Compare commits

...

19 Commits

Author SHA1 Message Date
Maarten L. Hekkelman
5fc965789d messages updated 2025-09-24 15:11:33 +02:00
Maarten L. Hekkelman
b4596902aa Add compile features for Catch2, required on Windows 2025-09-24 14:15:19 +02:00
Maarten L. Hekkelman
cbf8b52f62 Update catch2 usage 2025-09-24 13:25:56 +02:00
Maarten L. Hekkelman
4e0fa1c916 No complete jthread on macOS/CLang 2025-09-24 13:00:50 +02:00
Maarten L. Hekkelman
95b007d38f Merge branch 'trunk' into develop 2025-09-24 11:38:01 +02:00
Maarten L. Hekkelman
b66f7a30ce Progress bar using WriteConsole on Windows 2025-09-24 11:36:37 +02:00
Maarten L. Hekkelman
ec7287c503 remove warning 2025-09-24 11:32:12 +02:00
Maarten L. Hekkelman
a41c591f0c Restore order of imports, avoid reordering by clang-format 2025-09-24 10:51:53 +02:00
Maarten L. Hekkelman
3a6527cdd5 yet another update on progress bar 2025-09-24 10:23:58 +02:00
Maarten L. Hekkelman
5f21a094c0 added flush to progress bar 2025-09-24 10:14:03 +02:00
Maarten L. Hekkelman
2203a1855d improved progress bar 2025-09-24 09:49:28 +02:00
Maarten L. Hekkelman
7edd2ecc18 new progress bar 2025-09-23 15:53:55 +02:00
Maarten L. Hekkelman
1d2953c850 Fix reconstruction, version bump 2025-09-22 13:51:18 +02:00
Maarten L. Hekkelman
dbf59ce622 reconstruct better when entity ID's are missing 2025-09-22 12:59:16 +02:00
Maarten L. Hekkelman
1596db8499 Merge branch 'develop' of github.com:pdb-redo/libcifpp into develop 2025-09-16 13:29:57 +02:00
Maarten L. Hekkelman
bd1fb5c5cd Added model::create_link 2025-09-16 13:29:51 +02:00
Maarten L. Hekkelman
da500025c3 swap atoms should swap type_symbol as well 2025-09-10 17:16:22 +02:00
Maarten L. Hekkelman
60eeea9a93 more resilient loading of dictionary data 2025-09-10 15:06:06 +02:00
Maarten L. Hekkelman
1220f01f1e change location of mmcif_ma.dic 2025-09-10 14:22:52 +02:00
14 changed files with 569 additions and 275 deletions

View File

@@ -32,7 +32,7 @@ endif()
# set the project name
project(
libcifpp
VERSION 9.0.2
VERSION 9.0.3
LANGUAGES CXX C)
list(PREPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake")
@@ -174,7 +174,7 @@ find_file(FMT NAME format)
if(FMT EQUAL "FMT-NOTFOUND")
if(NOT (fmt_FOUND OR TARGET fmt))
find_package(fmt REQUIRED)
message(FATAL_ERROR "fmt not found, compiler too old, you're out of luck")
message(FATAL_ERROR "cifpp: <format> not found and neither was libfmt, compiler too old, you're out of luck")
endif()
endif()
@@ -189,10 +189,6 @@ include(FindPkgConfig)
if(PKG_CONFIG_FOUND)
pkg_check_modules(PCRE2 IMPORTED_TARGET libpcre2-8)
if(PCRE2_FOUND)
message(STATUS "Using pcre2 found using pkg-config")
endif()
endif()
if(NOT PCRE2_FOUND)

View File

@@ -1,3 +1,6 @@
Version 9.0.3
- Reconstruction fixed when some entity ids are missing
Version 9.0.2
- Fix code that reconstructs sequences, could throw a map::at
- Many optimisations in validation and reconstruction code.

View File

@@ -1065,6 +1065,16 @@ class structure
/// \param atom The set of item data containing the data for the atoms.
void create_water(row_initializer atom);
/// \brief Create a link, a struct_conn record for two atoms.
///
/// \param a1 Atom 1
/// \param a2 Atom 2
/// \param link_type The struct_conn_type ID for the link
/// \param role The pdbx_role field value
/// \return The ID of the struct_conn record created
std::string create_link(atom a1, atom a2, const std::string &link_type, const std::string &role);
/// \brief Create a new and empty (sugar) branch
branch &create_branch();

View File

@@ -296,6 +296,11 @@ class progress_bar
*/
void message(const std::string &inMessage);
/**
* @brief Flush the progress bar to the output stream
*/
void flush();
private:
progress_bar(const progress_bar &) = delete;
progress_bar &operator=(const progress_bar &) = delete;

View File

@@ -784,7 +784,7 @@ void compound_factory::report_missing_compound(std::string_view compound_id)
<< "in /var/cache/libcifpp using the following commands:\n\n"
<< "curl -o " << CACHE_DIR << "/components.cif https://files.wwpdb.org/pub/pdb/data/monomers/components.cif\n"
<< "curl -o " << CACHE_DIR << "/mmcif_pdbx.dic https://mmcif.wwpdb.org/dictionaries/ascii/mmcif_pdbx_v50.dic\n"
<< "curl -o " << CACHE_DIR << "/mmcif_ma.dic https://github.com/ihmwg/ModelCIF/raw/master/dist/mmcif_ma.dic\n\n";
<< "curl -o " << CACHE_DIR << "/mmcif_ma.dic https://mmcif.wwpdb.org/dictionaries/ascii/mmcif_ma.dic\n\n";
#endif
if (m_impl)

View File

@@ -28,6 +28,9 @@
#include "cif++/dictionary_parser.hpp"
#include "cif++/file.hpp"
#include "cif++/parser.hpp"
#include <exception>
#include <iomanip>
#include <stdexcept>
namespace cif
{
@@ -46,7 +49,7 @@ class dictionary_parser : public parser
void load_dictionary()
{
std::unique_ptr<datablock> dict;
auto savedDatablock = m_datablock;
auto savedDatablock = std::exchange(m_datablock, nullptr);
try
{
@@ -75,6 +78,9 @@ class dictionary_parser : public parser
error(ex.what());
}
if (m_datablock == nullptr)
throw std::runtime_error("Dictionary file is empty?");
// store all validators
for (auto &ic : mCategoryValidators)
m_validator.add_category_validator(std::move(ic));

View File

@@ -28,6 +28,7 @@
#include <filesystem>
#include <fstream>
#include <initializer_list>
#include <iomanip>
#include <numeric>
#include <stack>
@@ -1275,28 +1276,28 @@ void structure::load_atoms_for_model(structure_open_options options)
else
{
std::vector<cif::mm::atom> atoms;
std::map<std::tuple<std::string,int>, std::map<std::string, float>> alts;
std::map<std::tuple<std::string, int>, std::map<std::string, float>> alts;
for (auto id : atom_site.find<std::string>(std::move(c), "id"))
{
auto a = atoms.emplace_back(std::make_shared<atom::atom_impl>(m_db, id));
if (a.is_alternate())
{
auto key = std::make_tuple(a.get_label_asym_id(), a.get_label_seq_id());
auto alt_id = a.get_label_alt_id();
if (auto i = alts.find(key); i != alts.end())
i->second[alt_id] += a.get_occupancy();
else
alts[key][alt_id] = a.get_occupancy();
}
}
for (auto &&[key, value] : alts)
{
// const auto &[asym_id, seq_id] = key;
// select highest occupancy for this residue's alternates
std::string alt_id;
float occupancy = options.occupancy_mode == occupancy_policy::MAX ? 0.f : std::numeric_limits<float>::max();
@@ -1319,11 +1320,11 @@ void structure::load_atoms_for_model(structure_open_options options)
}
}
}
value.clear();
value.emplace(alt_id, occupancy);
}
for (auto a : atoms)
{
if (a.is_alternate())
@@ -1335,10 +1336,8 @@ void structure::load_atoms_for_model(structure_open_options options)
}
else
emplace_atom(a);
}
}
}
void structure::load_data()
@@ -1898,13 +1897,12 @@ void structure::swap_atoms(atom a1, atom a2)
auto r1 = atomSites.find1(key("id") == a1.id());
auto r2 = atomSites.find1(key("id") == a2.id());
auto l1 = r1["label_atom_id"];
auto l2 = r2["label_atom_id"];
l1.swap(l2);
auto l3 = r1["auth_atom_id"];
auto l4 = r2["auth_atom_id"];
l3.swap(l4);
for (std::string fld : std::initializer_list<std::string>{ "label_atom_id", "auth_atom_id", "type_symbol" })
{
auto l1 = r1[fld];
auto l2 = r2[fld];
l1.swap(l2);
}
}
catch (const std::exception &ex)
{
@@ -2412,6 +2410,61 @@ void structure::create_water(row_initializer atom)
});
}
std::string structure::create_link(atom a1, atom a2, const std::string &link_type, const std::string &role)
{
using namespace literals;
auto &struct_conn = m_db["struct_conn"];
auto &struct_conn_type = m_db["struct_conn_type"];
// This will validate link_type :-)
if (not struct_conn_type.contains("id"_key == link_type))
struct_conn_type.emplace({ { "id", link_type } });
std::string link_id = struct_conn.get_unique_id(link_type + '_');
item label_seq_id_1("ptnr1_label_seq_id");
if (int nr = a1.get_label_seq_id(); nr != 0)
label_seq_id_1.value(std::to_string(nr));
item label_seq_id_2("ptnr2_label_seq_id");
if (int nr = a2.get_label_seq_id(); nr != 0)
label_seq_id_2.value(std::to_string(nr));
struct_conn.emplace(
{ //
{ "id", link_id },
{ "conn_type_id", link_type },
{ "pdbx_leaving_atom_flag", "one" },
{ "ptnr1_label_asym_id", a1.get_label_asym_id() },
{ "ptnr1_label_comp_id", a1.get_label_comp_id() },
label_seq_id_1,
{ "ptnr1_label_atom_id", a1.get_label_atom_id() },
{ "pdbx_ptnr1_label_alt_id", a1.get_label_alt_id() },
{ "pdbx_ptnr1_PDB_ins_code", a1.get_pdb_ins_code() },
{ "ptnr1_auth_asym_id", a1.get_auth_asym_id() },
{ "ptnr1_auth_comp_id", a1.get_auth_comp_id() },
{ "ptnr1_auth_seq_id", a1.get_auth_seq_id() },
{ "ptnr1_symmetry", a1.symmetry() },
{ "ptnr2_label_asym_id", a2.get_label_asym_id() },
{ "ptnr2_label_comp_id", a2.get_label_comp_id() },
label_seq_id_2,
{ "ptnr2_label_atom_id", a2.get_label_atom_id() },
{ "pdbx_ptnr2_label_alt_id", a2.get_label_alt_id() },
{ "pdbx_ptnr2_PDB_ins_code", a2.get_pdb_ins_code() },
{ "ptnr2_auth_asym_id", a2.get_auth_asym_id() },
{ "ptnr2_auth_comp_id", a2.get_auth_comp_id() },
{ "ptnr2_auth_seq_id", a2.get_auth_seq_id() },
{ "ptnr2_symmetry", a2.symmetry() },
{ "pdbx_dist_value", distance(a1.get_location(), a2.get_location()), 3 },
{ "pdbx_role", role } });
return link_id;
}
branch &structure::create_branch()
{
auto &entity = m_db["entity"];

View File

@@ -28,6 +28,9 @@
#include "cif++/compound.hpp"
#include "cif++/row.hpp"
#include <stdexcept>
#include <string>
// --------------------------------------------------------------------
namespace cif::pdb
@@ -187,6 +190,8 @@ void checkEntities(datablock &db)
void createEntityIDs(datablock &db)
{
using namespace literals;
// Suppose the file does not have entity ID's. We have to make up some
// walk the atoms. For each auth_asym_id we have a new struct_asym.
@@ -198,6 +203,7 @@ void createEntityIDs(datablock &db)
// that should cover it
auto &atom_site = db["atom_site"];
auto &entity = db["entity"];
auto &cf = compound_factory::instance();
std::vector<std::vector<row_handle>> entities;
@@ -206,16 +212,29 @@ void createEntityIDs(datablock &db)
int lastSeqID = -1;
std::vector<row_handle> waters;
for (auto rh : atom_site)
int nextEntityID;
try
{
if (entity.empty())
nextEntityID = 1;
else
nextEntityID = entity.find_max<int>("id") + 1;
}
catch (...)
{
nextEntityID = 1;
}
for (auto rh : atom_site.find("label_entity_id"_key == cif::null))
{
residue_key_type k = rh.get<std::optional<std::string>,
std::optional<int>,
std::optional<std::string>,
std::optional<std::string>,
std::optional<int>,
std::optional<std::string>>(
"auth_asym_id", "auth_seq_id", "auth_comp_id",
"label_asym_id", "label_seq_id", "label_comp_id");
std::optional<int>,
std::optional<std::string>,
std::optional<std::string>,
std::optional<int>,
std::optional<std::string>>(
"auth_asym_id", "auth_seq_id", "auth_comp_id",
"label_asym_id", "label_seq_id", "label_comp_id");
std::string comp_id = get_comp_id(k);
@@ -230,8 +249,8 @@ void createEntityIDs(datablock &db)
bool is_monomer = cf.is_monomer(comp_id);
if (lastAsymID == asym_id and lastSeqID == seq_id and not is_monomer)
continue;
// if (lastAsymID == asym_id and lastSeqID == seq_id and not is_monomer)
// continue;
if (asym_id != lastAsymID or (not is_monomer and lastSeqID != seq_id))
entities.push_back({});
@@ -243,6 +262,7 @@ void createEntityIDs(datablock &db)
}
std::map<std::size_t, std::string> entity_ids;
std::map<std::string, std::string> newEntitiesForCompound;
atom_site.add_item("label_entity_id");
@@ -251,7 +271,37 @@ void createEntityIDs(datablock &db)
if (entity_ids.contains(i))
continue;
auto entity_id = std::to_string(i + 1);
residue_key_type k = entities[i].front().get<std::optional<std::string>, std::optional<int>, std::optional<std::string>, std::optional<std::string>, std::optional<int>, std::optional<std::string>>(
"auth_asym_id", "auth_seq_id", "auth_comp_id",
"label_asym_id", "label_seq_id", "label_comp_id");
std::string comp_id = get_comp_id(k);
std::string entity_id;
if (auto v = db["pdbx_entity_nonpoly"].find_first("comp_id"_key == comp_id); v)
entity_id = v.get<std::string>("entity_id");
else if (auto i2 = newEntitiesForCompound.find(comp_id); i2 != newEntitiesForCompound.end())
entity_id = i2->second;
else
{
entity_id = std::to_string(nextEntityID++);
if (cf.is_monomer(comp_id))
entity.emplace({ //
{ "id", entity_id },
{ "type", "polymer" } });
else if (cf.is_water(comp_id))
entity.emplace({ //
{ "id", entity_id },
{ "type", "water" } });
else
entity.emplace({ //
{ "id", entity_id },
{ "type", "non-polymer" } });
newEntitiesForCompound[comp_id] = entity_id;
}
entity_ids[i] = entity_id;
for (std::size_t j = i + 1; j < entities.size(); ++j)
@@ -439,7 +489,7 @@ void checkAtomRecords(datablock &db)
// And negative seq_id values
if (atom_site.contains(key("label_seq_id") < 0))
fixNegativeSeqID(atom_site);
std::set<std::string> polymer_entities;
if (db["entity"].empty())
{
@@ -576,7 +626,7 @@ void checkAtomRecords(datablock &db)
row["label_atom_id"] = row["auth_atom_id"].text();
else if (row["auth_atom_id"].empty())
row["auth_atom_id"] = row["label_atom_id"].text();
// Rewrite the coordinates and other items that look better in a fixed format
// Be careful not to nuke invalidly formatted data here
for (auto [item_name, prec] : std::vector<std::tuple<std::string_view, int>>{
@@ -690,23 +740,53 @@ void checkAtomAnisotropRecords(datablock &db)
}
}
void createStructAsym(datablock &db)
void checkStructAsym(datablock &db)
{
auto &atom_site = db["atom_site"];
auto &struct_asym = db["struct_asym"];
for (const auto &[label_asym_id, entity_id] : atom_site.rows<std::string, std::string>("label_asym_id", "label_entity_id"))
if (struct_asym.empty())
{
if (label_asym_id.empty())
throw std::runtime_error("File contains atom_site records without a label_asym_id");
if (struct_asym.count(key("id") == label_asym_id) == 0)
for (const auto &[label_asym_id, entity_id] : atom_site.rows<std::string, std::string>("label_asym_id", "label_entity_id"))
{
struct_asym.emplace({
// clang-format off
{ "id", label_asym_id },
{ "entity_id", entity_id }
//clang-format on
});
if (label_asym_id.empty())
throw std::runtime_error("File contains atom_site records without a label_asym_id");
if (struct_asym.count(key("id") == label_asym_id) == 0)
{
struct_asym.emplace({
// clang-format off
{ "id", label_asym_id },
{ "entity_id", entity_id }
//clang-format on
});
}
}
}
else
{
for (const auto &[label_asym_id, entity_id] :
atom_site.rows<std::string, std::string>("label_asym_id", "label_entity_id"))
{
if (label_asym_id.empty())
throw std::runtime_error("File contains atom_site records without a label_asym_id");
auto sa = struct_asym.find_first(key("id") == label_asym_id);
if (sa)
{
if (sa["entity_id"].empty())
sa.assign("entity_id", entity_id, false, true);
else if (sa.get<std::string>("entity_id") != entity_id)
throw std::runtime_error("Inconsistent entity ID's in struct_asym");
}
else
{
struct_asym.emplace({
// clang-format off
{ "id", label_asym_id },
{ "entity_id", entity_id }
//clang-format on
});
}
}
}
}
@@ -863,11 +943,11 @@ void createEntityPoly(datablock &db)
else
letter = '(' + comp_id + ')';
}
else if (auto i = compound_factory::kAAMap.find(comp_id); i != compound_factory::kAAMap.end())
else if (auto i2 = compound_factory::kAAMap.find(comp_id); i2 != compound_factory::kAAMap.end())
{
c_type = "polypeptide(L)";
letter = letter_can = i->second;
letter = letter_can = i2->second;
}
else if (iequals(c->type(), "D-PEPTIDE LINKING"))
{
@@ -1425,7 +1505,7 @@ bool reconstruct_pdbx(file &file, const validator &validator)
checkChemCompRecords(db);
// If the data is really horrible, it might not contain entities
if (not db["atom_site"].find_first(key("label_entity_id") != null))
if (db["atom_site"].find_first(key("label_entity_id") == null))
createEntityIDs(db);
// Now see if atom records make sense at all
@@ -1569,9 +1649,8 @@ bool reconstruct_pdbx(file &file, const validator &validator)
checkAtomAnisotropRecords(db);
// Now create any missing categories
// Next make sure we have struct_asym records
if (auto cat = db.get("struct_asym"); cat == nullptr or cat->empty())
createStructAsym(db);
// Next make sure we have good struct_asym records
checkStructAsym(db);
if (auto cat = db.get("entity"); cat == nullptr or cat->empty())
createEntity(db);
@@ -1680,9 +1759,8 @@ void fixup_pdbx(file &file, const validator &validator)
db.set_validator(&validator);
// Now create any missing categories
// Next make sure we have struct_asym records
if (auto cat = db.get("struct_asym"); cat == nullptr or cat->empty())
createStructAsym(db);
// Next make sure we have good struct_asym records
checkStructAsym(db);
if (auto cat = db.get("entity"); cat == nullptr or cat->empty())
createEntity(db);

View File

@@ -34,14 +34,20 @@
#include <condition_variable>
#include <cstring>
#include <deque>
#include <format>
#include <fstream>
#include <functional>
#include <iomanip>
#include <iostream>
#include <map>
#include <mutex>
#include <sstream>
#include <string>
#include <thread>
#include <utility>
#if __cpp_lib_jthread >= 201911L
#include <stop_token>
#endif
namespace fs = std::filesystem;
@@ -65,27 +71,50 @@ std::string get_version_nr()
#if defined(_WIN32) or defined(__MINGW32__)
}
#include <windows.h>
#include <libloaderapi.h>
#include <wincon.h>
#include <codecvt>
// clang-format off
# include <windows.h>
# include <libloaderapi.h>
# include <wincon.h>
// clang-format on
namespace cif
{
uint32_t get_terminal_width()
{
CONSOLE_SCREEN_BUFFER_INFO csbi;
::GetConsoleScreenBufferInfo(::GetStdHandle(STD_OUTPUT_HANDLE), &csbi);
return csbi.srWindow.Right - csbi.srWindow.Left + 1;
CONSOLE_SCREEN_BUFFER_INFO csbi;
return ::GetConsoleScreenBufferInfo(::GetStdHandle(STD_OUTPUT_HANDLE), &csbi)
? csbi.srWindow.Right - csbi.srWindow.Left + 1
: 80;
}
void write_to_console(const std::string &s)
{
auto h = ::GetStdHandle(STD_OUTPUT_HANDLE);
CONSOLE_SCREEN_BUFFER_INFO csbi;
if (auto l = ::MultiByteToWideChar(CP_UTF8, 0, s.data(), s.length(), nullptr, 0);
l > 0 and ::GetConsoleScreenBufferInfo(::GetStdHandle(STD_OUTPUT_HANDLE), &csbi))
{
std::u16string ws(l, 0);
::MultiByteToWideChar(CP_UTF8, 0, s.data(), s.length(), (LPWSTR)ws.data(), l);
DWORD w;
::WriteConsoleW(h, ws.data(), ws.length(), &w, nullptr);
}
else
{
std::cout.write(s.data(), s.length());
std::cout.flush();
}
}
#else
#include <sys/ioctl.h>
#include <termios.h>
#include <limits.h>
# include <limits.h>
# include <sys/ioctl.h>
# include <termios.h>
uint32_t get_terminal_width()
{
@@ -100,59 +129,220 @@ uint32_t get_terminal_width()
return result;
}
inline void write_to_console(const std::string &s)
{
std::cout << s << std::flush;
}
#endif
// --------------------------------------------------------------------
struct progress_bar_impl
{
progress_bar_impl(int64_t inMax, const std::string &inAction)
: m_max_value(inMax)
progress_bar_impl(uint64_t max_value, const std::string &message)
: m_max_value(max_value)
, m_consumed(0)
, m_action(inAction)
, m_message(inAction)
, m_thread(std::bind(&progress_bar_impl::run, this))
, m_action(message)
, m_message(message)
{
}
progress_bar_impl(const progress_bar_impl&) = delete;
progress_bar_impl &operator=(const progress_bar_impl &) = delete;
virtual ~progress_bar_impl() {}
~progress_bar_impl();
void run();
void consumed(int64_t n);
void progress(int64_t p);
void message(const std::string &msg);
void print_progress();
void print_done();
virtual void consumed(uint64_t n);
virtual void progress(uint64_t p);
virtual void message(const std::string &msg);
virtual void print_done();
using time_point = std::chrono::time_point<std::chrono::system_clock>;
int64_t m_max_value;
std::atomic<int64_t> m_consumed;
int64_t m_last_consumed = 0;
int m_spinner_index = 0;
uint64_t m_max_value;
std::atomic<uint64_t> m_consumed;
std::string m_action, m_message;
std::mutex m_mutex;
std::thread m_thread;
time_point m_start = std::chrono::system_clock::now();
time_point m_last = std::chrono::system_clock::now();
bool m_stop = false;
};
progress_bar_impl::~progress_bar_impl()
void progress_bar_impl::consumed(uint64_t n)
{
m_consumed += n;
}
void progress_bar_impl::progress(uint64_t p)
{
m_consumed = p;
}
void progress_bar_impl::message(const std::string &msg)
{
m_message = msg;
}
void progress_bar_impl::print_done()
{
std::chrono::duration<double> elapsed = std::chrono::system_clock::now() - m_start;
std::string days, hours, minutes, seconds;
uint64_t s = static_cast<uint64_t>(std::trunc(elapsed.count()));
if (s > 24 * 60 * 60)
{
days = std::format("{:d}d ", s / (24 * 60 * 60));
s %= 24 * 60 * 60;
}
if (s > 60 * 60)
{
hours = std::format("{:2d}h ", s / (60 * 60));
s %= 60 * 60;
}
if (s > 60)
{
minutes = std::format("{:2d}m ", s / 60);
s %= 60;
}
std::string msg = std::format("{} done in {}{}{}{:.1f}s", m_action, days, hours, minutes, s + 1e-6 * (elapsed.count() - s));
uint32_t width = get_terminal_width();
if (msg.length() < width)
msg += std::string(width - msg.length(), ' ');
write_to_console(msg += '\n');
}
// --------------------------------------------------------------------
struct simple_progress_bar_impl : public progress_bar_impl
{
simple_progress_bar_impl(uint64_t max_value, const std::string &message)
: progress_bar_impl(max_value, message)
{
}
~simple_progress_bar_impl()
{
if (m_printed_any)
print_done();
}
void consumed(uint64_t n) override
{
using namespace std::literals;
progress_bar_impl::consumed(n);
// print at most 10 steps, but only if it took long enough
int percentile = static_cast<int>(std::floor(10.f * m_consumed / m_max_value));
if (percentile > m_last_percentile and (m_printed_any or std::chrono::system_clock::now() - m_start >= 1s))
{
if (not std::exchange(m_printed_any, true))
write_to_console(m_action + ": ");
write_to_console(std::format("...{:d}0%", percentile));
m_last_percentile = percentile;
}
}
void progress(uint64_t p) override
{
consumed(p - m_consumed);
}
void print_done() override
{
if (m_printed_any)
{
write_to_console("\n");
progress_bar_impl::print_done();
}
}
bool m_printed_any = false;
int m_last_percentile = 0;
};
// --------------------------------------------------------------------
struct fancy_progress_bar_impl : public progress_bar_impl
{
fancy_progress_bar_impl(uint64_t max_value, const std::string &message)
: progress_bar_impl(max_value, message)
, m_thread(
#if __cpp_lib_jthread >= 201911L
[this](std::stop_token stoken)
{ this->run(stoken); }
#else
[this]()
{ this->run(); }
#endif
)
{
}
~fancy_progress_bar_impl();
#if __cpp_lib_jthread >= 201911L
void run(std::stop_token stoken);
#else
void run();
#endif
void consumed(uint64_t n) override;
void progress(uint64_t p) override;
void message(const std::string &msg) override;
void print_progress();
std::mutex m_mutex;
std::condition_variable m_cv;
float m_progress;
uint32_t m_width, m_bar_width;
uint32_t m_steps, m_last_steps = 0;
uint64_t m_last_consumed = 0;
#if __cpp_lib_jthread >= 201911L
std::jthread m_thread;
#else
std::thread m_thread;
bool m_stop = false;
#endif
};
const char *kBlocks[] = {
" ",
"",
"",
"",
"",
"",
"",
"",
"",
};
const size_t kBlockCount = sizeof(kBlocks) / sizeof(void *) - 1;
fancy_progress_bar_impl::~fancy_progress_bar_impl()
{
using namespace std::literals;
assert(m_thread.joinable());
#if __cpp_lib_jthread >= 201911L
m_thread.request_stop();
#else
m_stop = true;
#endif
m_thread.join();
}
void progress_bar_impl::run()
#if __cpp_lib_jthread >= 201911L
void fancy_progress_bar_impl::run(std::stop_token stoken)
#else
void fancy_progress_bar_impl::run()
#endif
{
using namespace std::literals;
@@ -160,25 +350,44 @@ void progress_bar_impl::run()
try
{
while (not m_stop)
for (;;)
{
std::unique_lock lock(m_mutex);
m_cv.wait_for(lock, 25ms);
#if __cpp_lib_jthread >= 201911L
if (stoken.stop_requested())
break;
#else
if (m_stop)
break;
#endif
auto now = std::chrono::system_clock::now();
if (now - m_start < 2s or now - m_last < 100ms)
{
std::this_thread::sleep_for(10ms);
if (m_consumed == m_last_consumed or now - m_start < 1s)
continue;
}
std::lock_guard lock(m_mutex);
m_last_consumed = m_consumed;
if (not printedAny and isatty(STDOUT_FILENO))
std::cout << "\x1b[?25l";
// See if we need to do work
m_width = get_terminal_width();
m_progress = static_cast<float>(m_consumed) / m_max_value;
m_bar_width = 7 * m_width / 10; // 70% of the width of the terminal
m_steps = static_cast<uint32_t>(std::ceil(m_progress * m_bar_width * kBlockCount));
if (m_steps == m_last_steps)
continue;
m_last_steps = m_steps;
if (not printedAny)
write_to_console("\x1b[?25l");
print_progress();
printedAny = true;
m_last = std::chrono::system_clock::now();
}
}
catch (...)
@@ -187,161 +396,94 @@ void progress_bar_impl::run()
if (printedAny)
{
write_to_console("\r\x1b[?25h");
print_done();
if (isatty(STDOUT_FILENO))
std::cout << "\x1b[?25h";
}
}
void progress_bar_impl::consumed(int64_t n)
void fancy_progress_bar_impl::consumed(uint64_t n)
{
m_consumed += n;
progress_bar_impl::consumed(n);
// m_cv.notify_one();
}
void progress_bar_impl::progress(int64_t p)
void fancy_progress_bar_impl::progress(uint64_t p)
{
m_consumed = p;
progress_bar_impl::progress(p);
// m_cv.notify_one();
}
void progress_bar_impl::message(const std::string &msg)
void fancy_progress_bar_impl::message(const std::string &msg)
{
std::unique_lock lock(m_mutex);
m_message = msg;
progress_bar_impl::message(msg);
// m_cv.notify_one();
}
const char* kSpinner[] = {
// ".", "o", "O", "0", "O", "o", ".", " "
// "⢄", "⢂", "⢁", "⡁", "⡈", "⡐", "⡠"
".", "o", "O", "0", "@", "*", " "
};
const std::size_t kSpinnerCount = sizeof(kSpinner) / sizeof(char*);
const int kSpinnerTimeInterval = 100;
const uint32_t kMinBarWidth = 40, kMinMsgWidth = 12;
void progress_bar_impl::print_progress()
void fancy_progress_bar_impl::print_progress()
{
const char *kBlocks[] = {
// "▯", // 0
// "▮", // 1
"=",
"-"
};
const uint32_t pct_width = 5;
uint32_t msg_width = m_width - m_bar_width - pct_width - 1;
uint32_t width = get_terminal_width();
float progress = static_cast<float>(m_consumed) / m_max_value;
if (width < kMinBarWidth)
std::cout << (100 * progress) << "%\n";
else
if (msg_width < kMinMsgWidth)
{
uint32_t bar_width = 7 * width / 10;
uint32_t pct_width = 7;
uint32_t msg_width = width - bar_width - pct_width - 1;
m_bar_width += kMinMsgWidth - msg_width;
msg_width = kMinMsgWidth;
}
if (msg_width < kMinMsgWidth)
{
bar_width += kMinMsgWidth - msg_width;
msg_width = kMinMsgWidth;
}
std::string bar;
bar.reserve(m_bar_width * 4);
std::ostringstream msg;
if (m_message.length() <= msg_width)
{
msg << m_message;
if (m_message.length() < msg_width)
msg << std::string(msg_width - m_message.length(), ' ');
}
for (uint32_t i = 0; i < m_bar_width; ++i)
{
if (i * kBlockCount <= m_steps)
bar += kBlocks[kBlockCount];
else if (i * kBlockCount > m_steps + kBlockCount)
bar += kBlocks[0];
else
msg << m_message.substr(0, msg_width - 3) << "...";
msg << ' ';
uint32_t pi = static_cast<uint32_t>(std::ceil(progress * bar_width));
for (uint32_t i = 0; i < bar_width; ++i)
msg << kBlocks[i > pi ? 1 : 0];
msg << ' ';
msg << std::setw(3) << static_cast<int>(std::ceil(progress * 100)) << "% ";
auto now = std::chrono::system_clock::now();
m_spinner_index = (std::chrono::duration_cast<std::chrono::milliseconds>(now - m_start).count() / kSpinnerTimeInterval) % kSpinnerCount;
msg << kSpinner[m_spinner_index];
std::cout << '\r' << msg.str();
std::cout.flush();
bar += kBlocks[1 + m_steps % kBlockCount];
}
}
namespace
{
std::ostream &operator<<(std::ostream &os, const std::chrono::duration<double> &t)
// make the bar more colorfull
struct color_type
{
uint64_t s = static_cast<uint64_t>(std::trunc(t.count()));
if (s > 24 * 60 * 60)
{
auto days = s / (24 * 60 * 60);
os << days << "d ";
s %= 24 * 60 * 60;
}
uint8_t r, g, b;
} fg{ 0, 3, 5 }, bg{ 0, 1, 2 };
if (s > 60 * 60)
{
auto hours = s / (60 * 60);
os << hours << "h ";
s %= 60 * 60;
}
auto esc_1 = std::format("\x1b[38;5;{}m\x1b[48;5;{}m",
16 + (fg.r * 36) + (fg.g * 6) + fg.b,
16 + (bg.r * 36) + (bg.g * 6) + bg.b);
std::string esc_2("\x1b[0m");
if (s > 60)
{
auto minutes = s / 60;
os << minutes << "m ";
s %= 60;
}
bar = esc_1 + bar + esc_2;
double ss = s + 1e-6 * (t.count() - s);
std::string msg = m_message.length() <= msg_width
? m_message
: m_message.substr(0, msg_width - 3) + "...";
os << std::fixed << std::setprecision(1) << ss << 's';
return os;
}
} // namespace
void progress_bar_impl::print_done()
{
std::chrono::duration<double> elapsed = std::chrono::system_clock::now() - m_start;
std::ostringstream msgstr;
msgstr << m_action << " done in " << elapsed << " seconds";
auto msg = msgstr.str();
uint32_t width = get_terminal_width();
if (msg.length() < width)
msg += std::string(width - msg.length(), ' ');
std::cout << '\r' << msg << '\n';
write_to_console(std::format("{:{}} {} {:3d}%\r", msg, msg_width, bar,
static_cast<int>(std::ceil(m_progress * 100))));
}
progress_bar::progress_bar(int64_t inMax, const std::string &inAction)
// --------------------------------------------------------------------
progress_bar::progress_bar(int64_t max_value, const std::string &message)
: m_impl(nullptr)
{
if (isatty(STDOUT_FILENO) and VERBOSE >= 0)
m_impl = new progress_bar_impl(inMax, inAction);
if (VERBOSE >= 0)
{
if (isatty(STDOUT_FILENO) and get_terminal_width() > kMinBarWidth)
m_impl = new fancy_progress_bar_impl(max_value, message);
else
m_impl = new simple_progress_bar_impl(max_value, message);
}
}
progress_bar::~progress_bar()
{
delete m_impl;
flush();
}
void progress_bar::consumed(int64_t inConsumed)
@@ -350,16 +492,25 @@ void progress_bar::consumed(int64_t inConsumed)
m_impl->consumed(inConsumed);
}
void progress_bar::progress(int64_t inProgress)
void progress_bar::progress(int64_t value)
{
if (m_impl != nullptr)
m_impl->progress(inProgress);
m_impl->progress(value);
}
void progress_bar::message(const std::string &inMessage)
void progress_bar::message(const std::string &message)
{
if (m_impl != nullptr)
m_impl->message(inMessage);
m_impl->message(message);
}
void progress_bar::flush()
{
if (m_impl)
{
delete m_impl;
m_impl = nullptr;
}
}
} // namespace cif
@@ -387,13 +538,13 @@ struct rsrc_imp
#if _WIN32
#if __MINGW32__
# if __MINGW32__
extern "C" __attribute__((weak, alias("gResourceIndexDefault"))) const mrsrc::rsrc_imp gResourceIndex[];
extern "C" __attribute__((weak, alias("gResourceDataDefault"))) const char gResourceData[];
extern "C" __attribute__((weak, alias("gResourceNameDefault"))) const char gResourceName[];
#else
# else
extern "C" const mrsrc::rsrc_imp *gResourceIndexDefault[1] = {};
extern "C" const char *gResourceDataDefault[1] = {};
@@ -403,11 +554,11 @@ extern "C" const mrsrc::rsrc_imp gResourceIndex[];
extern "C" const char gResourceData[];
extern "C" const char gResourceName[];
#pragma comment(linker, "/alternatename:gResourceIndex=gResourceIndexDefault")
#pragma comment(linker, "/alternatename:gResourceData=gResourceDataDefault")
#pragma comment(linker, "/alternatename:gResourceName=gResourceNameDefault")
# pragma comment(linker, "/alternatename:gResourceIndex=gResourceIndexDefault")
# pragma comment(linker, "/alternatename:gResourceData=gResourceDataDefault")
# pragma comment(linker, "/alternatename:gResourceName=gResourceNameDefault")
#endif
# endif
#else
extern const __attribute__((weak)) mrsrc::rsrc_imp gResourceIndex[];

View File

@@ -604,7 +604,7 @@ validator validator_factory::construct_validator(std::string_view name, std::opt
not v.matches_audit_conform(category{ "audit_conform", //
{ { "dict_name", name }, { "dict_version", version } } }))
{
std::clog << "Invalid dictionary?\n";
std::clog << "Loaded dictionary does not match name=" << name << " and version=" << version.value_or("''") << "\n";
}
return v;

View File

@@ -1,7 +1,29 @@
# We're using the older version 2 of Catch2
# SPDX-License-Identifier: BSD-2-Clause
#
# Copyright (c) 2025 NKI/AVL, Netherlands Cancer Institute
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# 1. Redistributions of source code must retain the above copyright notice, this
# list of conditions and the following disclaimer
# 2. Redistributions in binary form must reproduce the above copyright notice,
# this list of conditions and the following disclaimer in the documentation
# and/or other materials provided with the distribution.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
# ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
# (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
# LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
# ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
if(NOT(Catch2_FOUND OR TARGET Catch2))
find_package(Catch2 QUIET)
if(NOT (Catch2_FOUND OR TARGET Catch2))
find_package(Catch2 3 QUIET)
if(NOT Catch2_FOUND)
include(FetchContent)
@@ -9,11 +31,11 @@ if(NOT(Catch2_FOUND OR TARGET Catch2))
FetchContent_Declare(
Catch2
GIT_REPOSITORY https://github.com/catchorg/Catch2.git
GIT_TAG v2.13.9)
GIT_TAG v3.8.0)
FetchContent_MakeAvailable(Catch2)
set(Catch2_VERSION "2.13.9")
target_compile_features(Catch2 PRIVATE cxx_std_20)
endif()
endif()
@@ -29,18 +51,12 @@ list(
spinner
reconstruction
validate-pdbx
)
)
add_library(test-main OBJECT "${CMAKE_CURRENT_SOURCE_DIR}/test-main.cpp")
target_link_libraries(test-main cifpp::cifpp Catch2::Catch2)
if("${Catch2_VERSION}" VERSION_LESS 3.0.0)
target_compile_definitions(test-main PUBLIC CATCH22=1)
else()
target_compile_definitions(test-main PUBLIC CATCH22=0)
endif()
foreach(CIFPP_TEST IN LISTS CIFPP_tests)
set(CIFPP_TEST "${CIFPP_TEST}-test")
set(CIFPP_TEST_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/${CIFPP_TEST}.cpp")
@@ -48,12 +64,6 @@ foreach(CIFPP_TEST IN LISTS CIFPP_tests)
add_executable(
${CIFPP_TEST} ${CIFPP_TEST_SOURCE} $<TARGET_OBJECTS:test-main>)
if(${Catch2_VERSION} VERSION_GREATER_EQUAL 3.0.0)
target_compile_definitions(${CIFPP_TEST} PUBLIC CATCH22=0)
else()
target_compile_definitions(${CIFPP_TEST} PUBLIC CATCH22=1)
endif()
target_link_libraries(${CIFPP_TEST} PRIVATE cifpp::cifpp Catch2::Catch2)
target_include_directories(${CIFPP_TEST} PRIVATE "${EIGEN_INCLUDE_DIR}")
@@ -62,15 +72,6 @@ foreach(CIFPP_TEST IN LISTS CIFPP_tests)
target_compile_options(${CIFPP_TEST} PRIVATE /EHsc)
endif()
add_custom_target(
"run-${CIFPP_TEST}"
DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/Run${CIFPP_TEST}.touch ${CIFPP_TEST})
add_custom_command(
OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/Run${CIFPP_TEST}.touch
COMMAND $<TARGET_FILE:${CIFPP_TEST}> --data-dir
${CMAKE_CURRENT_SOURCE_DIR})
add_test(NAME ${CIFPP_TEST} COMMAND $<TARGET_FILE:${CIFPP_TEST}> --data-dir
${CMAKE_CURRENT_SOURCE_DIR})
endforeach()
add_test(NAME ${CIFPP_TEST}
COMMAND $<TARGET_FILE:${CIFPP_TEST}> --data-dir ${CMAKE_CURRENT_SOURCE_DIR})
endforeach()

View File

@@ -11,12 +11,7 @@ int main(int argc, char *argv[])
Catch::Session session; // There must be exactly one instance
// Build a new parser on top of Catch2's
#if CATCH22
using namespace Catch::clara;
#else
// Build a new parser on top of Catch2's
using namespace Catch::Clara;
#endif
auto cli = session.cli() // Get Catch2's command line parser
| Opt(gTestDir, "data-dir") // bind variable to a new option, with a hint string

View File

@@ -26,11 +26,7 @@
#pragma once
#if CATCH22
#include <catch2/catch.hpp>
#else
#include <catch2/catch_all.hpp>
#endif
#include <filesystem>

View File

@@ -63,7 +63,7 @@ update_dictionary() {
update_dictionary "@CIFPP_CACHE_DIR@/components.cif" "https://files.wwpdb.org/pub/pdb/data/monomers/components.cif.gz"
update_dictionary "@CIFPP_CACHE_DIR@/mmcif_pdbx.dic" "https://mmcif.wwpdb.org/dictionaries/ascii/mmcif_pdbx_v50.dic.gz"
update_dictionary "@CIFPP_CACHE_DIR@/mmcif_ma.dic" "https://github.com/ihmwg/ModelCIF/raw/master/dist/mmcif_ma.dic"
update_dictionary "@CIFPP_CACHE_DIR@/mmcif_ma.dic" "https://mmcif.wwpdb.org/dictionaries/ascii/mmcif_ma.dic"
# notify subscribers, using find instead of run-parts to make it work on FreeBSD as well