Fixed TLS parser, and more

CMakeLists.txt

@@ -198,6 +198,7 @@ set(project_sources
 	${PROJECT_SOURCE_DIR}/src/pdb/cif2pdb.cpp
 	${PROJECT_SOURCE_DIR}/src/pdb/cif2pdb.cpp
 	${PROJECT_SOURCE_DIR}/src/pdb/pdb2cif.cpp
 	${PROJECT_SOURCE_DIR}/src/pdb/pdb2cif.cpp
 	${PROJECT_SOURCE_DIR}/src/pdb/pdb2cif_remark_3.cpp
 	${PROJECT_SOURCE_DIR}/src/pdb/pdb2cif_remark_3.cpp
+	${PROJECT_SOURCE_DIR}/src/pdb/tls.cpp
 )
 )
 
 
 set(project_headers
 set(project_headers
@@ -227,6 +228,7 @@ set(project_headers
 	${PROJECT_SOURCE_DIR}/include/cif++/pdb/io.hpp
 	${PROJECT_SOURCE_DIR}/include/cif++/pdb/io.hpp
 	${PROJECT_SOURCE_DIR}/include/cif++/pdb/pdb2cif.hpp
 	${PROJECT_SOURCE_DIR}/include/cif++/pdb/pdb2cif.hpp
 	${PROJECT_SOURCE_DIR}/include/cif++/pdb/pdb2cif_remark_3.hpp
 	${PROJECT_SOURCE_DIR}/include/cif++/pdb/pdb2cif_remark_3.hpp
+	${PROJECT_SOURCE_DIR}/include/cif++/pdb/tls.hpp
 	)
 	)
 
 
 add_library(cifpp ${project_sources} ${project_headers} ${PROJECT_SOURCE_DIR}/src/symop_table_data.hpp)
 add_library(cifpp ${project_sources} ${project_headers} ${PROJECT_SOURCE_DIR}/src/symop_table_data.hpp)

include/cif++/category.hpp

@@ -278,7 +278,7 @@ class category
 	{
 	{
 		auto h = find<T>(pos, std::forward<condition>(cond), column);
 		auto h = find<T>(pos, std::forward<condition>(cond), column);
 
 
-		return h.size() == 1 ? std::get<0>(*h.begin()) : T{};
+		return h.size() == 1 ? *h.begin() : T{};
 	}
 	}
 
 
 	template <typename... Ts, typename... Cs, typename U = std::enable_if_t<sizeof...(Ts) != 1>>
 	template <typename... Ts, typename... Cs, typename U = std::enable_if_t<sizeof...(Ts) != 1>>
@@ -474,7 +474,8 @@ class category
 	}
 	}
 
 
 	// --------------------------------------------------------------------
 	// --------------------------------------------------------------------
-	
+
+	void sort(std::function<int(row_handle,row_handle)> f);
 	void reorder_by_index();
 	void reorder_by_index();
 
 
 	// --------------------------------------------------------------------
 	// --------------------------------------------------------------------

include/cif++/item.hpp

@@ -107,7 +107,7 @@ class item
 
 
 	/// \brief constructor for an item with name \a name and as
 	/// \brief constructor for an item with name \a name and as
 	/// content a the formatted integral value \a value
 	/// content a the formatted integral value \a value
-	template <typename T, std::enable_if_t<std::is_integral_v<T>, int> = 0>
+	template <typename T, std::enable_if_t<std::is_integral_v<T> and not std::is_same_v<T,bool>, int> = 0>
 	item(const std::string_view name, const T &value)
 	item(const std::string_view name, const T &value)
 		: m_name(name)
 		: m_name(name)
 	{
 	{
@@ -123,6 +123,15 @@ class item
 	}
 	}
 
 
 	/// \brief constructor for an item with name \a name and as
 	/// \brief constructor for an item with name \a name and as
+	/// content a the formatted boolean value \a value
+	template <typename T, std::enable_if_t<std::is_same_v<T,bool>, int> = 0>
+	item(const std::string_view name, const T &value)
+		: m_name(name)
+	{
+		m_value.assign(value ? "y" : "n");
+	}
+
+	/// \brief constructor for an item with name \a name and as
 	/// content value \a value
 	/// content value \a value
 	item(const std::string_view name, const std::string_view value)
 	item(const std::string_view name, const std::string_view value)
 		: m_name(name)
 		: m_name(name)

include/cif++/iterator.hpp

@@ -47,11 +47,13 @@ class iterator_impl
 	using category_type = std::remove_cv_t<Category>;
 	using category_type = std::remove_cv_t<Category>;
 	using row_type = std::conditional_t<std::is_const_v<Category>, const row, row>;
 	using row_type = std::conditional_t<std::is_const_v<Category>, const row, row>;
 
 
+	using tuple_type = std::tuple<Ts...>;
+
 	using iterator_category = std::forward_iterator_tag;
 	using iterator_category = std::forward_iterator_tag;
-	using value_type = std::conditional_t<N == 0, row_handle, std::tuple<Ts...>>;
+	using value_type = tuple_type;
 	using difference_type = std::ptrdiff_t;
 	using difference_type = std::ptrdiff_t;
-	using pointer = std::conditional_t<N == 0, row_handle, value_type *>;
-	using reference = std::conditional_t<N == 0, row_handle, value_type &>;
+	using pointer = value_type *;
+	using reference = value_type &;
 
 
 	iterator_impl() = default;
 	iterator_impl() = default;
 
 
@@ -66,56 +68,269 @@ class iterator_impl
 	{
 	{
 	}
 	}
 
 
+	template <typename IRowType>
+	iterator_impl(iterator_impl<IRowType, Ts...> &rhs)
+		: m_category(rhs.m_category)
+		, m_current(const_cast<row_type *>(rhs.m_current))
+		, m_value(rhs.m_value)
+		, m_column_ix(rhs.m_column_ix)
+	{
+		m_value = get(std::make_index_sequence<N>());
+	}
+
+	template <typename IRowType>
+	iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<size_t, N> &cix)
+		: m_category(rhs.m_category)
+		, m_current(rhs.m_current)
+		, m_column_ix(cix)
+	{
+		m_value = get(std::make_index_sequence<N>());
+	}
+
+	iterator_impl &operator=(const iterator_impl &i)
+	{
+		m_category = i.m_category;
+		m_current = i.m_current;
+		m_column_ix = i.m_column_ix;
+		m_value = i.m_value;
+		return *this;
+	}
+
+	virtual ~iterator_impl() = default;
+
+	reference operator*()
+	{
+		return m_value;
+	}
+
+	pointer operator->()
+	{
+		return &m_value;
+	}
+
+	operator const row_handle() const
+	{
+		return { *m_category, *m_current };
+	}
+
+	operator row_handle()
+	{
+		return { *m_category, *m_current };
+	}
+
+	iterator_impl &operator++()
+	{
+		if (m_current != nullptr)
+			m_current = m_current->m_next;
+
+		m_value = get(std::make_index_sequence<N>());
+
+		return *this;
+	}
+
+	iterator_impl operator++(int)
+	{
+		iterator_impl result(*this);
+		this->operator++();
+		return result;
+	}
+
+	bool operator==(const iterator_impl &rhs) const { return m_current == rhs.m_current; }
+	bool operator!=(const iterator_impl &rhs) const { return m_current != rhs.m_current; }
+
+	template <typename IRowType, typename... ITs>
+	bool operator==(const iterator_impl<IRowType, ITs...> &rhs) const
+	{
+		return m_current == rhs.m_current;
+	}
+
+	template <typename IRowType, typename... ITs>
+	bool operator!=(const iterator_impl<IRowType, ITs...> &rhs) const
+	{
+		return m_current != rhs.m_current;
+	}
+
+  private:
+	template <std::size_t... Is>
+	tuple_type get(std::index_sequence<Is...>) const
+	{
+		if (m_current != nullptr)
+		{
+			row_handle rh{*m_category, *m_current};
+			return tuple_type{rh[m_column_ix[Is]].template as<Ts>()...};
+		}
+
+		return {};
+	}
+
+	category_type *m_category = nullptr;
+	row_type *m_current = nullptr;
+	value_type m_value;
+	std::array<size_t, N> m_column_ix;
+};
+
+template<typename Category>
+class iterator_impl<Category>
+{
+  public:
+	template <typename, typename...>
+	friend class iterator_impl;
+
+	friend class category;
+	using category_type = std::remove_cv_t<Category>;
+	using row_type = std::conditional_t<std::is_const_v<Category>, const row, row>;
+
+	using iterator_category = std::forward_iterator_tag;
+	using value_type = row_handle;
+	using difference_type = std::ptrdiff_t;
+	using pointer = row_handle;
+	using reference = row_handle;
+
+	iterator_impl() = default;
+
+	iterator_impl(const iterator_impl &rhs) = default;
+
+	template <typename C2>
+	iterator_impl(const iterator_impl<C2> &rhs)
+		: m_category(rhs.m_category)
+		, m_current(const_cast<row_type*>(rhs.m_current))
+	{
+	}
+
 	iterator_impl(Category &cat, row *current)
 	iterator_impl(Category &cat, row *current)
 		: m_category(const_cast<category_type *>(&cat))
 		: m_category(const_cast<category_type *>(&cat))
 		, m_current(current)
 		, m_current(current)
-		, m_value(*m_category, *current)
 	{
 	{
-		static_assert(N == 0, "Only valid if this is a row iterator, not a row<xxx> iterator");
 	}
 	}
 
 
-	// iterator_impl(ItemRow *data)
-	// 	: m_current(data)
-	// {
-	// 	static_assert(N == 0, "Only valid if this is a row iterator, not a row<xxx> iterator");
-	// }
+	template <typename IRowType>
+	iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<size_t, 0> &cix)
+		: m_category(rhs.m_category)
+		, m_current(rhs.m_current)
+	{
+	}
 
 
-	// iterator_impl(ItemRow *data, const std::array<size_t, N> &cix)
-	// 	: m_current(data)
-	// 	, m_column_ix(cix)
-	// {
-	// }
+	iterator_impl &operator=(const iterator_impl &i)
+	{
+		m_category = i.m_category;
+		m_current = i.m_current;
+		return *this;
+	}
+
+	virtual ~iterator_impl() = default;
+
+	reference operator*()
+	{
+		return {*m_category, *m_current};
+	}
+
+	pointer operator->()
+	{
+		return &m_current;
+	}
+
+	operator const row_handle() const
+	{
+		return { *m_category, *m_current };
+	}
+
+	operator row_handle()
+	{
+		return { *m_category, *m_current };
+	}
+
+	iterator_impl &operator++()
+	{
+		if (m_current != nullptr)
+			m_current = m_current->m_next;
+
+		return *this;
+	}
+
+	iterator_impl operator++(int)
+	{
+		iterator_impl result(*this);
+		this->operator++();
+		return result;
+	}
+
+	bool operator==(const iterator_impl &rhs) const { return m_current == rhs.m_current; }
+	bool operator!=(const iterator_impl &rhs) const { return m_current != rhs.m_current; }
+
+	template <typename IRowType, typename... ITs>
+	bool operator==(const iterator_impl<IRowType, ITs...> &rhs) const
+	{
+		return m_current == rhs.m_current;
+	}
+
+	template <typename IRowType, typename... ITs>
+	bool operator!=(const iterator_impl<IRowType, ITs...> &rhs) const
+	{
+		return m_current != rhs.m_current;
+	}
+
+  private:
+	category_type *m_category = nullptr;
+	row_type *m_current = nullptr;
+};
+
+
+template<typename Category, typename T>
+class iterator_impl<Category, T>
+{
+  public:
+	template <typename, typename...>
+	friend class iterator_impl;
+
+	friend class category;
+
+	using category_type = std::remove_cv_t<Category>;
+	using row_type = std::conditional_t<std::is_const_v<Category>, const row, row>;
+
+	using iterator_category = std::forward_iterator_tag;
+	using value_type = T;
+	using difference_type = std::ptrdiff_t;
+	using pointer = value_type *;
+	using reference = value_type &;
+
+	iterator_impl() = default;
+
+	iterator_impl(const iterator_impl &rhs) = default;
+
+	template <typename C2, typename T2>
+	iterator_impl(const iterator_impl<C2, T2> &rhs)
+		: m_category(rhs.m_category)
+		, m_current(rhs.m_current)
+		, m_value(rhs.m_value)
+		, m_column_ix(rhs.m_column_ix)
+	{
+	}
 
 
 	template <typename IRowType>
 	template <typename IRowType>
-	iterator_impl(iterator_impl<IRowType, Ts...> &rhs)
+	iterator_impl(iterator_impl<IRowType, T> &rhs)
 		: m_category(rhs.m_category)
 		: m_category(rhs.m_category)
 		, m_current(const_cast<row_type *>(rhs.m_current))
 		, m_current(const_cast<row_type *>(rhs.m_current))
 		, m_value(rhs.m_value)
 		, m_value(rhs.m_value)
 		, m_column_ix(rhs.m_column_ix)
 		, m_column_ix(rhs.m_column_ix)
 	{
 	{
-		if constexpr (N > 0)
-			m_value = get(m_current, std::make_index_sequence<N>());
+		m_value = get(m_current);
 	}
 	}
 
 
 	template <typename IRowType>
 	template <typename IRowType>
-	iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<size_t, N> &cix)
+	iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<size_t, 1> &cix)
 		: m_category(rhs.m_category)
 		: m_category(rhs.m_category)
 		, m_current(rhs.m_current)
 		, m_current(rhs.m_current)
-		, m_column_ix(cix)
+		, m_column_ix(cix[0])
 	{
 	{
-		if constexpr (N > 0)
-			m_value = get(std::make_index_sequence<N>());
+		m_value = get();
 	}
 	}
 
 
 	iterator_impl &operator=(const iterator_impl &i)
 	iterator_impl &operator=(const iterator_impl &i)
 	{
 	{
 		m_category = i.m_category;
 		m_category = i.m_category;
 		m_current = i.m_current;
 		m_current = i.m_current;
-		if constexpr (N != 0)
-		{
-			m_column_ix = i.m_column_ix;
-			m_value = i.m_value;
-		}
+		m_column_ix = i.m_column_ix;
+		m_value = i.m_value;
 		return *this;
 		return *this;
 	}
 	}
 
 
@@ -123,18 +338,12 @@ class iterator_impl
 
 
 	reference operator*()
 	reference operator*()
 	{
 	{
-		if constexpr (N == 0)
-			return {*m_category, *m_current};
-		else
-			return m_value;
+		return m_value;
 	}
 	}
 
 
 	pointer operator->()
 	pointer operator->()
 	{
 	{
-		if constexpr (N == 0)
-			return &m_current;
-		else
-			return &m_value;
+		return &m_value;
 	}
 	}
 
 
 	operator const row_handle() const
 	operator const row_handle() const
@@ -152,8 +361,7 @@ class iterator_impl
 		if (m_current != nullptr)
 		if (m_current != nullptr)
 			m_current = m_current->m_next;
 			m_current = m_current->m_next;
 
 
-		if constexpr (N != 0)
-			m_value = get(std::make_index_sequence<N>());
+		m_value = get();
 
 
 		return *this;
 		return *this;
 	}
 	}
@@ -181,13 +389,12 @@ class iterator_impl
 	}
 	}
 
 
   private:
   private:
-	template <std::size_t... Is>
-	std::tuple<Ts...> get(std::index_sequence<Is...>) const
+	value_type get() const
 	{
 	{
 		if (m_current != nullptr)
 		if (m_current != nullptr)
 		{
 		{
 			row_handle rh{*m_category, *m_current};
 			row_handle rh{*m_category, *m_current};
-			return std::tuple<Ts...>{rh[m_column_ix[Is]].template as<Ts>()...};
+			return rh[m_column_ix].template as<T>();
 		}
 		}
 
 
 		return {};
 		return {};
@@ -196,7 +403,7 @@ class iterator_impl
 	category_type *m_category = nullptr;
 	category_type *m_category = nullptr;
 	row_type *m_current = nullptr;
 	row_type *m_current = nullptr;
 	value_type m_value;
 	value_type m_value;
-	std::array<size_t, N> m_column_ix;
+	size_t m_column_ix;
 };
 };
 
 
 // --------------------------------------------------------------------
 // --------------------------------------------------------------------

include/cif++/model.hpp

@@ -270,6 +270,7 @@ class atom
 	std::string get_auth_seq_id() const { return get_property("auth_seq_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_auth_atom_id() const { return get_property("auth_atom_id"); }
 	std::string get_auth_alt_id() const { return get_property("auth_alt_id"); }
 	std::string get_auth_alt_id() const { return get_property("auth_alt_id"); }
+	std::string get_auth_comp_id() const { return get_property("auth_comp_id"); }
 	std::string get_pdb_ins_code() const { return get_property("pdbx_PDB_ins_code"); }
 	std::string get_pdb_ins_code() const { return get_property("pdbx_PDB_ins_code"); }
 
 
 	bool is_alternate() const { return not get_label_alt_id().empty(); }
 	bool is_alternate() const { return not get_label_alt_id().empty(); }

include/cif++/pdb/TlsParser.hpp → include/cif++/pdb/tls.hpp

 /*-
 /*-
  * SPDX-License-Identifier: BSD-2-Clause
  * SPDX-License-Identifier: BSD-2-Clause
- * 
+ *
  * Copyright (c) 2020 NKI/AVL, Netherlands Cancer Institute
  * Copyright (c) 2020 NKI/AVL, Netherlands Cancer Institute
- * 
+ *
  * Redistribution and use in source and binary forms, with or without
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions are met:
  * modification, are permitted provided that the following conditions are met:
- * 
+ *
  * 1. Redistributions of source code must retain the above copyright notice, this
  * 1. Redistributions of source code must retain the above copyright notice, this
  *    list of conditions and the following disclaimer
  *    list of conditions and the following disclaimer
  * 2. Redistributions in binary form must reproduce the above copyright notice,
  * 2. Redistributions in binary form must reproduce the above copyright notice,
  *    this list of conditions and the following disclaimer in the documentation
  *    this list of conditions and the following disclaimer in the documentation
  *    and/or other materials provided with the distribution.
  *    and/or other materials provided with the distribution.
- * 
+ *
  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
  * 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
  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
  * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
  * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
@@ -26,32 +26,30 @@
 
 
 #pragma once
 #pragma once
 
 
-#include <vector>
 #include <string>
 #include <string>
 #include <tuple>
 #include <tuple>
+#include <vector>
 
 
 #include <cif++.hpp>
 #include <cif++.hpp>
 
 
 namespace cif
 namespace cif
 {
 {
-	
+
 extern const int
 extern const int
 	kResidueNrWildcard,
 	kResidueNrWildcard,
 	kNoSeqNum;
 	kNoSeqNum;
 
 
-struct TLSSelection;
-typedef std::unique_ptr<TLSSelection> TLSSelectionPtr;
-
-struct TLSResidue;
+struct tls_selection;
+struct tls_residue;
 
 
-struct TLSSelection
+struct tls_selection
 {
 {
-	virtual ~TLSSelection() {}
-	virtual void CollectResidues(cif::datablock& db, std::vector<TLSResidue>& residues, std::size_t indentLevel = 0) const = 0;
-	std::vector<std::tuple<std::string,int,int>> GetRanges(cif::datablock& db, bool pdbNamespace) const;
+	virtual ~tls_selection() {}
+	virtual void collect_residues(cif::datablock &db, std::vector<tls_residue> &residues, std::size_t indentLevel = 0) const = 0;
+	std::vector<std::tuple<std::string, int, int>> get_ranges(cif::datablock &db, bool pdbNamespace) const;
 };
 };
 
 
 // Low level: get the selections
 // Low level: get the selections
-TLSSelectionPtr ParseSelectionDetails(const std::string& program, const std::string& selection);
+std::unique_ptr<tls_selection> parse_tls_selection_details(const std::string &program, const std::string &selection);
 
 
-}
+} // namespace cif

src/category.cpp

@@ -1679,6 +1679,33 @@ void category::swap_item(size_t column_ix, row_handle &a, row_handle &b)
 	std::swap(ra.at(column_ix), rb.at(column_ix));
 	std::swap(ra.at(column_ix), rb.at(column_ix));
 }
 }
 
 
+void category::sort(std::function<int(row_handle,row_handle)> f)
+{
+	if (m_head == nullptr)
+		return;
+
+	std::vector<row_handle> rows;
+	for (auto itemRow = m_head; itemRow != nullptr; itemRow = itemRow->m_next)
+		rows.emplace_back(*this, *itemRow);
+
+	std::stable_sort(rows.begin(), rows.end(),
+		[&f](row_handle ia, row_handle ib)
+		{
+			return f(ia, ib) < 0;
+		});
+
+	m_head = rows.front().get_row();
+	m_tail = rows.back().get_row();
+
+	auto r = m_head;
+	for (size_t i = 1; i < rows.size(); ++i)
+		r = r->m_next = rows[i].get_row();
+	r->m_next = nullptr;
+
+	assert(r == m_tail);
+	assert(size() == rows.size());	
+}
+
 void category::reorder_by_index()
 void category::reorder_by_index()
 {
 {
 	if (m_index)
 	if (m_index)

src/model.cpp

@@ -1107,7 +1107,7 @@ branch::branch(structure &structure, const std::string &asym_id)
 	auto &branch_scheme = db["pdbx_branch_scheme"];
 	auto &branch_scheme = db["pdbx_branch_scheme"];
 	auto &branch_link = db["pdbx_entity_branch_link"];
 	auto &branch_link = db["pdbx_entity_branch_link"];
 
 
-	for (const auto &[entity_id] : struct_asym.find<std::string>("id"_key == asym_id, "entity_id"))
+	for (const auto &entity_id : struct_asym.find<std::string>("id"_key == asym_id, "entity_id"))
 	{
 	{
 		for (const auto &[comp_id, num] : branch_scheme.find<std::string, int>(
 		for (const auto &[comp_id, num] : branch_scheme.find<std::string, int>(
 				 "asym_id"_key == asym_id, "mon_id", "pdb_seq_num"))
 				 "asym_id"_key == asym_id, "mon_id", "pdb_seq_num"))
@@ -1313,7 +1313,7 @@ void structure::load_data()
 
 
 	auto &branchScheme = m_db["pdbx_branch_scheme"];
 	auto &branchScheme = m_db["pdbx_branch_scheme"];
 
 
-	for (const auto &[asym_id] : branchScheme.rows<std::string>("asym_id"))
+	for (const auto &asym_id : branchScheme.rows<std::string>("asym_id"))
 	{
 	{
 		if (m_branches.empty() or m_branches.back().get_asym_id() != asym_id)
 		if (m_branches.empty() or m_branches.back().get_asym_id() != asym_id)
 			m_branches.emplace_back(*this, asym_id);
 			m_branches.emplace_back(*this, asym_id);
@@ -2523,14 +2523,14 @@ void structure::cleanup_empty_categories()
 
 
 		std::optional<size_t> count;
 		std::optional<size_t> count;
 		if (type == "polymer")
 		if (type == "polymer")
-			count = m_db["entity_poly"].find("entity_id"_key == id).size();
+			count = m_db["struct_asym"].find("entity_id"_key == id).size();
 		else if (type == "non-polymer" or type == "water")
 		else if (type == "non-polymer" or type == "water")
 			count = m_db["pdbx_nonpoly_scheme"].find("entity_id"_key == id).size();
 			count = m_db["pdbx_nonpoly_scheme"].find("entity_id"_key == id).size();
 		else if (type == "branched")
 		else if (type == "branched")
 		{
 		{
 			// is this correct?
 			// is this correct?
 			std::set<std::string> asym_ids;
 			std::set<std::string> asym_ids;
-			for (const auto &[asym_id] : m_db["pdbx_branch_scheme"].find<std::string>("entity_id"_key == id, "asym_id"))
+			for (const auto &asym_id : m_db["pdbx_branch_scheme"].find<std::string>("entity_id"_key == id, "asym_id"))
 				asym_ids.insert(asym_id);
 				asym_ids.insert(asym_id);
 			count = asym_ids.size();
 			count = asym_ids.size();
 		}
 		}

src/pdb/pdb2cif.cpp

@@ -3006,21 +3006,20 @@ void PDBFileParser::ParseRemark200()
 	if (inRM200({ "REJECTION CRITERIA (SIGMA(I))", "RESOLUTION RANGE HIGH (A)", "RESOLUTION RANGE LOW (A)", "NUMBER OF UNIQUE REFLECTIONS", "COMPLETENESS FOR RANGE (%)", "<I/SIGMA(I)> FOR THE DATA SET", "R MERGE (I)", "R SYM (I)", "DATA REDUNDANCY" }))
 	if (inRM200({ "REJECTION CRITERIA (SIGMA(I))", "RESOLUTION RANGE HIGH (A)", "RESOLUTION RANGE LOW (A)", "NUMBER OF UNIQUE REFLECTIONS", "COMPLETENESS FOR RANGE (%)", "<I/SIGMA(I)> FOR THE DATA SET", "R MERGE (I)", "R SYM (I)", "DATA REDUNDANCY" }))
 	{
 	{
 		auto cat = getCategory("reflns");
 		auto cat = getCategory("reflns");
-		if (cat->empty())
-			cat->emplace({});
-		auto r = cat->back();
-		r["entry_id"] = mStructureID;
-		r["observed_criterion_sigma_I"] = mRemark200["REJECTION CRITERIA (SIGMA(I))"];
-		r["d_resolution_high"] = mRemark200["RESOLUTION RANGE HIGH (A)"];
-		r["d_resolution_low"] = mRemark200["RESOLUTION RANGE LOW (A)"];
-		r["number_obs"] = mRemark200["NUMBER OF UNIQUE REFLECTIONS"];
-		r["percent_possible_obs"] = mRemark200["COMPLETENESS FOR RANGE (%)"];
-		r["pdbx_netI_over_sigmaI"] = mRemark200["<I/SIGMA(I)> FOR THE DATA SET"];
-		r["pdbx_Rmerge_I_obs"] = mRemark200["R MERGE (I)"];
-		r["pdbx_Rsym_value"] = mRemark200["R SYM (I)"];
-		r["pdbx_redundancy"] = mRemark200["DATA REDUNDANCY"];
-		r["pdbx_ordinal"] = 1;
-		r["pdbx_diffrn_id"] = 1;
+		cat->emplace({
+			{ "entry_id", mStructureID },
+			{ "observed_criterion_sigma_I", mRemark200["REJECTION CRITERIA (SIGMA(I))"] },
+			{ "d_resolution_high", mRemark200["RESOLUTION RANGE HIGH (A)"] },
+			{ "d_resolution_low", mRemark200["RESOLUTION RANGE LOW (A)"] },
+			{ "number_obs", mRemark200["NUMBER OF UNIQUE REFLECTIONS"] },
+			{ "percent_possible_obs", mRemark200["COMPLETENESS FOR RANGE (%)"] },
+			{ "pdbx_netI_over_sigmaI", mRemark200["<I/SIGMA(I)> FOR THE DATA SET"] },
+			{ "pdbx_Rmerge_I_obs", mRemark200["R MERGE (I)"] },
+			{ "pdbx_Rsym_value", mRemark200["R SYM (I)"] },
+			{ "pdbx_redundancy", mRemark200["DATA REDUNDANCY"] },
+			{ "pdbx_ordinal", 1 },
+			{ "pdbx_diffrn_id", 1 }
+		});
 	}
 	}
 
 
 	if (inRM200({ "HIGHEST RESOLUTION SHELL, RANGE HIGH (A)" })) // that one field is mandatory...
 	if (inRM200({ "HIGHEST RESOLUTION SHELL, RANGE HIGH (A)" })) // that one field is mandatory...
@@ -3987,12 +3986,12 @@ void PDBFileParser::ConstructEntities()
 				{
 				{
 					seqAlignBeg = pdbxPolySeqScheme.find1<int>(key("pdb_strand_id") == std::string { dbref.chainID } and
 					seqAlignBeg = pdbxPolySeqScheme.find1<int>(key("pdb_strand_id") == std::string { dbref.chainID } and
 													key("pdb_seq_num") == dbref.seqBegin and
 													key("pdb_seq_num") == dbref.seqBegin and
-													key("pdb_ins_code") == insToStr(dbref.insertBegin),
+													(key("pdb_ins_code") == insToStr(dbref.insertBegin) or key("pdb_ins_code") == cif::null),
 													"seq_id");
 													"seq_id");
 
 
 					seqAlignEnd = pdbxPolySeqScheme.find1<int>(key("pdb_strand_id") == std::string { dbref.chainID } and
 					seqAlignEnd = pdbxPolySeqScheme.find1<int>(key("pdb_strand_id") == std::string { dbref.chainID } and
 													key("pdb_seq_num") == dbref.seqEnd and
 													key("pdb_seq_num") == dbref.seqEnd and
-													key("pdb_ins_code") == insToStr(dbref.insertEnd),
+													(key("pdb_ins_code") == insToStr(dbref.insertEnd) or key("pdb_ins_code") == cif::null),
 													"seq_id");
 													"seq_id");
 				}
 				}
 				catch (...)
 				catch (...)
@@ -4001,20 +4000,20 @@ void PDBFileParser::ConstructEntities()
 
 
 				getCategory("struct_ref_seq")->emplace({
 				getCategory("struct_ref_seq")->emplace({
 					{ "align_id", structRefSeqAlignID },
 					{ "align_id", structRefSeqAlignID },
-				{ "ref_id", structRefID },
-				{ "pdbx_PDB_id_code", dbref.PDBIDCode },
-				{ "pdbx_strand_id", std::string{ chain.mDbref.chainID } },
-				{ "seq_align_beg", seqAlignBeg },
-				{ "pdbx_seq_align_beg_ins_code", insToStr(dbref.insertBegin) },
-				{ "seq_align_end", seqAlignEnd },
-				{ "pdbx_seq_align_end_ins_code", insToStr(dbref.insertEnd) },
-				{ "pdbx_db_accession", dbref.dbAccession },
-				{ "db_align_beg", dbref.dbSeqBegin },
-				{ "pdbx_db_align_beg_ins_code", insToStr(dbref.dbinsBeg) },
-				{ "db_align_end", dbref.dbSeqEnd },
-				{ "pdbx_db_align_end_ins_code", insToStr(dbref.dbinsEnd) },
-				{ "pdbx_auth_seq_align_beg", dbref.seqBegin },
-				{ "pdbx_auth_seq_align_end", dbref.seqEnd } });
+					{ "ref_id", structRefID },
+					{ "pdbx_PDB_id_code", dbref.PDBIDCode },
+					{ "pdbx_strand_id", std::string{ chain.mDbref.chainID } },
+					{ "seq_align_beg", seqAlignBeg },
+					{ "pdbx_seq_align_beg_ins_code", insToStr(dbref.insertBegin) },
+					{ "seq_align_end", seqAlignEnd },
+					{ "pdbx_seq_align_end_ins_code", insToStr(dbref.insertEnd) },
+					{ "pdbx_db_accession", dbref.dbAccession },
+					{ "db_align_beg", dbref.dbSeqBegin },
+					{ "pdbx_db_align_beg_ins_code", insToStr(dbref.dbinsBeg) },
+					{ "db_align_end", dbref.dbSeqEnd },
+					{ "pdbx_db_align_end_ins_code", insToStr(dbref.dbinsEnd) },
+					{ "pdbx_auth_seq_align_beg", dbref.seqBegin },
+					{ "pdbx_auth_seq_align_end", dbref.seqEnd } });
 
 
 				// write the struct_ref_seq_dif
 				// write the struct_ref_seq_dif
 				for (auto &seqadv : mSeqadvs)
 				for (auto &seqadv : mSeqadvs)
@@ -5696,6 +5695,8 @@ void PDBFileParser::Parse(std::istream &is, cif::file &result)
 {
 {
 	try
 	try
 	{
 	{
+		mDatablock.set_validator(result.get_validator());
+
 		PreParseInput(is);
 		PreParseInput(is);
 
 
 		mRec = mData;
 		mRec = mData;
@@ -6164,10 +6165,10 @@ void ReadPDBFile(std::istream &pdbFile, cif::file &cifFile)
 {
 {
 	PDBFileParser p;
 	PDBFileParser p;
 
 
-	p.Parse(pdbFile, cifFile);
-
 	cifFile.load_dictionary("mmcif_pdbx");
 	cifFile.load_dictionary("mmcif_pdbx");
 
 
+	p.Parse(pdbFile, cifFile);
+
 	if (not cifFile.is_valid() and cif::VERBOSE >= 0)
 	if (not cifFile.is_valid() and cif::VERBOSE >= 0)
 		std::cerr << "Resulting mmCIF file is not valid!" << std::endl;
 		std::cerr << "Resulting mmCIF file is not valid!" << std::endl;
 }
 }

src/pdb/pdb2cif_remark_3.cpp

@@ -971,6 +971,7 @@ Remark3Parser::Remark3Parser(const std::string &name, const std::string &expMeth
 	, mTemplateCount(templateLineCount)
 	, mTemplateCount(templateLineCount)
 	, mProgramVersion(programversion)
 	, mProgramVersion(programversion)
 {
 {
+	mDb.set_validator(db.get_validator());
 }
 }
 
 
 std::string Remark3Parser::nextLine()
 std::string Remark3Parser::nextLine()