Fix drop_empty_items

don't use get_item_indices
version bump

.github/workflows/build-documentation.yml

@@ -42,7 +42,7 @@ jobs:
 
     - name: Run Sphinx
       run: |
-        cmake --build ${{ steps.strings.outputs.build-output-dir }} --target Sphinx-libcifpp
+        cmake --build ${{ steps.strings.outputs.build-output-dir }} --target Sphinx-cifpp
         ls -l ${{ steps.strings.outputs.build-output-dir }}
         ls -l ${{ steps.strings.outputs.build-output-dir }}/docs/sphinx
 

CMakeLists.txt

@@ -34,7 +34,7 @@ endif()
 # set the project name
 project(
     libcifpp
-    VERSION 10.0.3
+    VERSION 10.0.4
     LANGUAGES CXX C)
 
 list(PREPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake")

changelog

@@ -1,5 +1,12 @@
+Version 10.0.4
+- Fixed find_by_value in the index of a category,
+  avoid swapping columns in the search keys
+
 Version 10.0.3
 - Clear pdbx_nonpoly_scheme before filling it in reconstruction
+- Changed handling of numbers with a preceding plus character,
+  these are now stored as strings to avoid inadvertently
+  mutilating phone numbers.
 
 Version 10.0.2
 - Fixed regression in reconstruction introduced in 10.0.1

include/cif++/item.hpp

@@ -331,8 +331,11 @@ class item_value
 			case TEXT:
 			{
 				auto sv = m_data.sv();
+				auto sp = sv.data();
+				if (*sp == '+')
+					++sp;
 				int64_t v;
-				auto &&[ptr, ec] = from_chars(sv.data(), sv.data() + sv.length(), v);
+				auto &&[ptr, ec] = from_chars(sp, sv.data() + sv.length(), v);
 				if (ec != std::errc{})
 					throw std::system_error(std::make_error_code(ec));
 				if (ptr != sv.data() + sv.length())
@@ -361,8 +364,11 @@ class item_value
 			case TEXT:
 			{
 				auto sv = m_data.sv();
+				auto sp = sv.data();
+				if (*sp == '+')
+					++sp;
 				double v;
-				auto &&[ptr, ec] = from_chars(sv.data(), sv.data() + sv.length(), v);
+				auto &&[ptr, ec] = from_chars(sp, sv.data() + sv.length(), v);
 				if (ec != std::errc{})
 					throw std::system_error(std::make_error_code(ec));
 				if (ptr != sv.data() + sv.length())

include/cif++/model.hpp

@@ -122,7 +122,7 @@ class atom
 
 		// const compound *compound() const;
 
-		[[nodiscard]] const item_value &get_property(std::string_view name) const;
+		[[nodiscard]] const item_handle get_property(std::string_view name) const;
 		void set_property(const std::string_view name, item_value value);
 
 		row_handle row()
@@ -227,7 +227,7 @@ class atom
 	explicit operator bool() const { return m_impl.operator bool(); }
 
 	/// \brief Return the item named @a name in the _atom_site category for this atom
-	[[nodiscard]] const item_value &get_property_value(std::string_view name) const
+	[[nodiscard]] const item_handle get_property_value(std::string_view name) const
 	{
 		if (not m_impl)
 			throw std::logic_error("Error trying to fetch a property from an uninitialized atom");

src/category.cpp

@@ -118,6 +118,7 @@ class row_comparator
 
 		for (const auto &[k, f] : m_comparator)
 		{
+			assert(cat.get_item_name(k) == ai->name);
 			d = f(ai->value, rhb[k].value());
 
 			if (d != 0)
@@ -363,10 +364,9 @@ row *category_index::find_by_value(const category &cat, const category::key_type
 	// sort the values in k first
 
 	category::key_type k2;
-	for (auto &f : cat.key_item_indices())
+	auto cv = cat.get_cat_validator();
+	for (auto &fld : cv->m_keys)
 	{
-		auto fld = cat.get_item_name(f);
-
 		auto ki = std::ranges::find_if(k, [&fld](auto &i)
 			{ return i.name == fld; });
 		if (ki == k.end())
@@ -614,11 +614,11 @@ void category::drop_empty_items()
 {
 	std::vector<bool> is_empty(get_item_count(), true);
 
-	for (auto row : *this)
+	for (size_t ix = 0; ix < get_item_count(); ++ix)
 	{
-		for (size_t ix = 0; ix < get_item_count(); ++ix)
+		for (auto row : *this)
 		{
-			if (is_empty[ix] and not row[ix].empty())
+			if (not row[ix].empty())
 			{
 				is_empty[ix] = false;
 				break;
@@ -759,14 +759,10 @@ void category::set_validator(const validator *v, datablock &db)
 				if (not v.is_number())
 				{
 					// Try cast the value to a number and throw in case of failure
-					try
-					{
+					if (auto sv = v.sv(); sv.find_first_of(".eE") == std::string_view::npos)
 						v.cast_to_int();
-					}
-					catch (...)
-					{
+					else
 						v.cast_to_float();
-					}
 				}
 			}
 		}
@@ -1616,7 +1612,6 @@ void category::update_value(row *row, uint16_t item, item_value value, bool upda
 		}
 		else if (ec)
 			throw validation_exception(ec, m_name, m_items[item].m_name);
-
 	}
 
 	// If the item is part of the Key for this category, remove it from the index
@@ -2204,7 +2199,7 @@ void category::write_cif(std::ostream &os, const std::vector<uint16_t> &order, b
 					offset = 0;
 				}
 
-				offset = detail::write_value(os, s, offset, w, right_aligned[cix]/*  iv->is_number() */);
+				offset = detail::write_value(os, s, offset, w, right_aligned[cix] /*  iv->is_number() */);
 
 				if (offset > 132)
 				{
@@ -2726,70 +2721,20 @@ bool category::operator==(const category &rhs) const
 
 	using namespace std::placeholders;
 
-	//	set<std::string> item_namesA(a.items()), item_namesB(b.items());
-	//
-	//	if (item_namesA != item_namesB)
-	//		std::cout << "Unequal number of items\n";
+	std::set<std::string> keys;
 
-	const category_validator *catValidator = nullptr;
-
-	auto validator = a.get_validator();
-	if (validator != nullptr)
-		catValidator = validator->get_validator_for_category(a.name());
-
-	using compType = std::function<int(std::string_view, std::string_view)>;
-	std::vector<std::tuple<std::string, compType>> item_names;
-	std::vector<std::string> keys;
-	std::vector<std::size_t> keyIx;
-
-	if (catValidator == nullptr)
+	for (const auto &items : { a.get_items(), b.get_items()})
 	{
-		for (auto &item_name : a.get_items())
-		{
-			item_names.emplace_back(item_name, [](std::string_view va, std::string_view vb)
-				{ return va.compare(vb); });
-			keyIx.push_back(keys.size());
-			keys.push_back(item_name);
-		}
+		for (auto &item_name : items)
+			keys.insert(item_name);
 	}
-	else
-	{
-		keys = catValidator->m_keys;
-
-		for (auto &item_name : a.key_items())
-		{
-			auto iv = catValidator->get_validator_for_item(item_name);
-			if (iv == nullptr)
-				throw std::runtime_error("missing item validator");
-			auto tv = iv->m_type;
-			if (tv == nullptr)
-				throw std::runtime_error("missing type validator");
-			item_names.emplace_back(item_name, [tv](auto &&a1, auto &&a2)
-				{ return tv->compare(std::forward<decltype(a1)>(a1), std::forward<decltype(a2)>(a2)); });
-
-			auto pred = [item_name](const std::string &s) -> bool
-			{
-				return cif::iequals(item_name, s) == 0;
-			};
-			if (std::ranges::find_if(keys, pred) == keys.end())
-				keyIx.push_back(item_names.size() - 1);
-		}
-	}
-
-	// a.reorderByIndex();
-	// b.reorderByIndex();
 
 	auto rowEqual = [&](const_row_handle &a, const_row_handle &b)
 	{
 		int d = 0;
 
-		for (auto kix : keyIx)
+		for (const auto &item_name : keys)
 		{
-			std::string item_name;
-			compType compare;
-
-			std::tie(item_name, compare) = item_names[kix];
-
 			d = a[item_name].compare(b[item_name]);
 
 			if (d != 0)
@@ -2810,28 +2755,6 @@ bool category::operator==(const category &rhs) const
 		if (not rowEqual(ra, rb))
 			return false;
 
-		std::vector<std::string> missingA, missingB, different;
-
-		for (auto &tt : item_names)
-		{
-			std::string item_name;
-			compType compare;
-
-			std::tie(item_name, compare) = tt;
-
-			// make it an option to compare unapplicable to empty or something
-
-			// auto ta = ra[item_name].text();
-			// if (ta == "." or ta == "?")
-			// 	ta = "";
-			// auto tb = rb[item_name].text();
-			// if (tb == "." or tb == "?")
-			// 	tb = "";
-
-			if (ra[item_name].compare(rb[item_name]) != 0)
-				return false;
-		}
-
 		++ai;
 		++bi;
 	}

src/compound.cpp

@@ -799,8 +799,8 @@ void compound_factory::report_missing_compound(std::string_view compound_id)
 		std::clog << "\n"
 				  << cif::coloured("Configuration error:", white, red) << "\n\n"
 				  << "The attempt to retrieve compound information for " << std::quoted(compound_id) << " failed.\n\n"
-				  << "This information is searched for in a CCD file called components.cif or\n"
-				  << "components.cif.gz which should be located in one of the following directories:\n\n";
+				  << "This information is searched for in a CCD file called components.cif\n"
+				  << "which should be located in one of the following directories:\n\n";
 
 		cif::list_data_directories(std::clog);
 

src/condition.cpp

@@ -129,9 +129,9 @@ namespace detail
 			m_item_ix = *ix;
 			m_icase = is_item_type_uchar(c, m_item_name);
 
-			if (c.get_cat_validator() != nullptr and
-				c.key_item_indices().contains(m_item_ix) and
-				c.key_item_indices().size() == 1)
+			if (auto cv = c.get_cat_validator();
+				cv != nullptr and cv->m_keys.size() == 1 and
+				cv->m_keys.front() == m_item_name)
 			{
 				m_single_hit = c[{ { m_item_name, m_value } }];
 			}

src/item.cpp

@@ -239,7 +239,12 @@ void item_value::cast_to_int()
 		{
 			auto s = sv();
 			int64_t v;
-			auto [ptr, ec] = cif::from_chars(s.data(), s.data() + s.size(), v);
+
+			auto sp = s.data();
+			if (*sp == '+')
+				++sp;
+
+			auto [ptr, ec] = cif::from_chars(sp, s.data() + s.size(), v);
 			if (ec != std::errc{})
 				throw std::system_error(std::make_error_code(ec), "attempt to cast value to integer failed");
 			if (ptr != s.data() + s.size())

src/model.cpp

@@ -26,6 +26,7 @@
 
 #include "cif++/cif++.hpp"
 #include "cif++/item.hpp"
+#include "cif++/row.hpp"
 
 #include <algorithm>
 #include <cassert>
@@ -76,10 +77,10 @@ void atom::atom_impl::moveTo(const point &p)
 
 // const compound *compound() const;
 
-const item_value &atom::atom_impl::get_property(std::string_view name) const
+const item_handle atom::atom_impl::get_property(std::string_view name) const
 {
 	if (auto rh = row(); rh)
-		return rh[name].value();
+		return rh[name];
 	throw std::runtime_error(std::format("Missing property {} for atom", name));
 }
 
@@ -1911,7 +1912,13 @@ void structure::change_residue(residue &res, const std::string &newCompound,
 	for (const auto &[a1, a2] : remappedAtoms)
 	{
 		auto i = std::ranges::find_if(atoms, [id = a1](const atom &a)
-			{ return a.get_label_atom_id() == id; });
+			{
+				if (a.get_row())
+				{
+					auto ih = a.get_property_value("label_atom_id");
+					return not ih.empty() and ih.get<std::string>() == id;
+				}
+				return false; });
 		if (i == atoms.end())
 		{
 			if (VERBOSE >= 0)
@@ -2170,29 +2177,50 @@ std::string structure::create_non_poly(const std::string &entity_id, const std::
 	{
 		auto atom_id = atom_site.get_unique_id("");
 
-		auto row = atom_site.emplace({ { "group_PDB", atom.get_property("group_PDB") },
-			{ "id", atom_id },
-			{ "type_symbol", atom.get_property_value("type_symbol") },
-			{ "label_atom_id", atom.get_property_value("label_atom_id") },
-			{ "label_alt_id", atom.get_property_value("label_alt_id") },
-			{ "label_comp_id", comp_id },
-			{ "label_asym_id", asym_id },
-			{ "label_entity_id", entity_id },
-			{ "label_seq_id", cif::item_value_type::INAPPLICABLE },
-			{ "pdbx_PDB_ins_code", cif::item_value_type::MISSING },
-			{ "Cartn_x", atom.get_property_value("Cartn_x") },
-			{ "Cartn_y", atom.get_property_value("Cartn_y") },
-			{ "Cartn_z", atom.get_property_value("Cartn_z") },
-			{ "occupancy", atom.get_property_value("occupancy") },
-			{ "B_iso_or_equiv", atom.get_property_value("B_iso_or_equiv") },
-			{ "pdbx_formal_charge", atom.get_property_value("pdbx_formal_charge") },
-			{ "auth_seq_id", "1" },
-			{ "auth_comp_id", comp_id },
-			{ "auth_asym_id", asym_id },
-			{ "auth_atom_id", atom.get_property_value("label_atom_id") },
-			{ "pdbx_PDB_model_num", m_model_nr } });
+		cif::row_initializer data //
+			{
+				{ "id", atom_id },
+				{ "label_comp_id", comp_id },
+				{ "label_asym_id", asym_id },
+				{ "label_entity_id", entity_id },
+				{ "label_seq_id", cif::item_value_type::INAPPLICABLE },
+				{ "pdbx_PDB_ins_code", cif::item_value_type::MISSING },
+				{ "auth_seq_id", "1" },
+				{ "auth_comp_id", comp_id },
+				{ "auth_asym_id", asym_id },
+				{ "pdbx_PDB_model_num", m_model_nr } //
+			};
+
+		for (auto item : std::initializer_list<std::string>{
+				 // clang-format off
+				 "group_PDB",
+				 "type_symbol",
+				 "label_atom_id",
+				 "label_alt_id",
+				 "auth_atom_id",
+				 "Cartn_x",
+				 "Cartn_y",
+				 "Cartn_z",
+				 "occupancy",
+				 "B_iso_or_equiv",
+				 "pdbx_formal_charge"
+				 // clang-format on
+			 })
+		{
+			auto v = atom.get_property_value(item);
+			if (not v.empty())
+				data.push_back({ item, v.value() });
+			else
+				data.push_back({ item, cif::item_value_type::INAPPLICABLE });
+		}
+
+		auto row = atom_site.emplace(std::move(data));
 
 		auto &newAtom = emplace_atom(std::make_shared<atom::atom_impl>(m_db, atom_id));
+
+		if (newAtom.get_property_value("auth_atom_id").empty() and not newAtom.get_property_value("label_atom_id").empty())
+			newAtom.set_property("auth_atom_id", newAtom.get_property("label_atom_id"));
+
 		res.add_atom(newAtom);
 	}
 

src/parser.cpp

@@ -25,6 +25,7 @@
  */
 
 #include "cif++/cif++.hpp"
+#include "cif++/utilities.hpp"
 
 #include <cassert>
 #include <cctype>
@@ -635,15 +636,28 @@ sac_parser::CIFToken sac_parser::get_next_token()
 
 	if (result == CIFToken::VALUE_NUMERIC_INTEGER)
 	{
+		// Avoid interpreting phone numbers as integers, TODO: check if this is an issue
 		auto [ptr, ec] = from_chars(m_token_buffer.data(), m_token_buffer.data() + m_token_buffer.size(), m_token_value_int);
 		if (ec != std::errc{})
-			error("Invalid integer value: " + std::make_error_code(ec).message());
+		{
+			if (cif::VERBOSE > 0)
+				std::clog << "Invalid integer value: " << std::make_error_code(ec).message() << '\n';
+
+			result = CIFToken::VALUE_CHARSTRING;
+			m_token_value = std::string_view(m_token_buffer.data(), m_token_buffer.size());
+		}
 	}
 	else if (result == CIFToken::VALUE_NUMERIC_FLOAT)
 	{
 		auto [ptr, ec] = from_chars(m_token_buffer.data(), m_token_buffer.data() + m_token_buffer.size(), m_token_value_float);
 		if (ec != std::errc{})
-			error("Invalid integer value: " + std::make_error_code(ec).message());
+		{
+			if (cif::VERBOSE > 0)
+				std::clog << "Invalid floating point value: " << std::make_error_code(ec).message() << '\n';
+
+			result = CIFToken::VALUE_CHARSTRING;
+			m_token_value = std::string_view(m_token_buffer.data(), m_token_buffer.size());
+		}
 	}
 
 	return result;

src/pdb/cif2pdb.cpp

@@ -1811,13 +1811,10 @@ void WriteRemark3Phenix(std::ostream &pdbFile, const datablock &db)
 void WriteRemark3XPlor(std::ostream &pdbFile, const datablock &db)
 {
 	auto refine = db["refine"].front();
-	auto ls_shell = db["refine_ls_shell"].front();
 	auto hist = db["refine_hist"].front();
 	auto reflns = db["reflns"].front();
 	auto analyze = db["refine_analyze"].front();
 	auto &ls_restr = db["refine_ls_restr"];
-	auto ls_restr_ncs = db["refine_ls_restr_ncs"].front();
-	auto pdbx_xplor_file = db["pdbx_xplor_file"].front();
 
 	pdbFile << RM3("") << '\n'
 			<< RM3(" DATA USED IN REFINEMENT.") << '\n'
@@ -1837,7 +1834,11 @@ void WriteRemark3XPlor(std::ostream &pdbFile, const datablock &db)
 			<< RM3("  FREE R VALUE                     : ", 7, 3) << Ff(refine, "ls_R_factor_R_free") << '\n'
 			<< RM3("  FREE R VALUE TEST SET SIZE   (%) : ", 7, 3) << Ff(refine, "ls_percent_reflns_R_free") << '\n'
 			<< RM3("  FREE R VALUE TEST SET COUNT      : ", 12, 6) << Fi(refine, "ls_number_reflns_R_free") << '\n'
-			<< RM3("  ESTIMATED ERROR OF FREE R VALUE  : ", 7, 3) << Ff(refine, "ls_R_factor_R_free_error") << '\n'
+			<< RM3("  ESTIMATED ERROR OF FREE R VALUE  : ", 7, 3) << Ff(refine, "ls_R_factor_R_free_error") << '\n';
+	if (not db["refine_ls_shell"].empty())
+	{
+		auto ls_shell = db["refine_ls_shell"].front();
+		pdbFile
 
 			<< RM3("") << '\n'
 			<< RM3(" FIT IN THE HIGHEST RESOLUTION BIN.") << '\n'
@@ -1850,60 +1851,68 @@ void WriteRemark3XPlor(std::ostream &pdbFile, const datablock &db)
 			<< RM3("  BIN FREE R VALUE                    : ", 7, 3) << Ff(ls_shell, "R_factor_R_free") << '\n'
 			<< RM3("  BIN FREE R VALUE TEST SET SIZE  (%) : ", 5, 1) << Ff(ls_shell, "percent_reflns_R_free") << '\n'
 			<< RM3("  BIN FREE R VALUE TEST SET COUNT     : ", 12, 6) << Fi(ls_shell, "number_reflns_R_free") << '\n'
-			<< RM3("  ESTIMATED ERROR OF BIN FREE R VALUE : ", 7, 3) << Ff(ls_shell, "R_factor_R_free_error") << '\n'
+			<< RM3("  ESTIMATED ERROR OF BIN FREE R VALUE : ", 7, 3) << Ff(ls_shell, "R_factor_R_free_error") << '\n';
+	}
 
-			<< RM3("") << '\n'
-			<< RM3(" NUMBER OF NON-HYDROGEN ATOMS USED IN REFINEMENT.") << '\n'
-			<< RM3("  PROTEIN ATOMS            : ", 12, 6) << Fi(hist, "pdbx_number_atoms_protein") << '\n'
-			<< RM3("  NUCLEIC ACID ATOMS       : ", 12, 6) << Fi(hist, "pdbx_number_atoms_nucleic_acid") << '\n'
-			<< RM3("  HETEROGEN ATOMS          : ", 12, 6) << Fi(hist, "pdbx_number_atoms_ligand") << '\n'
-			<< RM3("  SOLVENT ATOMS            : ", 12, 6) << Fi(hist, "number_atoms_solvent") << '\n'
+	pdbFile
+		<< RM3("") << '\n'
+		<< RM3(" NUMBER OF NON-HYDROGEN ATOMS USED IN REFINEMENT.") << '\n'
+		<< RM3("  PROTEIN ATOMS            : ", 12, 6) << Fi(hist, "pdbx_number_atoms_protein") << '\n'
+		<< RM3("  NUCLEIC ACID ATOMS       : ", 12, 6) << Fi(hist, "pdbx_number_atoms_nucleic_acid") << '\n'
+		<< RM3("  HETEROGEN ATOMS          : ", 12, 6) << Fi(hist, "pdbx_number_atoms_ligand") << '\n'
+		<< RM3("  SOLVENT ATOMS            : ", 12, 6) << Fi(hist, "number_atoms_solvent") << '\n'
 
-			<< RM3("") << '\n'
-			<< RM3(" B VALUES.") << '\n'
-			<< RM3("  FROM WILSON PLOT           (A**2) : ", 7, 2) << Ff(reflns, "B_iso_Wilson_estimate") << '\n'
-			<< RM3("  MEAN B VALUE      (OVERALL, A**2) : ", 7, 2) << Ff(refine, "B_iso_mean") << '\n'
+		<< RM3("") << '\n'
+		<< RM3(" B VALUES.") << '\n'
+		<< RM3("  FROM WILSON PLOT           (A**2) : ", 7, 2) << Ff(reflns, "B_iso_Wilson_estimate") << '\n'
+		<< RM3("  MEAN B VALUE      (OVERALL, A**2) : ", 7, 2) << Ff(refine, "B_iso_mean") << '\n'
 
-			<< RM3("  OVERALL ANISOTROPIC B VALUE.") << '\n'
-			<< RM3("   B11 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[1][1]") << '\n'
-			<< RM3("   B22 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[2][2]") << '\n'
-			<< RM3("   B33 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[3][3]") << '\n'
-			<< RM3("   B12 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[1][2]") << '\n'
-			<< RM3("   B13 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[1][3]") << '\n'
-			<< RM3("   B23 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[2][3]") << '\n'
+		<< RM3("  OVERALL ANISOTROPIC B VALUE.") << '\n'
+		<< RM3("   B11 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[1][1]") << '\n'
+		<< RM3("   B22 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[2][2]") << '\n'
+		<< RM3("   B33 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[3][3]") << '\n'
+		<< RM3("   B12 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[1][2]") << '\n'
+		<< RM3("   B13 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[1][3]") << '\n'
+		<< RM3("   B23 (A**2) : ", -7, 2) << Ff(refine, "aniso_B[2][3]") << '\n'
 
-			<< RM3("") << '\n'
-			<< RM3(" ESTIMATED COORDINATE ERROR.") << '\n'
-			<< RM3("  ESD FROM LUZZATI PLOT        (A) : ", 7, 2) << Ff(analyze, "Luzzati_coordinate_error_obs") << '\n'
-			<< RM3("  ESD FROM SIGMAA              (A) : ", 7, 2) << Ff(analyze, "Luzzati_sigma_a_obs") << '\n'
-			<< RM3("  LOW RESOLUTION CUTOFF        (A) : ", 7, 2) << Ff(analyze, "Luzzati_d_res_low_obs") << '\n'
+		<< RM3("") << '\n'
+		<< RM3(" ESTIMATED COORDINATE ERROR.") << '\n'
+		<< RM3("  ESD FROM LUZZATI PLOT        (A) : ", 7, 2) << Ff(analyze, "Luzzati_coordinate_error_obs") << '\n'
+		<< RM3("  ESD FROM SIGMAA              (A) : ", 7, 2) << Ff(analyze, "Luzzati_sigma_a_obs") << '\n'
+		<< RM3("  LOW RESOLUTION CUTOFF        (A) : ", 7, 2) << Ff(analyze, "Luzzati_d_res_low_obs") << '\n'
 
-			<< RM3("") << '\n'
-			<< RM3(" CROSS-VALIDATED ESTIMATED COORDINATE ERROR.") << '\n'
-			<< RM3("  ESD FROM C-V LUZZATI PLOT    (A) : ", 7, 2) << Ff(analyze, "Luzzati_coordinate_error_free") << '\n'
-			<< RM3("  ESD FROM C-V SIGMAA          (A) : ", 7, 2) << Ff(analyze, "Luzzati_sigma_a_free") << '\n'
+		<< RM3("") << '\n'
+		<< RM3(" CROSS-VALIDATED ESTIMATED COORDINATE ERROR.") << '\n'
+		<< RM3("  ESD FROM C-V LUZZATI PLOT    (A) : ", 7, 2) << Ff(analyze, "Luzzati_coordinate_error_free") << '\n'
+		<< RM3("  ESD FROM C-V SIGMAA          (A) : ", 7, 2) << Ff(analyze, "Luzzati_sigma_a_free") << '\n'
 
-			<< RM3("") << '\n'
-			<< RM3(" RMS DEVIATIONS FROM IDEAL VALUES.") << '\n'
-			<< RM3("  BOND LENGTHS                 (A) : ", 7, 3) << Ff(ls_restr, key("type") == "x_bond_d", "dev_ideal") << '\n'
-			<< RM3("  BOND ANGLES            (DEGREES) : ", 7, 2) << Ff(ls_restr, key("type") == "x_angle_deg", "dev_ideal") << '\n'
-			<< RM3("  DIHEDRAL ANGLES        (DEGREES) : ", 7, 2) << Ff(ls_restr, key("type") == "x_dihedral_angle_d", "dev_ideal") << '\n'
-			<< RM3("  IMPROPER ANGLES        (DEGREES) : ", 7, 2) << Ff(ls_restr, key("type") == "x_improper_angle_d", "dev_ideal") << '\n'
+		<< RM3("") << '\n'
+		<< RM3(" RMS DEVIATIONS FROM IDEAL VALUES.") << '\n'
+		<< RM3("  BOND LENGTHS                 (A) : ", 7, 3) << Ff(ls_restr, key("type") == "x_bond_d", "dev_ideal") << '\n'
+		<< RM3("  BOND ANGLES            (DEGREES) : ", 7, 2) << Ff(ls_restr, key("type") == "x_angle_deg", "dev_ideal") << '\n'
+		<< RM3("  DIHEDRAL ANGLES        (DEGREES) : ", 7, 2) << Ff(ls_restr, key("type") == "x_dihedral_angle_d", "dev_ideal") << '\n'
+		<< RM3("  IMPROPER ANGLES        (DEGREES) : ", 7, 2) << Ff(ls_restr, key("type") == "x_improper_angle_d", "dev_ideal") << '\n'
 
-			<< RM3("") << '\n'
-			<< RM3(" ISOTROPIC THERMAL MODEL : ") << Fs(refine, "pdbx_isotropic_thermal_model") << '\n'
+		<< RM3("") << '\n'
+		<< RM3(" ISOTROPIC THERMAL MODEL : ") << Fs(refine, "pdbx_isotropic_thermal_model") << '\n'
 
-			<< RM3("") << '\n'
-			<< RM3(" ISOTROPIC THERMAL FACTOR RESTRAINTS.    RMS    SIGMA") << '\n'
-			<< RM3("  MAIN-CHAIN BOND              (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_mcbond_it", "dev_ideal") << SEP("; ", 6, 2)
-			<< Ff(ls_restr, key("type") == "x_mcbond_it", "dev_ideal_target") << '\n'
-			<< RM3("  MAIN-CHAIN ANGLE             (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_mcangle_it", "dev_ideal") << SEP("; ", 6, 2)
-			<< Ff(ls_restr, key("type") == "x_mcangle_it", "dev_ideal_target") << '\n'
-			<< RM3("  SIDE-CHAIN BOND              (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_scbond_it", "dev_ideal") << SEP("; ", 6, 2)
-			<< Ff(ls_restr, key("type") == "x_scbond_it", "dev_ideal_target") << '\n'
-			<< RM3("  SIDE-CHAIN ANGLE             (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_scangle_it", "dev_ideal") << SEP("; ", 6, 2)
-			<< Ff(ls_restr, key("type") == "x_scangle_it", "dev_ideal_target") << '\n'
-			<< RM3("") << '\n'
+		<< RM3("") << '\n'
+		<< RM3(" ISOTROPIC THERMAL FACTOR RESTRAINTS.    RMS    SIGMA") << '\n'
+		<< RM3("  MAIN-CHAIN BOND              (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_mcbond_it", "dev_ideal") << SEP("; ", 6, 2)
+		<< Ff(ls_restr, key("type") == "x_mcbond_it", "dev_ideal_target") << '\n'
+		<< RM3("  MAIN-CHAIN ANGLE             (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_mcangle_it", "dev_ideal") << SEP("; ", 6, 2)
+		<< Ff(ls_restr, key("type") == "x_mcangle_it", "dev_ideal_target") << '\n'
+		<< RM3("  SIDE-CHAIN BOND              (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_scbond_it", "dev_ideal") << SEP("; ", 6, 2)
+		<< Ff(ls_restr, key("type") == "x_scbond_it", "dev_ideal_target") << '\n'
+		<< RM3("  SIDE-CHAIN ANGLE             (A**2) : ", 6, 2) << Ff(ls_restr, key("type") == "x_scangle_it", "dev_ideal") << SEP("; ", 6, 2)
+		<< Ff(ls_restr, key("type") == "x_scangle_it", "dev_ideal_target") << '\n'
+		<< RM3("") << '\n';
+
+	if (not db["refine_ls_restr_ncs"].empty())
+	{
+		auto ls_restr_ncs = db["refine_ls_restr_ncs"].front();
+
+		pdbFile
 			<< RM3(" NCS MODEL : ") << Fs(ls_restr_ncs, "ncs_model_details") << '\n'
 
 			<< RM3("") << '\n'
@@ -1913,7 +1922,14 @@ void WriteRemark3XPlor(std::ostream &pdbFile, const datablock &db)
 			<< RM3("  GROUP  1  POSITIONAL            (A) : ", 4, 2) << Ff(ls_restr_ncs, "rms_dev_position") << SEP("; ", 6, 2)
 			<< Ff(ls_restr_ncs, "weight_position") << SEP("; ", 6, 2) << '\n'
 			<< RM3("  GROUP  1  B-FACTOR           (A**2) : ", 4, 2) << Ff(ls_restr_ncs, "rms_dev_B_iso") << SEP("; ", 6, 2)
-			<< Ff(ls_restr_ncs, "weight_B_iso") << SEP("; ", 6, 2) << '\n'
+			<< Ff(ls_restr_ncs, "weight_B_iso") << SEP("; ", 6, 2) << '\n';
+	}
+
+	if (not db["pdbx_xplor_file"].empty())
+	{
+		auto pdbx_xplor_file = db["pdbx_xplor_file"].front();
+
+		pdbFile
 
 			// TODO: using only files from serial_no 1 here
 			<< RM3("") << '\n'
@@ -1921,6 +1937,7 @@ void WriteRemark3XPlor(std::ostream &pdbFile, const datablock &db)
 			<< RM3(" TOPOLOGY FILE   1   : ") << Fs(pdbx_xplor_file, "topol_file") << '\n'
 
 			<< RM3("") << '\n';
+	}
 }
 
 void WriteRemark3NuclSQ(std::ostream &pdbFile, const datablock &db)
@@ -2258,25 +2275,28 @@ void WriteRemark200(std::ostream &pdbFile, const datablock &db)
 			std::string iis = cifSoftware(db, eDataReduction);
 			std::string dss = cifSoftware(db, eDataScaling);
 
-			auto source = diffrn_source["source"].get<std::string>();
-			std::string synchrotron, type;
-
-			if (source.empty())
-				synchrotron = "NULL";
-			else if (iequals(source, "SYNCHROTRON"))
+			std::string source, synchrotron, type;
+			if (not diffrn_source.empty())
 			{
-				synchrotron = "Y";
-				source = diffrn_source["pdbx_synchrotron_site"].get<std::string>();
+				source = diffrn_source["source"].get<std::string>();
+	
 				if (source.empty())
-					source = "NULL";
-				type = "NULL";
-			}
-			else
-			{
-				synchrotron = "N";
-				type = diffrn_source["type"].get<std::string>();
-				if (type.empty())
+					synchrotron = "NULL";
+				else if (iequals(source, "SYNCHROTRON"))
+				{
+					synchrotron = "Y";
+					source = diffrn_source["pdbx_synchrotron_site"].get<std::string>();
+					if (source.empty())
+						source = "NULL";
 					type = "NULL";
+				}
+				else
+				{
+					synchrotron = "N";
+					type = diffrn_source["type"].get<std::string>();
+					if (type.empty())
+						type = "NULL";
+				}
 			}
 
 			if (source.empty())
@@ -2343,7 +2363,7 @@ void WriteRemark200(std::ostream &pdbFile, const datablock &db)
 
 			for (auto &t : kTail)
 			{
-				auto s = t.r[t.field].get<std::string>();
+				auto s = t.r.empty() ? "" : t.r[t.field].get<std::string>();
 
 				if (s.empty())
 				{
@@ -2384,6 +2404,9 @@ void WriteRemark280(std::ostream &pdbFile, const datablock &db)
 				<< RM("MATTHEWS COEFFICIENT, VM (ANGSTROMS**3/DA): ", 6, 2) << Ff(exptl_crystal, "density_Matthews") << '\n'
 				<< RM("") << '\n';
 
+			if (exptl_crystal_grow.empty())
+				continue;
+
 			std::vector<std::string> conditions;
 			auto add = [&conditions](const std::string c)
 			{
@@ -3341,9 +3364,9 @@ void WriteCrystallographic(std::ostream &pdbFile, const datablock &db)
 	if (r)
 	{
 		auto symmetry = r["space_group_name_H-M"].get<std::string>();
-	
+
 		r = db["cell"].find_first(key("entry_id") == db.name());
-	
+
 		pdbFile << std::format("CRYST1{:9.3f}{:9.3f}{:9.3f}{:7.2f}{:7.2f}{:7.2f} {:<11.11s}{:4}", r["length_a"].get<double>(), r["length_b"].get<double>(), r["length_c"].get<double>(), r["angle_alpha"].get<double>(), r["angle_beta"].get<double>(), r["angle_gamma"].get<double>(), symmetry, r["Z_PDB"].get<int>()) << '\n';
 	}
 }

src/pdb/reconstruct.cpp

@@ -25,6 +25,7 @@
  */
 
 #include "cif++/cif++.hpp"
+#include "cif++/validate.hpp"
 
 #include <algorithm>
 #include <cstddef>
@@ -1662,7 +1663,7 @@ bool reconstruct_pdbx(file &file, const validator &validator)
 				if (not iv)
 				{
 					// Drop this item
-					cat.remove_item(item_name);
+					// cat.remove_item(item_name);
 					continue;
 				}
 
@@ -1691,6 +1692,13 @@ bool reconstruct_pdbx(file &file, const validator &validator)
 								continue;
 						}
 
+						if (ec == cif::make_error_code(cif::validation_error::value_is_not_in_enumeration_list))
+						{
+							if (VERBOSE > 0)
+								std::clog << "Value (" << std::quoted(row[ix].str()) << ") for item " << item_name << " in category " << cat.name() << " is not valid since it is not in the list of allowed values\n";
+							continue;
+						}
+
 						if (VERBOSE > 0)
 							std::clog << "Replacing value (" << std::quoted(row[ix].str()) << ") for item " << item_name << " in category " << cat.name() << " since it does not validate: " << ec.message() << "\n";
 

test/unit-v2-test.cpp

@@ -3664,5 +3664,63 @@ HETATM 2 O O . HOH A 1 . ? 10.518 -1.781 0 1 37.22 ? O HOH 2 D 1
 )");
 }
 
+// --------------------------------------------------------------------
 
+TEST_CASE("number-test-1")
+{
+	auto data = R"(data_test
+_pdbx_contact_author.id      1
+_pdbx_contact_author.name_mi +98765432109
+)"_cf;
 
+	auto &db = data.front();
+	db.load_dictionary("mmcif_pdbx.dic");
+
+	auto r = db["pdbx_contact_author"].front();
+	CHECK(r["name_mi"].str() == "+98765432109");
+	CHECK(r["name_mi"].get<int64_t>() == 98765432109);
+	CHECK(r["name_mi"].get<double>() == 98765432109.0);
+}
+
+TEST_CASE("number-test-2")
+{
+	auto data = R"(data_test
+_pdbx_contact_author.id      1
+_pdbx_contact_author.name_mi '+98765432109'
+)"_cf;
+
+	auto &db = data.front();
+	db.load_dictionary("mmcif_pdbx.dic");
+
+	auto r = db["pdbx_contact_author"].front();
+	CHECK(r["name_mi"].str() == "+98765432109");
+	CHECK(r["name_mi"].get<int64_t>() == 98765432109);
+	CHECK(r["name_mi"].get<double>() == 98765432109.0);
+}// --------------------------------------------------------------------
+
+TEST_CASE("q-1")
+{
+	auto data = R"(data_test
+_test.s
+;1234567890
+1234567890
+;
+	)"_cf;
+
+	auto r = data.front()["test"].find(cif::key("s") == "1234567890\n1234567890");
+	CHECK(r.size() == 1);
+}
+
+TEST_CASE("large-int-1")
+{
+	auto data = R"(data_test
+_entry.id     82E4475FF8B27F36
+)"_cf;
+
+	auto &db = data.front();
+	db.load_dictionary("mmcif_pdbx.dic");
+
+	auto r = db["entry"].front();
+
+	CHECK(r["id"].get<std::string>() == "82E4475FF8B27F36");
+}