void construct(matching_index<t_wt, t_bv>& idx, const std::string& file, sdsl::cache_config& config, uint8_t num_bytes)
{
sdsl::int_vector<0> text;
{
//auto event = memory_monitor::event("text");
load_vector_from_file(text, file, num_bytes);
}
sdsl::csa_wt<sdsl::wt_int<>> csa;
{
//auto event = memory_monitor::event("csa");
construct(csa, file, config, num_bytes);
}
t_wt wts;
{
//auto event = memory_monitor::event("wt");
construct(wts, cache_file_name(sdsl::conf::KEY_SA, config));
}
sdsl::util::delete_all_files(config.file_map);
{
//auto event = memory_monitor::event("compose"); // contains rank support initialization
idx = std::move(matching_index<t_wt, t_bv>(text, wts));
}
}
int main(int argc, char** argv)
{
::testing::InitGoogleTest(&argc, argv);
if (argc!=2) {
cout << "Usage: " << argv[0] << " test_file" << endl;
cout << " (1) Reverses test_file; stores it in test_file_rev." << endl;
cout << " (2) Performs tests." << endl;
cout << " (3) Deletes test_file_reverse." << endl;
return 1;
}
test_file = argv[1];
test_file_rev = test_file + "_rev";
{
//reverse input
int_vector<8> text;
load_vector_from_file(text, test_file, 1);
size_type n = text.size();
int_vector<8> text_rev(n);
for (size_type i=0; i<n; i++) {
text_rev[n-1-i] = text[i];
}
char* text2 = (char*)text_rev.data();
ofstream of(test_file_rev, ofstream::binary);
of.write(text2, n);
of.close();
}
int result = RUN_ALL_TESTS();
sdsl::remove(test_file_rev);
return result;
}
void construct(t_index& idx, const std::string& file, cache_config& config, uint8_t num_bytes, lcp_tag)
{
auto event = memory_monitor::event("construct compressed LCP");
const char* KEY_TEXT = key_text_trait<t_width>::KEY_TEXT;
typedef int_vector<t_width> text_type;
{
// (2) check, if the longest common prefix array is cached
auto event = memory_monitor::event("LCP");
if (!cache_file_exists(conf::KEY_LCP, config)) {
{
auto event = memory_monitor::event("parse input text");
// (1) check, if the text is cached
if (!cache_file_exists(KEY_TEXT, config)) {
text_type text;
load_vector_from_file(text, file, num_bytes);
if (contains_no_zero_symbol(text, file)) {
append_zero_symbol(text);
store_to_cache(text,KEY_TEXT, config);
}
}
register_cache_file(KEY_TEXT, config);
}
{
// (2) check, if the suffix array is cached
auto event = memory_monitor::event("SA");
if (!cache_file_exists(conf::KEY_SA, config)) {
construct_sa<t_width>(config);
}
register_cache_file(conf::KEY_SA, config);
}
if (t_width==8) {
construct_lcp_semi_extern_PHI(config);
} else {
construct_lcp_PHI<t_width>(config);
}
}
register_cache_file(conf::KEY_LCP, config);
}
{
auto event = memory_monitor::event("compressed LCP");
t_index tmp(config);
tmp.swap(idx);
}
if (config.delete_files) {
auto event = memory_monitor::event("delete temporary files");
util::delete_all_files(config.file_map);
}
}
void construct(t_index& idx, const std::string& file, cache_config& config, uint8_t num_bytes, wt_tag)
{
auto event = memory_monitor::event("construct wavelet tree");
int_vector<t_index::alphabet_category::WIDTH> text;
load_vector_from_file(text, file, num_bytes);
std::string tmp_key = util::to_string(util::pid())+"_"+util::to_string(util::id());
std::string tmp_file_name = cache_file_name(tmp_key, config);
store_to_file(text, tmp_file_name);
util::clear(text);
{
int_vector_buffer<t_index::alphabet_category::WIDTH> text_buf(tmp_file_name);
t_index tmp(text_buf, text_buf.size());
idx.swap(tmp);
}
sdsl::remove(tmp_file_name);
}
void construct(t_index& idx, const std::string& file, cache_config& config, uint8_t num_bytes, csa_tag)
{
auto event = memory_monitor::event("construct CSA");
const char* KEY_TEXT = key_text_trait<t_index::alphabet_category::WIDTH>::KEY_TEXT;
const char* KEY_BWT = key_bwt_trait<t_index::alphabet_category::WIDTH>::KEY_BWT;
typedef int_vector<t_index::alphabet_category::WIDTH> text_type;
{
auto event = memory_monitor::event("parse input text");
// (1) check, if the text is cached
if (!cache_file_exists(KEY_TEXT, config)) {
text_type text;
load_vector_from_file(text, file, num_bytes);
if (contains_no_zero_symbol(text, file)) {
append_zero_symbol(text);
store_to_cache(text,KEY_TEXT, config);
}
}
register_cache_file(KEY_TEXT, config);
}
{
// (2) check, if the suffix array is cached
auto event = memory_monitor::event("SA");
if (!cache_file_exists(conf::KEY_SA, config)) {
construct_sa<t_index::alphabet_category::WIDTH>(config);
}
register_cache_file(conf::KEY_SA, config);
}
{
// (3) construct BWT
auto event = memory_monitor::event("BWT");
if (!cache_file_exists(KEY_BWT, config)) {
construct_bwt<t_index::alphabet_category::WIDTH>(config);
}
register_cache_file(KEY_BWT, config);
}
{
// (4) use BWT to construct the CSA
auto event = memory_monitor::event("construct CSA");
t_index tmp(config);
idx.swap(tmp);
}
if (config.delete_files) {
auto event = memory_monitor::event("delete temporary files");
util::delete_all_files(config.file_map);
}
}
int main(int argc, char** argv)
{
::testing::InitGoogleTest(&argc, argv);
if (argc < 4) {
// LCOV_EXCL_START
cout << "Usage: " << argv[0] << " test_file num_bytes temp_file tmp_dir" << endl;
cout << " (1) Generates a CSA out of test_file; stores it in temp_file." << endl;
cout << " Temporary files (SA/BWT/TEXT) are stored in tmp_dir." << endl;
cout << " num_bytes specifies who many bytes make a symbol in the"<< endl;
cout << " input sequence" << endl;
cout << " If `in-memory` is specified, the in-memory construction is tested." << endl;
cout << " (2) Performs tests." << endl;
cout << " (3) Deletes temp_file." << endl;
return 1;
// LCOV_EXCL_STOP
}
test_file = argv[1];
num_bytes = atoi(argv[2]);
temp_file = argv[3];
temp_dir = argv[4];
in_memory = argc > 5;
if (in_memory) {
temp_dir = "@";
int_vector<> data;
load_vector_from_file(data, test_file, num_bytes);
test_file = ram_file_name(test_file);
switch (num_bytes) {
case 0: store_to_file(data, test_file); break;
case 1: store_to_plain_array<uint8_t>(data, test_file); break;
case 2: store_to_plain_array<uint16_t>(data, test_file); break;
case 3: store_to_plain_array<uint32_t>(data, test_file); break;
case 4: store_to_plain_array<uint64_t>(data, test_file); break;
}
temp_file = ram_file_name(temp_file);
}
return RUN_ALL_TESTS();
}
//! @brief Read path from file.
void XC::PathSeries::readFromFile(const std::string &fileName)
{
// determine the number of data points .. open file and count num entries
int numDataPoints= getNumDataPointsOnFile(fileName);
// create a vector and read in the data
if(numDataPoints)
{
// first open the file
std::ifstream theFile1;
theFile1.open(fileName.c_str(), std::ios::in);
if(theFile1.bad() || !theFile1.is_open())
{
std::cerr << getClassName() << "::" << __FUNCTION__
<< "; could not open file: " << fileName << std::endl;
}
else
{
// now create the vector
thePath.resize(numDataPoints);
load_vector_from_file(thePath,theFile1);
theFile1.close(); // finally close the file
}
}
}
