r360: allow to set integer max occ

index.c

@@ -104,13 +104,13 @@ int mm_idx_getseq(const mm_idx_t *mi, uint32_t rid, uint32_t st, uint32_t en, ui
 	return en - st;
 }
 
-uint32_t mm_idx_cal_max_occ(const mm_idx_t *mi, float f)
+int32_t mm_idx_cal_max_occ(const mm_idx_t *mi, float f)
 {
 	int i;
 	size_t n = 0;
 	uint32_t thres;
 	khint_t *a, k;
-	if (f <= 0.) return UINT32_MAX;
+	if (f <= 0.) return INT32_MAX;
 	for (i = 0; i < 1<<mi->b; ++i)
 		if (mi->B[i].h) n += kh_size((idxhash_t*)mi->B[i].h);
 	a = (uint32_t*)malloc(n * 4);

main.c

@@ -6,7 +6,7 @@
 #include "mmpriv.h"
 #include "getopt.h"
 
-#define MM_VERSION "2.1.1-r347-dirty"
+#define MM_VERSION "2.1.1-r360-dirty"
 
 #ifdef __linux__
 #include <sys/resource.h>
@@ -78,7 +78,6 @@ int main(int argc, char *argv[])
 		else if (c == 'H') is_hpc = 1, idx_par_set = 1;
 		else if (c == 'd') fnw = optarg; // the above are indexing related options, except -I
 		else if (c == 'r') opt.bw = (int)mm_parse_num(optarg);
-		else if (c == 'f') opt.mid_occ_frac = atof(optarg);
 		else if (c == 't') n_threads = atoi(optarg);
 		else if (c == 'v') mm_verbose = atoi(optarg);
 		else if (c == 'g') opt.max_gap = (int)mm_parse_num(optarg);
@@ -117,6 +116,11 @@ int main(int argc, char *argv[])
 		else if (c == 'V') {
 			puts(MM_VERSION);
 			return 0;
+		} else if (c == 'f') {
+			double x;
+			x = atof(optarg);
+			if (x < 1.0) opt.mid_occ_frac = x, opt.mid_occ = 0;
+			else opt.mid_occ = (int)(x + .499);
 		} else if (c == 'u') {
 			if (*optarg == 'b') opt.flag |= MM_F_SPLICE_FOR|MM_F_SPLICE_REV;
 			else if (*optarg == 'B') opt.flag |= MM_F_SPLICE_BOTH;
@@ -168,7 +172,7 @@ int main(int argc, char *argv[])
 				opt.min_dp_max = 40;
 				opt.best_n = 20;
 				opt.bw = 50;
-				opt.mid_occ_frac = 2e-5f;
+				opt.mid_occ = 1000;
 			} else if (strcmp(optarg, "splice") == 0 || strcmp(optarg, "cdna") == 0) {
 				k = 15, w = 5;
 				opt.flag |= MM_F_SPLICE | MM_F_SPLICE_FOR | MM_F_SPLICE_REV;

map.c

@@ -40,7 +40,9 @@ void mm_mapopt_update(mm_mapopt_t *opt, const mm_idx_t *mi)
 {
 	if (opt->flag & MM_F_SPLICE_BOTH)
 		opt->flag &= ~(MM_F_SPLICE_FOR|MM_F_SPLICE_REV);
+	if (opt->max_occ <= 0)
 		opt->max_occ = mm_idx_cal_max_occ(mi, opt->max_occ_frac);
+	if (opt->mid_occ <= 0)
 		opt->mid_occ = mm_idx_cal_max_occ(mi, opt->mid_occ_frac);
 	if (mm_verbose >= 3)
 		fprintf(stderr, "[M::%s::%.3f*%.2f] mid_occ = %d; max_occ = %d\n", __func__, realtime() - mm_realtime0, cputime() / (realtime() - mm_realtime0),

minimap.h

@@ -105,8 +105,7 @@ typedef struct {
 	int min_dp_max;
 	int min_ksw_len;
 
-	int max_occ;
+	int32_t max_occ, mid_occ;
-	int mid_occ;
 } mm_mapopt_t;
 
 extern int mm_verbose, mm_dbg_flag;
@@ -127,7 +126,6 @@ void mm_sketch(void *km, const char *str, int len, int w, int k, uint32_t rid, i
 mm_idx_t *mm_idx_init(int w, int k, int b, int is_hpc);
 void mm_idx_destroy(mm_idx_t *mi);
 mm_idx_t *mm_idx_gen(struct mm_bseq_file_s *fp, int w, int k, int b, int is_hpc, int mini_batch_size, int n_threads, uint64_t batch_size, int keep_name);
-uint32_t mm_idx_cal_max_occ(const mm_idx_t *mi, float f);
 void mm_idx_stat(const mm_idx_t *idx);
 const uint64_t *mm_idx_get(const mm_idx_t *mi, uint64_t minier, int *n);
 int mm_idx_getseq(const mm_idx_t *mi, uint32_t rid, uint32_t st, uint32_t en, uint8_t *seq);

mmpriv.h

@@ -44,6 +44,8 @@ void mm_write_sam_SQ(const mm_idx_t *idx);
 void mm_write_sam_hdr_no_SQ(const char *rg, const char *ver, int argc, char *argv[]);
 void mm_write_paf(kstring_t *s, const mm_idx_t *mi, const mm_bseq1_t *t, const mm_reg1_t *r, void *km, int opt_flag);
 void mm_write_sam(kstring_t *s, const mm_idx_t *mi, const mm_bseq1_t *t, const mm_reg1_t *r, int n_regs, const mm_reg1_t *regs);
+
+int32_t mm_idx_cal_max_occ(const mm_idx_t *mi, float f);
 int mm_chain_dp(int max_dist_x, int max_dist_y, int bw, int max_skip, int min_cnt, int min_sc, int is_cdna, int64_t n, mm128_t *a, uint64_t **_u, void *km);
 mm_reg1_t *mm_align_skeleton(void *km, const mm_mapopt_t *opt, const mm_idx_t *mi, int qlen, const char *qstr, int *n_regs_, mm_reg1_t *regs, mm128_t *a);