scarlin-cloudera opened a new pull request, #4442:
URL: https://github.com/apache/hive/pull/4442
This commit adds implementation for lateral views using CBO. Previously, the
CBO step was skipped.
Among the changes that needed to be done:
One highlight is that the HiveTableFunctionScan needs to know if it is a
"lateral view". A non-lateral view HiveTableFunctionScan consists of just the
udtf which was supported before this commit. In this case, the generated row
type for the HiveTableFunctionScan consists only of the UDTF fields. In
"lateral view" mode, the HiveTableFunctionScan does a join with the base table
so the row type contains the base input RelNode table fields followed by the
UDTF fields.
The ASTConverter needed to be changed to handle conversion of the lateral
view back into an AST that contains the LATERAL_VIEW node. The non-"lateral
view" HiveTableFunctionScan still generates an AST without the LATERAL_VIEW
node which is more optimal. This commit also allows for the optimization of
lateral views to create an AST without the LATERAL_VIEW node. This can be done
when the RelNode above the HiveTableFunctionScan does not need any base table
fields, only the UDTF fields.
The OpProcFactory code needed to be changed in order to continue to allow a
filter optimization. There is pushdown (ppd) logic that was being hit when
there was no CBO for lateral views. CBO runs a different path for the ppd logic
which is in the OpProcFactory. The lateral view contains 2 node paths that are
joined together. The filter can be pushed down the path that contains the base
table, but not the path that contains the UDTF. By adding this logic, it
prevents a performance regression.
A new rule was also needed for CBO. There is a
HiveFilter-HiveTableFunctionScan transpose rule. These RelNodes can be
transposed when there is at least one AND condition that can be pushed through
the HiveTableFunctionScan. A condition can be pushed through if there are no
inputRefs that rely on the output of the UDTF function. Also, we do not want to
pass through the UDTF "inline" function which has a dummy table underneath.
There are various q.out changes that are in this commit because of this
push. The first is that we want to make sure the CBO tests are run, so many
"EXPLAIN CBO" queries were added to the "q" files.
Other changes to the q.out files include:
As mentioned above, there are various places where the Lateral View Operator
is optimized out because the above operator only depends on UDTF fields. This
is verified as working since the output of the query did not change, only the
plan.
In some places, an additional "Select Operator" was added. This should not
have any performance impact.
There should be no changes in the results of any query.
<!--
Thanks for sending a pull request! Here are some tips for you:
1. If this is your first time, please read our contributor guidelines:
https://cwiki.apache.org/confluence/display/Hive/HowToContribute
2. Ensure that you have created an issue on the Hive project JIRA:
https://issues.apache.org/jira/projects/HIVE/summary
3. Ensure you have added or run the appropriate tests for your PR:
4. If the PR is unfinished, add '[WIP]' in your PR title, e.g.,
'[WIP]HIVE-XXXXX: Your PR title ...'.
5. Be sure to keep the PR description updated to reflect all changes.
6. Please write your PR title to summarize what this PR proposes.
7. If possible, provide a concise example to reproduce the issue for a
faster review.
-->
### What changes were proposed in this pull request?
<!--
Please clarify what changes you are proposing. The purpose of this section
is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster
reviews in your PR. See the examples below.
1. If you refactor some codes with changing classes, showing the class
hierarchy will help reviewers.
2. If you fix some SQL features, you can provide some references of other
DBMSes.
3. If there is design documentation, please add the link.
4. If there is a discussion in the mailing list, please add the link.
-->
### Why are the changes needed?
<!--
Please clarify why the changes are needed. For instance,
1. If you propose a new API, clarify the use case for a new API.
2. If you fix a bug, you can clarify why it is a bug.
-->
### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such as
the documentation fix.
If yes, please clarify the previous behavior and the change this PR proposes
- provide the console output, description, screenshot and/or a reproducable
example to show the behavior difference if possible.
If possible, please also clarify if this is a user-facing change compared to
the released Hive versions or within the unreleased branches such as master.
If no, write 'No'.
-->
### Is the change a dependency upgrade?
<!--
If yes, please attach a file with output from mvn dependency:tree to
validate a complete upgrade of dependency.
If no, write 'No'.
-->
### How was this patch tested?
<!--
If tests were added, say they were added here. Please make sure to add some
test cases that check the changes thoroughly including negative and positive
cases if possible.
If it was tested in a way different from regular unit tests, please clarify
how you tested step by step, ideally copy and paste-able, so that other
reviewers can test and check, and descendants can verify in the future.
If tests were not added, please describe why they were not added and/or why
it was difficult to add.
-->
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
