int main(int argc, char **argv) {
bool iflag, dflag, oflag, qflag;
iflag = dflag = oflag = qflag = false;
std::string ffindex_header_db_prefix;
std::string ffindex_sequence_db_prefix;
std::string ffindex_ca3m_db_prefix;
std::string ffindex_a3m_db_prefix;
int c;
while ((c = getopt(argc, argv, "i:d:o:q:h")) != -1) {
switch (c) {
case 'i':
iflag = 1;
ffindex_ca3m_db_prefix = optarg;
break;
case 'd':
dflag = 1;
ffindex_sequence_db_prefix = optarg;
break;
case 'o':
oflag = 1;
ffindex_a3m_db_prefix = optarg;
break;
case 'q':
qflag = 1;
ffindex_header_db_prefix = optarg;
break;
case 'h':
usage();
exit(0);
case '?':
if (optopt == 'c')
fprintf(stderr, "Option -%c requires an argument.\n", optopt);
else if (isprint(optopt))
fprintf(stderr, "Unknown option `-%c'.\n", optopt);
else
fprintf(stderr, "Unknown option character `\\x%x'.\n", optopt);
return 1;
default:
abort();
}
}
if(!iflag || !dflag || !oflag || !qflag) {
std::cerr << "Missing arguments!" << std::endl;
usage();
exit(0);
}
//prepare ffindex a3m database
std::string a3mDataFile = ffindex_a3m_db_prefix+".ffdata";
std::string a3mIndexFile = ffindex_a3m_db_prefix+".ffindex";
FILE *a3m_data_fh = fopen(a3mDataFile.c_str(), "w");
FILE *a3m_index_fh = fopen(a3mIndexFile.c_str(), "w");
if (a3m_data_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex a3m data file! (" << a3mDataFile << ")!" << std::endl;
exit(1);
}
if(a3m_index_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex a3m index file! (" << a3mIndexFile << ")!" << std::endl;
exit(1);
}
size_t a3m_offset = 0;
//prepare ffindex ca3m database
std::string ca3mDataFile = ffindex_ca3m_db_prefix+".ffdata";
std::string ca3mIndexFile = ffindex_ca3m_db_prefix+".ffindex";
FILE *ca3m_data_fh = fopen(ca3mDataFile.c_str(), "r");
FILE *ca3m_index_fh = fopen(ca3mIndexFile.c_str(), "r");
if (ca3m_data_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex a3m data file! (" << ca3mDataFile << ")!" << std::endl;
exit(1);
}
if(ca3m_index_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex a3m index file! (" << ca3mIndexFile << ")!" << std::endl;
exit(1);
}
size_t ca3m_offset;
char* ca3m_data = ffindex_mmap_data(ca3m_data_fh, &ca3m_offset);
ffindex_index_t* ca3m_index = ffindex_index_parse(ca3m_index_fh, 0);
if(ca3m_index == NULL) {
std::cerr << "ERROR: CA3M index (" << ca3mIndexFile << ") could not be loaded!" << std::endl;
exit(1);
}
//prepare ffindex sequence database
std::string sequenceDataFile = ffindex_sequence_db_prefix+".ffdata";
std::string sequenceIndexFile = ffindex_sequence_db_prefix+".ffindex";
FILE *sequence_data_fh = fopen(sequenceDataFile.c_str(), "r");
FILE *sequence_index_fh = fopen(sequenceIndexFile.c_str(), "r");
if (sequence_data_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex sequence data file! (" << sequenceDataFile << ")!" << std::endl;
exit(1);
}
if(sequence_index_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex sequence index file! (" << sequenceIndexFile << ")!" << std::endl;
exit(1);
}
size_t sequence_data_size;
char* sequence_data = ffindex_mmap_data(sequence_data_fh, &sequence_data_size);
ffindex_index_t* sequence_index = ffindex_index_parse(sequence_index_fh, 80000000);
if(sequence_index == NULL) {
std::cerr << "ERROR: Sequence index could not be loaded!" << std::endl;
exit(1);
}
//prepare ffindex header database
std::string headerDataFile = ffindex_header_db_prefix + ".ffdata";
std::string headerIndexFile = ffindex_header_db_prefix + ".ffindex";
FILE *header_data_fh = fopen(headerDataFile.c_str(), "r");
FILE *header_index_fh = fopen(headerIndexFile.c_str(), "r");
if (header_data_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex sequence data file! ("
<< headerDataFile << ")!" << std::endl;
exit(1);
}
if (header_index_fh == NULL) {
std::cerr << "ERROR: Could not open ffindex header index file! ("
<< headerIndexFile << ")!" << std::endl;
exit(1);
}
size_t header_data_size;
char* header_data = ffindex_mmap_data(header_data_fh,
&header_data_size);
ffindex_index_t* header_index = ffindex_index_parse(header_index_fh, 1E8);
if (header_index == NULL) {
std::cerr << "ERROR: Header index could not be loaded!" << std::endl;
exit(1);
}
//prepare input stream
size_t ca3m_range_start = 0;
size_t ca3m_range_end = ca3m_index->n_entries;
// Foreach entry
#pragma omp parallel for shared(ca3m_index, ca3m_data, a3m_data_fh, a3m_index_fh, a3m_offset)
for(size_t entry_index = ca3m_range_start; entry_index < ca3m_range_end; entry_index++)
{
ffindex_entry_t* entry = ffindex_get_entry_by_index(ca3m_index, entry_index);
if(entry == NULL) { perror(entry->name); continue; }
char* data = ffindex_get_data_by_entry(ca3m_data, entry);
std::stringstream* out_buffer = new std::stringstream();
compressed_a3m::extract_a3m(data, entry->length, sequence_index, sequence_data, header_index, header_data, out_buffer);
std::string out_string = out_buffer->str();
#pragma omp critical
{
ffindex_insert_memory(a3m_data_fh, a3m_index_fh, &a3m_offset, const_cast<char*>(out_string.c_str()), out_string.size(), entry->name);
}
delete out_buffer;
}
fclose(a3m_data_fh);
fclose(a3m_index_fh);
ffsort_index(a3mIndexFile.c_str());
}
int main(int argn, char **argv)
{
int sort = 0, version = 0;
int opt, err = EXIT_SUCCESS;
while ((opt = getopt(argn, argv, "sv")) != -1)
{
switch (opt)
{
case 's':
sort = 1;
break;
case 'v':
version = 1;
break;
default:
usage(argv[0]);
return EXIT_FAILURE;
}
}
if(version == 1)
{
/* Don't you dare running it on a platform where byte != 8 bits */
printf("%s version %.2f, off_t = %zd bits\n", argv[0], FFINDEX_VERSION, sizeof(off_t) * 8);
return EXIT_SUCCESS;
}
if(argn - optind < 3)
{
usage(argv[0]);
return EXIT_FAILURE;
}
char *data_header_filename = argv[optind++];
char *index_header_filename = argv[optind++];
char *data_sequence_filename = argv[optind++];
char *index_sequence_filename = argv[optind++];
char *fasta_filename = argv[optind++];
printf("data header file: %s\n", data_header_filename);
printf("index header file: %s\n", index_header_filename);
printf("data sequence file: %s\n", data_sequence_filename);
printf("index sequence file: %s\n", index_sequence_filename);
printf("fasta file: %s\n", fasta_filename);
FILE *data_header_file, *index_header_file, *data_sequence_file, *index_sequence_file, *fasta_file;
size_t header_offset = 0;
size_t sequence_offset = 0;
struct stat st;
// open header ffindex
if(stat(data_header_filename, &st) == 0) { errno = EEXIST; perror(data_header_filename); return EXIT_FAILURE; }
data_header_file = fopen(data_header_filename, "w");
if( data_header_file == NULL) { perror(data_header_filename); return EXIT_FAILURE; }
if(stat(index_header_filename, &st) == 0) { errno = EEXIST; perror(index_header_filename); return EXIT_FAILURE; }
index_header_file = fopen(index_header_filename, "w+");
if(index_header_file == NULL) { perror(index_header_filename); return EXIT_FAILURE; }
//open sequence ffindex
if(stat(data_sequence_filename, &st) == 0) { errno = EEXIST; perror(data_sequence_filename); return EXIT_FAILURE; }
data_sequence_file = fopen(data_sequence_filename, "w");
if( data_sequence_file == NULL) { perror(data_sequence_filename); return EXIT_FAILURE; }
if(stat(index_sequence_filename, &st) == 0) { errno = EEXIST; perror(index_sequence_filename); return EXIT_FAILURE; }
index_sequence_file = fopen(index_sequence_filename, "w+");
if(index_sequence_file == NULL) { perror(index_sequence_filename); return EXIT_FAILURE; }
fasta_file = fopen(fasta_filename, "r");
if(fasta_file == NULL) { perror(fasta_filename); return EXIT_FAILURE; }
size_t fasta_size;
char *fasta_data = ffindex_mmap_data(fasta_file, &fasta_size);
// size_t from_length = 0;
char name[FFINDEX_MAX_ENTRY_NAME_LENTH];
int seq_id = 1;
size_t seq_id_length = 0;
size_t count_ws = 0;
char header[MAX_ENTRY_LENGTH];
header[0] = '>';
size_t header_length = 1;
char is_header = 1;
char sequence[MAX_ENTRY_LENGTH];
size_t sequence_length = 0;
for(size_t fasta_offset = 1; fasta_offset < fasta_size; fasta_offset++) // position after first ">"
{
seq_id_length = 0;
count_ws = 0;
is_header = 1;
header_length = 1;
sequence_length = 0;
while(fasta_offset < fasta_size && !(*(fasta_data + fasta_offset) == '>' && *(fasta_data + fasta_offset - 1) == '\n'))
{
char input = *(fasta_data + fasta_offset);
//get fasta name
if(isspace(input))
{
count_ws++;
name[seq_id_length] = '\0';
}
else if(count_ws == 0)
{
name[seq_id_length++] = *(fasta_data + fasta_offset);
}
if(input == '\n') {
is_header = 0;
header[header_length] = '\0';
sequence[sequence_length] = '\0';
}
else {
if(is_header == 1) {
header[header_length++] = input;
}
else {
sequence[sequence_length++] = input;
}
}
fasta_offset++;
}
if(seq_id_length == 0) {
sprintf(name, "%d", seq_id);
}
seq_id++;
get_short_id(name, '|', 2);
ffindex_insert_memory(data_header_file, index_header_file, &header_offset, header, header_length, name);
ffindex_insert_memory(data_sequence_file, index_sequence_file, &sequence_offset, sequence, sequence_length, name);
}
fclose(data_header_file);
fclose(data_sequence_file);
fclose(index_header_file);
fclose(index_sequence_file);
/* Sort the index entries and write back */
if(sort) {
ffsort_index(index_header_filename);
ffsort_index(index_sequence_filename);
}
return err;
}
