Support table handles with columns in JDBC applyProjection #14770
Conversation
This file contains 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
findepi
approved these changes
Oct 26, 2022
ssheikin
reviewed
Oct 27, 2022
Comment on lines
265
to
269
| Set<JdbcColumnHandle> newColumnSet = newColumns.stream() | ||
| // It may happen fresh table handle comes with a columns prepared already. | ||
| // In such case it may happen that applyProjection may want to add UPDATE_ROW_ID id, which is created later during the planning. | ||
| .filter(column -> !column.getColumnName().equals(UPDATE_ROW_ID)) | ||
| .collect(toImmutableSet()); |
There was a problem hiding this comment.
Move your changes just before verify.
Why?
2022-10-26T13:13:38.567-0500 ERROR pool-3-thread-1 io.trino.testng.services.ProgressLoggingListener [TEST FAILURE] io.trino.faulttolerant.postgresql.TestPostgresTaskFailureRecoveryTest.testDelete; (took: 3 minutes and 1 seconds)
io.trino.testing.QueryFailedException: The optimizer exhausted the time limit of 180000 ms: Top rules: {
io.trino.sql.planner.iterative.rule.PushProjectionIntoTableScan@3ed486ed: 51209 ms, 2133950 invocations, 2133950 applications,
io.trino.sql.planner.iterative.rule.PruneTableScanColumns@54c3ba96: 5112 ms, 2133950 invocations, 0 applications,
io.trino.sql.planner.iterative.rule.PushProjectionThroughUnion@16d94047: 2772 ms, 2133950 invocations, 0 applications,
io.trino.sql.planner.iterative.rule.PruneAggregationColumns@53236a7d: 2770 ms, 2133950 invocations, 0 applications,
io.trino.sql.planner.iterative.rule.PushDownDereferenceThroughJoin@c0241fd: 2327 ms, 2133950 invocations, 0 applications }
at io.trino.testing.AbstractTestingTrinoClient.execute(AbstractTestingTrinoClient.java:122)
at io.trino.testing.DistributedQueryRunner.executeWithQueryId(DistributedQueryRunner.java:496)
at io.trino.faulttolerant.BaseFailureRecoveryTest$FailureRecoveryAssert.execute(BaseFailureRecoveryTest.java:599)
at io.trino.faulttolerant.BaseFailureRecoveryTest$FailureRecoveryAssert.executeExpected(BaseFailureRecoveryTest.java:560)
at io.trino.faulttolerant.BaseFailureRecoveryTest$FailureRecoveryAssert.isCoordinatorOnly(BaseFailureRecoveryTest.java:638)
at io.trino.faulttolerant.jdbc.BaseJdbcFailureRecoveryTest.testDelete(BaseJdbcFailureRecoveryTest.java:73)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:568)
at org.testng.internal.MethodInvocationHelper.invokeMethod(MethodInvocationHelper.java:104)
at org.testng.internal.Invoker.invokeMethod(Invoker.java:645)
at org.testng.internal.Invoker.invokeTestMethod(Invoker.java:851)
at org.testng.internal.Invoker.invokeTestMethods(Invoker.java:1177)
at org.testng.internal.TestMethodWorker.invokeTestMethods(TestMethodWorker.java:129)
at org.testng.internal.TestMethodWorker.run(TestMethodWorker.java:112)
at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
at java.base/java.lang.Thread.run(Thread.java:833)
Caused by: io.trino.spi.TrinoException: The optimizer exhausted the time limit of 180000 ms: Top rules: {
io.trino.sql.planner.iterative.rule.PushProjectionIntoTableScan@3ed486ed: 51209 ms, 2133950 invocations, 2133950 applications,
io.trino.sql.planner.iterative.rule.PruneTableScanColumns@54c3ba96: 5112 ms, 2133950 invocations, 0 applications,
io.trino.sql.planner.iterative.rule.PushProjectionThroughUnion@16d94047: 2772 ms, 2133950 invocations, 0 applications,
io.trino.sql.planner.iterative.rule.PruneAggregationColumns@53236a7d: 2770 ms, 2133950 invocations, 0 applications,
io.trino.sql.planner.iterative.rule.PushDownDereferenceThroughJoin@c0241fd: 2327 ms, 2133950 invocations, 0 applications }
at io.trino.sql.planner.iterative.IterativeOptimizer$Context.checkTimeoutNotExhausted(IterativeOptimizer.java:380)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreNode(IterativeOptimizer.java:157)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreGroup(IterativeOptimizer.java:133)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreChildren(IterativeOptimizer.java:252)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreGroup(IterativeOptimizer.java:135)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreChildren(IterativeOptimizer.java:252)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreGroup(IterativeOptimizer.java:135)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreChildren(IterativeOptimizer.java:252)
at io.trino.sql.planner.iterative.IterativeOptimizer.exploreGroup(IterativeOptimizer.java:135)
at io.trino.sql.planner.iterative.IterativeOptimizer.optimize(IterativeOptimizer.java:118)
at io.trino.sql.planner.LogicalPlanner.plan(LogicalPlanner.java:247)
at io.trino.sql.planner.LogicalPlanner.plan(LogicalPlanner.java:222)
at io.trino.sql.planner.LogicalPlanner.plan(LogicalPlanner.java:217)
at io.trino.execution.SqlQueryExecution.doPlanQuery(SqlQueryExecution.java:475)
at io.trino.execution.SqlQueryExecution.planQuery(SqlQueryExecution.java:456)
at io.trino.execution.SqlQueryExecution.start(SqlQueryExecution.java:397)
at io.trino.execution.SqlQueryManager.createQuery(SqlQueryManager.java:249)
at io.trino.dispatcher.LocalDispatchQuery.lambda$startExecution$7(LocalDispatchQuery.java:143)
at io.trino.$gen.Trino_testversion____20221026_180846_142.run(Unknown Source)
at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
at java.base/java.lang.Thread.run(Thread.java:833)
| @@ -100,6 +101,7 @@ | |||
| implements JdbcMetadata | |||
| { | |||
| private static final String SYNTHETIC_COLUMN_NAME_PREFIX = "_pfgnrtd_"; | |||
| private static final String UPDATE_ROW_ID = "_trino_artificial_column_handle_for_update_row_id_"; | |||
There was a problem hiding this comment.
I don't think so. It is that long to make sure that we don't hit a conflict.
plugin/trino-base-jdbc/src/main/java/io/trino/plugin/jdbc/DefaultJdbcMetadata.java
Outdated
Show resolved
Hide resolved
hashhar
approved these changes
Oct 27, 2022
Comment on lines
267
to
268
| // In such case it may happen that applyProjection may want to add UPDATE_ROW_ID id, which is created later during the planning. | ||
| .filter(column -> !column.getColumnName().equals(UPDATE_ROW_ID)) |
There was a problem hiding this comment.
What about other synthetic handles like DELETE_ROW_ID, MERGE_ROW_ID? We don't care yet since DefaultJdbcMetadata doesn't provide them (via getDeleteRowIdColumnHandle)?
There was a problem hiding this comment.
There is no constant name for them, because they don't have implementation.
There was a problem hiding this comment.
It is not yet implemented yet.
ssheikin
reviewed
Oct 27, 2022
| Set<JdbcColumnHandle> newColumnSet = newColumns.stream() | ||
| // It may happen fresh table handle comes with a columns prepared already. | ||
| // In such case it may happen that applyProjection may want to add UPDATE_ROW_ID id, which is created later during the planning. | ||
| .filter(column -> !column.getColumnName().equals(UPDATE_ROW_ID)) |
There was a problem hiding this comment.
UPDATE_ROW_ID could be different for different connectors. So this will not work for some, which have getDeleteRowIdColumnHandle overriden.
Please consider
Set<JdbcColumnHandle> tableSyntheticColumnSet = ImmutableSet.<JdbcColumnHandle>builder()
.addAll(tableColumnSet)
.add((JdbcColumnHandle) getDeleteRowIdColumnHandle(session, table))
.add((JdbcColumnHandle) getMergeRowIdColumnHandle(session, table))
.add((JdbcColumnHandle) getUpdateRowIdColumnHandle(session, table, ImmutableList.of())) // not sure about ImmutableList.of()
.build();
verify(tableSyntheticColumnSet.containsAll(newColumnSet), "applyProjection called with columns %s and some are not available in existing query: %s", newColumnSet, tableSyntheticColumnSet);
There was a problem hiding this comment.
and actually these will fail:
.add((JdbcColumnHandle) getMergeRowIdColumnHandle(session, table))
.add((JdbcColumnHandle) getUpdateRowIdColumnHandle(session, table, ImmutableList.of())) // not sure about ImmutableList.of()
because they throw new TrinoException(NOT_SUPPORTED
There was a problem hiding this comment.
So may be add a new public method List<String> getSyntacticColumnNames()
and .filter(column -> !getSyntacticColumnNames().contains(column.getColumnName())
There was a problem hiding this comment.
Let's not overcomplicate things
It may happen fresh table handle comes with a columns prepared already. The BaseJdbcClient#getTableHandle may return table handle with columns included. That way it is possible to skip another remote metadata call in BaseJdbcClient#getColumns. However, now typically JDBC connectors do not provide table columns in table handle by default so the verify in applyProjection was very strict and assumed that applyProjection is not going to add any new column to table handle. However it may happen that applyProjection may want to add UPDATE_ROW_ID id, which is created later during the planning for UPDATE and DELETE statements. This column is artificial, and it is used only in planning. Co-authored-by: Sasha Sheikin <myminitrue@gmail.com>
kokosing
force-pushed
the
origin/master/156_verify
branch
from
2eea471 to
814d8cf
Compare
ssheikin
approved these changes
Oct 27, 2022
|
Does this require a release note? |
|
No release notes. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Support table handles with columns in JDBC applyProjection
It may happen fresh table handle comes with a columns prepared
already. The BaseJdbcClient#getTableHandle may return table handle with
columns included. That way it is possible to skip another remote metadata call
in BaseJdbcClient#getColumns.
However, now typically JDBC connectors do not provide table columns in
table handle by default so the verify in applyProjection was very
strict and assumed that applyProjection is not going to add any new
column to table handle. However it may happen that applyProjection
may want to add UPDATE_ROW_ID id, which is created later during the
planning for UPDATE and DELETE statements. This column is artificial,
and it is used only in planning.