diff --git a/.github/workflows/unit-test-cpp.yml b/.github/workflows/unit-test-cpp.yml index 08c138018..d4172f8b4 100644 --- a/.github/workflows/unit-test-cpp.yml +++ b/.github/workflows/unit-test-cpp.yml @@ -104,12 +104,14 @@ jobs: core.setOutput('platform_suffix', ``) } - - name: Install clang-format + # Install dependencies + - name: Install dependencies shell: bash run: | if [[ "$RUNNER_OS" == "Linux" ]]; then sudo update-alternatives --install /usr/bin/clang-format clang-format /usr/bin/clang-format-17 100 sudo update-alternatives --set clang-format /usr/bin/clang-format-17 + sudo apt-get install -y uuid-dev elif [[ "$RUNNER_OS" == "Windows" ]]; then choco install llvm --version 17.0.6 --force else diff --git a/.github/workflows/unit-test-python.yml b/.github/workflows/unit-test-python.yml index c99935976..72be32363 100644 --- a/.github/workflows/unit-test-python.yml +++ b/.github/workflows/unit-test-python.yml @@ -75,6 +75,22 @@ jobs: core.setOutput('platform_suffix', ``) } + # Install dependencies + - name: Install dependencies + shell: bash + run: | + if [[ "$RUNNER_OS" == "Linux" ]]; then + sudo update-alternatives --install /usr/bin/clang-format clang-format /usr/bin/clang-format-17 100 + sudo update-alternatives --set clang-format /usr/bin/clang-format-17 + sudo apt-get install -y uuid-dev + elif [[ "$RUNNER_OS" == "Windows" ]]; then + choco install llvm --version 17.0.6 --force + else + brew install llvm@17 + ln -sf $(brew --prefix llvm@17)/bin/clang-format /opt/homebrew/bin/clang-format + fi + + # Run the actual maven build including all tests. - name: Build and test with Maven shell: bash diff --git a/cpp/CMakeLists.txt b/cpp/CMakeLists.txt index 60f5dc861..3128c9968 100755 --- a/cpp/CMakeLists.txt +++ b/cpp/CMakeLists.txt @@ -19,9 +19,17 @@ under the License. cmake_minimum_required(VERSION 3.11) project(TsFile_CPP) +if (DEFINED ToolChain) + include(${CMAKE_SOURCE_DIR}/cmake/ToolChain.cmake) + message(STATUS "Using ToolChain: ${CMAKE_TOOLCHAIN_FILE}") +else() + message(STATUS "Not using ToolChain") +endif () + cmake_policy(SET CMP0079 NEW) set(TsFile_CPP_VERSION 2.2.0.dev) set(CMAKE_CXX_FLAGS "$ENV{CXXFLAGS} -Wall") + if (CMAKE_CXX_COMPILER_ID MATCHES "GNU") set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wunused -Wuninitialized -D__STDC_FORMAT_MACROS") endif () diff --git a/cpp/README.md b/cpp/README.md index 9e13cdd11..dced6c8c1 100644 --- a/cpp/README.md +++ b/cpp/README.md @@ -42,23 +42,6 @@ The source code can be found in the `./src` directory. C/C++ examples are locate We use `clang-format` to ensure that our C++ code adheres to a consistent set of rules defined in `./clang-format`. This is similar to the Google style. -**Feature List**: - -- [ ] Add unit tests for the reader, writer, compression, etc. -- [ ] Add unit tests for the C wrapper. -- [ ] Support multiple data flushes. -- [ ] Support aligned timeseries. -- [ ] Support table description in tsfile. -- [ ] Retrieve all table schemas/names. -- [ ] Implement automatic flush. -- [ ] Support out-of-order data writing. -- [ ] Support TsFile V4. Note: TsFile CPP does not implement support for the table model, therefore there are differences in file output compared to the Java version. - -**Bug List**: - -- [ ] Flushing without writing after registering a timeseries will cause a core dump. -- [ ] Misalignment in memory may lead to a bus error. - We welcome any bug reports. You can open an issue with a title starting with [CPP] to describe the bug, like: https://github.com/apache/tsfile/issues/94 ## Build @@ -67,7 +50,7 @@ We welcome any bug reports. You can open an issue with a title starting with [CP ```bash sudo apt-get update -sudo apt-get install -y cmake make g++ clang-format +sudo apt-get install -y cmake make g++ clang-format libuuid-dev ``` To build tsfile, you can run: `bash build.sh`. If you have Maven tools, you can run: `mvn package -P with-cpp clean verify`. Then, you can find the shared object at `./build`. @@ -81,6 +64,21 @@ If you compile using MinGW on windows and encounter an error, you can try replac * GCC 12.2.0 + LLVM/Clang/LLD/LLDB 16.0.0 + MinGW-w64 10.0.0 (MSVCRT) - release 5 * GCC 11.2.0 + MinGW-w64 10.0.0 (MSVCRT) - release 1 +### configure the cross-compilation toolchain + +Modify the Toolchain File `cmake/ToolChain.cmake`, define the following variables: + +- `CMAKE_C_COMPILER`: Specify the path to the C compiler. +- `CMAKE_CXX_COMPILER`: Specify the path to the C++ compiler. +- `CMAKE_FIND_ROOT_PATH`: Set the root path for the cross-compilation environment (e.g., the directory of the cross-compilation toolchain). + +In the `cpp/` directory, run the following commands to create the build directory and start the compilation: +``` +mkdir build && cd build +cmake .. -DToolChian=ON +make +``` + ## Use TsFile You can find examples on how to read and write data in `demo_read.cpp` and `demo_write.cpp` located under `./examples/cpp_examples`. There are also examples under `./examples/c_examples`on how to use a C-style API to read and write data in a C environment. You can run `bash build.sh` under `./examples` to generate an executable output under `./examples/build`. \ No newline at end of file diff --git a/cpp/pom.xml b/cpp/pom.xml index d2bd2ab6e..f2644a96e 100644 --- a/cpp/pom.xml +++ b/cpp/pom.xml @@ -171,30 +171,6 @@ Linux - - - - org.codehaus.mojo - exec-maven-plugin - - - install-uuid-dev - validate - - exec - - - - - bash - - -c - sudo apt-get update && sudo apt-get install -y uuid-dev - - - - - diff --git a/cpp/test/cwrapper/c_release_test.cc b/cpp/test/cwrapper/c_release_test.cc index 692f10435..85583d526 100644 --- a/cpp/test/cwrapper/c_release_test.cc +++ b/cpp/test/cwrapper/c_release_test.cc @@ -90,7 +90,7 @@ TEST_F(CReleaseTest, TsFileWriterNew) { table_schema.table_name = strdup("test_table"); table_schema.column_num = 2; table_schema.column_schemas = - static_cast(malloc(sizeof(ColumnSchema) * 2)); + static_cast(malloc(sizeof(ColumnSchema) * 2)); table_schema.column_schemas[0] = (ColumnSchema){.column_name = strdup("col1"), .data_type = TS_DATATYPE_STRING, @@ -121,7 +121,7 @@ TEST_F(CReleaseTest, TsFileWriterWriteDataAbnormalColumn) { abnormal_schema.table_name = strdup("!@#$%^*()_+-="); abnormal_schema.column_num = 3; abnormal_schema.column_schemas = - static_cast(malloc(sizeof(ColumnSchema) * 4)); + static_cast(malloc(sizeof(ColumnSchema) * 4)); abnormal_schema.column_schemas[0] = (ColumnSchema){.column_name = strdup("!@#$%^*()_+-="), .data_type = TS_DATATYPE_STRING, @@ -163,12 +163,12 @@ TEST_F(CReleaseTest, TsFileWriterWriteDataAbnormalColumn) { writer = tsfile_writer_new(file, &abnormal_schema, &error_code); ASSERT_EQ(RET_OK, error_code); - char **column_list = static_cast(malloc(sizeof(char *) * 3)); + char** column_list = static_cast(malloc(sizeof(char*) * 3)); column_list[0] = strdup("!@#$%^*()_+-="); column_list[1] = strdup("TAG2"); column_list[2] = strdup("!@#$%^*()_+-=1"); - TSDataType *type_list = - static_cast(malloc(sizeof(TSDataType) * 3)); + TSDataType* type_list = + static_cast(malloc(sizeof(TSDataType) * 3)); type_list[0] = TS_DATATYPE_STRING; type_list[1] = TS_DATATYPE_STRING; type_list[2] = TS_DATATYPE_DOUBLE; @@ -196,7 +196,7 @@ TEST_F(CReleaseTest, TsFileWriterWriteDataAbnormalColumn) { tsfile_result_set_get_value_by_name_int64_t(result_set, "time"); ASSERT_EQ(timestamp * 100.0, tsfile_result_set_get_value_by_name_double( result_set, "!@#$%^*()_+-=1")); - char *value_str = + char* value_str = tsfile_result_set_get_value_by_index_string(result_set, 2); ASSERT_EQ("device1", std::string(value_str)); free(value_str); @@ -226,7 +226,7 @@ TEST_F(CReleaseTest, TsFileWriterMultiDataType) { all_type_schema.table_name = strdup("All_Datatype"); all_type_schema.column_num = 6; all_type_schema.column_schemas = - static_cast(malloc(sizeof(ColumnSchema) * 6)); + static_cast(malloc(sizeof(ColumnSchema) * 6)); all_type_schema.column_schemas[0] = (ColumnSchema){.column_name = strdup("TAG"), .data_type = TS_DATATYPE_STRING, @@ -257,15 +257,15 @@ TEST_F(CReleaseTest, TsFileWriterMultiDataType) { ASSERT_EQ(RET_OK, error_code); free_table_schema(all_type_schema); - char **column_list = static_cast(malloc(sizeof(char *) * 6)); + char** column_list = static_cast(malloc(sizeof(char*) * 6)); column_list[0] = strdup("TAG"); column_list[1] = strdup("INT32"); column_list[2] = strdup("INT64"); column_list[3] = strdup("FLOAT"); column_list[4] = strdup("DOUBLE"); column_list[5] = strdup("BOOLEAN"); - TSDataType *type_list = - static_cast(malloc(sizeof(TSDataType) * 6)); + TSDataType* type_list = + static_cast(malloc(sizeof(TSDataType) * 6)); type_list[0] = TS_DATATYPE_STRING; type_list[1] = TS_DATATYPE_INT32; type_list[2] = TS_DATATYPE_INT64; @@ -302,7 +302,7 @@ TEST_F(CReleaseTest, TsFileWriterMultiDataType) { Timestamp timestamp = tsfile_result_set_get_value_by_name_int64_t(result_set, "time"); int64_t value = timestamp + 10; - char *str_value = + char* str_value = tsfile_result_set_get_value_by_name_string(result_set, "TAG"); ASSERT_EQ("device1", std::string(str_value)); free(str_value); @@ -336,4 +336,69 @@ TEST_F(CReleaseTest, TsFileWriterMultiDataType) { remove("TsFileWriterMultiDataType.tsfile"); } +TEST_F(CReleaseTest, TsFileWriterConfTest) { + ERRNO err_no = RET_OK; + WriteFile file = write_file_new("plain_file.tsfile", &err_no); + TableSchema plain_schema; + plain_schema.table_name = strdup("plain_table"); + plain_schema.column_num = 2; + + plain_schema.column_schemas = + (ColumnSchema*)malloc(sizeof(ColumnSchema) * 2); + plain_schema.column_schemas[0].column_name = strdup("id"); + plain_schema.column_schemas[0].data_type = TS_DATATYPE_STRING; + plain_schema.column_schemas[0].column_category = ColumnCategory::TAG; + plain_schema.column_schemas[1].column_name = strdup("value"); + plain_schema.column_schemas[1].data_type = TS_DATATYPE_INT64; + plain_schema.column_schemas[1].column_category = ColumnCategory::FIELD; + + set_global_compression(TS_COMPRESSION_UNCOMPRESSED); + set_datatype_encoding(TS_DATATYPE_INT64, TS_ENCODING_PLAIN); + TsFileWriter writer = tsfile_writer_new(file, &plain_schema, &err_no); + free_table_schema(plain_schema); + + char** column_list = static_cast(malloc(sizeof(char*) * 2)); + column_list[0] = strdup("id"); + column_list[1] = strdup("value"); + + TSDataType* type_list = + static_cast(malloc(sizeof(TSDataType) * 2)); + type_list[0] = TS_DATATYPE_STRING; + type_list[1] = TS_DATATYPE_INT64; + + Tablet tablet = tablet_new(column_list, type_list, 2, 10); + for (int i = 0; i < 10; i++) { + tablet_add_timestamp(tablet, i, static_cast(i)); + tablet_add_value_by_name_string(tablet, i, "id", "device1"); + tablet_add_value_by_name_int32_t(tablet, i, "value", i); + } + + ASSERT_EQ(RET_OK, tsfile_writer_write(writer, tablet)); + ASSERT_EQ(RET_OK, tsfile_writer_close(writer)); + free_write_file(&file); + TsFileReader reader = tsfile_reader_new("plain_file.tsfile", &err_no); + ASSERT_EQ(RET_OK, err_no); + TableSchema schema = tsfile_reader_get_table_schema(reader, "plain_table"); + ASSERT_EQ(schema.column_num, 2); + uint32_t size = 0; + DeviceSchema* device_schema = + tsfile_reader_get_all_timeseries_schemas(reader, &size); + ASSERT_EQ(1, size); + ASSERT_EQ(1, device_schema->timeseries_num); + ASSERT_EQ(device_schema->timeseries_schema[0].encoding, TS_ENCODING_PLAIN); + ASSERT_EQ(device_schema->timeseries_schema[0].compression, + TS_COMPRESSION_UNCOMPRESSED); + tsfile_reader_close(reader); + free_table_schema(schema); + free_device_schema(*device_schema); + free(device_schema); + free(column_list[0]); + free(column_list[1]); + free(column_list); + free(type_list); + free_tablet(&tablet); + + remove("plain_file.tsfile"); +} + } // namespace CReleaseTest \ No newline at end of file