Test case for min-max statistics on binary column built with parquet-mr

cpp/src/parquet/metadata.cc

@@ -48,7 +48,7 @@ const ApplicationVersion& ApplicationVersion::PARQUET_CPP_FIXED_STATS_VERSION()
 }
 
 const ApplicationVersion& ApplicationVersion::PARQUET_MR_FIXED_STATS_VERSION() {
-  static ApplicationVersion version("parquet-cpp", 1, 10, 0);
+  static ApplicationVersion version("parquet-mr", 1, 10, 0);
   return version;
 }
 
@@ -554,8 +554,8 @@ bool ApplicationVersion::HasCorrectStatistics(Type::type col_type,
                                               SortOrder::type sort_order) const {
   // parquet-cpp version 1.3.0 and parquet-mr 1.10.0 onwards stats are computed
   // correctly for all types
-  if ((application_ == "parquet-cpp" && VersionLt(PARQUET_CPP_FIXED_STATS_VERSION()))
-      || (application_ == "parquet-mr" && VersionLt(PARQUET_MR_FIXED_STATS_VERSION()))) {
+  if ((application_ == "parquet-cpp" && VersionLt(PARQUET_CPP_FIXED_STATS_VERSION())) ||
+      (application_ == "parquet-mr" && VersionLt(PARQUET_MR_FIXED_STATS_VERSION()))) {
     // Only SIGNED are valid unless max and min are the same
     // (in which case the sort order does not matter)
     bool max_equals_min = statistics.has_min && statistics.has_max

cpp/src/parquet/statistics-test.cc

@@ -772,5 +772,33 @@ TEST(TestStatisticsDoubleNaN, NaNValues) {
   ASSERT_EQ(min, -3.0);
   ASSERT_EQ(max, 4.0);
 }
+
+// Test statistics for binary column with UNSIGNED sort order
+TEST(TestStatisticsMinMax, Unsigned) {
+  std::string dir_string(test::get_data_dir());
+  std::stringstream ss;
+  ss << dir_string << "/binary.parquet";
+  auto path = ss.str();
+
+  // The file is generated by parquet-mr 1.10.0, the first version that
+  // supports correct statistics for binary data (see PARQUET-1025). It
+  // contains a single column of binary type. Data is just single byte values
+  // from 0x00 to 0x0B.
+  auto file_reader = ParquetFileReader::OpenFile(path);
+  auto rg_reader = file_reader->RowGroup(0);
+  auto metadata = rg_reader->metadata();
+  auto column_schema = metadata->schema()->Column(0);
+  ASSERT_EQ(SortOrder::UNSIGNED, column_schema->sort_order());
+
+  auto column_chunk = metadata->ColumnChunk(0);
+  ASSERT_TRUE(column_chunk->is_stats_set());
+
+  std::shared_ptr<RowGroupStatistics> stats = column_chunk->statistics();
+  ASSERT_TRUE(stats != NULL);
+  ASSERT_EQ(0, stats->null_count());
+  ASSERT_EQ(12, stats->num_values());
+  ASSERT_EQ(0x00, stats->EncodeMin()[0]);
+  ASSERT_EQ(0x0b, stats->EncodeMax()[0]);
+}
 }  // namespace test
 }  // namespace parquet