fix: Fix partition filters with timestamp value #12368
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -202,19 +202,8 @@ class TableScanTest : public virtual HiveConnectorTestBase { | |
| .planNode(); | ||
|
|
||
| std::string partitionValueStr; | ||
|
Collaborator
There was a problem hiding this comment. I'm not entirely sure why this piece needs to be removed.
Contributor
Author
There was a problem hiding this comment.
This code was introduced by https://github.com/facebookincubator/velox/pull/11754/files#diff-c62590f29a2bf22e50e276f2665e0a7666968880de73a6b4db745cbc0cfde6feL202 and is only called by |
||
| partitionValueStr = | ||
| partitionValue.has_value() ? "'" + *partitionValue + "'" : "null"; | ||
| assertQuery( | ||
| op, split, fmt::format("SELECT {}, * FROM tmp", partitionValueStr)); | ||
|
|
||
|
|
@@ -2001,48 +1990,187 @@ TEST_F(TableScanTest, partitionedTableTimestampKey) { | |
| writeToFile(filePath->getPath(), vectors); | ||
| createDuckDbTable(vectors); | ||
| const std::string partitionValue = "2023-10-27 00:12:35"; | ||
|
|
||
| auto partitionType = TIMESTAMP(); | ||
| // Test partition value is null. | ||
| testPartitionedTable(filePath->getPath(), partitionType, std::nullopt); | ||
|
|
||
| auto split = exec::test::HiveConnectorSplitBuilder(filePath->getPath()) | ||
| .partitionKey("pkey", partitionValue) | ||
| .build(); | ||
|
|
||
| ColumnHandleMap assignments = { | ||
| {"pkey", partitionKey("pkey", TIMESTAMP())}, | ||
| {"c0", regularColumn("c0", BIGINT())}, | ||
| {"c1", regularColumn("c1", DOUBLE())}}; | ||
|
|
||
| Timestamp ts = | ||
| util::fromTimestampString( | ||
| StringView(partitionValue), util::TimestampParseMode::kPrestoCast) | ||
| .thenOrThrow(folly::identity, [&](const Status& status) { | ||
| VELOX_USER_FAIL("{}", status.message()); | ||
| }); | ||
| // Read timestamp partition value as UTC. | ||
| std::string tsValue = "'" + ts.toString() + "'"; | ||
|
|
||
| Timestamp tsAsLocalTime = ts; | ||
| tsAsLocalTime.toGMT(Timestamp::defaultTimezone()); | ||
| // Read timestamp partition value as local time. | ||
| std::string tsValueAsLocal = "'" + tsAsLocalTime.toString() + "'"; | ||
|
|
||
| { | ||
| auto plan = | ||
| PlanBuilder() | ||
| .startTableScan() | ||
| .tableName("hive_table") | ||
| .outputType( | ||
| ROW({"pkey", "c0", "c1"}, {partitionType, BIGINT(), DOUBLE()})) | ||
| .assignments(assignments) | ||
| .endTableScan() | ||
| .planNode(); | ||
|
|
||
| auto expect = [&](bool asLocalTime) { | ||
| AssertQueryBuilder(plan, duckDbQueryRunner_) | ||
| .connectorSessionProperty( | ||
| kHiveConnectorId, | ||
| connector::hive::HiveConfig:: | ||
| kReadTimestampPartitionValueAsLocalTimeSession, | ||
| asLocalTime ? "true" : "false") | ||
| .splits({split}) | ||
| .assertResults(fmt::format( | ||
| "SELECT {}, * FROM tmp", asLocalTime ? tsValueAsLocal : tsValue)); | ||
| }; | ||
|
|
||
| expect(true); | ||
| expect(false); | ||
| } | ||
|
|
||
| { | ||
| auto plan = | ||
| PlanBuilder() | ||
| .startTableScan() | ||
| .tableName("hive_table") | ||
| .outputType( | ||
| ROW({"c0", "pkey", "c1"}, {BIGINT(), partitionType, DOUBLE()})) | ||
| .assignments(assignments) | ||
| .endTableScan() | ||
| .planNode(); | ||
|
|
||
| auto expect = [&](bool asLocalTime) { | ||
| AssertQueryBuilder(plan, duckDbQueryRunner_) | ||
| .connectorSessionProperty( | ||
| kHiveConnectorId, | ||
| connector::hive::HiveConfig:: | ||
| kReadTimestampPartitionValueAsLocalTimeSession, | ||
| asLocalTime ? "true" : "false") | ||
| .splits({split}) | ||
| .assertResults(fmt::format( | ||
| "SELECT c0, {}, c1 FROM tmp", | ||
| asLocalTime ? tsValueAsLocal : tsValue)); | ||
| }; | ||
| expect(true); | ||
| expect(false); | ||
| } | ||
|
|
||
| { | ||
| auto plan = | ||
| PlanBuilder() | ||
| .startTableScan() | ||
| .tableName("hive_table") | ||
| .outputType( | ||
| ROW({"c0", "c1", "pkey"}, {BIGINT(), DOUBLE(), partitionType})) | ||
| .assignments(assignments) | ||
| .endTableScan() | ||
| .planNode(); | ||
|
|
||
| auto expect = [&](bool asLocalTime) { | ||
| AssertQueryBuilder(plan, duckDbQueryRunner_) | ||
| .connectorSessionProperty( | ||
| kHiveConnectorId, | ||
| connector::hive::HiveConfig:: | ||
| kReadTimestampPartitionValueAsLocalTimeSession, | ||
| asLocalTime ? "true" : "false") | ||
| .splits({split}) | ||
| .assertResults(fmt::format( | ||
| "SELECT c0, c1, {} FROM tmp", | ||
| asLocalTime ? tsValueAsLocal : tsValue)); | ||
| }; | ||
| expect(true); | ||
| expect(false); | ||
| } | ||
|
|
||
| { | ||
| // Select only partition key. | ||
| auto plan = | ||
| PlanBuilder() | ||
| .startTableScan() | ||
| .tableName("hive_table") | ||
| .outputType(ROW({"pkey"}, {partitionType})) | ||
| .assignments({{"pkey", partitionKey("pkey", partitionType)}}) | ||
| .endTableScan() | ||
| .planNode(); | ||
|
|
||
| auto expect = [&](bool asLocalTime) { | ||
| AssertQueryBuilder(plan, duckDbQueryRunner_) | ||
| .connectorSessionProperty( | ||
| kHiveConnectorId, | ||
| connector::hive::HiveConfig:: | ||
| kReadTimestampPartitionValueAsLocalTimeSession, | ||
| asLocalTime ? "true" : "false") | ||
| .splits({split}) | ||
| .assertResults(fmt::format( | ||
| "SELECT {} FROM tmp", asLocalTime ? tsValueAsLocal : tsValue)); | ||
| }; | ||
| expect(true); | ||
| expect(false); | ||
| } | ||
|
|
||
| // Test partition filter on TIMESTAMP column. | ||
| { | ||
| auto planWithSubfilter = [&](bool asLocalTime) { | ||
| auto outputType = | ||
| ROW({"pkey", "c0", "c1"}, {TIMESTAMP(), BIGINT(), DOUBLE()}); | ||
| common::SubfieldFilters filters; | ||
| // pkey = 2023-10-27 00:12:35. | ||
| auto lower = | ||
| util::fromTimestampString( | ||
| StringView(partitionValue), util::TimestampParseMode::kPrestoCast) | ||
| .value(); | ||
| if (asLocalTime) { | ||
| lower.toGMT(Timestamp::defaultTimezone()); | ||
| } | ||
| filters[common::Subfield("pkey")] = | ||
| std::make_unique<common::TimestampRange>(lower, lower, false); | ||
| auto tableHandle = std::make_shared<HiveTableHandle>( | ||
| "test-hive", | ||
| "hive_table", | ||
| true, | ||
| std::move(filters), | ||
| nullptr, | ||
| nullptr); | ||
|
|
||
| return PlanBuilder() | ||
| .startTableScan() | ||
| .tableHandle(tableHandle) | ||
| .outputType(outputType) | ||
| .assignments(assignments) | ||
| .endTableScan() | ||
| .planNode(); | ||
| }; | ||
|
|
||
| auto expect = [&](bool asLocalTime) { | ||
| AssertQueryBuilder(planWithSubfilter(asLocalTime), duckDbQueryRunner_) | ||
| .connectorSessionProperty( | ||
| kHiveConnectorId, | ||
| connector::hive::HiveConfig:: | ||
| kReadTimestampPartitionValueAsLocalTimeSession, | ||
| asLocalTime ? "true" : "false") | ||
| .splits({split}) | ||
| .assertResults(fmt::format( | ||
| "SELECT {}, * FROM tmp", asLocalTime ? tsValueAsLocal : tsValue)); | ||
| }; | ||
| expect(true); | ||
| expect(false); | ||
| } | ||
| } | ||
|
|
||
|
|
||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It would be nice if we could add test for this change on the iceberg read.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@kecookier are you planning to address this comment?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@rui-mo @majetideepak I am not familiar with Iceberg now, and I need some time to add this test case. This may be done in another PR. Perhaps this PR can be merged first?