Skip to content

Commit 8568708

Browse files
scwattsscwatts
committed
Hello
0 parents  commit 8568708

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

91 files changed

+45612
-0
lines changed

‎awasm/Makefile

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,6 @@
1+
all:
2+
$(MAKE) -C src
3+
4+
clean:
5+
$(MAKE) -C src clean
6+
$(MAKE) -C lib/velvet clean

‎awasm/include/disk_streamer.h

Lines changed: 46 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,46 @@
1+
#ifndef __DISK_STREAMER_H__
2+
#define __DISK_STREAMER_H__
3+
4+
5+
#include <assert.h>
6+
#include <stdlib.h>
7+
#include <stdio.h>
8+
#include <string.h>
9+
10+
11+
#include "kseq.h"
12+
#include "zlib.h"
13+
14+
15+
#include <emscripten.h>
16+
17+
18+
#include "utility.h"
19+
20+
21+
#define BUFFER_SZ 1024 * 512
22+
23+
24+
// Defined in external.js
25+
extern void read_input_file_slice(int slice_start, int slice_end, char *reader_name);
26+
27+
28+
struct FileReader {
29+
char *name;
30+
size_t position;
31+
size_t buffer_sz;
32+
size_t file_sz;
33+
unsigned char *buffer_shared;
34+
z_stream *decomper;
35+
};
36+
37+
38+
int inflate_slice(z_stream *stream);
39+
int get_reader_data(char *reader_name, char *property);
40+
struct FileReader * init_file_reader(char *reader_name, size_t buffer_sz);
41+
int destroy_file_reader(struct FileReader *file_reader);
42+
int get_slice(struct FileReader *file_reader);
43+
int read_data(struct FileReader *file_reader, void *buffer, unsigned int length);
44+
45+
46+
#endif

‎awasm/include/kseq.h

Lines changed: 242 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,242 @@
1+
/* The MIT License
2+
3+
Copyright (c) 2008, 2009, 2011 Attractive Chaos <[email protected]>
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+
/* Last Modified: 05MAR2012 */
27+
28+
#ifndef AC_KSEQ_H
29+
#define AC_KSEQ_H
30+
31+
#include <ctype.h>
32+
#include <string.h>
33+
#include <stdlib.h>
34+
35+
#define KS_SEP_SPACE 0 // isspace(): \t, \n, \v, \f, \r
36+
#define KS_SEP_TAB 1 // isspace() && !' '
37+
#define KS_SEP_LINE 2 // line separator: "\n" (Unix) or "\r\n" (Windows)
38+
#define KS_SEP_MAX 2
39+
40+
#define __KS_TYPE(type_t) \
41+
typedef struct __kstream_t { \
42+
unsigned char *buf; \
43+
int begin, end, is_eof; \
44+
type_t f; \
45+
} kstream_t;
46+
47+
#define ks_err(ks) ((ks)->end == -1)
48+
#define ks_eof(ks) ((ks)->is_eof && (ks)->begin >= (ks)->end)
49+
#define ks_rewind(ks) ((ks)->is_eof = (ks)->begin = (ks)->end = 0)
50+
51+
#define __KS_BASIC(type_t, __bufsize) \
52+
static inline kstream_t *ks_init(type_t f) \
53+
{ \
54+
kstream_t *ks = (kstream_t*)calloc(1, sizeof(kstream_t)); \
55+
ks->f = f; \
56+
ks->buf = (unsigned char*)malloc(__bufsize); \
57+
return ks; \
58+
} \
59+
static inline void ks_destroy(kstream_t *ks) \
60+
{ \
61+
if (ks) { \
62+
free(ks->buf); \
63+
free(ks); \
64+
} \
65+
}
66+
67+
#define __KS_GETC(__read, __bufsize) \
68+
static inline int ks_getc(kstream_t *ks) \
69+
{ \
70+
if (ks_err(ks)) return -3; \
71+
if (ks->is_eof && ks->begin >= ks->end) return -1; \
72+
if (ks->begin >= ks->end) { \
73+
ks->begin = 0; \
74+
ks->end = __read(&ks->f, ks->buf, __bufsize); \
75+
if (ks->end == 0) { ks->is_eof = 1; return -1;} \
76+
if (ks->end == -1) { ks->is_eof = 1; return -3;}\
77+
} \
78+
return (int)ks->buf[ks->begin++]; \
79+
}
80+
81+
#ifndef KSTRING_T
82+
#define KSTRING_T kstring_t
83+
typedef struct __kstring_t {
84+
size_t l, m;
85+
char *s;
86+
} kstring_t;
87+
#endif
88+
89+
#ifndef kroundup32
90+
#define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x))
91+
#endif
92+
93+
#define __KS_GETUNTIL(__read, __bufsize) \
94+
static int ks_getuntil2(kstream_t *ks, int delimiter, kstring_t *str, int *dret, int append) \
95+
{ \
96+
int gotany = 0; \
97+
if (dret) *dret = 0; \
98+
str->l = append? str->l : 0; \
99+
for (;;) { \
100+
int i; \
101+
if (ks_err(ks)) return -3; \
102+
if (ks->begin >= ks->end) { \
103+
if (!ks->is_eof) { \
104+
ks->begin = 0; \
105+
ks->end = __read(&ks->f, ks->buf, __bufsize); \
106+
if (ks->end == 0) { ks->is_eof = 1; break; } \
107+
if (ks->end == -1) { ks->is_eof = 1; return -3; } \
108+
} else break; \
109+
} \
110+
if (delimiter == KS_SEP_LINE) { \
111+
for (i = ks->begin; i < ks->end; ++i) \
112+
if (ks->buf[i] == '\n') break; \
113+
} else if (delimiter > KS_SEP_MAX) { \
114+
for (i = ks->begin; i < ks->end; ++i) \
115+
if (ks->buf[i] == delimiter) break; \
116+
} else if (delimiter == KS_SEP_SPACE) { \
117+
for (i = ks->begin; i < ks->end; ++i) \
118+
if (isspace(ks->buf[i])) break; \
119+
} else if (delimiter == KS_SEP_TAB) { \
120+
for (i = ks->begin; i < ks->end; ++i) \
121+
if (isspace(ks->buf[i]) && ks->buf[i] != ' ') break; \
122+
} else i = 0; /* never come to here! */ \
123+
if (str->m - str->l < (size_t)(i - ks->begin + 1)) { \
124+
str->m = str->l + (i - ks->begin) + 1; \
125+
kroundup32(str->m); \
126+
str->s = (char*)realloc(str->s, str->m); \
127+
} \
128+
gotany = 1; \
129+
memcpy(str->s + str->l, ks->buf + ks->begin, i - ks->begin); \
130+
str->l = str->l + (i - ks->begin); \
131+
ks->begin = i + 1; \
132+
if (i < ks->end) { \
133+
if (dret) *dret = ks->buf[i]; \
134+
break; \
135+
} \
136+
} \
137+
if (!gotany && ks_eof(ks)) return -1; \
138+
if (str->s == 0) { \
139+
str->m = 1; \
140+
str->s = (char*)calloc(1, 1); \
141+
} else if (delimiter == KS_SEP_LINE && str->l > 1 && str->s[str->l-1] == '\r') --str->l; \
142+
str->s[str->l] = '\0'; \
143+
return str->l; \
144+
} \
145+
static inline int ks_getuntil(kstream_t *ks, int delimiter, kstring_t *str, int *dret) \
146+
{ return ks_getuntil2(ks, delimiter, str, dret, 0); }
147+
148+
#define KSTREAM_INIT(type_t, __read, __bufsize) \
149+
__KS_TYPE(type_t) \
150+
__KS_BASIC(type_t, __bufsize) \
151+
__KS_GETC(__read, __bufsize) \
152+
__KS_GETUNTIL(__read, __bufsize)
153+
154+
#define kseq_rewind(ks) ((ks)->last_char = (ks)->f->is_eof = (ks)->f->begin = (ks)->f->end = 0)
155+
156+
#define __KSEQ_BASIC(SCOPE, type_t) \
157+
SCOPE kseq_t *kseq_init(type_t fd) \
158+
{ \
159+
kseq_t *s = (kseq_t*)calloc(1, sizeof(kseq_t)); \
160+
s->f = ks_init(fd); \
161+
return s; \
162+
} \
163+
SCOPE void kseq_destroy(kseq_t *ks) \
164+
{ \
165+
if (!ks) return; \
166+
free(ks->name.s); free(ks->comment.s); free(ks->seq.s); free(ks->qual.s); \
167+
ks_destroy(ks->f); \
168+
free(ks); \
169+
}
170+
171+
/* Return value:
172+
>=0 length of the sequence (normal)
173+
-1 end-of-file
174+
-2 truncated quality string
175+
-3 error reading stream
176+
*/
177+
#define __KSEQ_READ(SCOPE) \
178+
SCOPE int kseq_read(kseq_t *seq) \
179+
{ \
180+
int c,r; \
181+
kstream_t *ks = seq->f; \
182+
if (seq->last_char == 0) { /* then jump to the next header line */ \
183+
while ((c = ks_getc(ks)) >= 0 && c != '>' && c != '@'); \
184+
if (c < 0) return c; /* end of file or error*/ \
185+
seq->last_char = c; \
186+
} /* else: the first header char has been read in the previous call */ \
187+
seq->comment.l = seq->seq.l = seq->qual.l = 0; /* reset all members */ \
188+
if ((r=ks_getuntil(ks, 0, &seq->name, &c)) < 0) return r; /* normal exit: EOF or error */ \
189+
if (c != '\n') ks_getuntil(ks, KS_SEP_LINE, &seq->comment, 0); /* read FASTA/Q comment */ \
190+
if (seq->seq.s == 0) { /* we can do this in the loop below, but that is slower */ \
191+
seq->seq.m = 256; \
192+
seq->seq.s = (char*)malloc(seq->seq.m); \
193+
} \
194+
while ((c = ks_getc(ks)) >= 0 && c != '>' && c != '+' && c != '@') { \
195+
if (c == '\n') continue; /* skip empty lines */ \
196+
seq->seq.s[seq->seq.l++] = c; /* this is safe: we always have enough space for 1 char */ \
197+
ks_getuntil2(ks, KS_SEP_LINE, &seq->seq, 0, 1); /* read the rest of the line */ \
198+
} \
199+
if (c == '>' || c == '@') seq->last_char = c; /* the first header char has been read */ \
200+
if (seq->seq.l + 1 >= seq->seq.m) { /* seq->seq.s[seq->seq.l] below may be out of boundary */ \
201+
seq->seq.m = seq->seq.l + 2; \
202+
kroundup32(seq->seq.m); /* rounded to the next closest 2^k */ \
203+
seq->seq.s = (char*)realloc(seq->seq.s, seq->seq.m); \
204+
} \
205+
seq->seq.s[seq->seq.l] = 0; /* null terminated string */ \
206+
if (c != '+') return seq->seq.l; /* FASTA */ \
207+
if (seq->qual.m < seq->seq.m) { /* allocate memory for qual in case insufficient */ \
208+
seq->qual.m = seq->seq.m; \
209+
seq->qual.s = (char*)realloc(seq->qual.s, seq->qual.m); \
210+
} \
211+
while ((c = ks_getc(ks)) >= 0 && c != '\n'); /* skip the rest of '+' line */ \
212+
if (c == -1) return -2; /* error: no quality string */ \
213+
while ((c = ks_getuntil2(ks, KS_SEP_LINE, &seq->qual, 0, 1) >= 0 && seq->qual.l < seq->seq.l)); \
214+
if (c == -3) return -3; /* stream error */ \
215+
seq->last_char = 0; /* we have not come to the next header line */ \
216+
if (seq->seq.l != seq->qual.l) return -2; /* error: qual string is of a different length */ \
217+
return seq->seq.l; \
218+
}
219+
220+
#define __KSEQ_TYPE(type_t) \
221+
typedef struct { \
222+
kstring_t name, comment, seq, qual; \
223+
int last_char; \
224+
kstream_t *f; \
225+
} kseq_t;
226+
227+
#define KSEQ_INIT2(SCOPE, type_t, __read) \
228+
KSTREAM_INIT(type_t, __read, 16384) \
229+
__KSEQ_TYPE(type_t) \
230+
__KSEQ_BASIC(SCOPE, type_t) \
231+
__KSEQ_READ(SCOPE)
232+
233+
#define KSEQ_INIT(type_t, __read) KSEQ_INIT2(static, type_t, __read)
234+
235+
#define KSEQ_DECLARE(type_t) \
236+
__KS_TYPE(type_t) \
237+
__KSEQ_TYPE(type_t) \
238+
extern kseq_t *kseq_init(type_t fd); \
239+
void kseq_destroy(kseq_t *ks); \
240+
int kseq_read(kseq_t *seq);
241+
242+
#endif

‎awasm/include/utility.h

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
#ifndef __UTILITY_H__
2+
#define __UTILITY_H__
3+
4+
5+
#include <dirent.h>
6+
#include <errno.h>
7+
8+
9+
#include <emscripten.h>
10+
11+
12+
void fssync_block();
13+
int setup_idbfs();
14+
15+
16+
#endif

‎awasm/include/velvetg.h

Lines changed: 23 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,23 @@
1+
#ifndef __VELVETG_H__
2+
#define __VELVETG_H__
3+
4+
5+
#include <stdlib.h>
6+
#include <stdio.h>
7+
#include <string.h>
8+
#include <unistd.h>
9+
10+
#include "velvet/src/run.h"
11+
12+
#include "velvet/src/binarySequences.h"
13+
#include "velvet/src/globals.h"
14+
15+
// Defined using compiler flags
16+
#define MAXKMERLENGTH 31
17+
#define CATEGORIES 2
18+
19+
20+
void velvetg(char *prefix);
21+
22+
23+
#endif

‎awasm/include/velveth.h

Lines changed: 37 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,37 @@
1+
#ifndef __VELVETH_H__
2+
#define __VELVETH_H__
3+
4+
5+
#include <dirent.h>
6+
#include <sys/stat.h>
7+
8+
9+
#include "zlib.h"
10+
#include "kseq.h"
11+
12+
13+
#include "velvet/src/run.h"
14+
#include "velvet/src/autoOpen.h"
15+
16+
17+
#include "disk_streamer.h"
18+
19+
20+
// Defined in velvet/src/readSet.c
21+
typedef struct {
22+
gzFile gzFile;
23+
AutoFile *autoFile;
24+
} FileGZOrAuto;
25+
26+
27+
// Defined in velvet/src/readSet.c
28+
extern void writeSeqName(char*seq_name, SequencesWriter *seqWriteInfo, Category cat, IDnum *sequenceIndex);
29+
extern void writeSequence(char*seq, SequencesWriter *seqWriteInfo);
30+
extern void initFastX(SequencesWriter *seqWriteInfo, Category cat);
31+
extern void cleanupFastX(SequencesWriter *seqWriteInfo, Category cat);
32+
33+
34+
void velveth(char *prefix);
35+
36+
37+
#endif

0 commit comments

Comments
 (0)