minimap2 0.2.22.0 → 0.2.24.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/README.md +60 -76
- data/ext/Rakefile +55 -0
- data/ext/cmappy/cmappy.c +129 -0
- data/ext/cmappy/cmappy.h +44 -0
- data/ext/minimap2/FAQ.md +46 -0
- data/ext/minimap2/LICENSE.txt +24 -0
- data/ext/minimap2/MANIFEST.in +10 -0
- data/ext/minimap2/Makefile +132 -0
- data/ext/minimap2/Makefile.simde +97 -0
- data/ext/minimap2/NEWS.md +821 -0
- data/ext/minimap2/README.md +403 -0
- data/ext/minimap2/align.c +1020 -0
- data/ext/minimap2/bseq.c +169 -0
- data/ext/minimap2/bseq.h +64 -0
- data/ext/minimap2/code_of_conduct.md +30 -0
- data/ext/minimap2/cookbook.md +243 -0
- data/ext/minimap2/esterr.c +64 -0
- data/ext/minimap2/example.c +63 -0
- data/ext/minimap2/format.c +559 -0
- data/ext/minimap2/hit.c +466 -0
- data/ext/minimap2/index.c +775 -0
- data/ext/minimap2/kalloc.c +205 -0
- data/ext/minimap2/kalloc.h +76 -0
- data/ext/minimap2/kdq.h +132 -0
- data/ext/minimap2/ketopt.h +120 -0
- data/ext/minimap2/khash.h +615 -0
- data/ext/minimap2/krmq.h +474 -0
- data/ext/minimap2/kseq.h +256 -0
- data/ext/minimap2/ksort.h +153 -0
- data/ext/minimap2/ksw2.h +184 -0
- data/ext/minimap2/ksw2_dispatch.c +96 -0
- data/ext/minimap2/ksw2_extd2_sse.c +402 -0
- data/ext/minimap2/ksw2_exts2_sse.c +416 -0
- data/ext/minimap2/ksw2_extz2_sse.c +313 -0
- data/ext/minimap2/ksw2_ll_sse.c +152 -0
- data/ext/minimap2/kthread.c +159 -0
- data/ext/minimap2/kthread.h +15 -0
- data/ext/minimap2/kvec.h +105 -0
- data/ext/minimap2/lchain.c +369 -0
- data/ext/minimap2/main.c +459 -0
- data/ext/minimap2/map.c +714 -0
- data/ext/minimap2/minimap.h +410 -0
- data/ext/minimap2/minimap2.1 +725 -0
- data/ext/minimap2/misc/README.md +179 -0
- data/ext/minimap2/misc/mmphase.js +335 -0
- data/ext/minimap2/misc/paftools.js +3149 -0
- data/ext/minimap2/misc.c +162 -0
- data/ext/minimap2/mmpriv.h +132 -0
- data/ext/minimap2/options.c +234 -0
- data/ext/minimap2/pe.c +177 -0
- data/ext/minimap2/python/README.rst +196 -0
- data/ext/minimap2/python/cmappy.h +152 -0
- data/ext/minimap2/python/cmappy.pxd +153 -0
- data/ext/minimap2/python/mappy.pyx +273 -0
- data/ext/minimap2/python/minimap2.py +39 -0
- data/ext/minimap2/sdust.c +213 -0
- data/ext/minimap2/sdust.h +25 -0
- data/ext/minimap2/seed.c +131 -0
- data/ext/minimap2/setup.py +55 -0
- data/ext/minimap2/sketch.c +143 -0
- data/ext/minimap2/splitidx.c +84 -0
- data/ext/minimap2/sse2neon/emmintrin.h +1689 -0
- data/ext/minimap2/test/MT-human.fa +278 -0
- data/ext/minimap2/test/MT-orang.fa +276 -0
- data/ext/minimap2/test/q-inv.fa +4 -0
- data/ext/minimap2/test/q2.fa +2 -0
- data/ext/minimap2/test/t-inv.fa +127 -0
- data/ext/minimap2/test/t2.fa +2 -0
- data/ext/minimap2/tex/Makefile +21 -0
- data/ext/minimap2/tex/bioinfo.cls +930 -0
- data/ext/minimap2/tex/blasr-mc.eval +17 -0
- data/ext/minimap2/tex/bowtie2-s3.sam.eval +28 -0
- data/ext/minimap2/tex/bwa-s3.sam.eval +52 -0
- data/ext/minimap2/tex/bwa.eval +55 -0
- data/ext/minimap2/tex/eval2roc.pl +33 -0
- data/ext/minimap2/tex/graphmap.eval +4 -0
- data/ext/minimap2/tex/hs38-simu.sh +10 -0
- data/ext/minimap2/tex/minialign.eval +49 -0
- data/ext/minimap2/tex/minimap2.bib +460 -0
- data/ext/minimap2/tex/minimap2.tex +724 -0
- data/ext/minimap2/tex/mm2-s3.sam.eval +62 -0
- data/ext/minimap2/tex/mm2-update.tex +240 -0
- data/ext/minimap2/tex/mm2.approx.eval +12 -0
- data/ext/minimap2/tex/mm2.eval +13 -0
- data/ext/minimap2/tex/natbib.bst +1288 -0
- data/ext/minimap2/tex/natbib.sty +803 -0
- data/ext/minimap2/tex/ngmlr.eval +38 -0
- data/ext/minimap2/tex/roc.gp +60 -0
- data/ext/minimap2/tex/snap-s3.sam.eval +62 -0
- data/ext/minimap2.patch +19 -0
- data/lib/minimap2/aligner.rb +4 -4
- data/lib/minimap2/alignment.rb +11 -11
- data/lib/minimap2/ffi/constants.rb +20 -16
- data/lib/minimap2/ffi/functions.rb +5 -0
- data/lib/minimap2/ffi.rb +4 -5
- data/lib/minimap2/version.rb +2 -2
- data/lib/minimap2.rb +51 -15
- metadata +97 -79
- data/lib/minimap2/ffi_helper.rb +0 -53
- data/vendor/libminimap2.so +0 -0
@@ -0,0 +1,615 @@
|
|
1
|
+
/* The MIT License
|
2
|
+
|
3
|
+
Copyright (c) 2008, 2009, 2011 by Attractive Chaos <attractor@live.co.uk>
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
6
|
+
a copy of this software and associated documentation files (the
|
7
|
+
"Software"), to deal in the Software without restriction, including
|
8
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
9
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
10
|
+
permit persons to whom the Software is furnished to do so, subject to
|
11
|
+
the following conditions:
|
12
|
+
|
13
|
+
The above copyright notice and this permission notice shall be
|
14
|
+
included in all copies or substantial portions of the Software.
|
15
|
+
|
16
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
17
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
18
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
19
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
|
20
|
+
BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
|
21
|
+
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
|
22
|
+
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
23
|
+
SOFTWARE.
|
24
|
+
*/
|
25
|
+
|
26
|
+
/*
|
27
|
+
An example:
|
28
|
+
|
29
|
+
#include "khash.h"
|
30
|
+
KHASH_MAP_INIT_INT(32, char)
|
31
|
+
int main() {
|
32
|
+
int ret, is_missing;
|
33
|
+
khiter_t k;
|
34
|
+
khash_t(32) *h = kh_init(32);
|
35
|
+
k = kh_put(32, h, 5, &ret);
|
36
|
+
kh_value(h, k) = 10;
|
37
|
+
k = kh_get(32, h, 10);
|
38
|
+
is_missing = (k == kh_end(h));
|
39
|
+
k = kh_get(32, h, 5);
|
40
|
+
kh_del(32, h, k);
|
41
|
+
for (k = kh_begin(h); k != kh_end(h); ++k)
|
42
|
+
if (kh_exist(h, k)) kh_value(h, k) = 1;
|
43
|
+
kh_destroy(32, h);
|
44
|
+
return 0;
|
45
|
+
}
|
46
|
+
*/
|
47
|
+
|
48
|
+
/*
|
49
|
+
2013-05-02 (0.2.8):
|
50
|
+
|
51
|
+
* Use quadratic probing. When the capacity is power of 2, stepping function
|
52
|
+
i*(i+1)/2 guarantees to traverse each bucket. It is better than double
|
53
|
+
hashing on cache performance and is more robust than linear probing.
|
54
|
+
|
55
|
+
In theory, double hashing should be more robust than quadratic probing.
|
56
|
+
However, my implementation is probably not for large hash tables, because
|
57
|
+
the second hash function is closely tied to the first hash function,
|
58
|
+
which reduce the effectiveness of double hashing.
|
59
|
+
|
60
|
+
Reference: http://research.cs.vt.edu/AVresearch/hashing/quadratic.php
|
61
|
+
|
62
|
+
2011-12-29 (0.2.7):
|
63
|
+
|
64
|
+
* Minor code clean up; no actual effect.
|
65
|
+
|
66
|
+
2011-09-16 (0.2.6):
|
67
|
+
|
68
|
+
* The capacity is a power of 2. This seems to dramatically improve the
|
69
|
+
speed for simple keys. Thank Zilong Tan for the suggestion. Reference:
|
70
|
+
|
71
|
+
- http://code.google.com/p/ulib/
|
72
|
+
- http://nothings.org/computer/judy/
|
73
|
+
|
74
|
+
* Allow to optionally use linear probing which usually has better
|
75
|
+
performance for random input. Double hashing is still the default as it
|
76
|
+
is more robust to certain non-random input.
|
77
|
+
|
78
|
+
* Added Wang's integer hash function (not used by default). This hash
|
79
|
+
function is more robust to certain non-random input.
|
80
|
+
|
81
|
+
2011-02-14 (0.2.5):
|
82
|
+
|
83
|
+
* Allow to declare global functions.
|
84
|
+
|
85
|
+
2009-09-26 (0.2.4):
|
86
|
+
|
87
|
+
* Improve portability
|
88
|
+
|
89
|
+
2008-09-19 (0.2.3):
|
90
|
+
|
91
|
+
* Corrected the example
|
92
|
+
* Improved interfaces
|
93
|
+
|
94
|
+
2008-09-11 (0.2.2):
|
95
|
+
|
96
|
+
* Improved speed a little in kh_put()
|
97
|
+
|
98
|
+
2008-09-10 (0.2.1):
|
99
|
+
|
100
|
+
* Added kh_clear()
|
101
|
+
* Fixed a compiling error
|
102
|
+
|
103
|
+
2008-09-02 (0.2.0):
|
104
|
+
|
105
|
+
* Changed to token concatenation which increases flexibility.
|
106
|
+
|
107
|
+
2008-08-31 (0.1.2):
|
108
|
+
|
109
|
+
* Fixed a bug in kh_get(), which has not been tested previously.
|
110
|
+
|
111
|
+
2008-08-31 (0.1.1):
|
112
|
+
|
113
|
+
* Added destructor
|
114
|
+
*/
|
115
|
+
|
116
|
+
|
117
|
+
#ifndef __AC_KHASH_H
|
118
|
+
#define __AC_KHASH_H
|
119
|
+
|
120
|
+
/*!
|
121
|
+
@header
|
122
|
+
|
123
|
+
Generic hash table library.
|
124
|
+
*/
|
125
|
+
|
126
|
+
#define AC_VERSION_KHASH_H "0.2.8"
|
127
|
+
|
128
|
+
#include <stdlib.h>
|
129
|
+
#include <string.h>
|
130
|
+
#include <limits.h>
|
131
|
+
#include "kalloc.h"
|
132
|
+
|
133
|
+
/* compiler specific configuration */
|
134
|
+
|
135
|
+
#if UINT_MAX == 0xffffffffu
|
136
|
+
typedef unsigned int khint32_t;
|
137
|
+
#elif ULONG_MAX == 0xffffffffu
|
138
|
+
typedef unsigned long khint32_t;
|
139
|
+
#endif
|
140
|
+
|
141
|
+
#if ULONG_MAX == ULLONG_MAX
|
142
|
+
typedef unsigned long khint64_t;
|
143
|
+
#else
|
144
|
+
typedef unsigned long long khint64_t;
|
145
|
+
#endif
|
146
|
+
|
147
|
+
#ifndef kh_inline
|
148
|
+
#ifdef _MSC_VER
|
149
|
+
#define kh_inline __inline
|
150
|
+
#else
|
151
|
+
#define kh_inline inline
|
152
|
+
#endif
|
153
|
+
#endif /* kh_inline */
|
154
|
+
|
155
|
+
#ifndef klib_unused
|
156
|
+
#if (defined __clang__ && __clang_major__ >= 3) || (defined __GNUC__ && __GNUC__ >= 3)
|
157
|
+
#define klib_unused __attribute__ ((__unused__))
|
158
|
+
#else
|
159
|
+
#define klib_unused
|
160
|
+
#endif
|
161
|
+
#endif /* klib_unused */
|
162
|
+
|
163
|
+
typedef khint32_t khint_t;
|
164
|
+
typedef khint_t khiter_t;
|
165
|
+
|
166
|
+
#define __ac_isempty(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&2)
|
167
|
+
#define __ac_isdel(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&1)
|
168
|
+
#define __ac_iseither(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&3)
|
169
|
+
#define __ac_set_isdel_false(flag, i) (flag[i>>4]&=~(1ul<<((i&0xfU)<<1)))
|
170
|
+
#define __ac_set_isempty_false(flag, i) (flag[i>>4]&=~(2ul<<((i&0xfU)<<1)))
|
171
|
+
#define __ac_set_isboth_false(flag, i) (flag[i>>4]&=~(3ul<<((i&0xfU)<<1)))
|
172
|
+
#define __ac_set_isdel_true(flag, i) (flag[i>>4]|=1ul<<((i&0xfU)<<1))
|
173
|
+
|
174
|
+
#define __ac_fsize(m) ((m) < 16? 1 : (m)>>4)
|
175
|
+
|
176
|
+
#ifndef kroundup32
|
177
|
+
#define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x))
|
178
|
+
#endif
|
179
|
+
|
180
|
+
static const double __ac_HASH_UPPER = 0.77;
|
181
|
+
|
182
|
+
#define __KHASH_TYPE(name, khkey_t, khval_t) \
|
183
|
+
typedef struct kh_##name##_s { \
|
184
|
+
khint_t n_buckets, size, n_occupied, upper_bound; \
|
185
|
+
khint32_t *flags; \
|
186
|
+
khkey_t *keys; \
|
187
|
+
khval_t *vals; \
|
188
|
+
} kh_##name##_t;
|
189
|
+
|
190
|
+
#define __KHASH_PROTOTYPES(name, khkey_t, khval_t) \
|
191
|
+
extern kh_##name##_t *kh_init_##name(void); \
|
192
|
+
extern void kh_destroy_##name(kh_##name##_t *h); \
|
193
|
+
extern void kh_clear_##name(kh_##name##_t *h); \
|
194
|
+
extern khint_t kh_get_##name(const kh_##name##_t *h, khkey_t key); \
|
195
|
+
extern int kh_resize_##name(kh_##name##_t *h, khint_t new_n_buckets); \
|
196
|
+
extern khint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret); \
|
197
|
+
extern void kh_del_##name(kh_##name##_t *h, khint_t x);
|
198
|
+
|
199
|
+
#define __KHASH_IMPL(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \
|
200
|
+
SCOPE kh_##name##_t *kh_init_##name(void) { \
|
201
|
+
return (kh_##name##_t*)kcalloc(0, 1, sizeof(kh_##name##_t)); \
|
202
|
+
} \
|
203
|
+
SCOPE void kh_destroy_##name(kh_##name##_t *h) \
|
204
|
+
{ \
|
205
|
+
if (h) { \
|
206
|
+
kfree(0, (void *)h->keys); kfree(0, h->flags); \
|
207
|
+
kfree(0, (void *)h->vals); \
|
208
|
+
kfree(0, h); \
|
209
|
+
} \
|
210
|
+
} \
|
211
|
+
SCOPE void kh_clear_##name(kh_##name##_t *h) \
|
212
|
+
{ \
|
213
|
+
if (h && h->flags) { \
|
214
|
+
memset(h->flags, 0xaa, __ac_fsize(h->n_buckets) * sizeof(khint32_t)); \
|
215
|
+
h->size = h->n_occupied = 0; \
|
216
|
+
} \
|
217
|
+
} \
|
218
|
+
SCOPE khint_t kh_get_##name(const kh_##name##_t *h, khkey_t key) \
|
219
|
+
{ \
|
220
|
+
if (h->n_buckets) { \
|
221
|
+
khint_t k, i, last, mask, step = 0; \
|
222
|
+
mask = h->n_buckets - 1; \
|
223
|
+
k = __hash_func(key); i = k & mask; \
|
224
|
+
last = i; \
|
225
|
+
while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \
|
226
|
+
i = (i + (++step)) & mask; \
|
227
|
+
if (i == last) return h->n_buckets; \
|
228
|
+
} \
|
229
|
+
return __ac_iseither(h->flags, i)? h->n_buckets : i; \
|
230
|
+
} else return 0; \
|
231
|
+
} \
|
232
|
+
SCOPE int kh_resize_##name(kh_##name##_t *h, khint_t new_n_buckets) \
|
233
|
+
{ /* This function uses 0.25*n_buckets bytes of working space instead of [sizeof(key_t+val_t)+.25]*n_buckets. */ \
|
234
|
+
khint32_t *new_flags = 0; \
|
235
|
+
khint_t j = 1; \
|
236
|
+
{ \
|
237
|
+
kroundup32(new_n_buckets); \
|
238
|
+
if (new_n_buckets < 4) new_n_buckets = 4; \
|
239
|
+
if (h->size >= (khint_t)(new_n_buckets * __ac_HASH_UPPER + 0.5)) j = 0; /* requested size is too small */ \
|
240
|
+
else { /* hash table size to be changed (shrink or expand); rehash */ \
|
241
|
+
new_flags = (khint32_t*)kmalloc(0, __ac_fsize(new_n_buckets) * sizeof(khint32_t)); \
|
242
|
+
if (!new_flags) return -1; \
|
243
|
+
memset(new_flags, 0xaa, __ac_fsize(new_n_buckets) * sizeof(khint32_t)); \
|
244
|
+
if (h->n_buckets < new_n_buckets) { /* expand */ \
|
245
|
+
khkey_t *new_keys = (khkey_t*)krealloc(0, (void *)h->keys, new_n_buckets * sizeof(khkey_t)); \
|
246
|
+
if (!new_keys) { kfree(0, new_flags); return -1; } \
|
247
|
+
h->keys = new_keys; \
|
248
|
+
if (kh_is_map) { \
|
249
|
+
khval_t *new_vals = (khval_t*)krealloc(0, (void *)h->vals, new_n_buckets * sizeof(khval_t)); \
|
250
|
+
if (!new_vals) { kfree(0, new_flags); return -1; } \
|
251
|
+
h->vals = new_vals; \
|
252
|
+
} \
|
253
|
+
} /* otherwise shrink */ \
|
254
|
+
} \
|
255
|
+
} \
|
256
|
+
if (j) { /* rehashing is needed */ \
|
257
|
+
for (j = 0; j != h->n_buckets; ++j) { \
|
258
|
+
if (__ac_iseither(h->flags, j) == 0) { \
|
259
|
+
khkey_t key = h->keys[j]; \
|
260
|
+
khval_t val; \
|
261
|
+
khint_t new_mask; \
|
262
|
+
new_mask = new_n_buckets - 1; \
|
263
|
+
if (kh_is_map) val = h->vals[j]; \
|
264
|
+
__ac_set_isdel_true(h->flags, j); \
|
265
|
+
while (1) { /* kick-out process; sort of like in Cuckoo hashing */ \
|
266
|
+
khint_t k, i, step = 0; \
|
267
|
+
k = __hash_func(key); \
|
268
|
+
i = k & new_mask; \
|
269
|
+
while (!__ac_isempty(new_flags, i)) i = (i + (++step)) & new_mask; \
|
270
|
+
__ac_set_isempty_false(new_flags, i); \
|
271
|
+
if (i < h->n_buckets && __ac_iseither(h->flags, i) == 0) { /* kick out the existing element */ \
|
272
|
+
{ khkey_t tmp = h->keys[i]; h->keys[i] = key; key = tmp; } \
|
273
|
+
if (kh_is_map) { khval_t tmp = h->vals[i]; h->vals[i] = val; val = tmp; } \
|
274
|
+
__ac_set_isdel_true(h->flags, i); /* mark it as deleted in the old hash table */ \
|
275
|
+
} else { /* write the element and jump out of the loop */ \
|
276
|
+
h->keys[i] = key; \
|
277
|
+
if (kh_is_map) h->vals[i] = val; \
|
278
|
+
break; \
|
279
|
+
} \
|
280
|
+
} \
|
281
|
+
} \
|
282
|
+
} \
|
283
|
+
if (h->n_buckets > new_n_buckets) { /* shrink the hash table */ \
|
284
|
+
h->keys = (khkey_t*)krealloc(0, (void *)h->keys, new_n_buckets * sizeof(khkey_t)); \
|
285
|
+
if (kh_is_map) h->vals = (khval_t*)krealloc(0, (void *)h->vals, new_n_buckets * sizeof(khval_t)); \
|
286
|
+
} \
|
287
|
+
kfree(0, h->flags); /* free the working space */ \
|
288
|
+
h->flags = new_flags; \
|
289
|
+
h->n_buckets = new_n_buckets; \
|
290
|
+
h->n_occupied = h->size; \
|
291
|
+
h->upper_bound = (khint_t)(h->n_buckets * __ac_HASH_UPPER + 0.5); \
|
292
|
+
} \
|
293
|
+
return 0; \
|
294
|
+
} \
|
295
|
+
SCOPE khint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret) \
|
296
|
+
{ \
|
297
|
+
khint_t x; \
|
298
|
+
if (h->n_occupied >= h->upper_bound) { /* update the hash table */ \
|
299
|
+
if (h->n_buckets > (h->size<<1)) { \
|
300
|
+
if (kh_resize_##name(h, h->n_buckets - 1) < 0) { /* clear "deleted" elements */ \
|
301
|
+
*ret = -1; return h->n_buckets; \
|
302
|
+
} \
|
303
|
+
} else if (kh_resize_##name(h, h->n_buckets + 1) < 0) { /* expand the hash table */ \
|
304
|
+
*ret = -1; return h->n_buckets; \
|
305
|
+
} \
|
306
|
+
} /* TODO: to implement automatically shrinking; resize() already support shrinking */ \
|
307
|
+
{ \
|
308
|
+
khint_t k, i, site, last, mask = h->n_buckets - 1, step = 0; \
|
309
|
+
x = site = h->n_buckets; k = __hash_func(key); i = k & mask; \
|
310
|
+
if (__ac_isempty(h->flags, i)) x = i; /* for speed up */ \
|
311
|
+
else { \
|
312
|
+
last = i; \
|
313
|
+
while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \
|
314
|
+
if (__ac_isdel(h->flags, i)) site = i; \
|
315
|
+
i = (i + (++step)) & mask; \
|
316
|
+
if (i == last) { x = site; break; } \
|
317
|
+
} \
|
318
|
+
if (x == h->n_buckets) { \
|
319
|
+
if (__ac_isempty(h->flags, i) && site != h->n_buckets) x = site; \
|
320
|
+
else x = i; \
|
321
|
+
} \
|
322
|
+
} \
|
323
|
+
} \
|
324
|
+
if (__ac_isempty(h->flags, x)) { /* not present at all */ \
|
325
|
+
h->keys[x] = key; \
|
326
|
+
__ac_set_isboth_false(h->flags, x); \
|
327
|
+
++h->size; ++h->n_occupied; \
|
328
|
+
*ret = 1; \
|
329
|
+
} else if (__ac_isdel(h->flags, x)) { /* deleted */ \
|
330
|
+
h->keys[x] = key; \
|
331
|
+
__ac_set_isboth_false(h->flags, x); \
|
332
|
+
++h->size; \
|
333
|
+
*ret = 2; \
|
334
|
+
} else *ret = 0; /* Don't touch h->keys[x] if present and not deleted */ \
|
335
|
+
return x; \
|
336
|
+
} \
|
337
|
+
SCOPE void kh_del_##name(kh_##name##_t *h, khint_t x) \
|
338
|
+
{ \
|
339
|
+
if (x != h->n_buckets && !__ac_iseither(h->flags, x)) { \
|
340
|
+
__ac_set_isdel_true(h->flags, x); \
|
341
|
+
--h->size; \
|
342
|
+
} \
|
343
|
+
}
|
344
|
+
|
345
|
+
#define KHASH_DECLARE(name, khkey_t, khval_t) \
|
346
|
+
__KHASH_TYPE(name, khkey_t, khval_t) \
|
347
|
+
__KHASH_PROTOTYPES(name, khkey_t, khval_t)
|
348
|
+
|
349
|
+
#define KHASH_INIT2(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \
|
350
|
+
__KHASH_TYPE(name, khkey_t, khval_t) \
|
351
|
+
__KHASH_IMPL(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal)
|
352
|
+
|
353
|
+
#define KHASH_INIT(name, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \
|
354
|
+
KHASH_INIT2(name, static kh_inline klib_unused, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal)
|
355
|
+
|
356
|
+
/* --- BEGIN OF HASH FUNCTIONS --- */
|
357
|
+
|
358
|
+
/*! @function
|
359
|
+
@abstract Integer hash function
|
360
|
+
@param key The integer [khint32_t]
|
361
|
+
@return The hash value [khint_t]
|
362
|
+
*/
|
363
|
+
#define kh_int_hash_func(key) (khint32_t)(key)
|
364
|
+
/*! @function
|
365
|
+
@abstract Integer comparison function
|
366
|
+
*/
|
367
|
+
#define kh_int_hash_equal(a, b) ((a) == (b))
|
368
|
+
/*! @function
|
369
|
+
@abstract 64-bit integer hash function
|
370
|
+
@param key The integer [khint64_t]
|
371
|
+
@return The hash value [khint_t]
|
372
|
+
*/
|
373
|
+
#define kh_int64_hash_func(key) (khint32_t)((key)>>33^(key)^(key)<<11)
|
374
|
+
/*! @function
|
375
|
+
@abstract 64-bit integer comparison function
|
376
|
+
*/
|
377
|
+
#define kh_int64_hash_equal(a, b) ((a) == (b))
|
378
|
+
/*! @function
|
379
|
+
@abstract const char* hash function
|
380
|
+
@param s Pointer to a null terminated string
|
381
|
+
@return The hash value
|
382
|
+
*/
|
383
|
+
static kh_inline khint_t __ac_X31_hash_string(const char *s)
|
384
|
+
{
|
385
|
+
khint_t h = (khint_t)*s;
|
386
|
+
if (h) for (++s ; *s; ++s) h = (h << 5) - h + (khint_t)*s;
|
387
|
+
return h;
|
388
|
+
}
|
389
|
+
/*! @function
|
390
|
+
@abstract Another interface to const char* hash function
|
391
|
+
@param key Pointer to a null terminated string [const char*]
|
392
|
+
@return The hash value [khint_t]
|
393
|
+
*/
|
394
|
+
#define kh_str_hash_func(key) __ac_X31_hash_string(key)
|
395
|
+
/*! @function
|
396
|
+
@abstract Const char* comparison function
|
397
|
+
*/
|
398
|
+
#define kh_str_hash_equal(a, b) (strcmp(a, b) == 0)
|
399
|
+
|
400
|
+
static kh_inline khint_t __ac_Wang_hash(khint_t key)
|
401
|
+
{
|
402
|
+
key += ~(key << 15);
|
403
|
+
key ^= (key >> 10);
|
404
|
+
key += (key << 3);
|
405
|
+
key ^= (key >> 6);
|
406
|
+
key += ~(key << 11);
|
407
|
+
key ^= (key >> 16);
|
408
|
+
return key;
|
409
|
+
}
|
410
|
+
#define kh_int_hash_func2(key) __ac_Wang_hash((khint_t)key)
|
411
|
+
|
412
|
+
/* --- END OF HASH FUNCTIONS --- */
|
413
|
+
|
414
|
+
/* Other convenient macros... */
|
415
|
+
|
416
|
+
/*!
|
417
|
+
@abstract Type of the hash table.
|
418
|
+
@param name Name of the hash table [symbol]
|
419
|
+
*/
|
420
|
+
#define khash_t(name) kh_##name##_t
|
421
|
+
|
422
|
+
/*! @function
|
423
|
+
@abstract Initiate a hash table.
|
424
|
+
@param name Name of the hash table [symbol]
|
425
|
+
@return Pointer to the hash table [khash_t(name)*]
|
426
|
+
*/
|
427
|
+
#define kh_init(name) kh_init_##name()
|
428
|
+
|
429
|
+
/*! @function
|
430
|
+
@abstract Destroy a hash table.
|
431
|
+
@param name Name of the hash table [symbol]
|
432
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
433
|
+
*/
|
434
|
+
#define kh_destroy(name, h) kh_destroy_##name(h)
|
435
|
+
|
436
|
+
/*! @function
|
437
|
+
@abstract Reset a hash table without deallocating memory.
|
438
|
+
@param name Name of the hash table [symbol]
|
439
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
440
|
+
*/
|
441
|
+
#define kh_clear(name, h) kh_clear_##name(h)
|
442
|
+
|
443
|
+
/*! @function
|
444
|
+
@abstract Resize a hash table.
|
445
|
+
@param name Name of the hash table [symbol]
|
446
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
447
|
+
@param s New size [khint_t]
|
448
|
+
*/
|
449
|
+
#define kh_resize(name, h, s) kh_resize_##name(h, s)
|
450
|
+
|
451
|
+
/*! @function
|
452
|
+
@abstract Insert a key to the hash table.
|
453
|
+
@param name Name of the hash table [symbol]
|
454
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
455
|
+
@param k Key [type of keys]
|
456
|
+
@param r Extra return code: -1 if the operation failed;
|
457
|
+
0 if the key is present in the hash table;
|
458
|
+
1 if the bucket is empty (never used); 2 if the element in
|
459
|
+
the bucket has been deleted [int*]
|
460
|
+
@return Iterator to the inserted element [khint_t]
|
461
|
+
*/
|
462
|
+
#define kh_put(name, h, k, r) kh_put_##name(h, k, r)
|
463
|
+
|
464
|
+
/*! @function
|
465
|
+
@abstract Retrieve a key from the hash table.
|
466
|
+
@param name Name of the hash table [symbol]
|
467
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
468
|
+
@param k Key [type of keys]
|
469
|
+
@return Iterator to the found element, or kh_end(h) if the element is absent [khint_t]
|
470
|
+
*/
|
471
|
+
#define kh_get(name, h, k) kh_get_##name(h, k)
|
472
|
+
|
473
|
+
/*! @function
|
474
|
+
@abstract Remove a key from the hash table.
|
475
|
+
@param name Name of the hash table [symbol]
|
476
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
477
|
+
@param k Iterator to the element to be deleted [khint_t]
|
478
|
+
*/
|
479
|
+
#define kh_del(name, h, k) kh_del_##name(h, k)
|
480
|
+
|
481
|
+
/*! @function
|
482
|
+
@abstract Test whether a bucket contains data.
|
483
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
484
|
+
@param x Iterator to the bucket [khint_t]
|
485
|
+
@return 1 if containing data; 0 otherwise [int]
|
486
|
+
*/
|
487
|
+
#define kh_exist(h, x) (!__ac_iseither((h)->flags, (x)))
|
488
|
+
|
489
|
+
/*! @function
|
490
|
+
@abstract Get key given an iterator
|
491
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
492
|
+
@param x Iterator to the bucket [khint_t]
|
493
|
+
@return Key [type of keys]
|
494
|
+
*/
|
495
|
+
#define kh_key(h, x) ((h)->keys[x])
|
496
|
+
|
497
|
+
/*! @function
|
498
|
+
@abstract Get value given an iterator
|
499
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
500
|
+
@param x Iterator to the bucket [khint_t]
|
501
|
+
@return Value [type of values]
|
502
|
+
@discussion For hash sets, calling this results in segfault.
|
503
|
+
*/
|
504
|
+
#define kh_val(h, x) ((h)->vals[x])
|
505
|
+
|
506
|
+
/*! @function
|
507
|
+
@abstract Alias of kh_val()
|
508
|
+
*/
|
509
|
+
#define kh_value(h, x) ((h)->vals[x])
|
510
|
+
|
511
|
+
/*! @function
|
512
|
+
@abstract Get the start iterator
|
513
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
514
|
+
@return The start iterator [khint_t]
|
515
|
+
*/
|
516
|
+
#define kh_begin(h) (khint_t)(0)
|
517
|
+
|
518
|
+
/*! @function
|
519
|
+
@abstract Get the end iterator
|
520
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
521
|
+
@return The end iterator [khint_t]
|
522
|
+
*/
|
523
|
+
#define kh_end(h) ((h)->n_buckets)
|
524
|
+
|
525
|
+
/*! @function
|
526
|
+
@abstract Get the number of elements in the hash table
|
527
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
528
|
+
@return Number of elements in the hash table [khint_t]
|
529
|
+
*/
|
530
|
+
#define kh_size(h) ((h)->size)
|
531
|
+
|
532
|
+
/*! @function
|
533
|
+
@abstract Get the number of buckets in the hash table
|
534
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
535
|
+
@return Number of buckets in the hash table [khint_t]
|
536
|
+
*/
|
537
|
+
#define kh_n_buckets(h) ((h)->n_buckets)
|
538
|
+
|
539
|
+
/*! @function
|
540
|
+
@abstract Iterate over the entries in the hash table
|
541
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
542
|
+
@param kvar Variable to which key will be assigned
|
543
|
+
@param vvar Variable to which value will be assigned
|
544
|
+
@param code Block of code to execute
|
545
|
+
*/
|
546
|
+
#define kh_foreach(h, kvar, vvar, code) { khint_t __i; \
|
547
|
+
for (__i = kh_begin(h); __i != kh_end(h); ++__i) { \
|
548
|
+
if (!kh_exist(h,__i)) continue; \
|
549
|
+
(kvar) = kh_key(h,__i); \
|
550
|
+
(vvar) = kh_val(h,__i); \
|
551
|
+
code; \
|
552
|
+
} }
|
553
|
+
|
554
|
+
/*! @function
|
555
|
+
@abstract Iterate over the values in the hash table
|
556
|
+
@param h Pointer to the hash table [khash_t(name)*]
|
557
|
+
@param vvar Variable to which value will be assigned
|
558
|
+
@param code Block of code to execute
|
559
|
+
*/
|
560
|
+
#define kh_foreach_value(h, vvar, code) { khint_t __i; \
|
561
|
+
for (__i = kh_begin(h); __i != kh_end(h); ++__i) { \
|
562
|
+
if (!kh_exist(h,__i)) continue; \
|
563
|
+
(vvar) = kh_val(h,__i); \
|
564
|
+
code; \
|
565
|
+
} }
|
566
|
+
|
567
|
+
/* More conenient interfaces */
|
568
|
+
|
569
|
+
/*! @function
|
570
|
+
@abstract Instantiate a hash set containing integer keys
|
571
|
+
@param name Name of the hash table [symbol]
|
572
|
+
*/
|
573
|
+
#define KHASH_SET_INIT_INT(name) \
|
574
|
+
KHASH_INIT(name, khint32_t, char, 0, kh_int_hash_func, kh_int_hash_equal)
|
575
|
+
|
576
|
+
/*! @function
|
577
|
+
@abstract Instantiate a hash map containing integer keys
|
578
|
+
@param name Name of the hash table [symbol]
|
579
|
+
@param khval_t Type of values [type]
|
580
|
+
*/
|
581
|
+
#define KHASH_MAP_INIT_INT(name, khval_t) \
|
582
|
+
KHASH_INIT(name, khint32_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal)
|
583
|
+
|
584
|
+
/*! @function
|
585
|
+
@abstract Instantiate a hash map containing 64-bit integer keys
|
586
|
+
@param name Name of the hash table [symbol]
|
587
|
+
*/
|
588
|
+
#define KHASH_SET_INIT_INT64(name) \
|
589
|
+
KHASH_INIT(name, khint64_t, char, 0, kh_int64_hash_func, kh_int64_hash_equal)
|
590
|
+
|
591
|
+
/*! @function
|
592
|
+
@abstract Instantiate a hash map containing 64-bit integer keys
|
593
|
+
@param name Name of the hash table [symbol]
|
594
|
+
@param khval_t Type of values [type]
|
595
|
+
*/
|
596
|
+
#define KHASH_MAP_INIT_INT64(name, khval_t) \
|
597
|
+
KHASH_INIT(name, khint64_t, khval_t, 1, kh_int64_hash_func, kh_int64_hash_equal)
|
598
|
+
|
599
|
+
typedef const char *kh_cstr_t;
|
600
|
+
/*! @function
|
601
|
+
@abstract Instantiate a hash map containing const char* keys
|
602
|
+
@param name Name of the hash table [symbol]
|
603
|
+
*/
|
604
|
+
#define KHASH_SET_INIT_STR(name) \
|
605
|
+
KHASH_INIT(name, kh_cstr_t, char, 0, kh_str_hash_func, kh_str_hash_equal)
|
606
|
+
|
607
|
+
/*! @function
|
608
|
+
@abstract Instantiate a hash map containing const char* keys
|
609
|
+
@param name Name of the hash table [symbol]
|
610
|
+
@param khval_t Type of values [type]
|
611
|
+
*/
|
612
|
+
#define KHASH_MAP_INIT_STR(name, khval_t) \
|
613
|
+
KHASH_INIT(name, kh_cstr_t, khval_t, 1, kh_str_hash_func, kh_str_hash_equal)
|
614
|
+
|
615
|
+
#endif /* __AC_KHASH_H */
|