#include <string.h>
#include <assert.h>
+#include <unistd.h>
#include "kstring.h"
#include "bam.h"
uint8_t *seq = bam1_seq(b);
// b->core.l_qseq gives length of the SEQ entry (including soft clips, S)
// We need the padded length after alignment from the CIGAR (excluding
- // soft clips S, but including pads)
+ // soft clips S, but including pads from CIGAR D operations)
length = 0;
for (k = 0; k < b->core.n_cigar; ++k) {
int op, ol;
op= bam_cigar_op(cigar[k]);
ol = bam_cigar_oplen(cigar[k]);
- if (op == BAM_CMATCH || op == BAM_CEQUAL || op == BAM_CDIFF || op == BAM_CDEL || op == BAM_CPAD)
+ if (op == BAM_CMATCH || op == BAM_CEQUAL || op == BAM_CDIFF || op == BAM_CDEL)
length += ol;
}
ks_resize(s, length);
j += ol;
} else if (op == BAM_CHARD_CLIP) {
/* do nothing */
- } else if (op == BAM_CDEL || op == BAM_CPAD) {
+ } else if (op == BAM_CDEL) {
for (i = 0; i < ol; ++i) s->s[s->l++] = 0;
} else {
fprintf(stderr, "[depad] ERROR: Didn't expect CIGAR op %c in read %s\n", BAM_CIGAR_STR[op], bam1_qname(b));
return 0;
}
+static int usage(int is_long_help);
+
int main_pad2unpad(int argc, char *argv[])
{
bamFile in, out;
+ int c, is_long_help = 0;
int result=0;
- if (argc == 1) {
- fprintf(stderr, "Usage: samtools depad <in.bam>\n");
- return 1;
- }
+
+ /* parse command-line options */
+ while ((c = getopt(argc, argv, "?")) >= 0) {
+ switch (c) {
+ case '?': is_long_help = 1; break;
+ default: return usage(is_long_help);
+ }
+ }
+ if (argc == optind) return usage(is_long_help);
+
in = strcmp(argv[1], "-")? bam_open(argv[1], "r") : bam_dopen(fileno(stdin), "r");
out = bam_dopen(fileno(stdout), "w");
result = bam_pad2unpad(in, out);
bam_close(in); bam_close(out);
return result;
}
+
+static int usage(int is_long_help)
+{
+ fprintf(stderr, "\n");
+ fprintf(stderr, "Usage: samtools depad <in.bam>\n\n");
+ fprintf(stderr, "Options: -? longer help\n");
+ //TODO - These are the arguments I think make sense to support:
+ //fprintf(stderr, "Usage: samtools depad [options] <in.bam>|<in.sam>\n\n");
+ //fprintf(stderr, "Options: -b output BAM\n");
+ //fprintf(stderr, " -S input is SAM\n");
+ //fprintf(stderr, " -u uncompressed BAM output (force -b)\n");
+ //fprintf(stderr, " -1 fast compression (force -b)\n");
+ //fprintf(stderr, " -@ INT number of BAM compression threads [0]\n");
+ //fprintf(stderr, " -T FILE reference sequence file (force -S) [null]\n");
+ //fprintf(stderr, " -o FILE output file name [stdout]\n");
+ //fprintf(stderr, " -? longer help\n");
+ fprintf(stderr, "\n");
+ if (is_long_help)
+ fprintf(stderr, "Notes:\n\
+\n\
+ 1. Requires embedded reference sequences (before the reads for that reference).\n\
+\n\
+ 2. The input padded alignment read's CIGAR strings must not use P or I operators.\n\
+\n");
+ return 1;
+}