vibhatha commented on code in PR #13401:
URL: https://github.com/apache/arrow/pull/13401#discussion_r918450190
##########
cpp/src/arrow/engine/substrait/serde_test.cc:
##########
@@ -1383,5 +1400,188 @@ TEST(Substrait, JoinPlanInvalidKeys) {
}
}
+TEST(Substrait, SerializeRelation) {
+#ifdef _WIN32
+ GTEST_SKIP() << "ARROW-16392: Substrait File URI not supported for Windows";
+#else
+ ExtensionSet ext_set;
+ auto dummy_schema = schema({field("foo", binary())});
+ // creating a dummy dataset using a dummy table
+ auto format = std::make_shared<arrow::dataset::ParquetFileFormat>();
+ auto filesystem = std::make_shared<fs::LocalFileSystem>();
+
+ std::vector<fs::FileInfo> files;
+ const std::vector<std::string> f_paths = {"/tmp/data1.parquet",
"/tmp/data2.parquet"};
+
+ for (const auto& f_path : f_paths) {
+ ASSERT_OK_AND_ASSIGN(auto f_file, filesystem->GetFileInfo(f_path));
+ files.push_back(std::move(f_file));
+ }
+
+ ASSERT_OK_AND_ASSIGN(auto ds_factory,
dataset::FileSystemDatasetFactory::Make(
+ std::move(filesystem),
std::move(files),
+ std::move(format), {}));
+ ASSERT_OK_AND_ASSIGN(auto dataset, ds_factory->Finish(dummy_schema));
+
+ auto options = std::make_shared<dataset::ScanOptions>();
+ options->projection = compute::project({}, {});
+ auto scan_node_options = dataset::ScanNodeOptions{dataset, options};
+
+ auto scan_declaration = compute::Declaration({"scan", scan_node_options});
+
+ ASSERT_OK_AND_ASSIGN(auto serialized_rel,
+ SerializeRelation(scan_declaration, &ext_set));
+ ASSERT_OK_AND_ASSIGN(auto deserialized_decl,
+ DeserializeRelation(*serialized_rel, ext_set));
+
+ auto dataset_comparator = [](std::shared_ptr<dataset::Dataset> ds_lhs,
+ std::shared_ptr<dataset::Dataset> ds_rhs) ->
bool {
+ const auto& fsd_lhs = checked_cast<const
dataset::FileSystemDataset&>(*ds_lhs);
+ const auto& fsd_rhs = checked_cast<const
dataset::FileSystemDataset&>(*ds_lhs);
+ const auto& files_lhs = fsd_lhs.files();
+ const auto& files_rhs = fsd_rhs.files();
+
+ if (files_lhs.size() != files_rhs.size()) {
+ return false;
+ }
+ uint64_t fidx = 0;
+ for (const auto& l_file : files_lhs) {
+ if (l_file != files_rhs[fidx]) {
+ return false;
+ }
+ fidx++;
+ }
+ bool cmp_file_format = fsd_lhs.format()->Equals(*fsd_lhs.format());
+ bool cmp_file_system = fsd_lhs.filesystem()->Equals(fsd_rhs.filesystem());
+ return cmp_file_format && cmp_file_system;
+ };
+
+ auto scan_option_comparator = [dataset_comparator](
+ const dataset::ScanNodeOptions& lhs,
+ const dataset::ScanNodeOptions& rhs) ->
bool {
+ bool cmp_rso = lhs.require_sequenced_output ==
rhs.require_sequenced_output;
+ bool cmp_ds = dataset_comparator(lhs.dataset, rhs.dataset);
+ return cmp_rso && cmp_ds;
+ };
+
+ EXPECT_EQ(deserialized_decl.factory_name, scan_declaration.factory_name);
+ const auto& lhs =
+ checked_cast<const
dataset::ScanNodeOptions&>(*deserialized_decl.options);
+ const auto& rhs =
+ checked_cast<const dataset::ScanNodeOptions&>(*scan_declaration.options);
+ ASSERT_TRUE(scan_option_comparator(lhs, rhs));
+#endif
+}
+
+TEST(Substrait, SerializeRelationEndToEnd) {
+#ifdef _WIN32
+ GTEST_SKIP() << "ARROW-16392: Substrait File URI not supported for Windows";
+#else
+ ExtensionSet ext_set;
+ compute::ExecContext exec_context;
+
+ auto dummy_schema = schema({field("a", int32()), field("b", int32())});
+ auto table = TableFromJSON(dummy_schema, {R"([
+ [1, 1],
+ [3, 4]
+ ])",
+ R"([
+ [0, 2],
+ [1, 3],
+ [4, 1],
+ [3, 1],
+ [1, 2]
+ ])",
+ R"([
+ [2, 2],
+ [5, 3],
+ [1, 3]
+ ])"});
+ // Note:: adding `/` to support the currently supported file read format
from file
+ // system for Substrait
Review Comment:
Ah instead of just saying 'test.parquet` we say `/test.parquet`
--
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: github-unsubscr...@arrow.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
