[GitHub] [arrow-site] wesm merged pull request #34: Update docs for 0.15.0
wesm merged pull request #34: Update docs for 0.15.0 URL: https://github.com/apache/arrow-site/pull/34 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [arrow-site] wesm opened a new pull request #34: Update docs for 0.15.0
wesm opened a new pull request #34: Update docs for 0.15.0 URL: https://github.com/apache/arrow-site/pull/34 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[arrow] branch master updated: ARROW-6860: [Python][C++] Do not link shared libraries monolithically to pyarrow.lib, add libarrow_python_flight.so
This is an automated email from the ASF dual-hosted git repository. wesm pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/arrow.git The following commit(s) were added to refs/heads/master by this push: new 102acc4 ARROW-6860: [Python][C++] Do not link shared libraries monolithically to pyarrow.lib, add libarrow_python_flight.so 102acc4 is described below commit 102acc47287c37a01ac11a5cb6bd1da3f1f0712d Author: Wes McKinney AuthorDate: Sat Oct 12 14:42:04 2019 -0500 ARROW-6860: [Python][C++] Do not link shared libraries monolithically to pyarrow.lib, add libarrow_python_flight.so Adding a new shared library libarrow_python_flight.so that allows us to link libarrow_flight and this new library to the Cython _flight extension. I initially tried moving the Flight Python bindings directly to libarrow_flight but realized this would create a transitive dependency on libpython which is not desirable. Any shared library that uses Python C APIs is expected to be loaded into a running Python interpreter and not linked explicitly to libpython Because Apache ORC also needs to statically link Protocol Buffers, I have disabled it in the manylinux wheels. Hopefully we can come up with a solution where projects like Apache Beam, TensorFlow, and others can all use Protocol Buffers together and not have these problems Closes #5627 from wesm/ARROW-6860 and squashes the following commits: d5d67f81c Revert libarrow_flight.pxd changes b31fbdf32 Build libarrow_python_flight that links to libarrow_python and libarrow_flight. Do not link all shared libraries to Cython "lib" extension Authored-by: Wes McKinney Signed-off-by: Wes McKinney --- cpp/cmake_modules/FindArrowFlight.cmake | 41 +-- cpp/src/arrow/python/CMakeLists.txt | 48 +++ cpp/src/arrow/python/flight.h | 59 +++-- python/CMakeLists.txt | 46 - python/manylinux1/build_arrow.sh| 11 +++--- python/manylinux2010/build_arrow.sh | 11 +++--- 6 files changed, 166 insertions(+), 50 deletions(-) diff --git a/cpp/cmake_modules/FindArrowFlight.cmake b/cpp/cmake_modules/FindArrowFlight.cmake index 193361a..048e153 100644 --- a/cpp/cmake_modules/FindArrowFlight.cmake +++ b/cpp/cmake_modules/FindArrowFlight.cmake @@ -66,7 +66,13 @@ find_library(ARROW_FLIGHT_LIB_PATH PATHS ${ARROW_SEARCH_LIB_PATH} PATH_SUFFIXES ${LIB_PATH_SUFFIXES} NO_DEFAULT_PATH) +find_library(ARROW_PYTHON_FLIGHT_LIB_PATH + NAMES arrow_python_flight + PATHS ${ARROW_SEARCH_LIB_PATH} + PATH_SUFFIXES ${LIB_PATH_SUFFIXES} + NO_DEFAULT_PATH) get_filename_component(ARROW_FLIGHT_LIBS ${ARROW_FLIGHT_LIB_PATH} DIRECTORY) +get_filename_component(ARROW_PYTHON_FLIGHT_LIBS ${ARROW_PYTHON_FLIGHT_LIB_PATH} DIRECTORY) if(MSVC) # Prioritize "/bin" over LIB_PATH_SUFFIXES - DLL files are installed @@ -77,7 +83,15 @@ if(MSVC) PATHS ${ARROW_HOME} PATH_SUFFIXES "bin" ${LIB_PATH_SUFFIXES} NO_DEFAULT_PATH) + find_library(ARROW_PYTHON_FLIGHT_SHARED_LIBRARIES + NAMES arrow_flight + PATHS ${ARROW_HOME} + PATH_SUFFIXES "bin" ${LIB_PATH_SUFFIXES} + NO_DEFAULT_PATH) + get_filename_component(ARROW_FLIGHT_SHARED_LIBS ${ARROW_FLIGHT_SHARED_LIBRARIES} DIRECTORY) + get_filename_component(ARROW_PYTHON_FLIGHT_SHARED_LIBS +${ARROW_PYTHON_FLIGHT_SHARED_LIBRARIES} DIRECTORY) endif() if(ARROW_FLIGHT_INCLUDE_DIR AND ARROW_FLIGHT_LIBS) @@ -117,9 +131,32 @@ else() set(ARROW_FLIGHT_FOUND FALSE) endif() +if(ARROW_PYTHON_FLIGHT_LIBS) + set(ARROW_PYTHON_FLIGHT_FOUND TRUE) + set(ARROW_PYTHON_FLIGHT_LIB_NAME arrow_python_flight) + if(MSVC) +set( + ARROW_PYTHON_FLIGHT_STATIC_LIB + ${ARROW_PYTHON_FLIGHT_LIBS}/${ARROW_PYTHON_FLIGHT_LIB_NAME}${ARROW_MSVC_STATIC_LIB_SUFFIX}${CMAKE_STATIC_LIBRARY_SUFFIX} + ) +set(ARROW_PYTHON_FLIGHT_SHARED_LIB + ${ARROW_PYTHON_FLIGHT_SHARED_LIBS}/${ARROW_PYTHON_FLIGHT_LIB_NAME}${CMAKE_SHARED_LIBRARY_SUFFIX}) +set(ARROW_PYTHON_FLIGHT_SHARED_IMP_LIB ${ARROW_PYTHON_FLIGHT_LIBS}/${ARROW_PYTHON_FLIGHT_LIB_NAME}.lib) + else() +set(ARROW_PYTHON_FLIGHT_STATIC_LIB ${ARROW_LIBS}/lib${ARROW_PYTHON_FLIGHT_LIB_NAME}.a) +set(ARROW_PYTHON_FLIGHT_SHARED_LIB + ${ARROW_LIBS}/lib${ARROW_PYTHON_FLIGHT_LIB_NAME}${CMAKE_SHARED_LIBRARY_SUFFIX}) + endif() + + message(STATUS "Found the Arrow Flight Python library: ${ARROW_PYTHON_FLIGHT_LIB_PATH}") +endif() + if(MSVC) mark_as_advanced(ARROW_FLIGHT_INCLUDE_DIR ARROW_FLIGHT_STATIC_LIB ARROW_FLIGHT_SHARED_LIB - ARROW_FLIGHT_SHARED_IMP_LIB) +ARROW_FLIGHT_SHARED_IMP_LIB +ARROW_PYTHON_FLIGHT_STATIC_LIB ARROW_PYTHON_FLIGHT_SHARED_LIB +
[arrow] branch master updated: ARROW-5680: [Rust] [DataFusion] GROUP BY sql tests are now deterministic
This is an automated email from the ASF dual-hosted git repository. agrove pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/arrow.git The following commit(s) were added to refs/heads/master by this push: new d47a40e ARROW-5680: [Rust] [DataFusion] GROUP BY sql tests are now deterministic d47a40e is described below commit d47a40e88292840b9c73bec59af1828b273445e8 Author: Andy Grove AuthorDate: Sat Oct 12 10:37:11 2019 -0600 ARROW-5680: [Rust] [DataFusion] GROUP BY sql tests are now deterministic DataFusion doesn't support `ORDER BY` yet, so I modified the aggregate tests to collect `Vec` and sort it before comparing against expected results, making these tests deterministic. The PR is a little noisy because I removed the final `\n` from the expected results in most of the tests due to the new approach of collecting `Vec` and then performing a `.join("\n")` to create the string. Closes #5622 from andygrove/ARROW-5680 and squashes the following commits: a3083c6cb GROUP BY sql tests are now deterministic Authored-by: Andy Grove Signed-off-by: Andy Grove --- rust/datafusion/tests/sql.rs | 141 +++ 1 file changed, 47 insertions(+), 94 deletions(-) diff --git a/rust/datafusion/tests/sql.rs b/rust/datafusion/tests/sql.rs index bb60603..b9741ce 100644 --- a/rust/datafusion/tests/sql.rs +++ b/rust/datafusion/tests/sql.rs @@ -87,8 +87,10 @@ fn parquet_query() { let mut ctx = ExecutionContext::new(); register_alltypes_parquet( ctx); let sql = "SELECT id, string_col FROM alltypes_plain"; -let actual = execute( ctx, sql); -let expected = "4\t\"0\"\n5\t\"1\"\n6\t\"0\"\n7\t\"1\"\n2\t\"0\"\n3\t\"1\"\n0\t\"0\"\n1\t\"1\"\n".to_string(); +let actual = execute( ctx, sql).join("\n"); +let expected = + "4\t\"0\"\n5\t\"1\"\n6\t\"0\"\n7\t\"1\"\n2\t\"0\"\n3\t\"1\"\n0\t\"0\"\n1\t\"1\"" +.to_string(); assert_eq!(expected, actual); } @@ -114,8 +116,8 @@ fn csv_count_star() { let mut ctx = ExecutionContext::new(); register_aggregate_csv( ctx); let sql = "SELECT COUNT(*), COUNT(1), COUNT(c1) FROM aggregate_test_100"; -let actual = execute( ctx, sql); -let expected = "100\t100\t100\n".to_string(); +let actual = execute( ctx, sql).join("\n"); +let expected = "100\t100\t100".to_string(); assert_eq!(expected, actual); } @@ -124,8 +126,8 @@ fn csv_query_with_predicate() { let mut ctx = ExecutionContext::new(); register_aggregate_csv( ctx); let sql = "SELECT c1, c12 FROM aggregate_test_100 WHERE c12 > 0.376 AND c12 < 0.4"; -let actual = execute( ctx, sql); -let expected = "\"e\"\t0.39144436569161134\n\"d\"\t0.38870280983958583\n".to_string(); +let actual = execute( ctx, sql).join("\n"); +let expected = "\"e\"\t0.39144436569161134\n\"d\"\t0.38870280983958583".to_string(); assert_eq!(expected, actual); } @@ -133,40 +135,39 @@ fn csv_query_with_predicate() { fn csv_query_group_by_int_min_max() { let mut ctx = ExecutionContext::new(); register_aggregate_csv( ctx); -//TODO add ORDER BY once supported, to make this test determistic let sql = "SELECT c2, MIN(c12), MAX(c12) FROM aggregate_test_100 GROUP BY c2"; -let actual = execute( ctx, sql); -let expected = "4\t0.02182578039211991\t0.9237877978193884\n5\t0.0147930530301\t0.9723580396501548\n2\t0.16301110515739792\t0.991517828651004\n3\t0.047343434291126085\t0.9293883502480845\n1\t0.05636955101974106\t0.9965400387585364\n".to_string(); -assert_eq!(expected, actual); +let mut actual = execute( ctx, sql); +actual.sort(); +let expected = "1\t0.05636955101974106\t0.9965400387585364\n2\t0.16301110515739792\t0.991517828651004\n3\t0.047343434291126085\t0.9293883502480845\n4\t0.02182578039211991\t0.9237877978193884\n5\t0.0147930530301\t0.9723580396501548".to_string(); +assert_eq!(expected, actual.join("\n")); } #[test] fn csv_query_avg() { let mut ctx = ExecutionContext::new(); register_aggregate_csv( ctx); -//TODO add ORDER BY once supported, to make this test determistic let sql = "SELECT avg(c12) FROM aggregate_test_100"; -let actual = execute( ctx, sql); -let expected = "0.5089725099127211\n".to_string(); -assert_eq!(expected, actual); +let mut actual = execute( ctx, sql); +actual.sort(); +let expected = "0.5089725099127211".to_string(); +assert_eq!(expected, actual.join("\n")); } #[test] fn csv_query_group_by_avg() { let mut ctx = ExecutionContext::new(); register_aggregate_csv( ctx); -//TODO add ORDER BY once supported, to make this test determistic let sql = "SELECT c1, avg(c12) FROM aggregate_test_100 GROUP BY c1"; -let actual = execute( ctx, sql); -let expected =
[arrow] branch master updated (1fc1015 -> b1d5d0d)
This is an automated email from the ASF dual-hosted git repository. agrove pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/arrow.git. from 1fc1015 ARROW-6859: [CI][Nightly] Disable docker layer caching for CircleCI tasks add b1d5d0d ARROW-6690: [Rust] [DataFusion] Optimize aggregates without GROUP BY to use SIMD No new revisions were added by this update. Summary of changes: .../src/execution/physical_plan/common.rs | 90 ++- .../src/execution/physical_plan/expressions.rs | 652 + .../src/execution/physical_plan/hash_aggregate.rs | 25 +- rust/datafusion/src/execution/physical_plan/mod.rs | 11 +- 4 files changed, 521 insertions(+), 257 deletions(-)
[arrow] branch master updated (8621a5c -> 1fc1015)
This is an automated email from the ASF dual-hosted git repository. npr pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/arrow.git. from 8621a5c ARROW-6464: [Java] Refactor FixedSizeListVector#splitAndTransfer with slice API (#5293) add 1fc1015 ARROW-6859: [CI][Nightly] Disable docker layer caching for CircleCI tasks No new revisions were added by this update. Summary of changes: dev/tasks/docker-tests/circle.linux.yml | 1 - 1 file changed, 1 deletion(-)
[arrow] branch master updated (ad85b11 -> 8621a5c)
This is an automated email from the ASF dual-hosted git repository. siddteotia pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/arrow.git. from ad85b11 ARROW-6661: [Java] Implement APIs like slice to enhance VectorSchemaRoot (#5470) add 8621a5c ARROW-6464: [Java] Refactor FixedSizeListVector#splitAndTransfer with slice API (#5293) No new revisions were added by this update. Summary of changes: .../src/main/codegen/templates/UnionVector.java| 1 + .../apache/arrow/vector/BaseFixedWidthVector.java | 2 +- .../arrow/vector/complex/FixedSizeListVector.java | 69 +- .../apache/arrow/vector/complex/ListVector.java| 2 +- .../apache/arrow/vector/complex/StructVector.java | 3 +- .../arrow/vector/TestFixedSizeListVector.java | 31 ++ 6 files changed, 102 insertions(+), 6 deletions(-)
[arrow] branch master updated (b9203a9 -> ad85b11)
This is an automated email from the ASF dual-hosted git repository. siddteotia pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/arrow.git. from b9203a9 ARROW-6074: [FlightRPC][Java] Middleware add ad85b11 ARROW-6661: [Java] Implement APIs like slice to enhance VectorSchemaRoot (#5470) No new revisions were added by this update. Summary of changes: .../org/apache/arrow/vector/VectorSchemaRoot.java | 91 +- .../apache/arrow/vector/TestVectorSchemaRoot.java | 85 2 files changed, 174 insertions(+), 2 deletions(-)
[arrow] branch master updated (c8bcd70 -> b9203a9)
This is an automated email from the ASF dual-hosted git repository. emkornfield pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/arrow.git. from c8bcd70 ARROW-6732: [Java] Implement quick sort in a non-recursive way to avoid stack overflow add b9203a9 ARROW-6074: [FlightRPC][Java] Middleware No new revisions were added by this update. Summary of changes: .../org/apache/arrow/flight/FlightGrpcUtils.java | 3 +- ...lightRuntimeException.java => CallHeaders.java} | 39 ++- .../flight/{FlightConstants.java => CallInfo.java} | 12 +- .../java/org/apache/arrow/flight/CallStatus.java | 9 + .../java/org/apache/arrow/flight/FlightClient.java | 34 +- .../arrow/flight/FlightClientMiddleware.java | 52 +++ .../java/org/apache/arrow/flight/FlightMethod.java | 61 .../org/apache/arrow/flight/FlightProducer.java| 12 + .../arrow/flight/FlightRuntimeException.java | 6 + .../java/org/apache/arrow/flight/FlightServer.java | 33 +- .../arrow/flight/FlightServerMiddleware.java | 99 ++ .../org/apache/arrow/flight/FlightService.java | 147 ++--- .../org/apache/arrow/flight/FlightStatusCode.java | 4 + .../java/org/apache/arrow/flight/FlightStream.java | 9 + .../java/org/apache/arrow/flight/StreamPipe.java | 30 +- .../flight/grpc/ClientInterceptorAdapter.java | 149 + .../grpc/ContextPropagatingExecutorService.java| 117 +++ .../apache/arrow/flight/grpc/MetadataAdapter.java | 72 + .../flight/grpc/ServerInterceptorAdapter.java | 142 .../org/apache/arrow/flight/grpc/StatusUtils.java | 4 + .../apache/arrow/flight/TestClientMiddleware.java | 211 .../apache/arrow/flight/TestServerMiddleware.java | 360 + .../org/apache/arrow/flight/perf/TestPerf.java | 7 +- 23 files changed, 1538 insertions(+), 74 deletions(-) copy java/flight/src/main/java/org/apache/arrow/flight/{FlightRuntimeException.java => CallHeaders.java} (56%) copy java/flight/src/main/java/org/apache/arrow/flight/{FlightConstants.java => CallInfo.java} (76%) create mode 100644 java/flight/src/main/java/org/apache/arrow/flight/FlightClientMiddleware.java create mode 100644 java/flight/src/main/java/org/apache/arrow/flight/FlightMethod.java create mode 100644 java/flight/src/main/java/org/apache/arrow/flight/FlightServerMiddleware.java create mode 100644 java/flight/src/main/java/org/apache/arrow/flight/grpc/ClientInterceptorAdapter.java create mode 100644 java/flight/src/main/java/org/apache/arrow/flight/grpc/ContextPropagatingExecutorService.java create mode 100644 java/flight/src/main/java/org/apache/arrow/flight/grpc/MetadataAdapter.java create mode 100644 java/flight/src/main/java/org/apache/arrow/flight/grpc/ServerInterceptorAdapter.java create mode 100644 java/flight/src/test/java/org/apache/arrow/flight/TestClientMiddleware.java create mode 100644 java/flight/src/test/java/org/apache/arrow/flight/TestServerMiddleware.java