GH-46371: [C++][Parquet] Parquet Variant decoding tools

cpp/src/parquet/CMakeLists.txt

@@ -189,7 +189,8 @@ set(PARQUET_SRCS
     statistics.cc
     stream_reader.cc
     stream_writer.cc
-    types.cc)
+    types.cc
+    variant.cc)
 
 if(ARROW_HAVE_RUNTIME_AVX2)
   # AVX2 is used as a proxy for BMI2.
@@ -382,7 +383,8 @@ add_parquet_test(internals-test
                  public_api_test.cc
                  size_statistics_test.cc
                  statistics_test.cc
-                 types_test.cc)
+                 types_test.cc
+                 variant_test.cc)
 
 set_source_files_properties(public_api_test.cc PROPERTIES SKIP_UNITY_BUILD_INCLUSION ON)
 

cpp/src/parquet/test_util.cc

@@ -53,6 +53,15 @@ std::string get_bad_data_dir() {
   return ss.str();
 }
 
+std::string get_variant_dir() {
+  // PARQUET_TEST_DATA should point to ARROW_HOME/cpp/submodules/parquet-testing/data
+  // so need to reach one folder up to access the "variant" folder.
+  std::string data_dir(get_data_dir());
+  std::stringstream ss;
+  ss << data_dir << "/../variant";
+  return ss.str();
+}
+
 std::string get_data_file(const std::string& filename, bool is_good) {
   std::stringstream ss;
 

cpp/src/parquet/test_util.h

@@ -71,6 +71,7 @@ class ParquetTestException : public parquet::ParquetException {
 
 const char* get_data_dir();
 std::string get_bad_data_dir();
+std::string get_variant_dir();
 
 std::string get_data_file(const std::string& filename, bool is_good = true);