Remove windows tz workarounds

rok · rok · commit 6328db3db3fb · 2026-01-05T19:00:28.000+01:00
diff --git a/.github/workflows/cpp.yml b/.github/workflows/cpp.yml
@@ -366,9 +366,6 @@ jobs:
         run: |
           export CMAKE_BUILD_PARALLEL_LEVEL=$NUMBER_OF_PROCESSORS
           ci/scripts/cpp_build.sh "$(pwd)" "$(pwd)/build"
-      - name: Download Timezone Database
-        shell: bash
-        run: ci/scripts/download_tz_database.sh
       - name: Download MinIO
         shell: msys2 {0}
         run: |
diff --git a/.github/workflows/cpp_extra.yml b/.github/workflows/cpp_extra.yml
@@ -365,9 +365,6 @@ jobs:
         with:
           fetch-depth: 0
           submodules: recursive
-      - name: Download Timezone Database
-        shell: bash
-        run: ci/scripts/download_tz_database.sh
       - name: Install cmake
         shell: bash
         run: |
diff --git a/.github/workflows/matlab.yml b/.github/workflows/matlab.yml
@@ -147,9 +147,6 @@ jobs:
         uses: matlab-actions/setup-matlab@v2
         with:
           release: R2025b
-      - name: Download Timezone Database
-        shell: bash
-        run: ci/scripts/download_tz_database.sh
       - name: Install ccache
         shell: bash
         run: ci/scripts/install_ccache.sh 4.6.3 /usr
diff --git a/.github/workflows/verify_rc.yml b/.github/workflows/verify_rc.yml
@@ -228,9 +228,6 @@ jobs:
         run: |
           choco install --no-progress --yes boost-msvc-14.1
           choco install --no-progress --yes wget
-      - name: Download Timezone Database
-        shell: bash
-        run: ci/scripts/download_tz_database.sh
       - name: Run verification
         env:
           GH_TOKEN: ${{ github.token }}
diff --git a/ci/scripts/download_tz_database.sh b/ci/scripts/download_tz_database.sh
diff --git a/cpp/src/arrow/compute/function_test.cc b/cpp/src/arrow/compute/function_test.cc
@@ -95,11 +95,9 @@ TEST(FunctionOptions, Equality) {
   options.emplace_back(new StrptimeOptions("%Y", TimeUnit::type::MILLI, true));
   options.emplace_back(new StrptimeOptions("%Y", TimeUnit::type::NANO));
   options.emplace_back(new StrftimeOptions("%Y-%m-%dT%H:%M:%SZ", "C"));
-#ifndef _WIN32
   options.emplace_back(new AssumeTimezoneOptions(
       "Europe/Amsterdam", AssumeTimezoneOptions::Ambiguous::AMBIGUOUS_RAISE,
       AssumeTimezoneOptions::Nonexistent::NONEXISTENT_RAISE));
-#endif
   options.emplace_back(new PadOptions(5, " "));
   options.emplace_back(new PadOptions(10, "A"));
   options.emplace_back(new PadOptions(10, "A", false));
diff --git a/cpp/src/arrow/compute/kernels/scalar_cast_test.cc b/cpp/src/arrow/compute/kernels/scalar_cast_test.cc
@@ -2358,15 +2358,7 @@ constexpr char kTimestampSecondsJson[] =
 constexpr char kTimestampExtremeJson[] =
     R"(["1677-09-20T00:00:59.123456", "2262-04-13T23:23:23.999999"])";
 
-class CastTimezone : public ::testing::Test {
- protected:
-  void SetUp() override {
-#ifdef _WIN32
-    // Initialize timezone database on Windows
-    ASSERT_OK(InitTestTimezoneDatabase());
-#endif
-  }
-};
+class CastTimezone : public ::testing::Test {};
 
 TEST(Cast, TimestampToDate) {
   // See scalar_temporal_test.cc
diff --git a/cpp/src/arrow/compute/kernels/scalar_temporal_test.cc b/cpp/src/arrow/compute/kernels/scalar_temporal_test.cc
@@ -412,14 +412,6 @@ class ScalarTemporalTest : public ::testing::Test {
   RoundTemporalOptions round_to_15_quarters =
       RoundTemporalOptions(15, CalendarUnit::QUARTER);
   RoundTemporalOptions round_to_15_years = RoundTemporalOptions(15, CalendarUnit::YEAR);
-
- protected:
-  void SetUp() override {
-#ifdef _WIN32
-    // Initialize timezone database on Windows
-    ASSERT_OK(InitTestTimezoneDatabase());
-#endif
-  }
 };
 
 class ScalarTemporalTestStrictCeil : public ScalarTemporalTest {
diff --git a/cpp/src/arrow/config.cc b/cpp/src/arrow/config.cc
@@ -64,8 +64,6 @@ std::string MakeSimdLevelString(QueryFlagFunction&& query_flag) {
   }
 }
 
-std::optional<std::string> timezone_db_path;
-
 };  // namespace
 
 const BuildInfo& GetBuildInfo() { return kBuildInfo; }
@@ -77,32 +75,7 @@ RuntimeInfo GetRuntimeInfo() {
       MakeSimdLevelString([&](int64_t flags) { return cpu_info->IsSupported(flags); });
   info.detected_simd_level =
       MakeSimdLevelString([&](int64_t flags) { return cpu_info->IsDetected(flags); });
-  info.using_os_timezone_db = USE_OS_TZDB;
-#if !USE_OS_TZDB
-  info.timezone_db_path = timezone_db_path;
-#else
-  info.timezone_db_path = std::optional<std::string>();
-#endif
   return info;
 }
 
-Status Initialize(const GlobalOptions& options) noexcept {
-  if (options.timezone_db_path.has_value()) {
-#if !USE_OS_TZDB
-    try {
-      arrow_vendored::date::set_install(options.timezone_db_path.value());
-      arrow_vendored::date::reload_tzdb();
-    } catch (const std::runtime_error& e) {
-      return Status::IOError(e.what());
-    }
-    timezone_db_path = options.timezone_db_path.value();
-#else
-    return Status::Invalid(
-        "Arrow was set to use OS timezone database at compile time, "
-        "so a downloaded database cannot be provided at runtime.");
-#endif  // !USE_OS_TZDB
-  }
-  return Status::OK();
-}
-
 }  // namespace arrow
diff --git a/cpp/src/arrow/config.h b/cpp/src/arrow/config.h
@@ -64,13 +64,6 @@ struct RuntimeInfo {
 
   /// The SIMD level available on the OS and CPU
   std::string detected_simd_level;
-
-  /// Whether using the OS-based timezone database
-  /// This is set at compile-time.
-  bool using_os_timezone_db;
-
-  /// The path to the timezone database; by default None.
-  std::optional<std::string> timezone_db_path;
 };
 
 /// \brief Get runtime build info.
@@ -86,13 +79,4 @@ const BuildInfo& GetBuildInfo();
 ARROW_EXPORT
 RuntimeInfo GetRuntimeInfo();
 
-struct GlobalOptions {
-  /// Path to text timezone database. This is only configurable on Windows,
-  /// which does not have a compatible OS timezone database.
-  std::optional<std::string> timezone_db_path;
-};
-
-ARROW_EXPORT
-Status Initialize(const GlobalOptions& options) noexcept;
-
 }  // namespace arrow
diff --git a/cpp/src/arrow/public_api_test.cc b/cpp/src/arrow/public_api_test.cc
@@ -122,46 +122,4 @@ TEST(Misc, BuildInfo) {
   ASSERT_THAT(info.full_so_version, ::testing::HasSubstr(info.so_version));
 }
 
-TEST(Misc, SetTimezoneConfig) {
-#ifndef _WIN32
-  GTEST_SKIP() << "Can only set the Timezone database on Windows";
-#elif !defined(ARROW_FILESYSTEM)
-  GTEST_SKIP() << "Need filesystem support to test timezone config.";
-#else
-  auto fs = std::make_shared<arrow::fs::LocalFileSystem>();
-
-  std::optional<std::string> tzdata_result = GetTestTimezoneDatabaseRoot();
-  std::string tzdata_dir;
-  if (tzdata_result.has_value()) {
-    tzdata_dir = tzdata_result.value();
-  } else {
-    auto home_raw = std::getenv("USERPROFILE");
-    std::string home = home_raw == nullptr ? "~" : std::string(home_raw);
-    ASSERT_OK_AND_ASSIGN(tzdata_dir, fs->NormalizePath(home + "\\Downloads\\tzdata"));
-  }
-  ASSERT_OK_AND_ASSIGN(tzdata_dir, fs->NormalizePath(tzdata_dir));
-  ASSERT_OK_AND_ASSIGN(auto tzdata_path,
-                       arrow::internal::PlatformFilename::FromString(tzdata_dir));
-
-  if (!arrow::internal::FileExists(tzdata_path).ValueOr(false)) {
-    GTEST_SKIP() << "Couldn't find timezone database in expected dir: " << tzdata_dir;
-  }
-  // Create a tmp directory
-  ASSERT_OK_AND_ASSIGN(auto tempdir, arrow::internal::TemporaryDir::Make("tzdata"));
-
-  // Validate that setting tzdb to that dir fails
-  arrow::GlobalOptions options = {std::make_optional(tempdir->path().ToString())};
-  ASSERT_NOT_OK(arrow::Initialize(options));
-
-  // Copy tzdb data from ~/Downloads
-  auto selector = arrow::fs::FileSelector();
-  selector.base_dir = tzdata_dir;
-  selector.recursive = true;
-  ASSERT_OK(arrow::fs::CopyFiles(fs, selector, fs, tempdir->path().ToString()));
-
-  // Validate that tzdb is working
-  ASSERT_OK(arrow::Initialize(options));
-#endif
-}
-
 }  // namespace arrow
diff --git a/cpp/src/arrow/testing/util.cc b/cpp/src/arrow/testing/util.cc
@@ -122,25 +122,6 @@ Status GetTestResourceRoot(std::string* out) {
   return Status::OK();
 }
 
-std::optional<std::string> GetTestTimezoneDatabaseRoot() {
-  const char* c_root = std::getenv("ARROW_TIMEZONE_DATABASE");
-  if (!c_root) {
-    return std::optional<std::string>();
-  }
-  return std::make_optional(std::string(c_root));
-}
-
-Status InitTestTimezoneDatabase() {
-  auto maybe_tzdata = GetTestTimezoneDatabaseRoot();
-  // If missing, timezone database will default to %USERPROFILE%\Downloads\tzdata
-  if (!maybe_tzdata.has_value()) return Status::OK();
-
-  auto tzdata_path = std::string(maybe_tzdata.value());
-  arrow::GlobalOptions options = {std::make_optional(tzdata_path)};
-  ARROW_RETURN_NOT_OK(arrow::Initialize(options));
-  return Status::OK();
-}
-
 int GetListenPort() {
   // Get a new available port number by binding a socket to an ephemeral port
   // and then closing it.  Since ephemeral port allocation tends to avoid
diff --git a/cpp/src/arrow/testing/util.h b/cpp/src/arrow/testing/util.h
@@ -112,13 +112,6 @@ UnionTypeFactories() {
 // Status
 ARROW_TESTING_EXPORT Status GetTestResourceRoot(std::string*);
 
-// Return the value of the ARROW_TIMEZONE_DATABASE environment variable
-ARROW_TESTING_EXPORT std::optional<std::string> GetTestTimezoneDatabaseRoot();
-
-// Set the Timezone database based on the ARROW_TIMEZONE_DATABASE env variable
-// This is only relevant on Windows, since other OSs have compatible databases built-in
-ARROW_TESTING_EXPORT Status InitTestTimezoneDatabase();
-
 // Get a TCP port number to listen on.  This is a different number every time,
 // as reusing the same port across tests can produce spurious bind errors on
 // Windows.
diff --git a/cpp/src/arrow/vendored/datetime/visibility.h b/cpp/src/arrow/vendored/datetime/visibility.h
@@ -17,9 +17,7 @@
 
 #pragma once
 
-#ifndef _WIN32
-#  define USE_OS_TZDB 1
-#endif
+#define USE_OS_TZDB 1
 
 #if defined(ARROW_STATIC)
 // intentially empty
diff --git a/dev/tasks/vcpkg-tests/github.windows.yml b/dev/tasks/vcpkg-tests/github.windows.yml
@@ -35,9 +35,6 @@ jobs:
         run: |
           arrow/ci/scripts/install_cmake.sh 3.29.0 /c/cmake
           echo "c:\\cmake\\bin" >> $GITHUB_PATH
-      - name: Download Timezone Database
-        shell: bash
-        run: arrow/ci/scripts/download_tz_database.sh
       - name: Remove and Reinstall vcpkg
         # When running vcpkg in GitHub Actions on Windows, remove the
         # preinstalled vcpkg and install the newest version from source.
diff --git a/dev/tasks/verify-rc/github.win.yml b/dev/tasks/verify-rc/github.win.yml
@@ -40,10 +40,6 @@ jobs:
           choco install boost-msvc-14.1
           choco install wget
 
-      - name: Download Timezone Database
-        shell: bash
-        run: arrow/ci/scripts/download_tz_database.sh
-
       - name: Run verification
         shell: cmd
         run: |
diff --git a/docs/source/cpp/build_system.rst b/docs/source/cpp/build_system.rst
@@ -228,26 +228,3 @@ can control the source of each dependency and whether it is statically or
 dynamically linked. See :doc:`/developers/cpp/building` for instructions. Or
 alternatively, use Arrow from a package manager such as Conda or vcpkg which
 will manage consistent versions of Arrow and its dependencies.
-
-
-.. _download-timezone-database:
-
-Runtime Dependencies
-====================
-
-While Arrow uses the OS-provided timezone database on Linux and macOS, it
-requires a user-provided database on Windows. You must download and extract the
-text version of the IANA timezone database and add the Windows timezone mapping
-XML. To download, you can use the following batch script:
-
-.. literalinclude:: ../../../ci/appveyor-cpp-setup.bat
-   :language: batch
-   :start-after: @rem (Doc section: Download timezone database)
-   :end-before: @rem (Doc section: Download timezone database)
-
-By default, the timezone database will be detected at ``%USERPROFILE%\Downloads\tzdata``,
-but you can set a custom path at runtime in :struct:`arrow::ArrowGlobalOptions`::
-
-   arrow::GlobalOptions options;
-   options.timezone_db_path = "path/to/tzdata";
-   ARROW_RETURN_NOT_OK(arrow::Initialize(options));
diff --git a/docs/source/developers/cpp/windows.rst b/docs/source/developers/cpp/windows.rst
@@ -381,15 +381,6 @@ be defined, and similarly for ``-DARROW_FLIGHT_SQL=ON``.
                               ARROW_FLIGHT_STATIC
                               ARROW_FLIGHT_SQL_STATIC)
 
-Downloading the Timezone Database
-=================================
-
-To run some of the compute unit tests on Windows, the IANA timezone database
-and the Windows timezone mapping need to be downloaded first. See
-:ref:`download-timezone-database` for download instructions. To set a non-default
-path for the timezone database while running the unit tests, set the
-``ARROW_TIMEZONE_DATABASE`` environment variable.
-
 Replicating Appveyor Builds
 ===========================
 
diff --git a/docs/source/python/install.rst b/docs/source/python/install.rst
@@ -80,39 +80,6 @@ Optional dependencies
 Additional packages PyArrow is compatible with are :ref:`fsspec <filesystem-fsspec>`
 and **pytz**, **dateutil** or **tzdata** package for timezones.
 
-tzdata on Windows
-^^^^^^^^^^^^^^^^^
-
-While Arrow uses the OS-provided timezone database on Linux and macOS, it requires a
-user-provided database on Windows. To download and extract the text version of
-the IANA timezone database follow the instructions in the C++
-:ref:`download-timezone-database` or use pyarrow utility function
-``pyarrow.util.download_tzdata_on_windows()`` that does the same.
-
-By default, the timezone database will be detected at ``%USERPROFILE%\Downloads\tzdata``.
-If the database has been downloaded in a different location, you will need to set
-a custom path to the database from Python:
-
-.. code-block:: python
-
-   >>> import pyarrow as pa
-   >>> pa.set_timezone_db_path("custom_path")
-
-You may encounter problems writing datetime data to an ORC file if you install
-pyarrow with pip. One possible solution to fix this problem:
-
-   1. Install tzdata with ``pip install tzdata``
-   2. Set the environment variable ``TZDIR = path\to\.venv\Lib\site-packages\tzdata\``
-
-You can find where ``tzdata`` is installed with the following python
-command:
-
-.. code-block:: python
-
-   >>> import tzdata
-   >>> print(tzdata.__file__)
-   path\to\.venv\Lib\site-packages\tzdata\__init__.py
-
 
 .. _python-conda-differences:
 
diff --git a/python/pyarrow/conftest.py b/python/pyarrow/conftest.py
@@ -22,7 +22,6 @@
 from pyarrow import Codec
 from pyarrow import fs
 from pyarrow.lib import is_threading_enabled
-from pyarrow.tests.util import windows_has_tzdata
 import sys
 
 
@@ -108,9 +107,7 @@
     defaults['processes'] = False
     defaults['sockets'] = False
 
-if sys.platform == "win32":
-    defaults['timezone_data'] = windows_has_tzdata()
-elif sys.platform == "emscripten":
+if sys.platform == "emscripten":
     defaults['timezone_data'] = os.path.exists("/usr/share/zoneinfo")
 
 try:
diff --git a/python/pyarrow/tests/conftest.py b/python/pyarrow/tests/conftest.py
diff --git a/python/pyarrow/tests/util.py b/python/pyarrow/tests/util.py
diff --git a/python/pyarrow/util.py b/python/pyarrow/util.py
