Function TEST_F

cpp/src/parquet/size_statistics_test.cc:243–305

};

TEST_F(SizeStatisticsRoundTripTest, EnableSizeStats) {
  auto schema = ::arrow::schema({
      ::arrow::field("a", ::arrow::list(::arrow::list(::arrow::int32()))),
      ::arrow::field("b", ::arrow::list(::arrow::list(::arrow::utf8()))),
  });
  // First two rows will be in one row group, and the other two rows in another row group.
  auto table = ::arrow::TableFromJSON(schema, {R"([
      [ [[1],[1,1],[1,1,1]], [["a"],["a","a"],["a","a","a"]] ],
      [ [[0,1,null]],        [["foo","bar",null]]            ],
      [ [],                  []                              ],
      [ [[],[null],null],    [[],[null],null]                ]
    ])"});

  for (auto size_stats_level :
       {SizeStatisticsLevel::None, SizeStatisticsLevel::ColumnChunk,
        SizeStatisticsLevel::PageAndColumnChunk}) {
    WriteFile(size_stats_level, table, /*max_row_group_length=*/2, /*page_size=*/1);
    ReadSizeStatistics();

    if (size_stats_level == SizeStatisticsLevel::None) {
      EXPECT_THAT(row_group_stats_,
                  ::testing::ElementsAre(kEmptyRowGroupStats, kEmptyRowGroupStats,
                                         kEmptyRowGroupStats, kEmptyRowGroupStats));
    } else {
      EXPECT_THAT(row_group_stats_, ::testing::ElementsAre(
                                        SizeStatistics{/*def_levels=*/{0, 0, 0, 0, 1, 8},
                                                       /*rep_levels=*/{2, 2, 5},
                                                       /*byte_array_bytes=*/std::nullopt},
                                        SizeStatistics{/*def_levels=*/{0, 0, 0, 0, 1, 8},
                                                       /*rep_levels=*/{2, 2, 5},
                                                       /*byte_array_bytes=*/12},
                                        SizeStatistics{/*def_levels=*/{0, 1, 1, 1, 1, 0},
                                                       /*rep_levels=*/{2, 2, 0},
                                                       /*byte_array_bytes=*/std::nullopt},
                                        SizeStatistics{/*def_levels=*/{0, 1, 1, 1, 1, 0},
                                                       /*rep_levels=*/{2, 2, 0},
                                                       /*byte_array_bytes=*/0}));
    }

    if (size_stats_level == SizeStatisticsLevel::PageAndColumnChunk) {
      EXPECT_THAT(
          page_stats_,
          ::testing::ElementsAre(
              PageSizeStatistics{/*def_levels=*/{0, 0, 0, 0, 0, 6, 0, 0, 0, 0, 1, 2},
                                 /*rep_levels=*/{1, 2, 3, 1, 0, 2},
                                 /*byte_array_bytes=*/{}},
              PageSizeStatistics{/*def_levels=*/{0, 0, 0, 0, 0, 6, 0, 0, 0, 0, 1, 2},
                                 /*rep_levels=*/{1, 2, 3, 1, 0, 2},
                                 /*byte_array_bytes=*/{6, 6}},
              PageSizeStatistics{/*def_levels=*/{0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0},
                                 /*rep_levels=*/{1, 0, 0, 1, 2, 0},
                                 /*byte_array_bytes=*/{}},
              PageSizeStatistics{/*def_levels=*/{0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0},
                                 /*rep_levels=*/{1, 0, 0, 1, 2, 0},
                                 /*byte_array_bytes=*/{0, 0}}));
    } else {
      EXPECT_THAT(page_stats_, ::testing::ElementsAre(kEmptyPageStats, kEmptyPageStats,
                                                      kEmptyPageStats, kEmptyPageStats));