// Read VCF file
void vcf_file::scan_file(const string &chr, const string &exclude_chr, bool force_write_index)
{
bool filter_by_chr = (chr != "");
bool exclude_by_chr = (exclude_chr != "");
string index_filename = filename + ".vcfidx";
bool could_read_index_file = false;
if (force_write_index == false)
could_read_index_file = read_index_file(index_filename);
string CHROM, last_CHROM="";
int POS, last_POS = -1;
if (could_read_index_file == false)
{
printLOG("Building new index file.\n");
string line, CHROM, last_CHROM = "";
streampos filepos;
char c;
N_entries=0;
N_indv = 0;
while (!feof())
{
filepos = get_filepos();
c = peek();
if ((c == '\n') || (c == '\r'))
{
read_line(line);
continue;
}
else if (c == EOF)
break;
if (c == '#')
{
read_line(line);
if (line[1] == '#')
{ // Meta information
parse_meta(line);
}
else
{ // Must be header information: #CHROM POS ID REF ALT QUAL FILTER INFO (FORMAT NA00001 NA00002 ... )
parse_header(line);
}
}
else
{ // Must be a data line
read_CHROM_and_POS_and_skip_remainder_of_line(CHROM, POS);
if (last_CHROM != CHROM)
{
printLOG("\tScanning Chromosome: " + CHROM + "\n");
last_CHROM = CHROM;
}
if (POS == last_POS)
{
one_off_warning("\tWarning - file contains entries with the same position. This is not supported by vcftools, and may cause unexpected behaviour.\n");
}
last_POS = POS;
entry_file_locations.push_back(filepos);
N_entries++;
}
}
write_index_file(index_filename);
}
printLOG("File contains " + int2str(N_entries) + " entries and " + int2str(N_indv) + " individuals.\n");
vector<string> meta_lines = meta; meta.resize(0);
for (unsigned int ui=0; ui<meta_lines.size(); ui++)
parse_meta(meta_lines[ui]);
has_genotypes = (N_indv > 0);
bool already_found_required_chr = false;
bool already_filtered_required_chr = false;
if ((exclude_by_chr == true) || (filter_by_chr == true))
{
printLOG("Filtering by chromosome.\n");
for (unsigned int ui=0; ui<N_entries; ui++)
{
if (already_found_required_chr == true)
{
printLOG("Skipping Remainder.\n");
entry_file_locations.erase(entry_file_locations.begin()+ui, entry_file_locations.end());
break;
}
if (already_filtered_required_chr == true)
{
printLOG("Skipping Remainder.\n");
break;
}
set_filepos(entry_file_locations[ui]);
read_CHROM_only(CHROM);
if (last_CHROM != CHROM)
{
printLOG("\tChromosome: " + CHROM + "\n");
if ((filter_by_chr == true) && (last_CHROM == chr))
already_found_required_chr = true;
if ((exclude_by_chr == true) && (last_CHROM == exclude_chr))
already_filtered_required_chr = true;
last_CHROM = CHROM;
}
if ((exclude_by_chr == true) && (CHROM == exclude_chr))
{
entry_file_locations[ui] = -1;
continue;
}
if ((filter_by_chr == true) && (CHROM != chr))
{
entry_file_locations[ui] = -1;
continue;
}
}
sort(entry_file_locations.begin(), entry_file_locations.end());
while((entry_file_locations.size() > 0) && (entry_file_locations[0] < 0))
entry_file_locations.pop_front();
N_entries = entry_file_locations.size();
printLOG("Keeping " + int2str(N_entries) + " entries on specified chromosomes.\n");
}
include_indv.clear();
include_indv.resize(N_indv, true);
include_entry.clear();
include_entry.resize(N_entries, true);
include_genotype.clear();
include_genotype.resize(N_entries, vector<bool>(N_indv, true));
}
/**
* Open has to be called before any of the other functions are called.
* Throws a string exception if it is unable to open the index file, or if
* there is a format error in the sarray index file.
*
* Will throw an exception if a file set is already open.
*/
void open(std::string sidx_file) {
index_file = sidx_file;
ASSERT_MSG(!array_open, "sarray already open");
index_info = read_index_file(index_file);
initialize();
}
