minimap2 0.0.4 → 0.2.23.1

data/ext/minimap2/misc.c

@@ -0,0 +1,162 @@
+#include <stdlib.h>
+#include "mmpriv.h"
+
+int mm_verbose = 1;
+int mm_dbg_flag = 0;
+double mm_realtime0;
+
+#if defined(WIN32) || defined(_WIN32)
+#include <windows.h>
+
+struct timezone
+{
+  __int32  tz_minuteswest; /* minutes W of Greenwich */
+  int  tz_dsttime;     /* type of dst correction */
+};
+
+/*
+ * gettimeofday.c
+ *    Win32 gettimeofday() replacement
+ *    taken from PostgreSQL, according to
+ *    https://stackoverflow.com/questions/1676036/what-should-i-use-to-replace-gettimeofday-on-windows
+ *
+ * src/port/gettimeofday.c
+ *
+ * Copyright (c) 2003 SRA, Inc.
+ * Copyright (c) 2003 SKC, Inc.
+ *
+ * Permission to use, copy, modify, and distribute this software and
+ * its documentation for any purpose, without fee, and without a
+ * written agreement is hereby granted, provided that the above
+ * copyright notice and this paragraph and the following two
+ * paragraphs appear in all copies.
+ *
+ * IN NO EVENT SHALL THE AUTHOR BE LIABLE TO ANY PARTY FOR DIRECT,
+ * INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, INCLUDING
+ * LOST PROFITS, ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS
+ * DOCUMENTATION, EVEN IF THE UNIVERSITY OF CALIFORNIA HAS BEEN ADVISED
+ * OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * THE AUTHOR SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE.  THE SOFTWARE PROVIDED HEREUNDER IS ON AN "AS
+ * IS" BASIS, AND THE AUTHOR HAS NO OBLIGATIONS TO PROVIDE MAINTENANCE,
+ * SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
+ */
+
+/* FILETIME of Jan 1 1970 00:00:00. */
+static const unsigned __int64 epoch = ((unsigned __int64) 116444736000000000ULL);
+
+/*
+ * timezone information is stored outside the kernel so tzp isn't used anymore.
+ *
+ * Note: this function is not for Win32 high precision timing purpose. See
+ * elapsed_time().
+ */
+int gettimeofday(struct timeval * tp, struct timezone *tzp)
+{
+    FILETIME    file_time;
+    SYSTEMTIME  system_time;
+    ULARGE_INTEGER ularge;
+
+    GetSystemTime(&system_time);
+    SystemTimeToFileTime(&system_time, &file_time);
+    ularge.LowPart = file_time.dwLowDateTime;
+    ularge.HighPart = file_time.dwHighDateTime;
+
+    tp->tv_sec = (long) ((ularge.QuadPart - epoch) / 10000000L);
+    tp->tv_usec = (long) (system_time.wMilliseconds * 1000);
+
+    return 0;
+}
+
+// taken from https://stackoverflow.com/questions/5272470/c-get-cpu-usage-on-linux-and-windows
+double cputime()
+{
+	HANDLE hProcess = GetCurrentProcess();
+	FILETIME ftCreation, ftExit, ftKernel, ftUser;
+	SYSTEMTIME stKernel;
+	SYSTEMTIME stUser;
+
+	GetProcessTimes(hProcess, &ftCreation, &ftExit, &ftKernel, &ftUser);
+	FileTimeToSystemTime(&ftKernel, &stKernel);
+	FileTimeToSystemTime(&ftUser, &stUser);
+
+	double kernelModeTime = ((stKernel.wHour * 60.) + stKernel.wMinute * 60.) + stKernel.wSecond * 1. + stKernel.wMilliseconds / 1000.;
+	double userModeTime = ((stUser.wHour * 60.) + stUser.wMinute * 60.) + stUser.wSecond * 1. + stUser.wMilliseconds / 1000.;
+
+	return kernelModeTime + userModeTime;
+}
+
+long peakrss(void) { return 0; }
+#else
+#include <sys/resource.h>
+#include <sys/time.h>
+
+double cputime(void)
+{
+	struct rusage r;
+	getrusage(RUSAGE_SELF, &r);
+	return r.ru_utime.tv_sec + r.ru_stime.tv_sec + 1e-6 * (r.ru_utime.tv_usec + r.ru_stime.tv_usec);
+}
+
+long peakrss(void)
+{
+	struct rusage r;
+	getrusage(RUSAGE_SELF, &r);
+#ifdef __linux__
+	return r.ru_maxrss * 1024;
+#else
+	return r.ru_maxrss;
+#endif
+}
+
+#endif /* WIN32 || _WIN32 */
+
+double realtime(void)
+{
+	struct timeval tp;
+	gettimeofday(&tp, NULL);
+	return tp.tv_sec + tp.tv_usec * 1e-6;
+}
+
+void mm_err_puts(const char *str)
+{
+	int ret;
+	ret = puts(str);
+	if (ret == EOF) {
+		perror("[ERROR] failed to write the results");
+		exit(EXIT_FAILURE);
+	}
+}
+
+void mm_err_fwrite(const void *p, size_t size, size_t nitems, FILE *fp)
+{
+	int ret;
+	ret = fwrite(p, size, nitems, fp);
+	if (ret == EOF) {
+		perror("[ERROR] failed to write data");
+		exit(EXIT_FAILURE);
+	}
+}
+
+void mm_err_fread(void *p, size_t size, size_t nitems, FILE *fp)
+{
+	int ret;
+	ret = fread(p, size, nitems, fp);
+	if (ret == EOF) {
+		perror("[ERROR] failed to read data");
+		exit(EXIT_FAILURE);
+	}
+}
+
+#include "ksort.h"
+
+#define sort_key_128x(a) ((a).x)
+KRADIX_SORT_INIT(128x, mm128_t, sort_key_128x, 8) 
+
+#define sort_key_64(x) (x)
+KRADIX_SORT_INIT(64, uint64_t, sort_key_64, 8)
+
+KSORT_INIT_GENERIC(uint32_t)
+KSORT_INIT_GENERIC(uint64_t)

data/ext/minimap2/mmpriv.h

@@ -0,0 +1,131 @@
+#ifndef MMPRIV2_H
+#define MMPRIV2_H
+
+#include <assert.h>
+#include "minimap.h"
+#include "bseq.h"
+#include "kseq.h"
+
+#define MM_PARENT_UNSET   (-1)
+#define MM_PARENT_TMP_PRI (-2)
+
+#define MM_DBG_NO_KALLOC     0x1
+#define MM_DBG_PRINT_QNAME   0x2
+#define MM_DBG_PRINT_SEED    0x4
+#define MM_DBG_PRINT_ALN_SEQ 0x8
+
+#define MM_SEED_LONG_JOIN  (1ULL<<40)
+#define MM_SEED_IGNORE     (1ULL<<41)
+#define MM_SEED_TANDEM     (1ULL<<42)
+#define MM_SEED_SELF       (1ULL<<43)
+
+#define MM_SEED_SEG_SHIFT  48
+#define MM_SEED_SEG_MASK   (0xffULL<<(MM_SEED_SEG_SHIFT))
+
+#ifndef kroundup32
+#define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x))
+#endif
+
+#define mm_seq4_set(s, i, c) ((s)[(i)>>3] |= (uint32_t)(c) << (((i)&7)<<2))
+#define mm_seq4_get(s, i)    ((s)[(i)>>3] >> (((i)&7)<<2) & 0xf)
+
+#define MALLOC(type, len) ((type*)malloc((len) * sizeof(type)))
+#define CALLOC(type, len) ((type*)calloc((len), sizeof(type)))
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef struct {
+	uint32_t n;
+	uint32_t q_pos;
+	uint32_t q_span:31, flt:1;
+	uint32_t seg_id:31, is_tandem:1;
+	const uint64_t *cr;
+} mm_seed_t;
+
+typedef struct {
+	int n_u, n_a;
+	uint64_t *u;
+	mm128_t *a;
+} mm_seg_t;
+
+double cputime(void);
+double realtime(void);
+long peakrss(void);
+
+void radix_sort_128x(mm128_t *beg, mm128_t *end);
+void radix_sort_64(uint64_t *beg, uint64_t *end);
+uint32_t ks_ksmall_uint32_t(size_t n, uint32_t arr[], size_t kk);
+
+void mm_sketch(void *km, const char *str, int len, int w, int k, uint32_t rid, int is_hpc, mm128_v *p);
+
+mm_seed_t *mm_collect_matches(void *km, int *_n_m, int qlen, int max_occ, int max_max_occ, int dist, const mm_idx_t *mi, const mm128_v *mv, int64_t *n_a, int *rep_len, int *n_mini_pos, uint64_t **mini_pos);
+void mm_seed_mz_flt(void *km, mm128_v *mv, int32_t q_occ_max, float q_occ_frac);
+
+double mm_event_identity(const mm_reg1_t *r);
+int mm_write_sam_hdr(const mm_idx_t *mi, 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, int64_t opt_flag);
+void mm_write_paf3(kstring_t *s, const mm_idx_t *mi, const mm_bseq1_t *t, const mm_reg1_t *r, void *km, int64_t opt_flag, int rep_len);
+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);
+void mm_write_sam2(kstring_t *s, const mm_idx_t *mi, const mm_bseq1_t *t, int seg_idx, int reg_idx, int n_seg, const int *n_regs, const mm_reg1_t *const* regs, void *km, int64_t opt_flag);
+void mm_write_sam3(kstring_t *s, const mm_idx_t *mi, const mm_bseq1_t *t, int seg_idx, int reg_idx, int n_seg, const int *n_regss, const mm_reg1_t *const* regss, void *km, int64_t opt_flag, int rep_len);
+
+void mm_idxopt_init(mm_idxopt_t *opt);
+const uint64_t *mm_idx_get(const mm_idx_t *mi, uint64_t minier, int *n);
+int32_t mm_idx_cal_max_occ(const mm_idx_t *mi, float f);
+int mm_idx_getseq2(const mm_idx_t *mi, int is_rev, uint32_t rid, uint32_t st, uint32_t en, uint8_t *seq);
+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);
+mm_reg1_t *mm_gen_regs(void *km, uint32_t hash, int qlen, int n_u, uint64_t *u, mm128_t *a, int is_qstrand);
+
+mm128_t *mm_chain_dp(int max_dist_x, int max_dist_y, int bw, int max_skip, int max_iter, int min_cnt, int min_sc, float gap_scale,
+					 int is_cdna, int n_segs, int64_t n, mm128_t *a, int *n_u_, uint64_t **_u, void *km);
+mm128_t *mg_lchain_dp(int max_dist_x, int max_dist_y, int bw, int max_skip, int max_iter, int min_cnt, int min_sc, float chn_pen_gap, float chn_pen_skip,
+					  int is_cdna, int n_segs, int64_t n, mm128_t *a, int *n_u_, uint64_t **_u, void *km);
+mm128_t *mg_lchain_rmq(int max_dist, int max_dist_inner, int bw, int max_chn_skip, int cap_rmq_size, int min_cnt, int min_sc, float chn_pen_gap, float chn_pen_skip,
+					   int64_t n, mm128_t *a, int *n_u_, uint64_t **_u, void *km);
+
+void mm_mark_alt(const mm_idx_t *mi, int n, mm_reg1_t *r);
+void mm_split_reg(mm_reg1_t *r, mm_reg1_t *r2, int n, int qlen, mm128_t *a, int is_qstrand);
+void mm_sync_regs(void *km, int n_regs, mm_reg1_t *regs);
+int mm_squeeze_a(void *km, int n_regs, mm_reg1_t *regs, mm128_t *a);
+int mm_set_sam_pri(int n, mm_reg1_t *r);
+void mm_set_parent(void *km, float mask_level, int mask_len, int n, mm_reg1_t *r, int sub_diff, int hard_mask_level, float alt_diff_frac);
+void mm_select_sub(void *km, float pri_ratio, int min_diff, int best_n, int check_strand, int min_strand_sc, int *n_, mm_reg1_t *r);
+void mm_select_sub_multi(void *km, float pri_ratio, float pri1, float pri2, int max_gap_ref, int min_diff, int best_n, int n_segs, const int *qlens, int *n_, mm_reg1_t *r);
+int mm_filter_strand_retained(int n_regs, mm_reg1_t *r);
+void mm_filter_regs(const mm_mapopt_t *opt, int qlen, int *n_regs, mm_reg1_t *regs);
+void mm_hit_sort(void *km, int *n_regs, mm_reg1_t *r, float alt_diff_frac);
+void mm_set_mapq(void *km, int n_regs, mm_reg1_t *regs, int min_chain_sc, int match_sc, int rep_len, int is_sr);
+void mm_update_dp_max(int qlen, int n_regs, mm_reg1_t *regs, float frac, int a, int b);
+
+void mm_est_err(const mm_idx_t *mi, int qlen, int n_regs, mm_reg1_t *regs, const mm128_t *a, int32_t n, const uint64_t *mini_pos);
+
+mm_seg_t *mm_seg_gen(void *km, uint32_t hash, int n_segs, const int *qlens, int n_regs0, const mm_reg1_t *regs0, int *n_regs, mm_reg1_t **regs, const mm128_t *a);
+void mm_seg_free(void *km, int n_segs, mm_seg_t *segs);
+void mm_pair(void *km, int max_gap_ref, int dp_bonus, int sub_diff, int match_sc, const int *qlens, int *n_regs, mm_reg1_t **regs);
+
+FILE *mm_split_init(const char *prefix, const mm_idx_t *mi);
+mm_idx_t *mm_split_merge_prep(const char *prefix, int n_splits, FILE **fp, uint32_t *n_seq_part);
+int mm_split_merge(int n_segs, const char **fn, const mm_mapopt_t *opt, int n_split_idx);
+void mm_split_rm_tmp(const char *prefix, int n_splits);
+
+void mm_err_puts(const char *str);
+void mm_err_fwrite(const void *p, size_t size, size_t nitems, FILE *fp);
+void mm_err_fread(void *p, size_t size, size_t nitems, FILE *fp);
+
+static inline float mg_log2(float x) // NB: this doesn't work when x<2
+{
+	union { float f; uint32_t i; } z = { x };
+	float log_2 = ((z.i >> 23) & 255) - 128;
+	z.i &= ~(255 << 23);
+	z.i += 127 << 23;
+	log_2 += (-0.34484843f * z.f + 2.02466578f) * z.f - 0.67487759f;
+	return log_2;
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif

data/ext/minimap2/options.c

@@ -0,0 +1,233 @@
+#include <stdio.h>
+#include <limits.h>
+#include "mmpriv.h"
+
+void mm_idxopt_init(mm_idxopt_t *opt)
+{
+	memset(opt, 0, sizeof(mm_idxopt_t));
+	opt->k = 15, opt->w = 10, opt->flag = 0;
+	opt->bucket_bits = 14;
+	opt->mini_batch_size = 50000000;
+	opt->batch_size = 4000000000ULL;
+}
+
+void mm_mapopt_init(mm_mapopt_t *opt)
+{
+	memset(opt, 0, sizeof(mm_mapopt_t));
+	opt->seed = 11;
+	opt->mid_occ_frac = 2e-4f;
+	opt->min_mid_occ = 10;
+	opt->max_mid_occ = 1000000;
+	opt->sdust_thres = 0; // no SDUST masking
+	opt->q_occ_frac = 0.01f;
+
+	opt->min_cnt = 3;
+	opt->min_chain_score = 40;
+	opt->bw = 500, opt->bw_long = 20000;
+	opt->max_gap = 5000;
+	opt->max_gap_ref = -1;
+	opt->max_chain_skip = 25;
+	opt->max_chain_iter = 5000;
+	opt->rmq_inner_dist = 1000;
+	opt->rmq_size_cap = 100000;
+	opt->rmq_rescue_size = 1000;
+	opt->rmq_rescue_ratio = 0.1f;
+	opt->chain_gap_scale = 0.8f;
+	opt->chain_skip_scale = 0.0f;
+	opt->max_max_occ = 4095;
+	opt->occ_dist = 500;
+
+	opt->mask_level = 0.5f;
+	opt->mask_len = INT_MAX;
+	opt->pri_ratio = 0.8f;
+	opt->best_n = 5;
+
+	opt->alt_drop = 0.15f;
+
+	opt->a = 2, opt->b = 4, opt->q = 4, opt->e = 2, opt->q2 = 24, opt->e2 = 1;
+	opt->sc_ambi = 1;
+	opt->zdrop = 400, opt->zdrop_inv = 200;
+	opt->end_bonus = -1;
+	opt->min_dp_max = opt->min_chain_score * opt->a;
+	opt->min_ksw_len = 200;
+	opt->anchor_ext_len = 20, opt->anchor_ext_shift = 6;
+	opt->max_clip_ratio = 1.0f;
+	opt->mini_batch_size = 500000000;
+	opt->max_sw_mat = 100000000;
+	opt->cap_kalloc = 1000000000;
+
+	opt->rank_min_len = 500;
+	opt->rank_frac = 0.9f;
+
+	opt->pe_ori = 0; // FF
+	opt->pe_bonus = 33;
+}
+
+void mm_mapopt_update(mm_mapopt_t *opt, const mm_idx_t *mi)
+{
+	if ((opt->flag & MM_F_SPLICE_FOR) || (opt->flag & MM_F_SPLICE_REV))
+		opt->flag |= MM_F_SPLICE;
+	if (opt->mid_occ <= 0) {
+		opt->mid_occ = mm_idx_cal_max_occ(mi, opt->mid_occ_frac);
+		if (opt->mid_occ < opt->min_mid_occ)
+			opt->mid_occ = opt->min_mid_occ;
+		if (opt->max_mid_occ > opt->min_mid_occ && opt->mid_occ > opt->max_mid_occ)
+			opt->mid_occ = opt->max_mid_occ;
+	}
+	if (mm_verbose >= 3)
+		fprintf(stderr, "[M::%s::%.3f*%.2f] mid_occ = %d\n", __func__, realtime() - mm_realtime0, cputime() / (realtime() - mm_realtime0), opt->mid_occ);
+}
+
+void mm_mapopt_max_intron_len(mm_mapopt_t *opt, int max_intron_len)
+{
+	if ((opt->flag & MM_F_SPLICE) && max_intron_len > 0)
+		opt->max_gap_ref = opt->bw = opt->bw_long = max_intron_len;
+}
+
+int mm_set_opt(const char *preset, mm_idxopt_t *io, mm_mapopt_t *mo)
+{
+	if (preset == 0) {
+		mm_idxopt_init(io);
+		mm_mapopt_init(mo);
+	} else if (strcmp(preset, "map-ont") == 0) { // this is the same as the default
+	} else if (strcmp(preset, "ava-ont") == 0) {
+		io->flag = 0, io->k = 15, io->w = 5;
+		mo->flag |= MM_F_ALL_CHAINS | MM_F_NO_DIAG | MM_F_NO_DUAL | MM_F_NO_LJOIN;
+		mo->min_chain_score = 100, mo->pri_ratio = 0.0f, mo->max_chain_skip = 25;
+		mo->bw = mo->bw_long = 2000;
+		mo->occ_dist = 0;
+	} else if (strcmp(preset, "map10k") == 0 || strcmp(preset, "map-pb") == 0) {
+		io->flag |= MM_I_HPC, io->k = 19;
+	} else if (strcmp(preset, "ava-pb") == 0) {
+		io->flag |= MM_I_HPC, io->k = 19, io->w = 5;
+		mo->flag |= MM_F_ALL_CHAINS | MM_F_NO_DIAG | MM_F_NO_DUAL | MM_F_NO_LJOIN;
+		mo->min_chain_score = 100, mo->pri_ratio = 0.0f, mo->max_chain_skip = 25;
+		mo->bw_long = mo->bw;
+		mo->occ_dist = 0;
+	} else if (strcmp(preset, "map-hifi") == 0 || strcmp(preset, "map-ccs") == 0) {
+		io->flag = 0, io->k = 19, io->w = 19;
+		mo->max_gap = 10000;
+		mo->a = 1, mo->b = 4, mo->q = 6, mo->q2 = 26, mo->e = 2, mo->e2 = 1;
+		mo->occ_dist = 500;
+		mo->min_mid_occ = 50, mo->max_mid_occ = 500;
+		mo->min_dp_max = 200;
+	} else if (strncmp(preset, "asm", 3) == 0) {
+		io->flag = 0, io->k = 19, io->w = 19;
+		mo->bw = mo->bw_long = 100000;
+		mo->max_gap = 10000;
+		mo->flag |= MM_F_RMQ;
+		mo->min_mid_occ = 50, mo->max_mid_occ = 500;
+		mo->min_dp_max = 200;
+		mo->best_n = 50;
+		if (strcmp(preset, "asm5") == 0) {
+			mo->a = 1, mo->b = 19, mo->q = 39, mo->q2 = 81, mo->e = 3, mo->e2 = 1, mo->zdrop = mo->zdrop_inv = 200;
+		} else if (strcmp(preset, "asm10") == 0) {
+			mo->a = 1, mo->b = 9, mo->q = 16, mo->q2 = 41, mo->e = 2, mo->e2 = 1, mo->zdrop = mo->zdrop_inv = 200;
+		} else if (strcmp(preset, "asm20") == 0) {
+			mo->a = 1, mo->b = 4, mo->q = 6, mo->q2 = 26, mo->e = 2, mo->e2 = 1, mo->zdrop = mo->zdrop_inv = 200;
+			io->w = 10;
+		} else return -1;
+	} else if (strcmp(preset, "short") == 0 || strcmp(preset, "sr") == 0) {
+		io->flag = 0, io->k = 21, io->w = 11;
+		mo->flag |= MM_F_SR | MM_F_FRAG_MODE | MM_F_NO_PRINT_2ND | MM_F_2_IO_THREADS | MM_F_HEAP_SORT;
+		mo->pe_ori = 0<<1|1; // FR
+		mo->a = 2, mo->b = 8, mo->q = 12, mo->e = 2, mo->q2 = 24, mo->e2 = 1;
+		mo->zdrop = mo->zdrop_inv = 100;
+		mo->end_bonus = 10;
+		mo->max_frag_len = 800;
+		mo->max_gap = 100;
+		mo->bw = mo->bw_long = 100;
+		mo->pri_ratio = 0.5f;
+		mo->min_cnt = 2;
+		mo->min_chain_score = 25;
+		mo->min_dp_max = 40;
+		mo->best_n = 20;
+		mo->mid_occ = 1000;
+		mo->max_occ = 5000;
+		mo->mini_batch_size = 50000000;
+	} else if (strncmp(preset, "splice", 6) == 0 || strcmp(preset, "cdna") == 0) {
+		io->flag = 0, io->k = 15, io->w = 5;
+		mo->flag |= MM_F_SPLICE | MM_F_SPLICE_FOR | MM_F_SPLICE_REV | MM_F_SPLICE_FLANK;
+		mo->max_sw_mat = 0;
+		mo->max_gap = 2000, mo->max_gap_ref = mo->bw = mo->bw_long = 200000;
+		mo->a = 1, mo->b = 2, mo->q = 2, mo->e = 1, mo->q2 = 32, mo->e2 = 0;
+		mo->noncan = 9;
+		mo->junc_bonus = 9;
+		mo->zdrop = 200, mo->zdrop_inv = 100; // because mo->a is halved
+		if (strcmp(preset, "splice:hq") == 0)
+			mo->junc_bonus = 5, mo->b = 4, mo->q = 6, mo->q2 = 24;
+	} else return -1;
+	return 0;
+}
+
+int mm_check_opt(const mm_idxopt_t *io, const mm_mapopt_t *mo)
+{
+	if (mo->bw > mo->bw_long) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m with '-rNUM1,NUM2', NUM1 (%d) can't be larger than NUM2 (%d)\033[0m\n", mo->bw, mo->bw_long);
+		return -8;
+	}
+	if ((mo->flag & MM_F_RMQ) && (mo->flag & (MM_F_SR|MM_F_SPLICE))) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m --rmq doesn't work with --sr or --splice\033[0m\n");
+		return -7;
+	}
+	if (mo->split_prefix && (mo->flag & (MM_F_OUT_CS|MM_F_OUT_MD))) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m --cs or --MD doesn't work with --split-prefix\033[0m\n");
+		return -6;
+	}
+	if (io->k <= 0 || io->w <= 0) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m -k and -w must be positive\033[0m\n");
+		return -5;
+	}
+	if (mo->best_n < 0) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m -N must be no less than 0\033[0m\n");
+		return -4;
+	}
+	if (mo->best_n == 0 && mm_verbose >= 2)
+		fprintf(stderr, "[WARNING]\033[1;31m '-N 0' reduces mapping accuracy. Please use '--secondary=no' instead.\033[0m\n");
+	if (mo->pri_ratio < 0.0f || mo->pri_ratio > 1.0f) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m -p must be within 0 and 1 (including 0 and 1)\033[0m\n");
+		return -4;
+	}
+	if ((mo->flag & MM_F_FOR_ONLY) && (mo->flag & MM_F_REV_ONLY)) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m --for-only and --rev-only can't be applied at the same time\033[0m\n");
+		return -3;
+	}
+	if (mo->e <= 0 || mo->q <= 0) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m -O and -E must be positive\033[0m\n");
+		return -1;
+	}
+	if ((mo->q != mo->q2 || mo->e != mo->e2) && !(mo->e > mo->e2 && mo->q + mo->e < mo->q2 + mo->e2)) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m dual gap penalties violating E1>E2 and O1+E1<O2+E2\033[0m\n");
+		return -2;
+	}
+	if ((mo->q + mo->e) + (mo->q2 + mo->e2) > 127) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m scoring system violating ({-O}+{-E})+({-O2}+{-E2}) <= 127\033[0m\n");
+		return -1;
+	}
+	if (mo->zdrop < mo->zdrop_inv) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m Z-drop should not be less than inversion-Z-drop\033[0m\n");
+		return -5;
+	}
+	if ((mo->flag & MM_F_NO_PRINT_2ND) && (mo->flag & MM_F_ALL_CHAINS)) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m -X/-P and --secondary=no can't be applied at the same time\033[0m\n");
+		return -5;
+	}
+	if ((mo->flag & MM_F_QSTRAND) && ((mo->flag & (MM_F_OUT_SAM|MM_F_SPLICE|MM_F_FRAG_MODE)) || (io->flag & MM_I_HPC))) {
+		if (mm_verbose >= 1)
+			fprintf(stderr, "[ERROR]\033[1;31m --qstrand doesn't work with -a, -H, --frag or --splice\033[0m\n");
+		return -5;
+	}
+	return 0;
+}

data/ext/minimap2/pe.c

@@ -0,0 +1,177 @@
+#include <stdlib.h>
+#include <math.h>
+#include "mmpriv.h"
+#include "kvec.h"
+
+void mm_select_sub_multi(void *km, float pri_ratio, float pri1, float pri2, int max_gap_ref, int min_diff, int best_n, int n_segs, const int *qlens, int *n_, mm_reg1_t *r)
+{
+	if (pri_ratio > 0.0f && *n_ > 0) {
+		int i, k, n = *n_, n_2nd = 0;
+		int max_dist = n_segs == 2? qlens[0] + qlens[1] + max_gap_ref : 0;
+		for (i = k = 0; i < n; ++i) {
+			int to_keep = 0;
+			if (r[i].parent == i) { // primary
+				to_keep = 1;
+			} else if (r[i].score + min_diff >= r[r[i].parent].score) {
+				to_keep = 1;
+			} else {
+				mm_reg1_t *p = &r[r[i].parent], *q = &r[i];
+				if (p->rev == q->rev && p->rid == q->rid && q->re - p->rs < max_dist && p->re - q->rs < max_dist) { // child and parent are close on the ref
+					if (q->score >= p->score * pri1)
+						to_keep = 1;
+				} else {
+					int is_par_both = (n_segs == 2 && p->qs < qlens[0] && p->qe > qlens[0]);
+					int is_chi_both = (n_segs == 2 && q->qs < qlens[0] && q->qe > qlens[0]);
+					if (is_chi_both || is_chi_both == is_par_both) {
+						if (q->score >= p->score * pri_ratio)
+							to_keep = 1;
+					} else { // the remaining case: is_chi_both == 0 && is_par_both == 1
+						if (q->score >= p->score * pri2)
+							to_keep = 1;
+					}
+				}
+			}
+			if (to_keep && r[i].parent != i) {
+				if (n_2nd++ >= best_n) to_keep = 0; // don't keep if there are too many secondary hits
+			}
+			if (to_keep) r[k++] = r[i];
+			else if (r[i].p) free(r[i].p);
+		}
+		if (k != n) mm_sync_regs(km, k, r); // removing hits requires sync()
+		*n_ = k;
+	}
+}
+
+void mm_set_pe_thru(const int *qlens, int *n_regs, mm_reg1_t **regs)
+{
+	int s, i, n_pri[2], pri[2];
+	n_pri[0] = n_pri[1] = 0;
+	pri[0] = pri[1] = -1;
+	for (s = 0; s < 2; ++s)
+		for (i = 0; i < n_regs[s]; ++i)
+			if (regs[s][i].id == regs[s][i].parent)
+				++n_pri[s], pri[s] = i;
+	if (n_pri[0] == 1 && n_pri[1] == 1) {
+		mm_reg1_t *p = &regs[0][pri[0]];
+		mm_reg1_t *q = &regs[1][pri[1]];
+		if (p->rid == q->rid && p->rev == q->rev && abs(p->rs - q->rs) < 3 && abs(p->re - q->re) < 3
+			&& ((p->qs == 0 && qlens[1] - q->qe == 0) || (q->qs == 0 && qlens[0] - p->qe == 0)))
+		{
+			p->pe_thru = q->pe_thru = 1;
+		}
+	}
+}
+
+#include "ksort.h"
+
+typedef struct {
+	int s, rev;
+	uint64_t key;
+	mm_reg1_t *r;
+} pair_arr_t;
+
+#define sort_key_pair(a) ((a).key)
+KRADIX_SORT_INIT(pair, pair_arr_t, sort_key_pair, 8)
+
+void mm_pair(void *km, int max_gap_ref, int pe_bonus, int sub_diff, int match_sc, const int *qlens, int *n_regs, mm_reg1_t **regs)
+{
+	int i, j, s, n, last[2], dp_thres, segs = 0, max_idx[2];
+	int64_t max;
+	pair_arr_t *a;
+	kvec_t(uint64_t) sc = {0,0,0};
+
+	a = (pair_arr_t*)kmalloc(km, (n_regs[0] + n_regs[1]) * sizeof(pair_arr_t));
+	for (s = n = 0, dp_thres = 0; s < 2; ++s) {
+		int max = 0;
+		for (i = 0; i < n_regs[s]; ++i) {
+			a[n].s = s;
+			a[n].r = &regs[s][i];
+			a[n].rev = a[n].r->rev;
+			a[n].key = (uint64_t)a[n].r->rid << 32 | a[n].r->rs<<1 | (s^a[n].rev);
+			max = max > a[n].r->p->dp_max? max : a[n].r->p->dp_max;
+			++n;
+			segs |= 1<<s;
+		}
+		dp_thres += max;
+	}
+	if (segs != 3) {
+		kfree(km, a); // only one end is mapped
+		return;
+	}
+	dp_thres -= pe_bonus;
+	if (dp_thres < 0) dp_thres = 0;
+	radix_sort_pair(a, a + n);
+
+	max = -1;
+	max_idx[0] = max_idx[1] = -1;
+	last[0] = last[1] = -1;
+	kv_resize(uint64_t, km, sc, (size_t)n);
+	for (i = 0; i < n; ++i) {
+		if (a[i].key & 1) { // reverse first read or forward second read
+			mm_reg1_t *q, *r;
+			if (last[a[i].rev] < 0) continue;
+			r = a[i].r;
+			q = a[last[a[i].rev]].r;
+			if (r->rid != q->rid || r->rs - q->re > max_gap_ref) continue;
+			for (j = last[a[i].rev]; j >= 0; --j) {
+				int64_t score;
+				if (a[j].rev != a[i].rev || a[j].s == a[i].s) continue;
+				q = a[j].r;
+				if (r->rid != q->rid || r->rs - q->re > max_gap_ref) break;
+				if (r->p->dp_max + q->p->dp_max < dp_thres) continue;
+				score = (int64_t)(r->p->dp_max + q->p->dp_max) << 32 | (r->hash + q->hash);
+				if (score > max)
+					max = score, max_idx[a[j].s] = j, max_idx[a[i].s] = i;
+				kv_push(uint64_t, km, sc, score);
+			}
+		} else { // forward first read or reverse second read
+			last[a[i].rev] = i;
+		}
+	}
+	if (sc.n > 1)
+		radix_sort_64(sc.a, sc.a + sc.n);
+
+	if (sc.n > 0 && max > 0) { // found at least one pair
+		int n_sub = 0, mapq_pe;
+		mm_reg1_t *r[2];
+		r[0] = a[max_idx[0]].r, r[1] = a[max_idx[1]].r;
+		r[0]->proper_frag = r[1]->proper_frag = 1;
+		for (s = 0; s < 2; ++s) {
+			if (r[s]->id != r[s]->parent) { // then lift to primary and update parent
+				mm_reg1_t *p = &regs[s][r[s]->parent];
+				for (i = 0; i < n_regs[s]; ++i)
+					if (regs[s][i].parent == p->id)
+						regs[s][i].parent = r[s]->id;
+				p->mapq = 0;
+			}
+			if (!r[s]->sam_pri) { // then sync sam_pri
+				for (i = 0; i < n_regs[s]; ++i)
+					regs[s][i].sam_pri = 0;
+				r[s]->sam_pri = 1;
+			}
+		}
+		mapq_pe = r[0]->mapq > r[1]->mapq? r[0]->mapq : r[1]->mapq;
+		for (i = 0; i < (int)sc.n; ++i)
+			if ((sc.a[i]>>32) + sub_diff >= (uint64_t)max>>32)
+				++n_sub;
+		if (sc.n > 1) {
+			int mapq_pe_alt;
+			mapq_pe_alt = (int)(6.02f * ((max>>32) - (sc.a[sc.n - 2]>>32)) / match_sc - 4.343f * logf(n_sub)); // n_sub > 0 because it counts the optimal, too
+			mapq_pe = mapq_pe < mapq_pe_alt? mapq_pe : mapq_pe_alt;
+		}
+		if (r[0]->mapq < mapq_pe) r[0]->mapq = (int)(.2f * r[0]->mapq + .8f * mapq_pe + .499f);
+		if (r[1]->mapq < mapq_pe) r[1]->mapq = (int)(.2f * r[1]->mapq + .8f * mapq_pe + .499f);
+		if (sc.n == 1) {
+			if (r[0]->mapq < 2) r[0]->mapq = 2;
+			if (r[1]->mapq < 2) r[1]->mapq = 2;
+		} else if ((uint64_t)max>>32 > sc.a[sc.n - 2]>>32) {
+			if (r[0]->mapq < 1) r[0]->mapq = 1;
+			if (r[1]->mapq < 1) r[1]->mapq = 1;
+		}
+	}
+
+	kfree(km, a);
+	kfree(km, sc.a);
+
+	mm_set_pe_thru(qlens, n_regs, regs);
+}