int pull_by_re(char *input_file, pcre *re, pcre_extra *re_extra, int min, int max, int length, int exclude, int convert, int just_count) {
gzFile fp;
int count=0,l;
int excluded = 0;
int is_fasta = 0; /* assume fastq */
kseq_t *seq;
/* open fasta file */
fp = gzopen(input_file,"r");
if (!fp) {
fprintf(stderr,"%s - Couldn't open fasta file %s\n",progname,input_file);
exit(EXIT_FAILURE);
}
seq = kseq_init(fp);
/* determine file type */
l = kseq_read(seq); /* read the first sequence */
is_fasta = seq->qual.s == NULL ? 1 : 0;
gzrewind(fp);
kseq_rewind(seq); /* rewind to beginning for main loop */
if (verbose_flag) {
if (is_fasta)
fprintf(stderr, "Input is FASTA format\n");
else
fprintf(stderr, "Input is FASTQ format\n");
}
/* search through list and see if this header matches */
while((l = kseq_read(seq)) >= 0) {
if (exclude) {
if (search_header(re, re_extra, seq->name.s) || search_header(re, re_extra, seq->comment.s))
excluded++;
else {
/* regex doesn't match, so check size/print */
count += size_filter(seq, is_fasta, min, max, length, convert, just_count);
}
} else {
if (search_header(re, re_extra, seq->name.s) || search_header(re, re_extra, seq->comment.s)) {
/* regex matches so check size/print */
count += size_filter(seq, is_fasta, min, max, length, convert, just_count);
} else
excluded++;
}
} /* end of seq traversal */
kseq_destroy(seq);
gzclose(fp); /* done reading file so close */
if (just_count) {
fprintf(stdout, "Total output: %i\n", count);
fprintf(stdout, "Total excluded: %i\n", excluded);
}
return count;
}
/* The gateway function */
void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[])
{
char * filename;
double max_gap_fraction;
gzFile fp;
kseq_t *seq;
int N, M;
double * N_ptr;
double * M_ptr;
double * Z_ptr;
double ** Z;
int * inds;
int * zinds;
/* check for proper number of arguments */
if (nrhs != 2) {
mexErrMsgIdAndTxt("read_alignemnt_fasta:nrhs", "Two inputs required: filename, max_gap_fraction.");
}
if (nlhs != 3) {
mexErrMsgIdAndTxt("read_alignemnt_fasta:nlhs", "Three outputs required: N, M, Z.");
}
/* get the value of the frequence matrices */
filename = mxArrayToString(prhs[0]);
fp = gzopen(filename, "r");
if (fp == Z_NULL) {
mexErrMsgIdAndTxt("read_alignemnt_fasta:open_file", "Error opening file");
}
seq = kseq_init(fp);
/* get the max_gap_fraction value */
max_gap_fraction = mxGetScalar(prhs[1]);
/* create the outputs N, M */
plhs[0] = mxCreateDoubleMatrix(1, 1, mxREAL);
plhs[1] = mxCreateDoubleMatrix(1, 1, mxREAL);
N_ptr = mxGetPr(plhs[0]);
M_ptr = mxGetPr(plhs[1]);
parse_seq_pass1(seq, &inds, &zinds, &N, &M, max_gap_fraction);
*N_ptr = (double) N;
*M_ptr = (double) M;
/* create the output matrix Z */
plhs[2] = mxCreateDoubleMatrix(M, N, mxREAL);
Z_ptr = mxGetPr(plhs[2]);
Z = malloc(N * sizeof(double));
{
int i;
for (i = 0; i < N; ++i) {
Z[i] = Z_ptr;
Z_ptr += M;
}
}
gzrewind(fp);
kseq_rewind(seq);
parse_seq_pass2(seq, Z, inds, zinds);
/* release memory */
kseq_destroy(seq);
gzclose(fp);
mxFree(filename);
free(Z);
free(inds);
}
int pull_by_name(char *input_file, FILE *names_fp, int min, int max, int length, int exclude, int convert, int just_count) {
gzFile fp;
int i,l,capacity=80;
int count=0,excluded=0;
int is_fasta = -1;
char *fasta_name;
char *line;
kseq_t *seq;
/* get some space for the line */
line = malloc(sizeof(char) * capacity); /* get memory allocated */
if (!line) {
fprintf(stderr, "%s - line malloc: %s\n",progname, strerror(errno));
exit(EXIT_FAILURE);
}
while((i = getl(&line,names_fp)) != -1) {
fasta_name = parse_name(line);
if (fasta_name) {
add_name(fasta_name); /* add fasta_name to hash */
}
}
free(line); /* free up line */
if (verbose_flag) {
fprintf(stderr,"\n");
fprintf(stderr,"done reading from input (%d entries)\n", hash_key_count());
}
/*print_hash();*/
/* open fasta file */
fp = gzopen(input_file,"r");
if (!fp) {
fprintf(stderr,"%s - Couldn't open fasta file %s\n",progname,input_file);
exit(EXIT_FAILURE);
}
seq = kseq_init(fp); /* initialize kseq */
/* determine file type */
l = kseq_read(seq); /* read the first sequence */
is_fasta = seq->qual.s == NULL ? 1 : 0;
gzrewind(fp); /* rewind to beginning for main loop */
kseq_rewind(seq);
if (verbose_flag) {
if (is_fasta)
fprintf(stderr, "Input is FASTA format\n");
else
fprintf(stderr, "Input is FASTQ format\n");
}
/* search through list and see if this header matches */
while((l = kseq_read(seq)) >= 0) {
if (exclude == 0) { /* INCLUDE names from names file */
if (find_name(seq->name.s)) { /* found name in list */
if (min > 0 && max > 0) { /* got a min and max */
if (seq->seq.l >= min && seq->seq.l <= max) {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
}
} else if (min > 0 || max > 0) { /* either min or max is 0 */
if (min > 0 && seq->seq.l >= min) {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
} else if (max > 0 && seq->seq.l <= max) {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
}
} else {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
}
}
} else { /* EXCLUDE names from names file */
if (find_name(seq->name.s)) { /* found name in list */
excluded++;
} else {
if (min > 0 && max > 0) { /* got a min and max */
if (seq->seq.l >= min && seq->seq.l <= max) {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
}
} else if (min > 0 || max > 0) { /* either min or max is 0 */
if (min > 0 && seq->seq.l >= min) {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
} else if (max > 0 && seq->seq.l <= max) {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
}
} else {
count++;
if (!just_count) {
if (convert)
is_fasta ? print_fastq_seq(seq) : print_fasta_seq(seq,length);
else
is_fasta ? print_fasta_seq(seq,length) : print_fastq_seq(seq);
}
}
}
}
}
kseq_destroy(seq);
gzclose(fp); /* done reading file */
delete_hash(); /* free the list nodes */
if (just_count) {
fprintf(stdout, "Total output: %i\n", count);
if (exclude)
fprintf(stdout, "Total excluded: %i\n", excluded);
}
if (verbose_flag) {
fprintf(stderr,"Processed %i entries\n",count);
if (exclude)
fprintf(stderr,"Excluded %i entries\n",excluded);
}
return count;
}
