mathyingzhou commented on a change in pull request #8648:
URL: https://github.com/apache/arrow/pull/8648#discussion_r563395657
##########
File path: cpp/src/arrow/adapters/orc/adapter_test.cc
##########
@@ -58,6 +73,81 @@ class MemoryOutputStream : public liborc::OutputStream {
uint64_t length_, natural_write_size_;
};
+class ORCMemWriter {
+ public:
+ Status Open(const std::shared_ptr<Schema>& schema,
+ ORC_UNIQUE_PTR<liborc::OutputStream>& outStream) {
+ orc_options_ = std::unique_ptr<liborc::WriterOptions>(new
liborc::WriterOptions());
+ outStream_ = std::move(outStream);
+ ARROW_EXPECT_OK(adapters::orc::GetORCType(*schema, &orcSchema_));
+ try {
+ writer_ = createWriter(*orcSchema_, outStream_.get(), *orc_options_);
+ } catch (const liborc::ParseError& e) {
+ return Status::IOError(e.what());
+ }
+ schema_ = schema;
+ num_cols_ = schema->num_fields();
+ return Status::OK();
+ }
+
+ Status Write(const std::shared_ptr<Table> table) {
+ int64_t numRows = table->num_rows();
+ int64_t batch_size = 1024; // Doesn't matter what it is
+ std::vector<int64_t> arrowIndexOffset(num_cols_, 0);
+ std::vector<int> arrowChunkOffset(num_cols_, 0);
+ ORC_UNIQUE_PTR<liborc::ColumnVectorBatch> batch =
writer_->createRowBatch(batch_size);
+ liborc::StructVectorBatch* root =
+ internal::checked_cast<liborc::StructVectorBatch*>(batch.get());
+ std::vector<liborc::ColumnVectorBatch*> fields = root->fields;
+ while (numRows > 0) {
+ for (int i = 0; i < num_cols_; i++) {
+ ARROW_EXPECT_OK(adapters::orc::FillBatch(
+ schema_->field(i)->type().get(), fields[i], arrowIndexOffset[i],
+ arrowChunkOffset[i], batch_size, table->column(i).get()));
+ }
+ root->numElements = fields[0]->numElements;
+ writer_->add(*batch);
+ batch->clear();
+ numRows -= batch_size;
+ }
+ writer_->close();
+ return Status::OK();
+ }
+
+ liborc::OutputStream* ReleaseOutStream() { return outStream_.release(); }
+
+ ORC_UNIQUE_PTR<liborc::Writer> writer_;
+ std::unique_ptr<liborc::WriterOptions> orc_options_;
+ std::shared_ptr<Schema> schema_;
+ ORC_UNIQUE_PTR<liborc::OutputStream> outStream_;
+ ORC_UNIQUE_PTR<liborc::Type> orcSchema_;
+ int num_cols_;
+};
+
+bool tableWriteReadEqual(const std::shared_ptr<Table>& tableIn,
+ const std::shared_ptr<Table>& predictedTableOut,
Review comment:
Fixed!
----------------------------------------------------------------
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
