Merge branch 'main' into remove-python-37-and-38

.github/workflows/test-lang-js.yml

@@ -64,7 +64,7 @@ jobs:
 
   interop:
     name: Node ${{ matrix.node }} Interop
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
     strategy:
       matrix:
         node:

doc/content/en/docs/++version++/Specification/_index.md

@@ -787,6 +787,8 @@ A logical type is always serialized using its underlying Avro type so that value
 Language implementations must ignore unknown logical types when reading, and should use the underlying Avro type. If a logical type is invalid, for example a decimal with scale greater than its precision, then implementations should ignore the logical type and use the underlying Avro type.
 
 ### Decimal
+
+#### Fixed precision
 The `decimal` logical type represents an arbitrary-precision signed decimal number of the form _unscaled × 10<sup>-scale</sup>_.
 
 A `decimal` logical type annotates Avro _bytes_ or _fixed_ types. The byte array must contain the two's-complement representation of the unscaled integer value in big-endian byte order. The scale is fixed, and is specified using an attribute.
@@ -810,19 +812,19 @@ Scale must be zero or a positive integer less than or equal to the precision.
 
 For the purposes of schema resolution, two schemas that are `decimal` logical types _match_ if their scales and precisions match.
 
-**alternative**
+#### Scalable precision
 
 As it's not always possible to fix scale and precision in advance for a decimal field, `big-decimal` is another `decimal` logical type restrict to Avro _bytes_.
 
-_Currently only available in Java and Rust_.
+_Currently only available in C++, Java and Rust_.
 
 ```json
 {
   "type": "bytes",
   "logicalType": "big-decimal"
 }
 ```
-Here, as scale property is stored in value itself it needs more bytes than preceding `decimal` type, but it allows more flexibility.
+Here, bytes array contains two serialized properties. First part is an Avro byte arrays which is the two's-complement representation of the unscaled integer value in big-endian byte order. The second part is the scale property stored as an Avro integer. Scale must be zero or a positive integer less than or equal to the precision. Value itself needs more bytes than preceding `decimal` type, but it allows more flexibility.
 
 ### UUID
 

doc/package.json

@@ -1,7 +1,7 @@
 {
   "devDependencies": {
     "autoprefixer": "^10.4.0",
-    "postcss": "^8.4.47",
+    "postcss": "^8.4.49",
     "postcss-cli": "^11.0.0"
   }
 }

lang/c++/CMakeLists.txt

@@ -55,6 +55,9 @@ list(GET AVRO_VERSION 2 AVRO_VERSION_PATCH)
 project (Avro-cpp)
 set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${CMAKE_SOURCE_DIR})
 
+option(AVRO_BUILD_EXECUTABLES "Build executables" ON)
+option(AVRO_BUILD_TESTS "Build tests" ON)
+
 if (WIN32 AND NOT CYGWIN AND NOT MSYS)
     add_definitions (/EHa)
     add_definitions (
@@ -78,9 +81,12 @@ if (AVRO_ADD_PROTECTOR_FLAGS)
 endif ()
 endif ()
 
-
-find_package (Boost 1.38 REQUIRED
-    COMPONENTS filesystem iostreams program_options regex system)
+if (AVRO_BUILD_TESTS OR AVRO_BUILD_EXECUTABLES)
+    find_package (Boost 1.38 REQUIRED
+        COMPONENTS filesystem iostreams program_options system)
+else ()
+    find_package (Boost 1.38 REQUIRED COMPONENTS iostreams)
+endif ()
 
 include(FetchContent)
 FetchContent_Declare(
@@ -149,86 +155,90 @@ set_target_properties (avrocpp_s PROPERTIES
 target_link_libraries (avrocpp ${Boost_LIBRARIES} ${SNAPPY_LIBRARIES} fmt::fmt-header-only)
 target_include_directories(avrocpp PRIVATE ${SNAPPY_INCLUDE_DIR})
 
-add_executable (precompile test/precompile.cc)
-
-target_link_libraries (precompile avrocpp_s)
-
-macro (gen file ns)
-    add_custom_command (OUTPUT ${file}.hh
-        COMMAND avrogencpp
-            -p -
-            -i ${CMAKE_CURRENT_SOURCE_DIR}/jsonschemas/${file}
-            -o ${file}.hh -n ${ns}
-        DEPENDS avrogencpp ${CMAKE_CURRENT_SOURCE_DIR}/jsonschemas/${file})
-    add_custom_target (${file}_hh DEPENDS ${file}.hh)
-endmacro (gen)
-
-gen (empty_record empty)
-gen (bigrecord testgen)
-gen (bigrecord_r testgen_r)
-gen (bigrecord2 testgen2)
-gen (tweet testgen3)
-gen (union_array_union uau)
-gen (union_map_union umu)
-gen (union_conflict uc)
-gen (union_empty_record uer)
-gen (recursive rec)
-gen (reuse ru)
-gen (circulardep cd)
-gen (tree1 tr1)
-gen (tree2 tr2)
-gen (crossref cr)
-gen (primitivetypes pt)
-gen (cpp_reserved_words cppres)
-gen (cpp_reserved_words_union_typedef cppres_union)
-gen (big_union big_union)
-gen (union_redundant_types redundant_types)
-
-add_executable (avrogencpp impl/avrogencpp.cc)
-target_link_libraries (avrogencpp avrocpp_s)
-
-target_include_directories(avrocpp_s PUBLIC
+target_include_directories(avrocpp PUBLIC
   $<BUILD_INTERFACE:${CMAKE_CURRENT_SOURCE_DIR}/include>
   $<INSTALL_INTERFACE:include>
 )
-target_include_directories(avrocpp PUBLIC
+target_include_directories(avrocpp_s PUBLIC
   $<BUILD_INTERFACE:${CMAKE_CURRENT_SOURCE_DIR}/include>
   $<INSTALL_INTERFACE:include>
 )
 
-enable_testing()
-
-macro (unittest name)
-    add_executable (${name} test/${name}.cc)
-    target_link_libraries (${name} avrocpp ${Boost_LIBRARIES} ${SNAPPY_LIBRARIES})
-    add_test (NAME ${name} WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}
-        COMMAND ${CMAKE_CURRENT_BINARY_DIR}/${name})
-endmacro (unittest)
-
-unittest (buffertest)
-unittest (unittest)
-unittest (SchemaTests)
-unittest (LargeSchemaTests)
-unittest (CodecTests)
-unittest (StreamTests)
-unittest (SpecificTests)
-unittest (DataFileTests)
-unittest (JsonTests)
-unittest (AvrogencppTests)
-unittest (CompilerTests)
-unittest (AvrogencppTestReservedWords)
-unittest (CommonsSchemasTests)
-
-add_dependencies (AvrogencppTestReservedWords cpp_reserved_words_hh)
-add_dependencies (AvrogencppTestReservedWords cpp_reserved_words_hh
-    cpp_reserved_words_union_typedef_hh)
-
-add_dependencies (AvrogencppTests bigrecord_hh bigrecord_r_hh bigrecord2_hh
-    tweet_hh
-    union_array_union_hh union_map_union_hh union_conflict_hh
-    recursive_hh reuse_hh circulardep_hh tree1_hh tree2_hh crossref_hh
-    primitivetypes_hh empty_record_hh cpp_reserved_words_union_typedef_hh
-    union_empty_record_hh big_union_hh union_redundant_types_hh)
+if (AVRO_BUILD_EXECUTABLES)
+    add_executable (precompile test/precompile.cc)
+
+    target_link_libraries (precompile avrocpp_s)
+
+    macro (gen file ns)
+        add_custom_command (OUTPUT ${file}.hh
+            COMMAND avrogencpp
+                -p -
+                -i ${CMAKE_CURRENT_SOURCE_DIR}/jsonschemas/${file}
+                -o ${file}.hh -n ${ns}
+            DEPENDS avrogencpp ${CMAKE_CURRENT_SOURCE_DIR}/jsonschemas/${file})
+        add_custom_target (${file}_hh DEPENDS ${file}.hh)
+    endmacro (gen)
+
+    gen (empty_record empty)
+    gen (bigrecord testgen)
+    gen (bigrecord_r testgen_r)
+    gen (bigrecord2 testgen2)
+    gen (tweet testgen3)
+    gen (union_array_union uau)
+    gen (union_map_union umu)
+    gen (union_conflict uc)
+    gen (union_empty_record uer)
+    gen (recursive rec)
+    gen (reuse ru)
+    gen (circulardep cd)
+    gen (tree1 tr1)
+    gen (tree2 tr2)
+    gen (crossref cr)
+    gen (primitivetypes pt)
+    gen (cpp_reserved_words cppres)
+    gen (cpp_reserved_words_union_typedef cppres_union)
+    gen (big_union big_union)
+    gen (union_redundant_types redundant_types)
+
+    add_executable (avrogencpp impl/avrogencpp.cc)
+    target_link_libraries (avrogencpp avrocpp_s ${Boost_LIBRARIES})
+endif ()
+
+if (AVRO_BUILD_TESTS)
+    enable_testing()
+
+    macro (unittest name)
+        add_executable (${name} test/${name}.cc)
+        target_link_libraries (${name} avrocpp_s ${Boost_LIBRARIES} ${SNAPPY_LIBRARIES})
+        add_test (NAME ${name} WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}
+            COMMAND ${CMAKE_CURRENT_BINARY_DIR}/${name})
+    endmacro (unittest)
+
+    unittest (buffertest)
+    unittest (unittest)
+    unittest (SchemaTests)
+    unittest (LargeSchemaTests)
+    unittest (CodecTests)
+    unittest (StreamTests)
+    unittest (SpecificTests)
+    unittest (DataFileTests)
+    unittest (JsonTests)
+    unittest (AvrogencppTests)
+    unittest (CompilerTests)
+    unittest (AvrogencppTestReservedWords)
+    unittest (CommonsSchemasTests)
+
+    add_dependencies (AvrogencppTestReservedWords cpp_reserved_words_hh)
+    add_dependencies (AvrogencppTestReservedWords cpp_reserved_words_hh
+        cpp_reserved_words_union_typedef_hh)
+
+    add_dependencies (AvrogencppTests bigrecord_hh bigrecord_r_hh bigrecord2_hh
+        tweet_hh
+        union_array_union_hh union_map_union_hh union_conflict_hh
+        recursive_hh reuse_hh circulardep_hh tree1_hh tree2_hh crossref_hh
+        primitivetypes_hh empty_record_hh cpp_reserved_words_union_typedef_hh
+        union_empty_record_hh big_union_hh union_redundant_types_hh)
+endif ()
 
 include (InstallRequiredSystemLibraries)
 
@@ -241,7 +251,9 @@ install (TARGETS avrocpp avrocpp_s
     ARCHIVE DESTINATION lib
     RUNTIME DESTINATION lib)
 
-install (TARGETS avrogencpp RUNTIME DESTINATION bin)
+if (AVRO_BUILD_EXECUTABLES)
+    install (TARGETS avrogencpp RUNTIME DESTINATION bin)
+endif ()
 
 install (DIRECTORY include/avro DESTINATION include
     FILES_MATCHING PATTERN *.hh)

lang/c++/examples/imaginary.hh

@@ -22,7 +22,6 @@
 #include "avro/Decoder.hh"
 #include "avro/Encoder.hh"
 #include "avro/Specific.hh"
-#include "boost/any.hpp"
 
 namespace i {
 struct cpx {

lang/c++/impl/Compiler.cc

@@ -359,7 +359,11 @@ static LogicalType makeLogicalType(const Entity &e, const Object &m) {
     }
 
     LogicalType::Type t = LogicalType::NONE;
-    if (typeField == "date")
+    if (typeField == "big-decimal"
+        && !containsField(m, "precision")
+        && !containsField(m, "scale"))
+        t = LogicalType::BIG_DECIMAL;
+    else if (typeField == "date")
         t = LogicalType::DATE;
     else if (typeField == "time-millis")
         t = LogicalType::TIME_MILLIS;

lang/c++/impl/CustomAttributes.cc

@@ -23,8 +23,8 @@
 
 namespace avro {
 
-boost::optional<std::string> CustomAttributes::getAttribute(const std::string &name) const {
-    boost::optional<std::string> result;
+std::optional<std::string> CustomAttributes::getAttribute(const std::string &name) const {
+    std::optional<std::string> result;
     std::map<std::string, std::string>::const_iterator iter =
         attributes_.find(name);
     if (iter == attributes_.end()) {

lang/c++/impl/DataFile.cc

@@ -61,6 +61,7 @@ boost::iostreams::zlib_params get_zlib_params() {
     ret.noheader = true;
     return ret;
 }
+
 } // namespace
 
 DataFileWriterBase::DataFileWriterBase(const char *filename, const ValidSchema &schema, size_t syncInterval,
@@ -442,6 +443,11 @@ void DataFileReaderBase::readDataBlock() {
 }
 
 void DataFileReaderBase::close() {
+    stream_.reset();
+    eof_ = true;
+    objectCount_ = 0;
+    blockStart_ = 0;
+    blockEnd_ = 0;
 }
 
 static string toString(const vector<uint8_t> &v) {

lang/c++/impl/LogicalType.cc

@@ -51,6 +51,9 @@ void LogicalType::setScale(int32_t scale) {
 void LogicalType::printJson(std::ostream &os) const {
     switch (type_) {
         case LogicalType::NONE: break;
+        case LogicalType::BIG_DECIMAL:
+            os << R"("logicalType": "big-decimal")";
+            break;
         case LogicalType::DECIMAL:
             os << R"("logicalType": "decimal")";
             os << ", \"precision\": " << precision_;

lang/c++/impl/Node.cc

@@ -139,6 +139,13 @@ void Node::setLogicalType(LogicalType logicalType) {
     // Check that the logical type is applicable to the node type.
     switch (logicalType.type()) {
         case LogicalType::NONE: break;
+        case LogicalType::BIG_DECIMAL: {
+            if (type_ != AVRO_BYTES) {
+                throw Exception("BIG_DECIMAL logical type can annotate "
+                                "only BYTES type");
+            }
+            break;
+        }
         case LogicalType::DECIMAL: {
             if (type_ != AVRO_BYTES && type_ != AVRO_FIXED) {
                 throw Exception("DECIMAL logical type can annotate "

lang/c++/impl/Resolver.cc

@@ -445,7 +445,7 @@ class FixedParser : public Resolver {
     size_t offset_;
 };
 
-class ResolverFactory : private boost::noncopyable {
+class ResolverFactory {
 
     template<typename T>
     unique_ptr<Resolver>
@@ -512,6 +512,10 @@ class ResolverFactory : private boost::noncopyable {
     }
 
 public:
+    ResolverFactory() = default;
+    ResolverFactory(const ResolverFactory &) = delete;
+    ResolverFactory &operator=(const ResolverFactory &) = delete;
+
     unique_ptr<Resolver>
     construct(const NodePtr &writer, const NodePtr &reader, const Layout &offset) {