Fix cif2pdb, somewhat...

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 10.0.2
+    VERSION 10.0.4
     LANGUAGES CXX C)
 
 list(PREPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake")

changelog

@@ -1,3 +1,13 @@
+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
 - Fixed symmetry operations

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())

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())
@@ -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();
-					}
 				}
 			}
 		}

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/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>
@@ -1340,6 +1341,8 @@ void createPdbxNonpolyScheme(datablock &db)
 	auto &pdbx_nonpoly_scheme = db["pdbx_nonpoly_scheme"];
 	auto &atom_site = db["atom_site"];
 
+	pdbx_nonpoly_scheme.clear();
+
 	for (const auto &[entity_id, comp_id] : pdbx_entity_nonpoly.rows<std::string, std::string>("entity_id", "comp_id"))
 	{
 		for (int ndb_nr = 1; auto row : atom_site.find("label_entity_id"_key == entity_id and "label_comp_id"_key == comp_id))
@@ -1660,7 +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;
 				}
 
@@ -1689,6 +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");
+}