messages updated

Updated pcre2s

.clang-format

@@ -0,0 +1,22 @@
+BasedOnStyle: LLVM
+UseTab: AlignWithSpaces
+IndentWidth: 4
+TabWidth: 4
+BreakBeforeBraces: Allman
+ColumnLimit: 0
+NamespaceIndentation: Inner
+FixNamespaceComments: true
+AccessModifierOffset: -2
+AllowShortCaseLabelsOnASingleLine: true
+IndentCaseLabels: true
+BreakConstructorInitializers: BeforeComma
+BraceWrapping:
+  BeforeLambdaBody: false
+AlignAfterOpenBracket: DontAlign
+Cpp11BracedListStyle: false
+IncludeBlocks: Regroup
+LambdaBodyIndentation: Signature
+AllowShortLambdasOnASingleLine: Inline
+EmptyLineBeforeAccessModifier: LogicalBlock
+IndentPPDirectives: AfterHash
+PPIndentWidth: 1

.github/workflows/cmake-multi-platform.yml

@@ -33,13 +33,18 @@ jobs:
 
     - name: Install dependencies Ubuntu
       if: matrix.os == 'ubuntu-latest'
-      run: sudo apt-get update && sudo apt-get install mrc
+      run: sudo apt-get update && sudo apt-get install mrc catch2
 
     - name: Install dependencies Window
       if: matrix.os == 'windows-latest'
       run: ./tools/depends.cmd
       shell: cmd
 
+    - name: Install Catch2 macOS
+      if: matrix.os == 'macos-latest'
+      run: >
+        brew install catch2
+
     - name: Configure CMake
       run: >
         cmake -B ${{ steps.strings.outputs.build-output-dir }}

CMakeLists.txt

@@ -32,7 +32,7 @@ endif()
 # set the project name
 project(
 	libcifpp
-	VERSION 9.0.0
+	VERSION 9.0.3
 	LANGUAGES CXX C)
 
 list(PREPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake")
@@ -174,18 +174,27 @@ find_file(FMT NAME format)
 if(FMT EQUAL "FMT-NOTFOUND")
 	if(NOT (fmt_FOUND OR TARGET fmt))
 		find_package(fmt REQUIRED)
-		message(FATAL_ERROR "fmt not found, compiler too old, you're out of luck")
+		message(FATAL_ERROR "cifpp: <format> not found and neither was libfmt, compiler too old, you're out of luck")
 	endif()
 endif()
 
 find_package(Threads)
-include(FindPCRE2)
 find_package(ZLIB QUIET)
 
 if(NOT ZLIB_FOUND)
 	message(FATAL_ERROR "cifpp: The zlib development files were not found you this system, please install them and try again (hint: on debian/ubuntu use apt-get install zlib1g-dev)")
 endif()
 
+include(FindPkgConfig)
+
+if(PKG_CONFIG_FOUND)
+	pkg_check_modules(PCRE2 IMPORTED_TARGET libpcre2-8)
+endif()
+
+if(NOT PCRE2_FOUND)
+	add_subdirectory(pcre2-simple)
+endif()
+
 # Using Eigen3 is a bit of a thing. We don't want to build it completely since
 # we only need a couple of header files. Nothing special. But often, eigen3 is
 # already installed and then we prefer that.
@@ -207,8 +216,6 @@ else()
 	set(EIGEN_INCLUDE_DIR ${SOURCE_DIR})
 endif()
 
-message(STATUS "cifpp: Eigen include dir is ${EIGEN_INCLUDE_DIR}")
-
 # Create a revision file, containing the current git version info
 include(VersionString)
 write_version_header(${CMAKE_CURRENT_SOURCE_DIR}/src/ LIB_NAME "LibCIFPP")
@@ -327,7 +334,14 @@ target_include_directories(
 	PRIVATE "${EIGEN_INCLUDE_DIR}")
 
 target_link_libraries(cifpp
-	PUBLIC Threads::Threads ZLIB::ZLIB $<$<TARGET_EXISTS:std::atomic>:std::atomic> $<BUILD_INTERFACE:pcre2-8>)
+	PUBLIC Threads::Threads ZLIB::ZLIB $<$<TARGET_EXISTS:std::atomic>:std::atomic>)
+
+if(PCRE2_FOUND)
+	target_include_directories(cifpp PRIVATE ${PCRE2_INCLUDE_DIRS})
+	target_link_libraries(cifpp PRIVATE ${PCRE2_LINK_LIBRARIES})
+else()
+	target_link_libraries(cifpp PRIVATE $<BUILD_INTERFACE:pcre2s>)
+endif()
 
 if(fmt_FOUND)
 	target_link_libraries(cifpp PUBLIC fmt)

changelog

@@ -1,3 +1,14 @@
+Version 9.0.3
+- Reconstruction fixed when some entity ids are missing
+
+Version 9.0.2
+- Fix code that reconstructs sequences, could throw a map::at
+- Many optimisations in validation and reconstruction code.
+
+Version 9.0.1
+- Use pcre2 from pkg-config if available, if not
+  build a version from the original code.
+
 Version 9.0.0
 - Rename fields of cif::mm::polymer to match the naming
   in mmcif_pdbx.dic. Also, related, fix building mm::structure

cmake/FindPCRE2.cmake

@@ -1,27 +1,12 @@
-set(PCRE2_USE_STATIC_LIBS ON)
+# The problem is, find_package(PCRE2) does not work
+# and using pkg-config results in linking to a shared library
+# causing all kinds of trouble later on
 
-# The cmake config files for pcre2 are broken
+find_path(PCRE2_INCLUDEDIR NAMES pcre2.h HINTS "C:/Program Files (x86)/PCRE2/include" REQUIRED)
+find_library(PCRE2_LIBRARY NAMES pcre2-8-static libpcre2-8.a HINTS "C:/Program Files (x86)/PCRE2/lib" REQUIRED)
 
-include(FindPkgConfig)
-
-if(PKG_CONFIG_FOUND)
-	pkg_check_modules(PCRE2 IMPORTED_TARGET libpcre2-8)
-
-	if(PCRE2_FOUND)
-		message(STATUS "Using pcre2 found using pkg-config")
-
-		add_library(pcre2-8 ALIAS PkgConfig::PCRE2)
-	endif()
-endif()
-
-if(NOT PCRE2_FOUND)
-	find_path(PCRE2_INCLUDEDIR NAMES pcre2.h HINTS "C:/Program Files (x86)/PCRE2/include" REQUIRED)
-	find_library(PCRE2_LIBRARY NAMES pcre2-8-static HINTS "C:/Program Files (x86)/PCRE2/lib" REQUIRED)
-
-	add_library(pcre2-8 IMPORTED STATIC)
-	target_include_directories(pcre2-8 INTERFACE ${PCRE2_INCLUDEDIR})
-	target_compile_definitions(pcre2-8 INTERFACE PCRE2_STATIC)
-	# target_link_libraries(pcre2-8 INTERFACE ${PCRE2_LIBRARY})
-	set_target_properties(pcre2-8 PROPERTIES IMPORTED_LOCATION ${PCRE2_LIBRARY})
-	set_target_properties(pcre2-8 PROPERTIES IMPORTED_IMPLIB ${PCRE2_LIBRARY})
-endif()
+add_library(pcre2-8 IMPORTED STATIC)
+target_include_directories(pcre2-8 INTERFACE ${PCRE2_INCLUDEDIR})
+target_compile_definitions(pcre2-8 INTERFACE PCRE2_STATIC)
+set_target_properties(pcre2-8 PROPERTIES IMPORTED_LOCATION ${PCRE2_LIBRARY})
+set_target_properties(pcre2-8 PROPERTIES IMPORTED_IMPLIB ${PCRE2_LIBRARY})

include/cif++/model.hpp

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

include/cif++/utilities.hpp

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

pcre2-simple/CMakeLists.txt

@@ -0,0 +1,316 @@
+# SPDX-License-Identifier: BSD-2-Clause
+# 
+# Copyright (c) 2025 Maarten L. Hekkelman
+# 
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are met:
+# 
+# 1. Redistributions of source code must retain the above copyright notice, this
+#    list of conditions and the following disclaimer
+# 2. Redistributions in binary form must reproduce the above copyright notice,
+#    this list of conditions and the following disclaimer in the documentation
+#    and/or other materials provided with the distribution.
+# 
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
+# ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+# (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+# LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+# ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+# A simplified wrapper CMakeLists.txt file for PCRE2
+#
+# This will generate an OBJECT library so it can be linked into another library
+
+cmake_minimum_required(VERSION 3.25)
+
+include(FetchContent)
+
+project(pcre2s VERSION 1.0.0 LANGUAGES C CXX)
+
+# The original code:
+
+file(DOWNLOAD https://github.com/PCRE2Project/pcre2/releases/download/pcre2-10.46/pcre2-10.46.tar.gz
+    ${CMAKE_CURRENT_BINARY_DIR}/pcre2-code.tgz
+    EXPECTED_HASH SHA256=8d28d7f2c3b970c3a4bf3776bcbb5adfc923183ce74bc8df1ebaad8c1985bd07)
+file(ARCHIVE_EXTRACT INPUT ${CMAKE_CURRENT_BINARY_DIR}/pcre2-code.tgz
+    DESTINATION ${CMAKE_CURRENT_BINARY_DIR})
+set(PCRE2_SOURCE_DIR ${CMAKE_CURRENT_BINARY_DIR}/pcre2-10.46)
+set(PCRE2_MAJOR 10)
+set(PCRE2_MINOR 46)
+set(PCRE2_VERSION "${PCRE2_MAJOR}.${PCRE2_MINOR}")
+set(PCRE2_DATE "2024-06-09")
+
+# Some needed configuration options
+
+# option(PCRE2_BUILD_PCRE2_8 "Build 8 bit PCRE2 library" ON)
+# option(PCRE2_BUILD_PCRE2_16 "Build 16 bit PCRE2 library" OFF)
+# option(PCRE2_BUILD_PCRE2_32 "Build 32 bit PCRE2 library" OFF)
+
+option(PCRE2_STATIC_PIC "Build the static library with the option position independent code enabled." OFF)
+
+set(PCRE2_NEWLINE "LF" CACHE STRING "What to recognize as a newline (one of CR, LF, CRLF, ANY, ANYCRLF, NUL)." FORCE)
+set_property(CACHE PCRE2_NEWLINE PROPERTY STRINGS "CR" "LF" "CRLF" "ANY" "ANYCRLF" "NUL")
+
+set(PCRE2_LINK_SIZE "2" CACHE STRING "Internal link size (2, 3 or 4 allowed). See LINK_SIZE in config.h.in for details.")
+set_property(CACHE PCRE2_LINK_SIZE PROPERTY STRINGS "2" "3" "4")
+
+set(PCRE2_PARENS_NEST_LIMIT "250" CACHE STRING "Default nested parentheses limit. See PARENS_NEST_LIMIT in config.h.in for details.")
+set(PCRE2_HEAP_LIMIT "20000000" CACHE STRING "Default limit on heap memory (kibibytes). See HEAP_LIMIT in config.h.in for details.")
+set(PCRE2_MAX_VARLOOKBEHIND "255" CACHE STRING "Default limit on variable lookbehinds.")
+set(PCRE2_MATCH_LIMIT "10000000" CACHE STRING "Default limit on internal looping. See MATCH_LIMIT in config.h.in for details.")
+set(PCRE2_MATCH_LIMIT_DEPTH "MATCH_LIMIT" CACHE STRING "Default limit on internal depth of search. See MATCH_LIMIT_DEPTH in config.h.in for details.")
+set(PCRE2GREP_BUFSIZE "20480" CACHE STRING "Buffer starting size parameter for pcre2grep. See PCRE2GREP_BUFSIZE in config.h.in for details.")
+set(PCRE2GREP_MAX_BUFSIZE "1048576" CACHE STRING "Buffer maximum size parameter for pcre2grep. See PCRE2GREP_MAX_BUFSIZE in config.h.in for details.")
+set(PCRE2_SUPPORT_JIT OFF CACHE BOOL "Enable support for Just-in-time compiling.")
+
+if(${CMAKE_SYSTEM_NAME} MATCHES Linux|NetBSD)
+    set(PCRE2_SUPPORT_JIT_SEALLOC OFF CACHE BOOL "Enable SELinux compatible execmem allocator in JIT (experimental).")
+else()
+    set(PCRE2_SUPPORT_JIT_SEALLOC IGNORE)
+endif()
+
+set(PCRE2GREP_SUPPORT_JIT ON CACHE BOOL "Enable use of Just-in-time compiling in pcre2grep.")
+set(PCRE2GREP_SUPPORT_CALLOUT ON CACHE BOOL "Enable callout string support in pcre2grep.")
+set(PCRE2GREP_SUPPORT_CALLOUT_FORK ON CACHE BOOL "Enable callout string fork support in pcre2grep.")
+set(PCRE2_SUPPORT_UNICODE ON CACHE BOOL "Enable support for Unicode and UTF-8/UTF-16/UTF-32 encoding.")
+set(PCRE2_SUPPORT_BSR_ANYCRLF OFF CACHE BOOL "ON=Backslash-R matches only LF CR and CRLF, OFF=Backslash-R matches all Unicode Linebreaks")
+set(PCRE2_NEVER_BACKSLASH_C OFF CACHE BOOL "If ON, backslash-C (upper case C) is locked out.")
+set(PCRE2_SUPPORT_VALGRIND OFF CACHE BOOL "Enable Valgrind support.")
+
+if(MINGW)
+    option(NON_STANDARD_LIB_PREFIX "ON=Shared libraries built in mingw will be named pcre2.dll, etc., instead of libpcre2.dll, etc." OFF)
+    option(NON_STANDARD_LIB_SUFFIX "ON=Shared libraries built in mingw will be named libpcre2-0.dll, etc., instead of libpcre2.dll, etc." OFF)
+endif()
+
+# 
+
+set(NEWLINE_DEFAULT "")
+
+if(PCRE2_NEWLINE STREQUAL "CR")
+    set(NEWLINE_DEFAULT "1")
+elseif(PCRE2_NEWLINE STREQUAL "LF")
+    set(NEWLINE_DEFAULT "2")
+elseif(PCRE2_NEWLINE STREQUAL "CRLF")
+    set(NEWLINE_DEFAULT "3")
+elseif(PCRE2_NEWLINE STREQUAL "ANY")
+    set(NEWLINE_DEFAULT "4")
+elseif(PCRE2_NEWLINE STREQUAL "ANYCRLF")
+    set(NEWLINE_DEFAULT "5")
+elseif(PCRE2_NEWLINE STREQUAL "NUL")
+    set(NEWLINE_DEFAULT "6")
+else()
+    message(FATAL_ERROR "The PCRE2_NEWLINE variable must be set to one of the following values: \"LF\", \"CR\", \"CRLF\", \"ANY\", \"ANYCRLF\".")
+endif()
+
+# Some tests
+
+include(CheckCSourceCompiles)
+include(CheckFunctionExists)
+include(CheckSymbolExists)
+include(CheckIncludeFile)
+
+check_include_file(assert.h HAVE_ASSERT_H)
+check_include_file(dirent.h HAVE_DIRENT_H)
+check_include_file(sys/stat.h HAVE_SYS_STAT_H)
+check_include_file(sys/types.h HAVE_SYS_TYPES_H)
+check_include_file(unistd.h HAVE_UNISTD_H)
+check_include_file(windows.h HAVE_WINDOWS_H)
+
+check_symbol_exists(bcopy "strings.h" HAVE_BCOPY)
+check_symbol_exists(memfd_create "sys/mman.h" HAVE_MEMFD_CREATE)
+check_symbol_exists(memmove "string.h" HAVE_MEMMOVE)
+check_symbol_exists(secure_getenv "stdlib.h" HAVE_SECURE_GETENV)
+check_symbol_exists(strerror "string.h" HAVE_STRERROR)
+
+check_c_source_compiles(
+  "int main(void) { char buf[128] __attribute__((uninitialized)); (void)buf; return 0; }"
+  HAVE_ATTRIBUTE_UNINITIALIZED
+)
+
+check_c_source_compiles(
+  [=[
+  extern __attribute__ ((visibility ("default"))) int f(void);
+  int main(void) { return f(); }
+  int f(void) { return 42; }
+  ]=]
+  HAVE_VISIBILITY
+)
+
+if(HAVE_VISIBILITY)
+  set(PCRE2_EXPORT [=[__attribute__ ((visibility ("default")))]=])
+else()
+  set(PCRE2_EXPORT)
+endif()
+
+check_c_source_compiles("int main(void) { __assume(1); return 0; }" HAVE_BUILTIN_ASSUME)
+
+check_c_source_compiles(
+  [=[
+  #include <stddef.h>
+  int main(void) { int a,b; size_t m; __builtin_mul_overflow(a,b,&m); return 0; }
+  ]=]
+  HAVE_BUILTIN_MUL_OVERFLOW
+)
+
+check_c_source_compiles(
+  "int main(int c, char *v[]) { if (c) __builtin_unreachable(); return (int)(*v[0]); }"
+  HAVE_BUILTIN_UNREACHABLE
+)
+
+# # Check whether Intel CET is enabled, and if so, adjust compiler flags. This
+# # code was written by PH, trying to imitate the logic from the autotools
+# # configuration.
+
+# check_c_source_compiles(
+#   [=[
+#   #ifndef __CET__
+#   #error CET is not enabled
+#   #endif
+#   int main() { return 0; }
+#   ]=]
+#   INTEL_CET_ENABLED
+# )
+
+# if(INTEL_CET_ENABLED)
+#   set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mshstk")
+# endif()
+
+# Set up some dependencies first
+
+configure_file(
+    ${PCRE2_SOURCE_DIR}/src/pcre2_chartables.c.dist
+    ${CMAKE_CURRENT_BINARY_DIR}/pcre2_chartables.c
+    COPYONLY
+)
+
+configure_file(
+    ${PCRE2_SOURCE_DIR}/config-cmake.h.in
+    ${CMAKE_CURRENT_BINARY_DIR}/interface/config.h
+    @ONLY
+)
+
+configure_file(
+    ${PCRE2_SOURCE_DIR}/src/pcre2.h.in
+    ${CMAKE_CURRENT_BINARY_DIR}/interface/pcre2.h
+    @ONLY
+)
+
+# Define our library
+
+list(APPEND PCRE2_HEADERS
+    ${CMAKE_CURRENT_BINARY_DIR}/interface/pcre2.h)
+
+list(APPEND PCRE2_SOURCES
+    ${PCRE2_SOURCE_DIR}/src/pcre2_auto_possess.c
+    ${CMAKE_CURRENT_BINARY_DIR}/pcre2_chartables.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_chkdint.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_compile.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_compile_class.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_config.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_context.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_convert.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_dfa_match.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_error.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_extuni.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_find_bracket.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_jit_compile.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_maketables.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_match.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_match_data.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_newline.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_ord2utf.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_pattern_info.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_script_run.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_serialize.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_string_utils.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_study.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_substitute.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_substring.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_tables.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_ucd.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_valid_utf.c
+    ${PCRE2_SOURCE_DIR}/src/pcre2_xclass.c
+)
+
+add_library(pcre2s OBJECT)
+
+target_sources(pcre2s
+    PRIVATE ${PCRE2_SOURCES}
+    PUBLIC
+    FILE_SET pcre2_headers TYPE HEADERS
+    BASE_DIRS ${PCRE2_SOURCE_DIR}/include ${CMAKE_CURRENT_BINARY_DIR}/interface
+    FILES ${PCRE2_HEADERS}
+)
+
+target_compile_definitions(pcre2s PUBLIC PCRE2_CODE_UNIT_WIDTH=8 HAVE_CONFIG_H)
+if(NOT BUILD_SHARED_LIBS)
+    target_compile_definitions(pcre2s PUBLIC PCRE2_STATIC)
+endif()
+
+target_include_directories(pcre2s PRIVATE ${CMAKE_CURRENT_BINARY_DIR}/interface ${PCRE2_SOURCE_DIR}/src)
+
+if(PCRE2_STATIC_PIC)
+    set_target_properties(pcre2s PROPERTIES POSITION_INDEPENDENT_CODE 1)
+endif()
+
+# # Installation and config files
+
+# include(CMakePackageConfigHelpers)
+# include(GenerateExportHeader)
+
+# # Install rules
+# install(TARGETS pcre2s
+#     EXPORT pcre2s
+#     FILE_SET pcre2_headers DESTINATION ${CMAKE_INSTALL_INCLUDEDIR})
+
+# if(MSVC AND BUILD_SHARED_LIBS)
+#     install(
+#         FILES $<TARGET_PDB_FILE:pcre2s>
+#         DESTINATION ${CMAKE_INSTALL_LIBDIR}
+#         OPTIONAL)
+# endif()
+
+# install(EXPORT pcre2s
+#     NAMESPACE pcre2s::
+#     FILE "pcre2s-targets.cmake"
+#     DESTINATION lib/cmake/pcre2s)
+
+# configure_package_config_file(
+#     ${CMAKE_CURRENT_SOURCE_DIR}/pcre2s-config.cmake.in ${CMAKE_CURRENT_BINARY_DIR}/pcre2s/pcre2s-config.cmake
+#     INSTALL_DESTINATION lib/cmake/pcre2s)
+
+# install(
+#     FILES "${CMAKE_CURRENT_BINARY_DIR}/pcre2s/pcre2s-config.cmake"
+#     "${CMAKE_CURRENT_BINARY_DIR}/pcre2s/pcre2s-config-version.cmake"
+#     DESTINATION lib/cmake/pcre2s)
+
+# set_target_properties(
+#     pcre2s
+#     PROPERTIES VERSION ${PCRE2_VERSION}
+#     SOVERSION ${PCRE2_VERSION}
+#     INTERFACE_pcre2s_MAJOR_VERSION ${PCRE2_MAJOR})
+
+# set_property(
+#     TARGET pcre2s
+#     APPEND
+#     PROPERTY COMPATIBLE_INTERFACE_STRING pcre2s_MAJOR_VERSION)
+
+# write_basic_package_version_file(
+#     "${CMAKE_CURRENT_BINARY_DIR}/pcre2s/pcre2s-config-version.cmake"
+#     VERSION "${PCRE2_VERSION}"
+#     COMPATIBILITY AnyNewerVersion)
+
+# # Testing
+
+# if(PROJECT_IS_TOP_LEVEL)
+#     include(CTest)
+
+#     if(BUILD_TESTING)
+#         add_subdirectory(test)
+#     endif()
+# endif()

src/compound.cpp

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

src/dictionary_parser.cpp

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

src/model.cpp

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

src/pdb/reconstruct.cpp

@@ -25,8 +25,12 @@
  */
 
 #include "cif++.hpp"
+#include "cif++/compound.hpp"
 #include "cif++/row.hpp"
 
+#include <stdexcept>
+#include <string>
+
 // --------------------------------------------------------------------
 
 namespace cif::pdb
@@ -186,6 +190,8 @@ void checkEntities(datablock &db)
 
 void createEntityIDs(datablock &db)
 {
+	using namespace literals;
+
 	// Suppose the file does not have entity ID's. We have to make up some
 
 	// walk the atoms. For each auth_asym_id we have a new struct_asym.
@@ -197,28 +203,44 @@ void createEntityIDs(datablock &db)
 	// that should cover it
 
 	auto &atom_site = db["atom_site"];
+	auto &entity = db["entity"];
 	auto &cf = compound_factory::instance();
 
-	std::vector<std::vector<residue_key_type>> entities;
+	std::vector<std::vector<row_handle>> entities;
 
 	std::string lastAsymID;
 	int lastSeqID = -1;
-	std::vector<residue_key_type> waters;
+	std::vector<row_handle> waters;
 
-	for (residue_key_type k : atom_site.rows<std::optional<std::string>,
-			 std::optional<int>,
-			 std::optional<std::string>,
-			 std::optional<std::string>,
-			 std::optional<int>,
-			 std::optional<std::string>>(
-			 "auth_asym_id", "auth_seq_id", "auth_comp_id",
-			 "label_asym_id", "label_seq_id", "label_comp_id"))
+	int nextEntityID;
+	try
 	{
+		if (entity.empty())
+			nextEntityID = 1;
+		else
+			nextEntityID = entity.find_max<int>("id") + 1;
+	}
+	catch (...)
+	{
+		nextEntityID = 1;
+	}
+
+	for (auto rh : atom_site.find("label_entity_id"_key == cif::null))
+	{
+		residue_key_type k = rh.get<std::optional<std::string>,
+			std::optional<int>,
+			std::optional<std::string>,
+			std::optional<std::string>,
+			std::optional<int>,
+			std::optional<std::string>>(
+			"auth_asym_id", "auth_seq_id", "auth_comp_id",
+			"label_asym_id", "label_seq_id", "label_comp_id");
+
 		std::string comp_id = get_comp_id(k);
 
 		if (cf.is_water(comp_id))
 		{
-			waters.emplace_back(k);
+			waters.emplace_back(rh);
 			continue;
 		}
 
@@ -227,19 +249,20 @@ void createEntityIDs(datablock &db)
 
 		bool is_monomer = cf.is_monomer(comp_id);
 
-		if (lastAsymID == asym_id and lastSeqID == seq_id and not is_monomer)
-			continue;
+		// if (lastAsymID == asym_id and lastSeqID == seq_id and not is_monomer)
+		// 	continue;
 
 		if (asym_id != lastAsymID or (not is_monomer and lastSeqID != seq_id))
 			entities.push_back({});
 
-		entities.back().emplace_back(k);
+		entities.back().emplace_back(rh);
 
 		lastAsymID = asym_id;
 		lastSeqID = seq_id;
 	}
 
 	std::map<std::size_t, std::string> entity_ids;
+	std::map<std::string, std::string> newEntitiesForCompound;
 
 	atom_site.add_item("label_entity_id");
 
@@ -248,7 +271,37 @@ void createEntityIDs(datablock &db)
 		if (entity_ids.contains(i))
 			continue;
 
-		auto entity_id = std::to_string(i + 1);
+		residue_key_type k = entities[i].front().get<std::optional<std::string>, std::optional<int>, std::optional<std::string>, std::optional<std::string>, std::optional<int>, std::optional<std::string>>(
+			"auth_asym_id", "auth_seq_id", "auth_comp_id",
+			"label_asym_id", "label_seq_id", "label_comp_id");
+
+		std::string comp_id = get_comp_id(k);
+
+		std::string entity_id;
+		if (auto v = db["pdbx_entity_nonpoly"].find_first("comp_id"_key == comp_id); v)
+			entity_id = v.get<std::string>("entity_id");
+		else if (auto i2 = newEntitiesForCompound.find(comp_id); i2 != newEntitiesForCompound.end())
+			entity_id = i2->second;
+		else
+		{
+			entity_id = std::to_string(nextEntityID++);
+
+			if (cf.is_monomer(comp_id))
+				entity.emplace({ //
+					{ "id", entity_id },
+					{ "type", "polymer" } });
+			else if (cf.is_water(comp_id))
+				entity.emplace({ //
+					{ "id", entity_id },
+					{ "type", "water" } });
+			else
+				entity.emplace({ //
+					{ "id", entity_id },
+					{ "type", "non-polymer" } });
+
+			newEntitiesForCompound[comp_id] = entity_id;
+		}
+
 		entity_ids[i] = entity_id;
 
 		for (std::size_t j = i + 1; j < entities.size(); ++j)
@@ -260,20 +313,17 @@ void createEntityIDs(datablock &db)
 
 	for (std::size_t ix = 0; auto &e : entities)
 	{
-		auto k = e.front();
 		const auto &entity_id = entity_ids[ix++];
 
-		std::string comp_id = get_comp_id(k);
-
-		for (auto &k2 : e)
-			atom_site.update_value(get_condition(k2), "label_entity_id", entity_id);
+		for (auto rh : e)
+			rh["label_entity_id"] = entity_id;
 	}
 
 	if (not waters.empty())
 	{
 		std::string waterEntityID = std::to_string(entities.size() + 1);
-		for (auto &k : waters)
-			atom_site.update_value(get_condition(k), "label_entity_id", waterEntityID);
+		for (auto rh : waters)
+			rh["label_entity_id"] = waterEntityID;
 	}
 }
 
@@ -440,9 +490,38 @@ void checkAtomRecords(datablock &db)
 	if (atom_site.contains(key("label_seq_id") < 0))
 		fixNegativeSeqID(atom_site);
 
-	std::set<int> polymer_entities;
-	for (int id : db["entity"].find<int>("type"_key == "polymer", "id"))
-		polymer_entities.insert(id);
+	std::set<std::string> polymer_entities;
+	if (db["entity"].empty())
+	{
+		// No entity, so we have to guess the types based on the content of atom_site
+
+		std::string last_entity_id;
+		std::optional<int> last_label_seq_id, last_auth_seq_id;
+
+		std::set<std::string> entityIDs;
+		for (auto &[entity_id, label_comp_id, label_seq_id, auth_comp_id, auth_seq_id] :
+			atom_site.rows<std::string, std::string, std::optional<int>, std::string, std::optional<int>>(
+				"label_entity_id", "label_comp_id", "label_seq_id", "auth_comp_id", "auth_seq_id"))
+		{
+			if (cf.is_water(label_comp_id) or cf.is_water(auth_comp_id))
+				continue;
+
+			if (polymer_entities.contains(entity_id))
+				continue;
+
+			if (last_entity_id == entity_id and (last_label_seq_id != label_seq_id or last_auth_seq_id != auth_seq_id))
+				polymer_entities.emplace(entity_id);
+
+			last_entity_id = entity_id;
+			last_label_seq_id = label_seq_id;
+			last_auth_seq_id = auth_seq_id;
+		}
+	}
+	else
+	{
+		for (std::string id : db["entity"].find<std::string>("type"_key == "polymer", "id"))
+			polymer_entities.insert(id);
+	}
 
 	std::set<std::string> missingCompounds;
 
@@ -479,7 +558,7 @@ void checkAtomRecords(datablock &db)
 		if (missingCompounds.contains(comp_id))
 			continue;
 
-		bool is_polymer = polymer_entities.contains(row["label_entity_id"].as<int>());
+		bool is_polymer = polymer_entities.contains(row["label_entity_id"].as<std::string>());
 		auto compound = cf.create(comp_id);
 
 		if (not compound)
@@ -533,14 +612,20 @@ void checkAtomRecords(datablock &db)
 		if (is_polymer and row["label_seq_id"].empty() and cf.is_monomer(comp_id))
 			row["label_seq_id"] = std::to_string(seq_id);
 
-		if (row["label_atom_id"].empty())
-			row["label_atom_id"] = row["auth_atom_id"].text();
 		if (row["label_asym_id"].empty())
 			row["label_asym_id"] = row["auth_asym_id"].text();
+		else if (row["auth_asym_id"].empty())
+			row["auth_asym_id"] = row["label_asym_id"].text();
+
 		if (row["label_comp_id"].empty())
 			row["label_comp_id"] = row["auth_comp_id"].text();
+		else if (row["auth_comp_id"].empty())
+			row["auth_comp_id"] = row["label_comp_id"].text();
+
 		if (row["label_atom_id"].empty())
 			row["label_atom_id"] = row["auth_atom_id"].text();
+		else if (row["auth_atom_id"].empty())
+			row["auth_atom_id"] = row["label_atom_id"].text();
 
 		// Rewrite the coordinates and other items that look better in a fixed format
 		// Be careful not to nuke invalidly formatted data here
@@ -605,19 +690,24 @@ void checkAtomAnisotropRecords(datablock &db)
 
 	std::vector<row_handle> to_be_deleted;
 
+	std::map<int, row_handle> atoms;
+	for (auto rh : atom_site)
+		atoms[rh.get<int>("id")] = rh;
+
 	bool warnReplaceTypeSymbol = true;
 	for (auto row : atom_site_anisotrop)
 	{
-		auto parents = atom_site_anisotrop.get_parents(row, atom_site);
-		if (parents.size() != 1)
+		auto ai = atoms.find(row.get<int>("id"));
+
+		if (ai == atoms.end())
 		{
 			to_be_deleted.emplace_back(row);
 			continue;
 		}
 
-		// this happens sometimes (Phenix):
+		auto parent = ai->second;
 
-		auto parent = parents.front();
+		// this happens sometimes (Phenix):
 
 		if (row["type_symbol"].empty())
 			row["type_symbol"] = parent["type_symbol"].text();
@@ -638,8 +728,6 @@ void checkAtomAnisotropRecords(datablock &db)
 			row["pdbx_label_atom_id"] = parent["label_atom_id"].text();
 		if (row["pdbx_label_comp_id"].empty() and not parent["label_comp_id"].empty())
 			row["pdbx_label_comp_id"] = parent["label_comp_id"].text();
-		// if (row["pdbx_PDB_model_num"].empty() and not parent["pdbx_PDB_model_num"].empty())
-		// 	row["pdbx_PDB_model_num"] = parent["pdbx_PDB_model_num"].text();
 	}
 
 	if (not to_be_deleted.empty())
@@ -652,23 +740,53 @@ void checkAtomAnisotropRecords(datablock &db)
 	}
 }
 
-void createStructAsym(datablock &db)
+void checkStructAsym(datablock &db)
 {
 	auto &atom_site = db["atom_site"];
 	auto &struct_asym = db["struct_asym"];
 
-	for (const auto &[label_asym_id, entity_id] : atom_site.rows<std::string, std::string>("label_asym_id", "label_entity_id"))
+	if (struct_asym.empty())
 	{
-		if (label_asym_id.empty())
-			throw std::runtime_error("File contains atom_site records without a label_asym_id");
-		if (struct_asym.count(key("id") == label_asym_id) == 0)
+		for (const auto &[label_asym_id, entity_id] : atom_site.rows<std::string, std::string>("label_asym_id", "label_entity_id"))
 		{
-			struct_asym.emplace({
-				// clang-format off
-				{ "id", label_asym_id },
-				{ "entity_id", entity_id }
-				//clang-format on
-			});
+			if (label_asym_id.empty())
+				throw std::runtime_error("File contains atom_site records without a label_asym_id");
+			if (struct_asym.count(key("id") == label_asym_id) == 0)
+			{
+				struct_asym.emplace({
+					// clang-format off
+					{ "id", label_asym_id },
+					{ "entity_id", entity_id }
+					//clang-format on
+				});
+			}
+		}
+	}
+	else
+	{
+		for (const auto &[label_asym_id, entity_id] :
+			atom_site.rows<std::string, std::string>("label_asym_id", "label_entity_id"))
+		{
+			if (label_asym_id.empty())
+				throw std::runtime_error("File contains atom_site records without a label_asym_id");
+			
+			auto sa = struct_asym.find_first(key("id") == label_asym_id);
+			if (sa)
+			{
+				if (sa["entity_id"].empty())
+					sa.assign("entity_id", entity_id, false, true);
+				else if (sa.get<std::string>("entity_id") != entity_id)
+					throw std::runtime_error("Inconsistent entity ID's in struct_asym");
+			}
+			else
+			{
+				struct_asym.emplace({
+					// clang-format off
+					{ "id", label_asym_id },
+					{ "entity_id", entity_id }
+					//clang-format on
+				});
+			}
 		}
 	}
 }
@@ -739,8 +857,11 @@ void createEntity(datablock &db)
 			auto c = cf.create(first_comp_id);
 
 			type = "non-polymer";
-			desc = c->name();
-			weight = c->formula_weight();
+			if (c)
+			{
+				desc = c->name();
+				weight = c->formula_weight();
+			}
 		}
 		else
 		{
@@ -811,28 +932,28 @@ void createEntityPoly(datablock &db)
 			if (type != "other")
 			{
 				std::string c_type;
-				if (cf.is_base(comp_id))
+				if (auto i = compound_factory::kBaseMap.find(comp_id); i != compound_factory::kBaseMap.end())
 				{
 					c_type = "polydeoxyribonucleotide";
-					letter_can = compound_factory::kBaseMap.at(comp_id);
+
+					letter_can = i->second;
+
 					if (comp_id.length() == 1)
 						letter = letter_can;
 					else
-						letter = '(' + letter_can + ')';
+						letter = '(' + comp_id + ')';
 				}
-				else if (cf.is_peptide(comp_id))
+				else if (auto i2 = compound_factory::kAAMap.find(comp_id); i2 != compound_factory::kAAMap.end())
 				{
 					c_type = "polypeptide(L)";
-					letter = letter_can = compound_factory::kAAMap.at(comp_id);
+
+					letter = letter_can = i2->second;
 				}
 				else if (iequals(c->type(), "D-PEPTIDE LINKING"))
 				{
 					c_type = "polypeptide(D)";
 
 					letter_can = c->one_letter_code();
-					if (letter_can == 0)
-						letter_can = 'X';
-
 					letter = '(' + comp_id + ')';
 
 					non_std_linkage = true;
@@ -843,9 +964,6 @@ void createEntityPoly(datablock &db)
 					c_type = "polypeptide(L)";
 
 					letter_can = c->one_letter_code();
-					if (letter_can == 0)
-						letter_can = 'X';
-
 					letter = '(' + comp_id + ')';
 
 					non_std_monomer = true;
@@ -855,9 +973,6 @@ void createEntityPoly(datablock &db)
 					// c_type = "other";
 
 					letter_can = c->one_letter_code();
-					if (letter_can == 0)
-						letter_can = 'X';
-
 					letter = '(' + comp_id + ')';
 
 					non_std_monomer = true;
@@ -870,7 +985,7 @@ void createEntityPoly(datablock &db)
 			}
 
 			seq[auth_asym_id] += letter;
-			seq_can[auth_asym_id] += letter_can;
+			seq_can[auth_asym_id] += letter_can ? letter_can : 'X';
 
 			if (find(pdb_strand_ids.begin(), pdb_strand_ids.end(), auth_asym_id) == pdb_strand_ids.end())
 				pdb_strand_ids.emplace_back(auth_asym_id);
@@ -1390,7 +1505,7 @@ bool reconstruct_pdbx(file &file, const validator &validator)
 	checkChemCompRecords(db);
 
 	// If the data is really horrible, it might not contain entities
-	if (not db["atom_site"].find_first(key("label_entity_id") != null))
+	if (db["atom_site"].find_first(key("label_entity_id") == null))
 		createEntityIDs(db);
 
 	// Now see if atom records make sense at all
@@ -1534,9 +1649,8 @@ bool reconstruct_pdbx(file &file, const validator &validator)
 		checkAtomAnisotropRecords(db);
 
 	// Now create any missing categories
-	// Next make sure we have struct_asym records
-	if (auto cat = db.get("struct_asym"); cat == nullptr or cat->empty())
-		createStructAsym(db);
+	// Next make sure we have good struct_asym records
+	checkStructAsym(db);
 
 	if (auto cat = db.get("entity"); cat == nullptr or cat->empty())
 		createEntity(db);
@@ -1645,9 +1759,8 @@ void fixup_pdbx(file &file, const validator &validator)
 	db.set_validator(&validator);
 
 	// Now create any missing categories
-	// Next make sure we have struct_asym records
-	if (auto cat = db.get("struct_asym"); cat == nullptr or cat->empty())
-		createStructAsym(db);
+	// Next make sure we have good struct_asym records
+	checkStructAsym(db);
 
 	if (auto cat = db.get("entity"); cat == nullptr or cat->empty())
 		createEntity(db);

src/pdb/validate-pdbx.cpp

@@ -69,7 +69,7 @@ bool is_valid_pdbx_file(const file &file, const validator &v)
 {
 	std::error_code ec;
 	bool result = is_valid_pdbx_file(file, v, ec);
-	return result and not (bool)ec;
+	return result and not(bool) ec;
 }
 
 bool is_valid_pdbx_file(const file &file, std::error_code &ec)
@@ -82,7 +82,7 @@ bool is_valid_pdbx_file(const file &file, std::error_code &ec)
 		result = is_valid_pdbx_file(file, validator_factory::instance().get(*ac), ec);
 	else
 		result = is_valid_pdbx_file(file, validator_factory::instance().get("mmcif_pdbx.dic"), ec);
-	
+
 	return result;
 }
 
@@ -167,7 +167,7 @@ bool is_valid_pdbx_file(const file &file, const validator &validator, std::error
 
 			const auto entity_poly_type = entity_poly.find1<std::string>("entity_id"_key == entity_id, "type");
 
-			std::map<int,std::set<std::string>> mon_per_seq_id;
+			std::map<int, std::set<std::string>> mon_per_seq_id;
 
 			for (const auto &[num, mon_id, hetero] : entity_poly_seq.find<int, std::string, bool>("entity_id"_key == entity_id, "num", "mon_id", "hetero"))
 			{
@@ -202,28 +202,37 @@ bool is_valid_pdbx_file(const file &file, const validator &validator, std::error
 					throw std::runtime_error("Mismatch between the hetero flag in the poly seq schemes and the number residues per seq_id");
 			}
 
-			for (const auto &[seq_id, mon_ids] : mon_per_seq_id)
-			{
-				for (auto asym_id : struct_asym.find<std::string>("entity_id"_key == entity_id, "id"))
-				{
-					condition cond;
-					
-					for (auto mon_id : mon_ids)
-						cond = std::move(cond) or "label_comp_id"_key == mon_id;
+			// This code proved to take too much time ...
 
-					cond = "label_entity_id"_key == entity_id and
-						"label_asym_id"_key == asym_id and
-						"label_seq_id"_key == seq_id and not std::move(cond);
-					
-					if (atom_site.contains(std::move(cond)))
-						throw std::runtime_error("An atom_site record exists that has no parent in the poly seq scheme categories");
-				}
+			// for (const auto &[seq_id, mon_ids] : mon_per_seq_id)
+			// {
+			// 	for (auto asym_id : struct_asym.find<std::string>("entity_id"_key == entity_id, "id"))
+			// 	{
+			// 		condition cond;
+
+			// 		for (auto mon_id : mon_ids)
+			// 			cond = std::move(cond) or "label_comp_id"_key == mon_id;
+
+			// 		cond = "label_entity_id"_key == entity_id and
+			// 			"label_asym_id"_key == asym_id and
+			// 			"label_seq_id"_key == seq_id and not std::move(cond);
+
+			// 		if (atom_site.contains(std::move(cond)))
+			// 			throw std::runtime_error("An atom_site record exists that has no parent in the poly seq scheme categories");
+			// 	}
+			// }
+
+			// ... so we're using this instead, should be almost the same...
+
+			for (const auto &[comp_id, seq_id] :
+				atom_site.find<std::string, int>("label_entity_id"_key == entity_id, "label_comp_id", "label_seq_id"))
+			{
+				if (not mon_per_seq_id[seq_id].contains(comp_id))
+					throw std::runtime_error("An atom_site record exists that has no parent in the poly seq scheme categories");
 			}
 
 			auto &&[seq, seq_can] = entity_poly.find1<std::optional<std::string>, std::optional<std::string>>("entity_id"_key == entity_id,
 				"pdbx_seq_one_letter_code", "pdbx_seq_one_letter_code_can");
-			
-			std::string::const_iterator si, sci, se, sce;
 
 			auto seq_match = [&](bool can, std::string::const_iterator si, std::string::const_iterator se)
 			{
@@ -260,8 +269,8 @@ bool is_valid_pdbx_file(const file &file, const validator &validator, std::error
 							else
 								letter = '(' + comp_id + ')';
 						}
-						
-						if (iequals(std::string{si, si + letter.length()}, letter))
+
+						if (iequals(std::string{ si, si + letter.length() }, letter))
 						{
 							match = true;
 							si += letter.length();
@@ -285,7 +294,9 @@ bool is_valid_pdbx_file(const file &file, const validator &validator, std::error
 			}
 			else
 			{
-				seq->erase(std::remove_if(seq->begin(), seq->end(), [](char ch) { return std::isspace(ch); }), seq->end());
+				seq->erase(std::remove_if(seq->begin(), seq->end(), [](char ch)
+							   { return std::isspace(ch); }),
+					seq->end());
 
 				if (not seq_match(false, seq->begin(), seq->end()))
 					throw std::runtime_error("Sequences do not match for entity " + entity_id);
@@ -298,7 +309,9 @@ bool is_valid_pdbx_file(const file &file, const validator &validator, std::error
 			}
 			else
 			{
-				seq_can->erase(std::remove_if(seq_can->begin(), seq_can->end(), [](char ch) { return std::isspace(ch); }), seq_can->end());
+				seq_can->erase(std::remove_if(seq_can->begin(), seq_can->end(), [](char ch)
+								   { return std::isspace(ch); }),
+					seq_can->end());
 
 				if (not seq_match(true, seq_can->begin(), seq_can->end()))
 					throw std::runtime_error("Canonical sequences do not match for entity " + entity_id);
@@ -315,11 +328,10 @@ bool is_valid_pdbx_file(const file &file, const validator &validator, std::error
 		ec = make_error_code(validation_error::not_valid_pdbx);
 	}
 
-	if (not result and (bool)ec)
+	if (not result and (bool) ec)
 		ec = make_error_code(validation_error::not_valid_pdbx);
 
 	return result;
 }
 
 } // namespace cif::pdb
-  

src/utilities.cpp

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

src/validate.cpp

@@ -62,14 +62,6 @@ validation_exception::validation_exception(std::error_code ec, std::string_view
 
 // --------------------------------------------------------------------
 
-// struct regex_impl : public regex
-// {
-// 	regex_impl(std::string_view rx)
-// 		: regex(rx.begin(), rx.end(), regex::extended | regex::optimize)
-// 	{
-// 	}
-// };
-
 struct regex_impl
 {
 	regex_impl(std::string_view rx);
@@ -81,7 +73,6 @@ struct regex_impl
 	bool match(std::string_view v) const;
 
   private:
-
 	pcre2_code *m_rx = nullptr;
 	pcre2_match_data *m_data = nullptr;
 };
@@ -613,7 +604,7 @@ validator validator_factory::construct_validator(std::string_view name, std::opt
 		not v.matches_audit_conform(category{ "audit_conform", //
 			{ { "dict_name", name }, { "dict_version", version } } }))
 	{
-		std::clog << "Invalid dictionary?\n";
+		std::clog << "Loaded dictionary does not match name=" << name << " and version=" << version.value_or("''") << "\n";
 	}
 
 	return v;

test/CMakeLists.txt

@@ -1,7 +1,29 @@
-# We're using the older version 2 of Catch2
+# SPDX-License-Identifier: BSD-2-Clause
+# 
+# Copyright (c) 2025  NKI/AVL, Netherlands Cancer Institute
+# 
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are met:
+# 
+# 1. Redistributions of source code must retain the above copyright notice, this
+#    list of conditions and the following disclaimer
+# 2. Redistributions in binary form must reproduce the above copyright notice,
+#    this list of conditions and the following disclaimer in the documentation
+#    and/or other materials provided with the distribution.
+# 
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
+# ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+# (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+# LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+# ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
-if(NOT(Catch2_FOUND OR TARGET Catch2))
-	find_package(Catch2 QUIET)
+if(NOT (Catch2_FOUND OR TARGET Catch2))
+	find_package(Catch2 3 QUIET)
 
 	if(NOT Catch2_FOUND)
 		include(FetchContent)
@@ -9,11 +31,11 @@ if(NOT(Catch2_FOUND OR TARGET Catch2))
 		FetchContent_Declare(
 			Catch2
 			GIT_REPOSITORY https://github.com/catchorg/Catch2.git
-			GIT_TAG v2.13.9)
+            GIT_TAG v3.8.0)
 
 		FetchContent_MakeAvailable(Catch2)
 
-		set(Catch2_VERSION "2.13.9")
+		target_compile_features(Catch2 PRIVATE cxx_std_20)
 	endif()
 endif()
 
@@ -27,20 +49,14 @@ list(
 	rename-compound
 	sugar
 	spinner
-	# reconstruction
+	reconstruction
 	validate-pdbx
-	)
+)
 
 add_library(test-main OBJECT "${CMAKE_CURRENT_SOURCE_DIR}/test-main.cpp")
 
 target_link_libraries(test-main cifpp::cifpp Catch2::Catch2)
 
-if("${Catch2_VERSION}" VERSION_LESS 3.0.0)
-	target_compile_definitions(test-main PUBLIC CATCH22=1)
-else()
-	target_compile_definitions(test-main PUBLIC CATCH22=0)
-endif()
-
 foreach(CIFPP_TEST IN LISTS CIFPP_tests)
 	set(CIFPP_TEST "${CIFPP_TEST}-test")
 	set(CIFPP_TEST_SOURCE "${CMAKE_CURRENT_SOURCE_DIR}/${CIFPP_TEST}.cpp")
@@ -48,12 +64,6 @@ foreach(CIFPP_TEST IN LISTS CIFPP_tests)
 	add_executable(
 		${CIFPP_TEST} ${CIFPP_TEST_SOURCE} $<TARGET_OBJECTS:test-main>)
 
-	if(${Catch2_VERSION} VERSION_GREATER_EQUAL 3.0.0)
-		target_compile_definitions(${CIFPP_TEST} PUBLIC CATCH22=0)
-	else()
-		target_compile_definitions(${CIFPP_TEST} PUBLIC CATCH22=1)
-	endif()
-
 	target_link_libraries(${CIFPP_TEST} PRIVATE cifpp::cifpp Catch2::Catch2)
 	target_include_directories(${CIFPP_TEST} PRIVATE "${EIGEN_INCLUDE_DIR}")
 
@@ -62,15 +72,6 @@ foreach(CIFPP_TEST IN LISTS CIFPP_tests)
 		target_compile_options(${CIFPP_TEST} PRIVATE /EHsc)
 	endif()
 
-	add_custom_target(
-		"run-${CIFPP_TEST}"
-		DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/Run${CIFPP_TEST}.touch ${CIFPP_TEST})
-
-	add_custom_command(
-		OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/Run${CIFPP_TEST}.touch
-		COMMAND $<TARGET_FILE:${CIFPP_TEST}> --data-dir
-		${CMAKE_CURRENT_SOURCE_DIR})
-
-	add_test(NAME ${CIFPP_TEST} COMMAND $<TARGET_FILE:${CIFPP_TEST}> --data-dir
-		${CMAKE_CURRENT_SOURCE_DIR})
-endforeach()
+	add_test(NAME ${CIFPP_TEST}
+		COMMAND $<TARGET_FILE:${CIFPP_TEST}> --data-dir ${CMAKE_CURRENT_SOURCE_DIR})
+endforeach()

test/reconstruction-test.cpp

@@ -28,6 +28,7 @@
 
 #include <cif++.hpp>
 
+#include <filesystem>
 #include <iostream>
 #include <fstream>
 

test/test-main.cpp

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

test/test-main.hpp

@@ -1,17 +1,17 @@
 /*-
  * SPDX-License-Identifier: BSD-2-Clause
- * 
+ *
  * Copyright (c) 2024 NKI/AVL, Netherlands Cancer Institute
- * 
+ *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions are met:
- * 
+ *
  * 1. Redistributions of source code must retain the above copyright notice, this
  *    list of conditions and the following disclaimer
  * 2. Redistributions in binary form must reproduce the above copyright notice,
  *    this list of conditions and the following disclaimer in the documentation
  *    and/or other materials provided with the distribution.
- * 
+ *
  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
  * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
@@ -26,11 +26,7 @@
 
 #pragma once
 
-#if CATCH22
-#include <catch2/catch.hpp>
-#else
 #include <catch2/catch_all.hpp>
-#endif
 
 #include <filesystem>
 

tools/depends.cmd

@@ -1,6 +1,5 @@
 @ECHO OFF
 SET ZLIB_VERSION=1.3
-SET PCRE2_VERSION=10.45
 
 IF NOT EXIST build_ci\libs (
   MKDIR build_ci\libs
@@ -22,19 +21,4 @@ cmake --build build --target ALL_BUILD --config Release || EXIT /b 1
 cmake --build build --target RUN_TESTS --config Release || EXIT /b 1
 cmake --build build --target INSTALL --config Release || EXIT /b 1
 
-@REM Install PCRE2
-IF NOT EXIST pcre2-%PCRE2_VERSION%.zip (
-  ECHO Downloading https://github.com/PCRE2Project/pcre2/releases/download/pcre2-%PCRE2_VERSION%/pcre2-%PCRE2_VERSION%.zip
-  curl -L -o pcre2-%PCRE2_VERSION%.zip https://github.com/PCRE2Project/pcre2/releases/download/pcre2-%PCRE2_VERSION%/pcre2-%PCRE2_VERSION%.zip || EXIT /b 1
-)
-IF NOT EXIST pcre2-%PCRE2_VERSION% (
-  ECHO Unpacking pcre2-%PCRE2_VERSION%.zip
-  C:\windows\system32\tar.exe -x -f pcre2-%PCRE2_VERSION%.zip || EXIT /b 1
-)
-CD pcre2-%PCRE2_VERSION%
-cmake -B build -DPCRE2_BUILD_PCRE2GREP=OFF || EXIT /b 1
-cmake --build build --target ALL_BUILD --config Release || EXIT /b 1
-@REM cmake --build build --target RUN_TESTS --config Release || EXIT /b 1
-cmake --build build --target INSTALL --config Release || EXIT /b 1
-
 @EXIT /b 0

tools/update-libcifpp-data.in

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