char *faidx_fetch_seq(const faidx_t *fai, char *c_name, int p_beg_i, int p_end_i, int *len)
{
int l;
char c;
khiter_t iter;
faidx1_t val;
char *seq=NULL;
// Adjust position
iter = kh_get(s, fai->hash, c_name);
if(iter == kh_end(fai->hash)) return 0;
val = kh_value(fai->hash, iter);
if(p_end_i < p_beg_i) p_beg_i = p_end_i;
if(p_beg_i < 0) p_beg_i = 0;
else if(val.len <= p_beg_i) p_beg_i = val.len - 1;
if(p_end_i < 0) p_end_i = 0;
else if(val.len <= p_end_i) p_end_i = val.len - 1;
// Now retrieve the sequence
l = 0;
seq = (char*)malloc(p_end_i - p_beg_i + 2);
razf_seek(fai->rz, val.offset + p_beg_i / val.line_blen * val.line_len + p_beg_i % val.line_blen, SEEK_SET);
while (razf_read(fai->rz, &c, 1) == 1 && l < p_end_i - p_beg_i + 1)
if (isgraph(c)) seq[l++] = c;
seq[l] = '\0';
*len = l;
return seq;
}
char *fai_fetch(const faidx_t *fai, const char *str, int *len)
{
char *s, *p, c;
int i, l, k;
khiter_t iter;
faidx1_t val;
khash_t(s) *h;
int beg, end;
beg = end = -1;
h = fai->hash;
l = strlen(str);
p = s = (char*)malloc(l+1);
/* squeeze out "," */
for (i = k = 0; i != l; ++i)
if (str[i] != ',' && !isspace(str[i])) s[k++] = str[i];
s[k] = 0;
for (i = 0; i != k; ++i) if (s[i] == ':') break;
s[i] = 0;
iter = kh_get(s, h, s); /* get the ref_id */
if (iter == kh_end(h)) {
*len = 0;
fprintf(stderr,"faidx fetch %s failed",str);
free(s); return 0;
}
val = kh_value(h, iter);
if (i == k) { /* dump the whole sequence */
beg = 0; end = val.len;
} else {
for (p = s + i + 1; i != k; ++i) if (s[i] == '-') break;
beg = atoi(p);
if (i < k) {
p = s + i + 1;
end = atoi(p);
} else end = val.len;
}
if (beg > 0) --beg;
if (beg >= val.len) beg = val.len;
if (end >= val.len) end = val.len;
if (beg > end) beg = end;
free(s);
// now retrieve the sequence
l = 0;
s = (char*)malloc(end - beg + 2);
razf_seek(fai->rz, val.offset + beg / val.line_blen * val.line_len + beg % val.line_blen, SEEK_SET);
while (razf_read(fai->rz, &c, 1) == 1 && l < end - beg && !fai->rz->z_err)
if (isgraph(c)) s[l++] = c;
s[l] = '\0';
*len = l;
return s;
}
char* faidx_fetch_seq_forced_lower( const faidx_t* fai, const char *c_name, int p_beg_i, int p_end_i )
{
int l;
char c;
khiter_t iter;
faidx1_t val;
char* seq0;
char* seq = NULL;
iter = kh_get(s, fai->hash, c_name);
if(iter == kh_end(fai->hash)) return 0;
seq0 = seq = (char*)malloc( (p_end_i - p_beg_i + 2) * sizeof(char) );
if( seq0 == NULL ) fail( "Out of memory.\n" );
seq0[p_end_i-p_beg_i+1] = '\0';
val = kh_value(fai->hash, iter);
/* entirely off the map: all Ns */
if( p_beg_i >= (int)val.len || p_end_i < 0 ) {
while( p_beg_i <= p_end_i ) {
*seq++ ='n';
}
return seq0;
}
/* beginning is off the map */
while( p_beg_i < 0 && p_beg_i <= p_end_i ) {
*seq++ = 'n';
p_beg_i++;
}
/* end is off the map */
while( p_end_i >= (int)val.len ) {
seq[p_end_i-p_beg_i] = 'n';
p_end_i--;
}
/* retrieve the sequence */
l = 0;
razf_seek(fai->rz, val.offset + p_beg_i / val.line_blen * val.line_len + p_beg_i % val.line_blen, SEEK_SET);
while (razf_read(fai->rz, &c, 1) == 1 && l < p_end_i - p_beg_i + 1)
if (isgraph(c)) seq[l++] = tolower(c);
while( p_beg_i+l <= p_end_i ) seq[l++] = 'n';
return seq0;
}
char *fai_fetch(const faidx_t *fai, const char *str, int *len)
{
char *s, c;
int i, l, k, name_end;
khiter_t iter;
faidx1_t val;
khash_t(s) *h;
int beg, end;
beg = end = -1;
h = fai->hash;
name_end = l = strlen(str);
s = (char*)malloc(l+1);
// remove space
for (i = k = 0; i < l; ++i)
if (!isspace(str[i])) s[k++] = str[i];
s[k] = 0; l = k;
// determine the sequence name
for (i = l - 1; i >= 0; --i) if (s[i] == ':') break; // look for colon from the end
if (i >= 0) name_end = i;
if (name_end < l) { // check if this is really the end
int n_hyphen = 0;
for (i = name_end + 1; i < l; ++i) {
if (s[i] == '-') ++n_hyphen;
else if (!isdigit(s[i]) && s[i] != ',') break;
}
if (i < l || n_hyphen > 1) name_end = l; // malformated region string; then take str as the name
s[name_end] = 0;
iter = kh_get(s, h, s);
if (iter == kh_end(h)) { // cannot find the sequence name
iter = kh_get(s, h, str); // try str as the name
if (iter == kh_end(h)) {
*len = 0;
free(s); return 0;
} else s[name_end] = ':', name_end = l;
}
} else iter = kh_get(s, h, str);
if(iter == kh_end(h)) {
fprintf(pysamerr, "[fai_fetch] Warning - Reference %s not found in FASTA file, returning empty sequence\n", str);
free(s);
return 0;
};
val = kh_value(h, iter);
// parse the interval
if (name_end < l) {
for (i = k = name_end + 1; i < l; ++i)
if (s[i] != ',') s[k++] = s[i];
s[k] = 0;
beg = atoi(s + name_end + 1);
for (i = name_end + 1; i != k; ++i) if (s[i] == '-') break;
end = i < k? atoi(s + i + 1) : val.len;
if (beg > 0) --beg;
} else beg = 0, end = val.len;
if (beg >= val.len) beg = val.len;
if (end >= val.len) end = val.len;
if (beg > end) beg = end;
free(s);
// now retrieve the sequence
l = 0;
s = (char*)malloc(end - beg + 2);
razf_seek(fai->rz, val.offset + beg / val.line_blen * val.line_len + beg % val.line_blen, SEEK_SET);
while (razf_read(fai->rz, &c, 1) == 1 && l < end - beg && !fai->rz->z_err)
if (isgraph(c)) s[l++] = c;
s[l] = '\0';
*len = l;
return s;
}
int main(int argc, char **argv)
{
int c, compress, pstdout, is_forced;
RAZF *rz;
void *buffer;
long start, end, size;
compress = 1; pstdout = 0; start = 0; size = -1; end = -1; is_forced = 0;
while((c = getopt(argc, argv, "cdlhfb:s:")) >= 0){
switch(c){
case 'h': return razf_main_usage();
case 'd': compress = 0; break;
case 'c': pstdout = 1; break;
case 'l': compress = 2; break;
case 'b': start = atol(optarg); break;
case 's': size = atol(optarg); break;
case 'f': is_forced = 1; break;
}
}
if (size >= 0) end = start + size;
if(end >= 0 && end < start){
fprintf(stderr, " -- Illegal region: [%ld, %ld] --\n", start, end);
return 1;
}
if(compress == 1){
int f_src, f_dst = -1;
if(argc > optind){
if((f_src = open(argv[optind], O_RDONLY)) < 0){
fprintf(stderr, " -- Cannot open file: %s --\n", argv[optind]);
return 1;
}
if(pstdout){
f_dst = fileno(stdout);
} else {
char *name = malloc(sizeof(strlen(argv[optind]) + 5));
strcpy(name, argv[optind]);
strcat(name, ".rz");
f_dst = write_open(name, is_forced);
if (f_dst < 0) return 1;
free(name);
}
} else if(pstdout){
f_src = fileno(stdin);
f_dst = fileno(stdout);
} else return razf_main_usage();
rz = razf_dopen(f_dst, "w");
buffer = malloc(WINDOW_SIZE);
while((c = read(f_src, buffer, WINDOW_SIZE)) > 0) razf_write(rz, buffer, c);
razf_close(rz); // f_dst will be closed here
if (argc > optind && !pstdout) unlink(argv[optind]);
free(buffer);
close(f_src);
return 0;
} else {
if(argc <= optind) return razf_main_usage();
if(compress == 2){
rz = razf_open(argv[optind], "r");
if(rz->file_type == FILE_TYPE_RZ) {
printf("%20s%20s%7s %s\n", "compressed", "uncompressed", "ratio", "name");
printf("%20lld%20lld%6.1f%% %s\n", (long long)rz->end, (long long)rz->src_end, rz->end * 100.0f / rz->src_end,
argv[optind]);
} else fprintf(stdout, "%s is not a regular rz file\n", argv[optind]);
} else {
int f_dst;
if (argc > optind && !pstdout) {
char *name;
if (strstr(argv[optind], ".rz") - argv[optind] != strlen(argv[optind]) - 3) {
printf("razip: %s: unknown suffix -- ignored\n", argv[optind]);
return 1;
}
name = strdup(argv[optind]);
name[strlen(name) - 3] = '\0';
f_dst = write_open(name, is_forced);
free(name);
} else f_dst = fileno(stdout);
rz = razf_open(argv[optind], "r");
buffer = malloc(WINDOW_SIZE);
razf_seek(rz, start, SEEK_SET);
while(1){
if(end < 0) c = razf_read(rz, buffer, WINDOW_SIZE);
else c = razf_read(rz, buffer, (end - start > WINDOW_SIZE)? WINDOW_SIZE:(end - start));
if(c <= 0) break;
start += c;
write(f_dst, buffer, c);
if(end >= 0 && start >= end) break;
}
free(buffer);
if (!pstdout) unlink(argv[optind]);
}
razf_close(rz);
return 0;
}
}