c) {
case 'S': is_bamin = 0; break;
case 'b': is_bamout = 1; break;
case 'x': of_type = BAM_OFHEX; break;
case 'X': of_type = BAM_OFSTR; break;
case '?': is_long_help = 1; break;
+ case 'T': fn_ref = strdup(optarg); is_bamin = 0; break;
default: return usage(is_long_help);
}
}
if (is_uncompressed) strcat(out_mode, "u");
if (argc == optind) return usage(is_long_help);
+ // generate the fn_list if necessary
+ if (fn_list && fn_ref) {
+ fprintf(stderr, "[main_samview] both -t and -T are applied. -T is ignored.\n");
+ } else fn_list = samfaipath(fn_ref);
// open file handlers
if ((in = samopen(argv[optind], in_mode, fn_list)) == 0) {
fprintf(stderr, "[main_samview] fail to open file for reading.\n");
goto view_end;
}
+ if (in->header == 0) {
+ fprintf(stderr, "[main_samview] fail to read SAM header.\n");
+ goto view_end;
+ }
if ((out = samopen(fn_out? fn_out : "-", out_mode, in->header)) == 0) {
fprintf(stderr, "[main_samview] fail to open file for writing.\n");
goto view_end;
view_end:
// close files, free and return
- free(fn_list); free(fn_out); free(g_library); free(g_rg);
+ free(fn_list); free(fn_ref); free(fn_out); free(g_library); free(g_rg);
samclose(in);
samclose(out);
return ret;
fprintf(stderr, " -x output FLAG in HEX (samtools-C specific)\n");
fprintf(stderr, " -X output FLAG in stirng (samtools-C specific)\n");
fprintf(stderr, " -t FILE list of reference names and lengths (force -S) [null]\n");
+ fprintf(stderr, " -T FILE reference sequence file (force -S) [null]\n");
fprintf(stderr, " -o FILE output file name [stdout]\n");
fprintf(stderr, " -f INT required flag, 0 for unset [0]\n");
fprintf(stderr, " -F INT filtering flag, 0 for unset [0]\n");
corresponding sequence length. The `.fai' file generated by `faidx'\n\
can be used here. This file may be empty if reads are unaligned.\n\
\n\
- 2. SAM->BAM conversion: `samtools view -bt ref.fa.fai in.sam.gz'.\n\
+ 2. SAM->BAM conversion: `samtools view -bT ref.fa in.sam.gz'.\n\
\n\
3. BAM->SAM conversion: `samtools view in.bam'.\n\
\n\