mapleFU commented on code in PR #38466:
URL: https://github.com/apache/arrow/pull/38466#discussion_r1392864197
##########
cpp/src/arrow/dataset/file_parquet_test.cc:
##########
@@ -834,5 +834,72 @@ TEST(TestParquetStatistics, NullMax) {
EXPECT_EQ(stat_expression->ToString(), "(x >= 1)");
}
+class DelayedBufferReader : public ::arrow::io::BufferReader {
+ public:
+ explicit DelayedBufferReader(const std::shared_ptr<::arrow::Buffer>& buffer)
+ : ::arrow::io::BufferReader(buffer) {}
+
+ ::arrow::Future<std::shared_ptr<Buffer>> ReadAsync(
+ const ::arrow::io::IOContext& io_context, int64_t position,
+ int64_t nbytes) override {
+ read_async_count.fetch_add(1);
+ auto self =
std::dynamic_pointer_cast<DelayedBufferReader>(shared_from_this());
+ return DeferNotOk(::arrow::io::internal::SubmitIO(
+ io_context, [self, position, nbytes]() ->
Result<std::shared_ptr<Buffer>> {
+ std::this_thread::sleep_for(std::chrono::seconds(1));
+ return self->DoReadAt(position, nbytes);
+ }));
+ }
+
+ std::atomic<int> read_async_count{0};
+};
+
+TEST_F(TestParquetFileFormat, MultithreadedScanUnsafe) {
+ auto reader = MakeGeneratedRecordBatch(schema({field("utf8", utf8())}),
10000, 100);
+
+ ASSERT_OK_AND_ASSIGN(auto buffer, ParquetFormatHelper::Write(reader.get()));
+
+ std::vector<Future<>> completes;
+ std::vector<std::shared_ptr<arrow::internal::ThreadPool>> pools;
+
+ for (int idx = 0; idx < 2; ++idx) {
Review Comment:
using 2 to make the state machine a bit complex, also make we use another
Executor
##########
cpp/src/arrow/dataset/file_parquet_test.cc:
##########
@@ -834,5 +834,72 @@ TEST(TestParquetStatistics, NullMax) {
EXPECT_EQ(stat_expression->ToString(), "(x >= 1)");
}
+class DelayedBufferReader : public ::arrow::io::BufferReader {
+ public:
+ explicit DelayedBufferReader(const std::shared_ptr<::arrow::Buffer>& buffer)
+ : ::arrow::io::BufferReader(buffer) {}
+
+ ::arrow::Future<std::shared_ptr<Buffer>> ReadAsync(
+ const ::arrow::io::IOContext& io_context, int64_t position,
+ int64_t nbytes) override {
+ read_async_count.fetch_add(1);
+ auto self =
std::dynamic_pointer_cast<DelayedBufferReader>(shared_from_this());
+ return DeferNotOk(::arrow::io::internal::SubmitIO(
+ io_context, [self, position, nbytes]() ->
Result<std::shared_ptr<Buffer>> {
+ std::this_thread::sleep_for(std::chrono::seconds(1));
+ return self->DoReadAt(position, nbytes);
+ }));
+ }
+
+ std::atomic<int> read_async_count{0};
+};
+
+TEST_F(TestParquetFileFormat, MultithreadedScanUnsafe) {
+ auto reader = MakeGeneratedRecordBatch(schema({field("utf8", utf8())}),
10000, 100);
+
+ ASSERT_OK_AND_ASSIGN(auto buffer, ParquetFormatHelper::Write(reader.get()));
+
+ std::vector<Future<>> completes;
+ std::vector<std::shared_ptr<arrow::internal::ThreadPool>> pools;
+
+ for (int idx = 0; idx < 2; ++idx) {
+ auto buffer_reader = std::make_shared<DelayedBufferReader>(buffer);
+ auto source = std::make_shared<FileSource>(buffer_reader, buffer->size());
+ auto fragment = MakeFragment(*source);
+ std::shared_ptr<Scanner> scanner;
+
+ {
+ auto options = std::make_shared<ScanOptions>();
+ ASSERT_OK_AND_ASSIGN(auto thread_pool,
arrow::internal::ThreadPool::Make(1));
+ pools.emplace_back(thread_pool);
+ options->io_context =
+ ::arrow::io::IOContext(::arrow::default_memory_pool(),
pools.back().get());
+ auto fragment_scan_options =
std::make_shared<ParquetFragmentScanOptions>();
+ fragment_scan_options->arrow_reader_properties->set_pre_buffer(true);
+
+ options->fragment_scan_options = fragment_scan_options;
+ options->use_threads = true;
+ ScannerBuilder builder(ArithmeticDatasetFixture::schema(), fragment,
options);
+
+ ASSERT_OK(builder.UseThreads(true));
Review Comment:
Thread is neccessary for testing
##########
cpp/src/arrow/dataset/file_parquet_test.cc:
##########
@@ -834,5 +834,72 @@ TEST(TestParquetStatistics, NullMax) {
EXPECT_EQ(stat_expression->ToString(), "(x >= 1)");
}
+class DelayedBufferReader : public ::arrow::io::BufferReader {
+ public:
+ explicit DelayedBufferReader(const std::shared_ptr<::arrow::Buffer>& buffer)
+ : ::arrow::io::BufferReader(buffer) {}
+
+ ::arrow::Future<std::shared_ptr<Buffer>> ReadAsync(
+ const ::arrow::io::IOContext& io_context, int64_t position,
+ int64_t nbytes) override {
+ read_async_count.fetch_add(1);
+ auto self =
std::dynamic_pointer_cast<DelayedBufferReader>(shared_from_this());
+ return DeferNotOk(::arrow::io::internal::SubmitIO(
+ io_context, [self, position, nbytes]() ->
Result<std::shared_ptr<Buffer>> {
+ std::this_thread::sleep_for(std::chrono::seconds(1));
+ return self->DoReadAt(position, nbytes);
+ }));
+ }
+
+ std::atomic<int> read_async_count{0};
+};
+
+TEST_F(TestParquetFileFormat, MultithreadedScanUnsafe) {
+ auto reader = MakeGeneratedRecordBatch(schema({field("utf8", utf8())}),
10000, 100);
+
+ ASSERT_OK_AND_ASSIGN(auto buffer, ParquetFormatHelper::Write(reader.get()));
+
+ std::vector<Future<>> completes;
+ std::vector<std::shared_ptr<arrow::internal::ThreadPool>> pools;
+
+ for (int idx = 0; idx < 2; ++idx) {
+ auto buffer_reader = std::make_shared<DelayedBufferReader>(buffer);
+ auto source = std::make_shared<FileSource>(buffer_reader, buffer->size());
+ auto fragment = MakeFragment(*source);
+ std::shared_ptr<Scanner> scanner;
+
+ {
+ auto options = std::make_shared<ScanOptions>();
+ ASSERT_OK_AND_ASSIGN(auto thread_pool,
arrow::internal::ThreadPool::Make(1));
+ pools.emplace_back(thread_pool);
Review Comment:
pool is for case mentioned in the description
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
