Enable test suite of libarrow

recipe/bld.bat

@@ -20,12 +20,13 @@ cmake -G "Ninja" ^
       -DARROW_AZURE=%READ_RECIPE_META_YAML_WHY_NOT% ^
       -DARROW_BOOST_USE_SHARED:BOOL=ON ^
       -DARROW_BUILD_STATIC:BOOL=OFF ^
-      -DARROW_BUILD_TESTS:BOOL=OFF ^
+      -DARROW_BUILD_TESTS:BOOL=ON ^
       -DARROW_BUILD_UTILITIES:BOOL=OFF ^
       -DARROW_COMPUTE:BOOL=ON ^
       -DARROW_CSV:BOOL=ON ^
       -DARROW_DATASET:BOOL=ON ^
       -DARROW_DEPENDENCY_SOURCE=SYSTEM ^
+      -DARROW_ENABLE_TIMING_TESTS=OFF ^
       -DARROW_FILESYSTEM:BOOL=ON ^
       -DARROW_FLIGHT:BOOL=ON ^
       -DARROW_FLIGHT_REQUIRE_TLSCREDENTIALSOPTIONS:BOOL=ON ^
@@ -66,4 +67,12 @@ if %ERRORLEVEL% neq 0 exit 1
 cmake --build . --config Release
 if %ERRORLEVEL% neq 0 exit 1
 
+if "%cuda_compiler_version%"=="None" (
+    npm install -g azurite
+    set ARROW_TEST_DATA=%SRC_DIR%\testing\data
+    set PARQUET_TEST_DATA=%SRC_DIR%\cpp\submodules\parquet-testing\data
+    ctest --progress --output-on-failure
+    if %ERRORLEVEL% neq 0 exit 1
+)
+
 popd

recipe/build.sh

@@ -57,13 +57,14 @@ cmake -GNinja \
     -DARROW_BOOST_USE_SHARED=ON \
     -DARROW_BUILD_BENCHMARKS=OFF \
     -DARROW_BUILD_STATIC=OFF \
-    -DARROW_BUILD_TESTS=OFF \
+    -DARROW_BUILD_TESTS=ON \
     -DARROW_BUILD_UTILITIES=OFF \
     -DARROW_COMPUTE=ON \
     -DARROW_CSV=ON \
     -DARROW_CXXFLAGS="${CXXFLAGS}" \
     -DARROW_DATASET=ON \
     -DARROW_DEPENDENCY_SOURCE=SYSTEM \
+    -DARROW_ENABLE_TIMING_TESTS=OFF \
     -DARROW_FILESYSTEM=ON \
     -DARROW_FLIGHT=ON \
     -DARROW_FLIGHT_REQUIRE_TLSCREDENTIALSOPTIONS=ON \
@@ -108,4 +109,11 @@ cmake -GNinja \
 # Do not install arrow, only build.
 cmake --build . --config Release
 
+if [[ "$CONDA_BUILD_CROSS_COMPILATION" != 1 && "$cuda_compiler_version" == "None" ]]; then
+    npm install -g azurite
+    export ARROW_TEST_DATA=$SRC_DIR/testing/data
+    export PARQUET_TEST_DATA=$SRC_DIR/cpp/submodules/parquet-testing/data
+    ctest --progress --output-on-failure
+fi
+
 popd

recipe/meta.yaml

@@ -21,10 +21,20 @@ source:
       - patches/0001-GH-44455-C-Update-vendored-date-to-3.0.3.patch
       # backport https://github.com/apache/arrow/pull/44507
       - patches/0002-GH-44448-C-Add-support-for-overriding-grpc_cpp_plugi.patch
-  # testing-submodule not part of release tarball
+      # backport https://github.com/apache/arrow/pull/44621
+      - patches/0003-GH-43808-C-skip-0117-in-StrptimeZoneOffset-for-old-g.patch
+      # skip gcsfs tests due to missing `storage-testbench`
+      - patches/0004-disable-gcsfs_test.patch
+      # upstream problems on with s3 tests on osx, see
+      # https://github.com/apache/arrow/issues/35587
+      - patches/0005-skip-NonExistentBucket-test-on-osx.patch
+  # testing-submodules not part of release tarball
   - git_url: https://github.com/apache/arrow-testing.git
     git_rev: 4d209492d514c2d3cb2d392681b9aa00e6d8da1c
     folder: testing
+  - git_url: https://github.com/apache/parquet-testing.git
+    git_rev: cb7a9674142c137367bf75a01b79c6e214a73199
+    folder: cpp/submodules/parquet-testing
 
 build:
   number: 8
@@ -50,6 +60,8 @@ requirements:
     # necessary for vendored jemalloc
     - autoconf     # [unix]
     - make         # [unix]
+    # where orc ends up looking (because during build, CONDA_PREFIX==BUILD_PREFIX)
+    - tzdata
   host:
     # for required dependencies, see
     # https://github.com/apache/arrow/blob/apache-arrow-11.0.0/cpp/cmake_modules/ThirdpartyToolchain.cmake#L46-L75
@@ -73,7 +85,7 @@ requirements:
     # arrow uses a customized jemalloc, see #944
     # - jemalloc
     - libabseil
-    - libboost-headers
+    - libboost-devel *
     - libgoogle-cloud-devel
     - libgoogle-cloud-storage-devel
     - libgrpc
@@ -88,6 +100,15 @@ requirements:
     - xsimd
     - zlib
     - zstd
+    # test requirements, c.f.
+    # https://github.com/apache/arrow/blob/apache-arrow-12.0.0/cpp/CMakeLists.txt#L842-L846
+    - gtest
+    # https://github.com/apache/arrow/blob/apache-arrow-12.0.0/cpp/src/arrow/flight/sql/CMakeLists.txt#L86-L87
+    - sqlite *
+    # required by s3fs tests
+    - minio-server
+    # for npm, to install azurite
+    - nodejs *
 
 outputs:
   - name: apache-arrow-proc

recipe/patches/0001-GH-44455-C-Update-vendored-date-to-3.0.3.patch

@@ -1,7 +1,7 @@
 From d774a80949599d58d4726aaa2d78cc64208ab300 Mon Sep 17 00:00:00 2001
 From: Sutou Kouhei <kou@clear-code.com>
 Date: Mon, 21 Oct 2024 16:41:27 +0900
-Subject: [PATCH 1/2] GH-44455: [C++] Update vendored date to 3.0.3
+Subject: [PATCH 1/5] GH-44455: [C++] Update vendored date to 3.0.3
 
 IANA tzdata changed its data format. So we need to update vendored
 date to parse it.
@@ -25,7 +25,7 @@ date to parse it.
  create mode 100755 cpp/src/arrow/vendored/datetime/update.sh
 
 diff --git a/ci/appveyor-cpp-build.bat b/ci/appveyor-cpp-build.bat
-index 08a052e82f..084117f387 100644
+index 08a052e82..084117f38 100644
 --- a/ci/appveyor-cpp-build.bat
 +++ b/ci/appveyor-cpp-build.bat
 @@ -139,7 +139,7 @@ set PARQUET_HOME=%CONDA_PREFIX%\Library
@@ -38,7 +38,7 @@ index 08a052e82f..084117f387 100644
  tar --extract --file tzdata.tar.gz --directory %USERPROFILE%\Downloads\test\tzdata
  curl https://raw.githubusercontent.com/unicode-org/cldr/master/common/supplemental/windowsZones.xml ^
 diff --git a/cpp/src/arrow/CMakeLists.txt b/cpp/src/arrow/CMakeLists.txt
-index c911f0f4e9..5f6b568460 100644
+index c911f0f4e..5f6b56846 100644
 --- a/cpp/src/arrow/CMakeLists.txt
 +++ b/cpp/src/arrow/CMakeLists.txt
 @@ -460,7 +460,7 @@ endif()
@@ -61,7 +61,7 @@ index c911f0f4e9..5f6b568460 100644
  arrow_add_object_library(ARROW_VENDORED ${ARROW_VENDORED_SRCS})
 diff --git a/cpp/src/arrow/vendored/datetime.cpp b/cpp/src/arrow/vendored/datetime.cpp
 new file mode 100644
-index 0000000000..0f0bd12c7e
+index 000000000..0f0bd12c7
 --- /dev/null
 +++ b/cpp/src/arrow/vendored/datetime.cpp
 @@ -0,0 +1,19 @@
@@ -85,7 +85,7 @@ index 0000000000..0f0bd12c7e
 +#include "datetime/visibility.h"
 +#include "datetime/tz.cpp"
 diff --git a/cpp/src/arrow/vendored/datetime.h b/cpp/src/arrow/vendored/datetime.h
-index e437cdcbc2..aea31ebe77 100644
+index e437cdcbc..aea31ebe7 100644
 --- a/cpp/src/arrow/vendored/datetime.h
 +++ b/cpp/src/arrow/vendored/datetime.h
 @@ -17,10 +17,11 @@
@@ -104,7 +104,7 @@ index e437cdcbc2..aea31ebe77 100644
 +#  undef NOEXCEPT
  #endif
 diff --git a/cpp/src/arrow/vendored/datetime/README.md b/cpp/src/arrow/vendored/datetime/README.md
-index 5a0993b7b4..89132d9cba 100644
+index 5a0993b7b..89132d9cb 100644
 --- a/cpp/src/arrow/vendored/datetime/README.md
 +++ b/cpp/src/arrow/vendored/datetime/README.md
 @@ -17,12 +17,16 @@ copies or substantial portions of the Software.
@@ -129,7 +129,7 @@ index 5a0993b7b4..89132d9cba 100644
 +$ ./update.sh 3.0.3
 +```
 diff --git a/cpp/src/arrow/vendored/datetime/date.h b/cpp/src/arrow/vendored/datetime/date.h
-index 75e2624296..c17d6f3f7a 100644
+index 75e262429..c17d6f3f7 100644
 --- a/cpp/src/arrow/vendored/datetime/date.h
 +++ b/cpp/src/arrow/vendored/datetime/date.h
 @@ -84,9 +84,7 @@
@@ -154,7 +154,7 @@ index 75e2624296..c17d6f3f7a 100644
  #ifdef _MSC_VER
  #   pragma warning(pop)
 diff --git a/cpp/src/arrow/vendored/datetime/ios.h b/cpp/src/arrow/vendored/datetime/ios.h
-index acad28d13b..d018e799a8 100644
+index acad28d13..d018e799a 100644
 --- a/cpp/src/arrow/vendored/datetime/ios.h
 +++ b/cpp/src/arrow/vendored/datetime/ios.h
 @@ -32,9 +32,7 @@
@@ -179,7 +179,7 @@ index acad28d13b..d018e799a8 100644
  # endif  // TARGET_OS_IPHONE
  #else   // !__APPLE__
 diff --git a/cpp/src/arrow/vendored/datetime/ios.mm b/cpp/src/arrow/vendored/datetime/ios.mm
-index 22b7ce6c30..70ba2adf0e 100644
+index 22b7ce6c3..70ba2adf0 100644
 --- a/cpp/src/arrow/vendored/datetime/ios.mm
 +++ b/cpp/src/arrow/vendored/datetime/ios.mm
 @@ -47,9 +47,7 @@
@@ -203,7 +203,7 @@ index 22b7ce6c30..70ba2adf0e 100644
 
  #endif  // TARGET_OS_IPHONE
 diff --git a/cpp/src/arrow/vendored/datetime/tz.cpp b/cpp/src/arrow/vendored/datetime/tz.cpp
-index 44c627775f..2cf6c62a84 100644
+index 44c627775..2cf6c62a8 100644
 --- a/cpp/src/arrow/vendored/datetime/tz.cpp
 +++ b/cpp/src/arrow/vendored/datetime/tz.cpp
 @@ -30,10 +30,6 @@
@@ -1009,7 +1009,7 @@ index 44c627775f..2cf6c62a84 100644
  #if defined(__GNUC__) && __GNUC__ < 5
  # pragma GCC diagnostic pop
 diff --git a/cpp/src/arrow/vendored/datetime/tz.h b/cpp/src/arrow/vendored/datetime/tz.h
-index df6d1a851a..61ab3df106 100644
+index df6d1a851..61ab3df10 100644
 --- a/cpp/src/arrow/vendored/datetime/tz.h
 +++ b/cpp/src/arrow/vendored/datetime/tz.h
 @@ -43,19 +43,13 @@
@@ -1124,7 +1124,7 @@ index df6d1a851a..61ab3df106 100644
 
  #endif  // TZ_H
 diff --git a/cpp/src/arrow/vendored/datetime/tz_private.h b/cpp/src/arrow/vendored/datetime/tz_private.h
-index a6bb8fd30a..1d7f858971 100644
+index a6bb8fd30..1d7f85897 100644
 --- a/cpp/src/arrow/vendored/datetime/tz_private.h
 +++ b/cpp/src/arrow/vendored/datetime/tz_private.h
 @@ -34,9 +34,7 @@
@@ -1150,7 +1150,7 @@ index a6bb8fd30a..1d7f858971 100644
  #include "tz.h"
 diff --git a/cpp/src/arrow/vendored/datetime/update.sh b/cpp/src/arrow/vendored/datetime/update.sh
 new file mode 100755
-index 0000000000..b4580c0426
+index 000000000..b4580c042
 --- /dev/null
 +++ b/cpp/src/arrow/vendored/datetime/update.sh
 @@ -0,0 +1,53 @@
@@ -1208,7 +1208,7 @@ index 0000000000..b4580c0426
 +rm *.bak
 +popd
 diff --git a/cpp/src/arrow/vendored/datetime/visibility.h b/cpp/src/arrow/vendored/datetime/visibility.h
-index ae031238d8..780c00d70b 100644
+index ae031238d..780c00d70 100644
 --- a/cpp/src/arrow/vendored/datetime/visibility.h
 +++ b/cpp/src/arrow/vendored/datetime/visibility.h
 @@ -17,10 +17,14 @@
@@ -1229,7 +1229,7 @@ index ae031238d8..780c00d70b 100644
 +#  define DATE_USE_DLL
  #endif
 diff --git a/cpp/src/gandiva/precompiled/CMakeLists.txt b/cpp/src/gandiva/precompiled/CMakeLists.txt
-index c2bc7fc027..e1427e25fb 100644
+index c2bc7fc02..e1427e25f 100644
 --- a/cpp/src/gandiva/precompiled/CMakeLists.txt
 +++ b/cpp/src/gandiva/precompiled/CMakeLists.txt
 @@ -63,7 +63,7 @@ add_gandiva_test(precompiled-test

recipe/patches/0002-GH-44448-C-Add-support-for-overriding-grpc_cpp_plugi.patch

@@ -1,7 +1,7 @@
 From 3d00b553ffd302ada8d74b5b77b71128c594c0b4 Mon Sep 17 00:00:00 2001
 From: Sutou Kouhei <kou@clear-code.com>
 Date: Wed, 23 Oct 2024 10:30:33 +0900
-Subject: [PATCH 2/2] GH-44448: [C++] Add support for overriding
+Subject: [PATCH 2/5] GH-44448: [C++] Add support for overriding
  grpc_cpp_plugin path for cross-compiling
 
 ---
@@ -10,7 +10,7 @@ Subject: [PATCH 2/2] GH-44448: [C++] Add support for overriding
  2 files changed, 13 insertions(+)
 
 diff --git a/cpp/cmake_modules/DefineOptions.cmake b/cpp/cmake_modules/DefineOptions.cmake
-index d823444cf7..a7bf9e59f8 100644
+index d823444cf..a7bf9e59f 100644
 --- a/cpp/cmake_modules/DefineOptions.cmake
 +++ b/cpp/cmake_modules/DefineOptions.cmake
 @@ -640,6 +640,11 @@ Always OFF if building binaries" OFF)
@@ -26,7 +26,7 @@ index d823444cf7..a7bf9e59f8 100644
    set_option_category("Advanced developer")
 
 diff --git a/cpp/cmake_modules/ThirdpartyToolchain.cmake b/cpp/cmake_modules/ThirdpartyToolchain.cmake
-index db151b4e0f..0b215b5b25 100644
+index db151b4e0..0b215b5b2 100644
 --- a/cpp/cmake_modules/ThirdpartyToolchain.cmake
 +++ b/cpp/cmake_modules/ThirdpartyToolchain.cmake
 @@ -4223,6 +4223,14 @@ if(ARROW_WITH_GRPC)

recipe/patches/0003-GH-43808-C-skip-0117-in-StrptimeZoneOffset-for-old-g.patch

@@ -0,0 +1,62 @@
+From acf75c664b813f49eff275e74c0026af5ba90e84 Mon Sep 17 00:00:00 2001
+From: h-vetinari <h.vetinari@gmx.com>
+Date: Fri, 15 Nov 2024 19:38:40 +1100
+Subject: [PATCH 3/5] GH-43808: [C++] skip `-0117` in StrptimeZoneOffset for
+ old glibc (#44621)
+
+### Rationale for this change
+
+Enable tests for libarrow in conda-forge: https://github.com/apache/arrow/issues/35587
+
+### What changes are included in this PR?
+
+old glibc does not actually support timezones like `-0117` (used in `StrptimeZoneOffset` test). The exact lower bound for glibc is hard for me to determine; I know that it passes with 2.28 and that it fails with 2.17. Anything in between is an open question. I went with the conservative option here.
+
+### Are these changes tested?
+
+Tested in https://github.com/conda-forge/arrow-cpp-feedstock/pull/1058
+
+### Are there any user-facing changes?
+
+* GitHub Issue: #43808
+
+Lead-authored-by: H. Vetinari <h.vetinari@gmx.com>
+Co-authored-by: Sutou Kouhei <kou@cozmixng.org>
+Signed-off-by: Sutou Kouhei <kou@clear-code.com>
+---
+ cpp/src/arrow/util/value_parsing_test.cc | 19 ++++++++++++++++---
+ 1 file changed, 16 insertions(+), 3 deletions(-)
+
+diff --git a/cpp/src/arrow/util/value_parsing_test.cc b/cpp/src/arrow/util/value_parsing_test.cc
+index 7cd1ab1e2..a833d266a 100644
+--- a/cpp/src/arrow/util/value_parsing_test.cc
++++ b/cpp/src/arrow/util/value_parsing_test.cc
+@@ -838,12 +838,25 @@ TEST(TimestampParser, StrptimeZoneOffset) {
+   std::string format = "%Y-%d-%m %H:%M:%S%z";
+   auto parser = TimestampParser::MakeStrptime(format);
+
++  std::vector<std::string> values = {
++    "2018-01-01 00:00:00+0000",
++    "2018-01-01 00:00:00+0100",
++#if defined(__GLIBC__) && defined(__GLIBC_MINOR__)
++// glibc < 2.28 doesn't support "-0117" timezone offset.
++// See also: https://github.com/apache/arrow/issues/43808
++#  if ((__GLIBC__ == 2) && (__GLIBC_MINOR__ >= 28)) || (__GLIBC__ >= 3)
++    "2018-01-01 00:00:00-0117",
++#  endif
++#else
++    "2018-01-01 00:00:00-0117",
++#endif
++    "2018-01-01 00:00:00+0130"
++  };
++
+   // N.B. GNU %z supports ISO8601 format while BSD %z supports only
+   // +HHMM or -HHMM and POSIX doesn't appear to define %z at all
+   for (auto unit : TimeUnit::values()) {
+-    for (const std::string value :
+-         {"2018-01-01 00:00:00+0000", "2018-01-01 00:00:00+0100",
+-          "2018-01-01 00:00:00+0130", "2018-01-01 00:00:00-0117"}) {
++    for (const std::string& value : values) {
+       SCOPED_TRACE(value);
+       int64_t converted = 0;
+       int64_t expected = 0;

recipe/patches/0004-disable-gcsfs_test.patch

@@ -0,0 +1,31 @@
+From a828c478993de3369729011e9f433953ea823622 Mon Sep 17 00:00:00 2001
+From: "H. Vetinari" <h.vetinari@gmx.com>
+Date: Sat, 2 Nov 2024 15:41:34 +1100
+Subject: [PATCH 4/5] disable gcsfs_test
+
+it cannot work unless we package https://github.com/googleapis/storage-testbench,
+which however has extremely tight dependencies on protobuf etc., making it very
+hard to fit this into our migration patterns
+---
+ cpp/src/arrow/filesystem/CMakeLists.txt | 8 --------
+ 1 file changed, 8 deletions(-)
+
+diff --git a/cpp/src/arrow/filesystem/CMakeLists.txt b/cpp/src/arrow/filesystem/CMakeLists.txt
+index 7afdf566f..b22a48a0c 100644
+--- a/cpp/src/arrow/filesystem/CMakeLists.txt
++++ b/cpp/src/arrow/filesystem/CMakeLists.txt
+@@ -42,14 +42,6 @@ if(ARROW_BUILD_BENCHMARKS)
+                       ${ARROW_BENCHMARK_LINK_LIBS})
+ endif()
+
+-if(ARROW_GCS)
+-  add_arrow_test(gcsfs_test
+-                 EXTRA_LABELS
+-                 filesystem
+-                 EXTRA_LINK_LIBS
+-                 google-cloud-cpp::storage)
+-endif()
+-
+ if(ARROW_AZURE)
+   add_arrow_test(azurefs_test
+                  EXTRA_LABELS

recipe/patches/0005-skip-NonExistentBucket-test-on-osx.patch

@@ -0,0 +1,24 @@
+From b249dd3a026d377c093be12d0e0ded0667996dfd Mon Sep 17 00:00:00 2001
+From: "H. Vetinari" <h.vetinari@gmx.com>
+Date: Sun, 24 Nov 2024 20:22:35 +1100
+Subject: [PATCH 5/5] skip NonExistentBucket test on osx
+
+---
+ cpp/src/arrow/filesystem/s3fs_test.cc | 4 ++++
+ 1 file changed, 4 insertions(+)
+
+diff --git a/cpp/src/arrow/filesystem/s3fs_test.cc b/cpp/src/arrow/filesystem/s3fs_test.cc
+index 43091aaa9..cf179551f 100644
+--- a/cpp/src/arrow/filesystem/s3fs_test.cc
++++ b/cpp/src/arrow/filesystem/s3fs_test.cc
+@@ -375,6 +375,10 @@ TEST_F(S3RegionResolutionTest, RestrictedBucket) {
+ }
+
+ TEST_F(S3RegionResolutionTest, NonExistentBucket) {
++#ifdef __APPLE__
++  // spuriously fails, see https://github.com/apache/arrow/issues/35587#issuecomment-2495637658
++  GTEST_SKIP();
++#endif
+   auto maybe_region = ResolveS3BucketRegion("ursa-labs-nonexistent-bucket");
+   ASSERT_RAISES(IOError, maybe_region);
+   ASSERT_THAT(maybe_region.status().message(),