size_t fastq_fread_bytes_aligner_pe(array_list_t *reads, size_t bytes, fastq_file_t *fq_file1, fastq_file_t *fq_file2) {
size_t accumulated_size = 0;
char header1[MAX_READ_ID_LENGTH];
char header2[MAX_READ_ID_LENGTH];
char read_separator[MAX_READ_ID_LENGTH];
char sequence1[MAX_READ_SEQUENCE_LENGTH];
char sequence2[MAX_READ_SEQUENCE_LENGTH];
char qualities1[MAX_READ_SEQUENCE_LENGTH];
char qualities2[MAX_READ_SEQUENCE_LENGTH];
int header_length1, sequence_length1, quality_length1;
int header_length2, sequence_length2, quality_length2;
fastq_read_t *read1, *read2;
while (accumulated_size < bytes && fgets(header1, MAX_READ_ID_LENGTH, fq_file1->fd) != NULL) {
fgets(sequence1, MAX_READ_SEQUENCE_LENGTH, fq_file1->fd);
fgets(read_separator, MAX_READ_ID_LENGTH, fq_file1->fd);
fgets(qualities1, MAX_READ_SEQUENCE_LENGTH, fq_file1->fd);
header_length1 = strlen(header1);
sequence_length1 = strlen(sequence1);
quality_length1 = strlen(qualities1);
// '\n' char is removed, but '\0' is left
chomp_at(header1, header_length1 - 1);
chomp_at(sequence1, sequence_length1 - 1);
chomp_at(qualities1, quality_length1 - 1);
// second file
fgets(header2, MAX_READ_ID_LENGTH, fq_file2->fd);
fgets(sequence2, MAX_READ_SEQUENCE_LENGTH, fq_file2->fd);
fgets(read_separator, MAX_READ_ID_LENGTH, fq_file2->fd);
fgets(qualities2, MAX_READ_SEQUENCE_LENGTH, fq_file2->fd);
header_length2 = strlen(header2);
sequence_length2 = strlen(sequence2);
quality_length2 = strlen(qualities2);
// '\n' char is removed, but '\0' is left
chomp_at(header2, header_length2 - 1);
chomp_at(sequence2, sequence_length2 - 1);
chomp_at(qualities2, quality_length2 - 1);
read1 = fastq_read_new(header1, sequence1, qualities1);
read2 = fastq_read_new(header2, sequence2, qualities2);
array_list_insert(read1, reads);
array_list_insert(read2, reads);
accumulated_size += header_length1 + sequence_length1 + quality_length1 + header_length2 + sequence_length2 + quality_length2;
}
return accumulated_size;
}
size_t fastq_fread_bytes_se(array_list_t *reads, size_t bytes, fastq_file_t *fq_file) {
size_t accumulated_size = 0;
char header1[MAX_READ_ID_LENGTH];
char sequence[MAX_READ_SEQUENCE_LENGTH];
char header2[MAX_READ_ID_LENGTH];
char qualities[MAX_READ_SEQUENCE_LENGTH];
int header_length, sequence_length, quality_length;
fastq_read_t *read;
while (accumulated_size < bytes && fgets(header1, MAX_READ_ID_LENGTH, fq_file->fd) != NULL) {
fgets(sequence, MAX_READ_SEQUENCE_LENGTH, fq_file->fd);
fgets(header2, MAX_READ_ID_LENGTH, fq_file->fd);
fgets(qualities, MAX_READ_SEQUENCE_LENGTH, fq_file->fd);
header_length = strlen(header1);
sequence_length = strlen(sequence);
quality_length = strlen(qualities);
// '\n' char is removed, but '\0' is left
chomp_at(header1, header_length - 1);
chomp_at(sequence, sequence_length - 1);
chomp_at(qualities, quality_length - 1);
read = fastq_read_new(header1, sequence, qualities);
array_list_insert(read, reads);
accumulated_size += header_length + sequence_length + quality_length;
}
return accumulated_size;
}
size_t fastq_fread_se_ex(array_list_t *reads, size_t num_reads, fastq_file_t *fq_file) {
size_t count = 0;
char *p;
char header1[MAX_READ_ID_LENGTH];
char sequence[MAX_READ_SEQUENCE_LENGTH];
char header2[MAX_READ_ID_LENGTH];
char qualities[MAX_READ_SEQUENCE_LENGTH];
int header_length, sequence_length, quality_length;
fastq_read_t *read;
while (count < num_reads && fgets(header1, MAX_READ_ID_LENGTH, fq_file->fd) != NULL) {
char *res = fgets(sequence, MAX_READ_SEQUENCE_LENGTH, fq_file->fd);
res = fgets(header2, MAX_READ_ID_LENGTH, fq_file->fd);
res = fgets(qualities, MAX_READ_SEQUENCE_LENGTH, fq_file->fd);
header_length = strlen(header1);
sequence_length = strlen(sequence);
quality_length = strlen(qualities);
// '\n' char is removed, but '\0' is left
chomp_at(header1, header_length - 1);
if ((p = strstr(header1, " ")) != NULL) {
*p = 0;
}
chomp_at(sequence, sequence_length - 1);
chomp_at(qualities, quality_length - 1);
read = fastq_read_new(&header1[1], sequence, qualities);
array_list_insert(read, reads);
count++;
}
return count;
}
size_t fastq_gzread_se(array_list_t *reads, size_t num_reads, fastq_gzfile_t *fq_gzfile) {
size_t count = 0;
char header1[MAX_READ_ID_LENGTH];
char sequence[MAX_READ_SEQUENCE_LENGTH];
char header2[MAX_READ_ID_LENGTH];
char qualities[MAX_READ_SEQUENCE_LENGTH];
int header_length, sequence_length, quality_length;
fastq_read_t *read;
size_t num_lines_to_read = 4 * num_reads; /* Each read consists of 4 lines */
int max_data_len = CHUNK;
int max_read_len = MAX_READ_SEQUENCE_LENGTH; /* Each read is supposed to be shorter than MAX_READ_SEQUENCE_LENGTH */
int eof_found = 0;
int c = 0;
int i = 0;
// fq_gzfile->i = 0;
size_t lines = 0;
char *aux;
// fq_gzfile->data = (char*) calloc (CHUNK, sizeof(char));
char *data; // = (char*) calloc (CHUNK, sizeof(char));
char *id = (char*) calloc (max_read_len, sizeof(char));
char *seq = (char*) calloc (max_read_len, sizeof(char));
char *qual = (char*) calloc (max_read_len, sizeof(char));
// ZLIB variables
unsigned have;
unsigned char in[CHUNK];
unsigned char out[CHUNK];
// If there is some data from before calls
if(fq_gzfile->data != NULL) {
if(fq_gzfile->data_size > max_data_len) {
data = (char*) calloc (fq_gzfile->data_size+max_data_len, sizeof(char));
max_data_len = fq_gzfile->data_size+max_data_len;
}else{
data = (char*) calloc (max_data_len, sizeof(char));
}
strncpy(data, fq_gzfile->data, fq_gzfile->data_size);
i = fq_gzfile->data_size;
}else {
// first time, no data has been saved before
data = (char*) calloc (max_data_len, sizeof(char));
}
do {
fq_gzfile->strm.avail_in = fread(in, 1, CHUNK, fq_gzfile->fd);
// printf("fq_gzfile->strm.avail_in: %i, CHUNK: %i\nnext_in: %s\n\n", fq_gzfile->strm.avail_in, CHUNK, fq_gzfile->strm.next_in);
if (ferror(fq_gzfile->fd)) {
(void)inflateEnd(&fq_gzfile->strm);
return Z_ERRNO;
}
if (fq_gzfile->strm.avail_in == 0)
break;
fq_gzfile->strm.next_in = in;
/* run inflate() on input until output buffer not full */
do {
fq_gzfile->strm.avail_out = CHUNK;
fq_gzfile->strm.next_out = out;
fq_gzfile->ret = inflate(&fq_gzfile->strm, Z_NO_FLUSH);
assert(fq_gzfile->ret != Z_STREAM_ERROR); /* state not clobbered */
switch (fq_gzfile->ret) {
case Z_NEED_DICT:
fq_gzfile->ret = Z_DATA_ERROR; /* and fall through */
case Z_DATA_ERROR:
case Z_MEM_ERROR:
(void)inflateEnd(&fq_gzfile->strm);
return fq_gzfile->ret;
}
have = CHUNK - fq_gzfile->strm.avail_out;
for (int j = 0; j < have && !eof_found; j++) {
c = out[j];
if (c != EOF) {
max_data_len = consume_input(c, &data, max_data_len, i);
if (c == '\n') {
lines++;
}
i++;
} else {
eof_found = 1;
}
}
} while (fq_gzfile->strm.avail_out == 0);
/* done when inflate() says it's done */
} while (lines < num_lines_to_read && fq_gzfile->ret != Z_STREAM_END);
// printf("data: %s\n", data);
// LOG_DEBUG_F("lines: %i, num_lines_to_read: %i\n", lines, num_lines_to_read);
// check if have read the expected number of lines
size_t parsed_chars;
size_t parsed_lines = 0;
size_t data_size;
// if(lines > 0) { //= num_lines_to_read
aux = data;
for(parsed_chars = 0; parsed_chars < i && parsed_lines < num_lines_to_read; parsed_chars++) {
if(data[parsed_chars] == '\n') {
// printf(">>i: %i, parsed_chars: %i, %i, aux: %s\n", i, parsed_chars, data[i-1], aux);
data[parsed_chars] = '\0';
if(count % 4 == 0) {
strcpy(id, aux); //printf("%s\n", id);
}
if(count % 4 == 1) {
strcpy(seq, aux); //printf("%s\n", seq);
}
if(count % 4 == 2) {
}
if(count % 4 == 3) {
strcpy(qual, aux); //printf("%s\n", qual);
read = fastq_read_new(id, seq, qual);
array_list_insert(read, reads);
}
count++;
aux = data + parsed_chars + 1;
parsed_lines++;
}
}
// LOG_DEBUG_F("i: %lu, parsed_lines: %lu\n", i, parsed_lines);
// LOG_DEBUG_F("parsed_chars: %lu, parsed_lines: %lu\n", parsed_chars, parsed_lines);
// lines = 0;
// LOG_DEBUG_F("BEFORE memcpy: fq_gzfile->data_size: %lu, new size: %lu\n", fq_gzfile->data_size, data_size);
data_size = i - parsed_chars;
if(fq_gzfile->data == NULL) {
fq_gzfile->data = (char*)malloc(data_size*sizeof(char));
}
if(fq_gzfile->data_size != 0 && fq_gzfile->data_size < data_size) {
fq_gzfile->data = realloc(fq_gzfile->data, data_size);
}
if(data_size > 0) {
memcpy(fq_gzfile->data, data+parsed_chars, data_size);
}
fq_gzfile->data_size = data_size;
// }
free(data);
free(id);
free(seq);
free(qual);
// if(fq_gzfile->ret == Z_STREAM_END) {
// (void)inflateEnd(&fq_gzfile->strm);
// }
// return fq_gzfile->ret == Z_STREAM_END ? Z_OK : Z_DATA_ERROR;
// printf(">>>>reads->size: %lu, num_reads: %lu\n", reads->size, num_reads);
return reads->size;
}
size_t fastq_gzread_bytes_se(array_list_t *reads, size_t bytes_to_read, fastq_gzfile_t *fq_gzfile) {
size_t count = 0;
fastq_read_t *read;
// size_t num_lines_to_read = bytes; /* Each read consists of 4 lines */
int max_data_len = CHUNK;
int max_read_len = MAX_READ_SEQUENCE_LENGTH_GZ; /* Each read is supposed to be shorter than MAX_READ_SEQUENCE_LENGTH */
int eof_found = 0;
int c = 0;
int i = 0;
size_t bytes_processed = 0;
char *aux;
char *data;
char *id = (char*) calloc (max_read_len, sizeof(char));
char *seq = (char*) calloc (max_read_len, sizeof(char));
char *qual = (char*) calloc (max_read_len, sizeof(char));
// ZLIB variables
unsigned have;
unsigned char in[CHUNK];
unsigned char out[CHUNK];
// If there is some data from before calls
if(fq_gzfile->data != NULL) {
if(fq_gzfile->data_size > max_data_len) {
data = (char*) calloc (fq_gzfile->data_size+max_data_len, sizeof(char));
max_data_len = fq_gzfile->data_size + max_data_len;
}else{
data = (char*) calloc (max_data_len, sizeof(char));
}
strncpy(data, fq_gzfile->data, fq_gzfile->data_size);
i = fq_gzfile->data_size;
}else {
// first time, no data has been saved before
data = (char*) calloc (max_data_len, sizeof(char));
}
do {
fq_gzfile->strm.avail_in = fread(in, 1, CHUNK, fq_gzfile->fd);
// printf("fq_gzfile->strm.avail_in: %i, CHUNK: %i\nnext_in: %s\n\n", fq_gzfile->strm.avail_in, CHUNK, fq_gzfile->strm.next_in);
if (ferror(fq_gzfile->fd)) {
(void)inflateEnd(&fq_gzfile->strm);
return Z_ERRNO;
}
if (fq_gzfile->strm.avail_in == 0)
break;
fq_gzfile->strm.next_in = in;
/* run inflate() on input until output buffer not full */
do {
fq_gzfile->strm.avail_out = CHUNK;
fq_gzfile->strm.next_out = out;
fq_gzfile->ret = inflate(&fq_gzfile->strm, Z_NO_FLUSH);
assert(fq_gzfile->ret != Z_STREAM_ERROR); /* state not clobbered */
switch (fq_gzfile->ret) {
case Z_NEED_DICT:
fq_gzfile->ret = Z_DATA_ERROR; /* and fall through */
case Z_DATA_ERROR:
case Z_MEM_ERROR:
(void)inflateEnd(&fq_gzfile->strm);
return fq_gzfile->ret;
}
have = CHUNK - fq_gzfile->strm.avail_out;
for (int j = 0; j < have && !eof_found; j++) {
c = out[j];
if (c != EOF) {
max_data_len = consume_input(c, &data, max_data_len, i);
// if (c == '\n') {
// bytes_processed++;
// }
i++;
bytes_processed++;
} else {
eof_found = 1;
}
}
} while (fq_gzfile->strm.avail_out == 0);
/* done when inflate() says it's done */
} while (i < bytes_to_read && fq_gzfile->ret != Z_STREAM_END);
// check if have read the expected number of lines
size_t parsed_chars;
size_t data_size;
aux = data;
for(parsed_chars = 0; parsed_chars < i; parsed_chars++) { //parsed_chars < bytes_to_read || parsed_lines % 4 == 0
if(data[parsed_chars] == '\n') {
data[parsed_chars] = '\0';
if(count % 4 == 0) {
strcpy(id, aux); //printf("%s\n", id);
}
if(count % 4 == 1) {
strcpy(seq, aux); //printf("%s\n", seq);
}
if(count % 4 == 2) {
}
if(count % 4 == 3) {
strcpy(qual, aux); //printf("%s\n", qual);
read = fastq_read_new(id, seq, qual);
array_list_insert(read, reads);
if(parsed_chars+1 > bytes_to_read) {
parsed_chars++;
break;
}
}
count++;
aux = data + parsed_chars + 1;
// parsed_lines++;
}
}
data_size = i - parsed_chars;
if(fq_gzfile->data == NULL) {
fq_gzfile->data = (char*)malloc(data_size*sizeof(char));
}
if(fq_gzfile->data_size != 0 && fq_gzfile->data_size < data_size) {
fq_gzfile->data = realloc(fq_gzfile->data, data_size);
}
if(data_size > 0) {
memcpy(fq_gzfile->data, data+parsed_chars, data_size);
}
fq_gzfile->data_size = data_size;
free(data);
free(id);
free(seq);
free(qual);
return parsed_chars;
}
