updates

CMakeLists.txt

@@ -370,7 +370,9 @@ if(ENABLE_TESTING)
 
 	find_package(Boost REQUIRED headers)
 
-	list(APPEND CIFPP_tests unit-v2 unit-3d format model rename-compound sugar)
+	list(APPEND CIFPP_tests unit-v2 unit-3d format model
+	# rename-compound sugar
+	)
 
 	foreach(CIFPP_TEST IN LISTS CIFPP_tests)
 		set(CIFPP_TEST "${CIFPP_TEST}-test")

include/cif++/category.hpp

@@ -286,14 +286,17 @@ class category
 	{
 		bool result = false;
 
-		cond.prepare(*this);
-
-		for (auto r : *this)
+		if (cond)
 		{
-			if (cond(r))
+			cond.prepare(*this);
+
+			for (auto r : *this)
 			{
-				result = true;
-				break;
+				if (cond(r))
+				{
+					result = true;
+					break;
+				}
 			}
 		}
 

include/cif++/model.hpp

@@ -28,6 +28,8 @@
 
 #include <numeric>
 
+#include <cif++/atom_type.hpp>
+
 #if __cpp_lib_format
 #include <format>
 #endif
@@ -80,7 +82,7 @@ class atom
 
 		// bool getAnisoU(float anisou[6]) const;
 
-		// int charge() const;
+		int charge() const;
 
 		void moveTo(const point &p)
 		{
@@ -128,6 +130,20 @@ class atom
 			return result;
 		}
 
+		float get_property_float(std::string_view name) const
+		{
+			float result = 0;
+			if (not m_row[name].empty())
+			{
+				auto s = get_property(name);
+
+				std::from_chars_result r = std::from_chars(s.data(), s.data() + s.length(), result);
+				if (r.ec != std::errc() and VERBOSE > 0)
+					std::cerr << "Error converting " << s << " to number for property " << name << std::endl;
+			}
+			return result;
+		}
+
 		void set_property(const std::string_view name, const std::string &value)
 		{
 			m_row.assign(name, value, true, true);
@@ -214,6 +230,13 @@ class atom
 		return m_impl->get_property_int(name);
 	}
 
+	float get_property_float(std::string_view name) const
+	{
+		if (not m_impl)
+			throw std::logic_error("Error trying to fetch a property from an uninitialized atom");
+		return m_impl->get_property_float(name);
+	}
+
 	void set_property(const std::string_view name, const std::string &value)
 	{
 		if (not m_impl)
@@ -228,7 +251,8 @@ class atom
 	}
 
 	const std::string &id() const { return impl().m_id; }
-	// AtomType type() const { return impl().mType; }
+
+	cif::atom_type get_type() const { return atom_type_traits(get_property("type_symbol")).type(); }
 
 	point get_location() const { return impl().m_location; }
 	void set_location(point p)
@@ -280,7 +304,7 @@ class atom
 
 	// const compound &compound() const;
 	// bool isWater() const { return impl().mCompID == "HOH" or impl().mCompID == "H2O" or impl().mCompID == "WAT"; }
-	// int charge() const;
+	int get_charge() const;
 
 	// float uIso() const;
 	// bool getAnisoU(float anisou[6]) const { return impl().getAnisoU(anisou); }
@@ -297,6 +321,7 @@ class atom
 
 	std::string get_auth_asym_id() const { return get_property("auth_asym_id"); }
 	std::string get_auth_seq_id() const { return get_property("auth_seq_id"); }
+	std::string get_auth_atom_id() const { return get_property("auth_atom_id"); }
 	std::string get_pdb_ins_code() const { return get_property("pdbx_PDB_ins_code"); }
 
 	// const std::string &labelAtomID() const { return impl().mAtomID; }
@@ -457,6 +482,11 @@ class residue
 	}
 	const std::string get_auth_seq_id() const { return m_auth_seq_id; }
 
+	std::string get_pdb_ins_code() const
+	{
+		return m_atoms.empty() ? "" : m_atoms.front().get_pdb_ins_code();
+	}
+
 	const std::string &get_compound_id() const { return m_compound_id; }
 	void set_compound_id(const std::string &id) { m_compound_id = id; }
 
@@ -476,8 +506,8 @@ class residue
 
 	void add_atom(atom &atom);
 
-	// /// \brief Unique atoms returns only the atoms without alternates and the first of each alternate atom id.
-	// std::vector<atom> unique_atoms() const;
+	/// \brief Unique atoms returns only the atoms without alternates and the first of each alternate atom id.
+	std::vector<atom> unique_atoms() const;
 
 	// /// \brief The alt ID used for the unique atoms
 	// std::string unique_alt_id() const;
@@ -773,12 +803,12 @@ class structure
 	const std::list<polymer> &polymers() const { return m_polymers; }
 	std::list<polymer> &polymers() { return m_polymers; }
 
-	// polymer &getPolymerByAsymID(const std::string &asymID);
+	polymer &get_polymer_by_asym_id(const std::string &asymID);
 
-	// const polymer &getPolymerByAsymID(const std::string &asymID) const
-	// {
-	// 	return const_cast<structure *>(this)->getPolymerByAsymID(asymID);
-	// }
+	const polymer &get_polymer_by_asym_id(const std::string &asymID) const
+	{
+		return const_cast<structure *>(this)->get_polymer_by_asym_id(asymID);
+	}
 
 	const std::list<branch> &branches() const { return m_branches; }
 	std::list<branch> &branches() { return m_branches; }
@@ -797,8 +827,8 @@ class structure
 	// /// \brief Return the atom closest to point \a p
 	// atom getAtomByPosition(point p) const;
 
-	// /// \brief Return the atom closest to point \a p with atom type \a type in a residue of type \a res_type
-	// atom getAtomByPositionAndType(point p, std::string_view type, std::string_view res_type) const;
+	/// \brief Return the atom closest to point \a p with atom type \a type in a residue of type \a res_type
+	atom get_atom_by_position_and_type(point p, std::string_view type, std::string_view res_type) const;
 
 	/// \brief Get a non-poly residue for an asym with id \a asymID
 	residue &get_residue(const std::string &asymID)
@@ -917,11 +947,11 @@ class structure
 
 	void cleanup_empty_categories();
 
-	// /// \brief Direct access to underlying data
-	// category &category(std::string_view name) const
-	// {
-	// 	return m_db[name];
-	// }
+	/// \brief Direct access to underlying data
+	category &get_category(std::string_view name) const
+	{
+		return m_db[name];
+	}
 
 	datablock &get_datablock() const
 	{

src/category.cpp

@@ -885,10 +885,7 @@ condition category::get_parents_condition(row_handle rh, const category &parentC
 			cond = std::move(cond) and key(link->m_parent_keys[ix]) == childValue.text();
 		}
 
-		if (result)
-			result = std::move(result) or std::move(cond);
-		else
-			result = std::move(cond);
+		result = std::move(result) or std::move(cond);
 	}
 
 	return result;

src/model.cpp

@@ -393,37 +393,37 @@ void residue::add_atom(atom &atom)
 	m_atoms.push_back(atom);
 }
 
-// std::vector<atom> residue::unique_atoms() const
-// {
-// 	if (m_structure == nullptr)
-// 		throw std::runtime_error("Invalid residue object");
+std::vector<atom> residue::unique_atoms() const
+{
+	// if (m_structure == nullptr)
+	// 	throw std::runtime_error("Invalid residue object");
 
-// 	std::vector<atom> result;
-// 	std::string firstAlt;
+	std::vector<atom> result;
+	std::string firstAlt;
 
-// 	for (auto &atom : m_atoms)
-// 	{
-// 		auto alt = atom.get_label_alt_id();
-// 		if (alt.empty())
-// 		{
-// 			result.push_back(atom);
-// 			continue;
-// 		}
+	for (auto &atom : m_atoms)
+	{
+		auto alt = atom.get_label_alt_id();
+		if (alt.empty())
+		{
+			result.push_back(atom);
+			continue;
+		}
 
-// 		if (firstAlt.empty())
-// 			firstAlt = alt;
-// 		else if (alt != firstAlt)
-// 		{
-// 			if (VERBOSE > 0)
-// 				std::cerr << "skipping alternate atom " << atom << std::endl;
-// 			continue;
-// 		}
+		if (firstAlt.empty())
+			firstAlt = alt;
+		else if (alt != firstAlt)
+		{
+			if (VERBOSE > 0)
+				std::cerr << "skipping alternate atom " << atom << std::endl;
+			continue;
+		}
 
-// 		result.push_back(atom);
-// 	}
+		result.push_back(atom);
+	}
 
-// 	return result;
-// }
+	return result;
+}
 
 // std::set<std::string> residue::getAlternateIDs() const
 // {
@@ -1489,47 +1489,47 @@ atom structure::get_atom_by_id(const std::string &id) const
 // 	return {};
 // }
 
-// atom structure::getAtomByPositionAndType(point p, std::string_view type, std::string_view res_type) const
-// {
-// 	double distance = std::numeric_limits<double>::max();
-// 	size_t index = std::numeric_limits<size_t>::max();
+atom structure::get_atom_by_position_and_type(point p, std::string_view type, std::string_view res_type) const
+{
+	double dist = std::numeric_limits<double>::max();
+	size_t index = std::numeric_limits<size_t>::max();
 
-// 	for (size_t i = 0; i < m_atoms.size(); ++i)
-// 	{
-// 		auto &a = m_atoms.at(i);
+	for (size_t i = 0; i < m_atoms.size(); ++i)
+	{
+		auto &a = m_atoms.at(i);
 
-// 		if (a.labelCompID() != res_type)
-// 			continue;
+		if (a.get_label_comp_id() != res_type)
+			continue;
 
-// 		if (a.get_label_atom_id() != type)
-// 			continue;
+		if (a.get_label_atom_id() != type)
+			continue;
 
-// 		auto d = Distance(a.location(), p);
-// 		if (d < distance)
-// 		{
-// 			distance = d;
-// 			index = i;
-// 		}
-// 	}
+		auto d = distance(a.get_location(), p);
+		if (dist > d)
+		{
+			dist = d;
+			index = i;
+		}
+	}
 
-// 	if (index < m_atoms.size())
-// 		return m_atoms.at(index);
+	if (index < m_atoms.size())
+		return m_atoms.at(index);
 
-// 	return {};
-// }
+	return {};
+}
 
-// polymer &structure::getPolymerByAsym_id(const std::string &asym_id)
-// {
-// 	for (auto &poly : m_polymers)
-// 	{
-// 		if (poly.get_asym_id() != asym_id)
-// 			continue;
+polymer &structure::get_polymer_by_asym_id(const std::string &asym_id)
+{
+	for (auto &poly : m_polymers)
+	{
+		if (poly.get_asym_id() != asym_id)
+			continue;
 
-// 		return poly;
-// 	}
+		return poly;
+	}
 
-// 	throw std::runtime_error("polymer with asym id " + asym_id + " not found");
-// }
+	throw std::runtime_error("polymer with asym id " + asym_id + " not found");
+}
 
 residue &structure::get_residue(const std::string &asym_id, int seqID, const std::string &authSeqID)
 {

test/unit-v2-test.cpp

@@ -1514,6 +1514,170 @@ _cat_2.parent_id3
 
 // --------------------------------------------------------------------
 
+// --------------------------------------------------------------------
+
+BOOST_AUTO_TEST_CASE(d6)
+{
+	const char dict[] = R"(
+data_test_dict.dic
+    _datablock.id	test_dict.dic
+    _datablock.description
+;
+    A test dictionary
+;
+    _dictionary.title           test_dict.dic
+    _dictionary.datablock_id    test_dict.dic
+    _dictionary.version         1.0
+
+     loop_
+    _item_type_list.code
+    _item_type_list.primitive_code
+    _item_type_list.construct
+               code      char
+               '[][_,.;:"&<>()/\{}'`~!@#$%A-Za-z0-9*|+-]*'
+
+               text      char
+               '[][ \n\t()_,.;:"&<>/\{}'`~!@#$%?+=*A-Za-z0-9|^-]*'
+
+               int       numb
+               '[+-]?[0-9]+'
+
+save_cat_1
+    _category.description     'A simple test category'
+    _category.id              cat_1
+    _category.mandatory_code  yes
+    _category_key.name        '_cat_1.id'
+    save_
+
+save__cat_1.id
+    _item.name                '_cat_1.id'
+    _item.category_id         cat_1
+    _item.mandatory_code      yes
+    _item_type.code           int
+    save_
+
+save__cat_1.id_2
+    _item.name                '_cat_1.id_2'
+    _item.category_id         cat_1
+    _item.mandatory_code      no
+    _item_type.code           int
+    save_
+
+save_cat_2
+    _category.description     'A second simple test category'
+    _category.id              cat_2
+    _category.mandatory_code  no
+    _category_key.name        '_cat_2.id'
+    save_
+
+save__cat_2.id
+    _item.name                '_cat_2.id'
+    _item.category_id         cat_2
+    _item.mandatory_code      yes
+    _item_type.code           int
+    save_
+
+save__cat_2.parent_id
+    _item.name                '_cat_2.parent_id'
+    _item.category_id         cat_2
+    _item.mandatory_code      yes
+    _item_type.code           int
+    save_
+
+save__cat_2.parent_id_2
+    _item.name                '_cat_2.parent_id_2'
+    _item.category_id         cat_2
+    _item.mandatory_code      no
+    _item_type.code           code
+    save_
+
+loop_
+_pdbx_item_linked_group_list.child_category_id
+_pdbx_item_linked_group_list.link_group_id
+_pdbx_item_linked_group_list.child_name
+_pdbx_item_linked_group_list.parent_name
+_pdbx_item_linked_group_list.parent_category_id
+cat_2 1 '_cat_2.parent_id'  '_cat_1.id' cat_1
+cat_2 1 '_cat_2.parent_id_2' '_cat_1.id_2' cat_1
+
+loop_
+_pdbx_item_linked_group.category_id
+_pdbx_item_linked_group.link_group_id
+_pdbx_item_linked_group.label
+cat_2 1 cat_2:cat_1:1
+    )";
+
+	struct membuf : public std::streambuf
+	{
+		membuf(char *text, size_t length)
+		{
+			this->setg(text, text, text + length);
+		}
+	} buffer(const_cast<char *>(dict), sizeof(dict) - 1);
+
+	std::istream is_dict(&buffer);
+
+	auto validator = cif::parse_dictionary("test", is_dict);
+
+	cif::file f;
+	f.set_validator(&validator);
+
+	// --------------------------------------------------------------------
+
+	const char data[] = R"(
+data_test
+loop_
+_cat_1.id
+_cat_1.id_2
+1 1
+2 2
+3 ?
+
+loop_
+_cat_2.id
+_cat_2.parent_id
+_cat_2.parent_id_2
+ 0 1 1
+ 1 1 ?
+ 2 ? 1
+ 3 ? ?
+ 4 2 2
+ 5 3 1
+ 6 3 ?
+    )";
+
+	// --------------------------------------------------------------------
+
+	struct data_membuf : public std::streambuf
+	{
+		data_membuf(char *text, size_t length)
+		{
+			this->setg(text, text, text + length);
+		}
+	} data_buffer(const_cast<char *>(data), sizeof(data) - 1);
+
+	std::istream is_data(&data_buffer);
+	f.load(is_data);
+
+	auto &cat1 = f.front()["cat_1"];
+	auto &cat2 = f.front()["cat_2"];
+
+	// f.front().validate_links();
+	// BOOST_CHECK(not );
+
+	using namespace cif::literals;
+
+	BOOST_CHECK(	cat2.has_parents(cat2.find1("id"_key == 0)));
+	BOOST_CHECK(	cat2.has_parents(cat2.find1("id"_key == 1)));
+	BOOST_CHECK(	cat2.has_parents(cat2.find1("id"_key == 2)));
+	BOOST_CHECK(not	cat2.has_parents(cat2.find1("id"_key == 3)));
+	BOOST_CHECK(	cat2.has_parents(cat2.find1("id"_key == 4)));
+	BOOST_CHECK(	cat2.has_parents(cat2.find1("id"_key == 5)));
+	BOOST_CHECK(	cat2.has_parents(cat2.find1("id"_key == 6)));
+}
+
+// --------------------------------------------------------------------
+
 BOOST_AUTO_TEST_CASE(c1)
 {
 	cif::VERBOSE = 1;