fix(ut,orc): fix orc read ut under debian and adjust some options

src/paimon/format/orc/complex_predicate_test.cpp

@@ -61,14 +61,6 @@ class ComplexPredicateTest : public ::testing::Test {
     void SetUp() override {
         pool_ = GetDefaultPool();
         batch_size_ = 10;
-
-        arrow::FieldVector fields = {
-            arrow::field("f1", arrow::int32()),
-            arrow::field("f2", arrow::int32()),
-            arrow::field("f3", arrow::date32()),
-            arrow::field("f4", arrow::timestamp(arrow::TimeUnit::NANO)),
-            arrow::field("f5", arrow::decimal128(23, 5)),
-        };
     }
     void TearDown() override {}
 
@@ -131,16 +123,33 @@ TEST_F(ComplexPredicateTest, TestSimple) {
         arrow::field("f5", arrow::decimal128(23, 5)),
     };
     auto read_schema = arrow::schema(fields);
-    auto expected_array = std::dynamic_pointer_cast<arrow::StructArray>(
-        arrow::ipc::internal::json::ArrayFromJSON(arrow::struct_({fields}), R"([
+    std::shared_ptr<arrow::StructArray> expected_array;
+    if (::paimon::test::OsReleaseDetector::IsDebian()) {
+        // refer: https://github.com/eggert/tz/blob/main/asia#L653
+        // When using the Asia/Shanghai timezone under Debian, timestamps prior to 1901 have an
+        // additional offset of 5 minutes and 43 seconds
+        expected_array = std::dynamic_pointer_cast<arrow::StructArray>(
+            arrow::ipc::internal::json::ArrayFromJSON(arrow::struct_({fields}), R"([
+            [10, 1, 1234,  "2033-05-18 03:33:20.0",         "123456789987654321.45678"],
+            [10, 1, 19909, "2033-05-18 03:33:20.000001001", "12.30000"],
+            [10, 1, 0,     "2008-12-28 00:00:00.000123456", null],
+            [10, 1, 100,   "2008-12-28 00:00:00.00012345",  "-123.45000"],
+            [10, 1, null,  "1899-01-01 01:05:03.001001001", "0.00000"],
+            [10, 1, 20006, "2024-10-10 10:10:10.100100100", "1728551410100.10010"]
+        ])")
+                .ValueOrDie());
+    } else {
+        expected_array = std::dynamic_pointer_cast<arrow::StructArray>(
+            arrow::ipc::internal::json::ArrayFromJSON(arrow::struct_({fields}), R"([
         [10, 1, 1234,  "2033-05-18 03:33:20.0",         "123456789987654321.45678"],
         [10, 1, 19909, "2033-05-18 03:33:20.000001001", "12.30000"],
         [10, 1, 0,     "2008-12-28 00:00:00.000123456", null],
         [10, 1, 100,   "2008-12-28 00:00:00.00012345",  "-123.45000"],
         [10, 1, null,  "1899-01-01 00:59:20.001001001", "0.00000"],
         [10, 1, 20006, "2024-10-10 10:10:10.100100100", "1728551410100.10010"]
     ])")
-            .ValueOrDie());
+                .ValueOrDie());
+    }
 
     //  date
     {

src/paimon/format/orc/orc_file_batch_reader.cpp

@@ -242,6 +242,8 @@ Result<::orc::RowReaderOptions> OrcFileBatchReader::CreateRowReaderOptions(
     }
     row_reader_options.include(include_fields);
     row_reader_options.searchArgument(std::move(search_arg));
+    // refer: https://github.com/apache/arrow/pull/34591
+    row_reader_options.setTimezoneName("GMT");
 
     PAIMON_ASSIGN_OR_RAISE(
         bool enable_lazy_decoding,

src/paimon/format/orc/orc_file_batch_reader_test.cpp

@@ -499,15 +499,31 @@ TEST_P(OrcFileBatchReaderTest, TestComplexType) {
     ASSERT_OK_AND_ASSIGN(auto result_array,
                          paimon::test::ReadResultCollector::CollectResult(orc_batch_reader.get()));
     std::shared_ptr<arrow::ChunkedArray> expected_array;
-    auto array_status = arrow::ipc::internal::json::ChunkedArrayFromJSON(arrow_data_type, {R"([
+    arrow::Status array_status;
+    if (::paimon::test::OsReleaseDetector::IsDebian()) {
+        // refer: https://github.com/eggert/tz/blob/main/asia#L653
+        // When using the Asia/Shanghai timezone under Debian, timestamps prior to 1901 have an
+        // additional offset of 5 minutes and 43 seconds
+        array_status = arrow::ipc::internal::json::ChunkedArrayFromJSON(arrow_data_type, {R"([
+        [10, 1, 1234,  "2033-05-18 03:33:20.0",         "123456789987654321.45678", "add"],
+        [10, 1, 19909, "2033-05-18 03:33:20.000001001", "12.30000", "cat"],
+        [10, 1, 0,     "2008-12-28 00:00:00.000123456", null, "dad"],
+        [10, 1, 100,   "2008-12-28 00:00:00.00012345",  "-123.45000", "eat"],
+        [10, 1, null,  "1899-01-01 01:05:03.001001001", "0.00000", "fat"],
+        [10, 1, 20006, "2024-10-10 10:10:10.1001001",   "1728551410100.10010", null]
+    ])"},
+                                                                        &expected_array);
+    } else {
+        array_status = arrow::ipc::internal::json::ChunkedArrayFromJSON(arrow_data_type, {R"([
         [10, 1, 1234,  "2033-05-18 03:33:20.0",         "123456789987654321.45678", "add"],
         [10, 1, 19909, "2033-05-18 03:33:20.000001001", "12.30000", "cat"],
         [10, 1, 0,     "2008-12-28 00:00:00.000123456", null, "dad"],
         [10, 1, 100,   "2008-12-28 00:00:00.00012345",  "-123.45000", "eat"],
         [10, 1, null,  "1899-01-01 00:59:20.001001001", "0.00000", "fat"],
         [10, 1, 20006, "2024-10-10 10:10:10.1001001",   "1728551410100.10010", null]
     ])"},
-                                                                         &expected_array);
+                                                                        &expected_array);
+    }
     ASSERT_TRUE(array_status.ok());
     ASSERT_TRUE(result_array->Equals(*expected_array));
 }

src/paimon/testing/utils/testharness.h

@@ -46,6 +46,7 @@
 #pragma once
 
 #include <filesystem>
+#include <fstream>
 #include <map>
 #include <memory>
 #include <string>
@@ -69,6 +70,26 @@ std::string GetJindoTestDir();
 
 int64_t RandomNumber(int64_t min, int64_t max);
 
+class OsReleaseDetector {
+ public:
+    static bool IsDebian() {
+        std::ifstream file("/etc/os-release");
+        if (!file.is_open()) {
+            return false;
+        }
+
+        std::string line;
+        while (std::getline(file, line)) {
+            if (line.find("ID=") == 0) {
+                if (line.find("debian") != std::string::npos) {
+                    return true;
+                }
+            }
+        }
+        return false;
+    }
+};
+
 ::testing::AssertionResult AssertStatus(const char* s_expr, const Status& s);
 
 #define ASSERT_OK(expr)                                                                  \

test/inte/read_inte_test.cpp

@@ -659,13 +659,26 @@ TEST_P(ReadInteTest, TestAppendReadWithComplexTypePredicate) {
         DataField::ConvertDataFieldsToArrowStructType(fields_with_row_kind);
 
     std::shared_ptr<arrow::ChunkedArray> expected_array;
-    auto array_status = arrow::ipc::internal::json::ChunkedArrayFromJSON(arrow_data_type, {R"([
+    arrow::Status array_status;
+    if (::paimon::test::OsReleaseDetector::IsDebian() && param.file_format == "orc") {
+        // refer: https://github.com/eggert/tz/blob/main/asia#L653
+        // When using the Asia/Shanghai timezone under Debian, timestamps prior to 1901 have an
+        // additional offset of 5 minutes and 43 seconds
+        array_status = arrow::ipc::internal::json::ChunkedArrayFromJSON(arrow_data_type, {R"([
+        [0, "add", 1, "2033-05-18 03:33:20.0",         1234,  "123456789987654321.45678"],
+        [0, "cat", 1, "2033-05-18 03:33:20.000001001", 19909, "12.30000"],
+        [0, "fat", 1, "1899-01-01 01:05:03.001001001", null,  "0.00000"]
+    ])"},
+                                                                        &expected_array);
+    } else {
+        array_status = arrow::ipc::internal::json::ChunkedArrayFromJSON(arrow_data_type, {R"([
         [0, "add", 1, "2033-05-18 03:33:20.0",         1234,  "123456789987654321.45678"],
         [0, "cat", 1, "2033-05-18 03:33:20.000001001", 19909, "12.30000"],
         [0, "fat", 1, "1899-01-01 00:59:20.001001001", null,  "0.00000"],
         [0, "bad", 1, "1899-01-01 00:59:20.001001001", -1234, "-123456789987654321.45678"]
     ])"},
-                                                                         &expected_array);
+                                                                        &expected_array);
+    }
     ASSERT_TRUE(array_status.ok());
     ASSERT_TRUE(result_array);
     ASSERT_TRUE(result_array->Equals(*expected_array));