exec: fix explain(vec) for queries with subqueries #40511
Merged
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
|
This change is |
jordanlewis
force-pushed
the
fix-explain-vec
branch
from
87dde4a
to
df55cb6
Compare
|
Maybe we should also include postqueries in the output then? |
|
Probably. We can do that later. |
rohany
suggested changes
Sep 5, 2019
There was a problem hiding this comment.
Reviewable status:
complete! 0 of 0 LGTMs obtained (waiting on @jordanlewis and @rohany)
pkg/sql/distsqlrun/column_exec_setup.go, line 777 at r1 (raw file):
return planProjectionExpr(ctx, t.Operator, t.ResolvedType(), t.TypedLeft(), t.TypedRight(), columnTypes, input) case *tree.CastExpr: expr := t.Expr.(tree.TypedExpr)
was this just a style change or did this have something to do with the type related panic?
pkg/sql/exec/cast_tmpl.go, line 143 at r1 (raw file):
sel = sel[:n] for _, i := range sel { if vecNulls.NullAt(i) {
why do you need to check that the element is actually null here?
pkg/sql/exec/cast_tmpl.go, line 146 at r1 (raw file):
projNulls.SetNull(i) } else { panic(fmt.Errorf("Unexpected non-null at index %d", i))
this should be an execerror panic
pkg/sql/exec/cast_tmpl.go, line 154 at r1 (raw file):
projNulls.SetNull(uint16(i)) } else { panic(fmt.Errorf("Unexpected non-null at index %d", i))
ditto
Also add logic tests that show the explain(vec) plans for all of the tpch queries. Release note: None
jordanlewis
force-pushed
the
fix-explain-vec
branch
from
df55cb6
to
3ed2af0
Compare
jordanlewis
commented
Sep 5, 2019
There was a problem hiding this comment.
Reviewable status:
pkg/sql/distsqlrun/column_exec_setup.go, line 777 at r1 (raw file):
Previously, rohany (Rohan Yadav) wrote…
was this just a style change or did this have something to do with the type related panic?
I reverted it, just a style thing.
pkg/sql/exec/cast_tmpl.go, line 143 at r1 (raw file):
Previously, rohany (Rohan Yadav) wrote…
why do you need to check that the element is actually null here?
Just a safety thing. Columns that are "unknown" type must only contain nulls. If they don't, something is very wrong.
pkg/sql/exec/cast_tmpl.go, line 146 at r1 (raw file):
Previously, rohany (Rohan Yadav) wrote…
this should be an execerror panic
Done.
pkg/sql/exec/cast_tmpl.go, line 154 at r1 (raw file):
Previously, rohany (Rohan Yadav) wrote…
ditto
Done.
rohany
approved these changes
Sep 5, 2019
|
Thanks for the reviews! |
|
bors r+ |
craig bot
pushed a commit
that referenced
this pull request
Sep 5, 2019
40436: sql: more test fixes for opt-driven foreign keys r=RaduBerinde a=RaduBerinde Fixing up some expected errors in tests and making sure we don't buffer the mutation input if we fall back to the legacy path (the bufferNode is unnecessary and interferes with an interleaved delete fast path). Release note: None 40451: testcluster: don't overwrite localities indiscriminately r=andreimatei a=andreimatei Before this patch, a TestCluster would set localities for all nodes to a static values. This was overwriting any values set throught the TestClusterArgs. This patch makes it so that, if any localities are set in the args, we don't overwrite them. Release note: None 40485: sql: Fix a bug with ordinal_position in information_schema.columns r=arulajmani a=arulajmani When a column other than the last is dropped, ordinal_position in information_schema.columns virtual table no longer matches attnum from the pg_attribute table. This PR fixes this issue. Fixes #39787 Release note (bug fix): ordinal_position in information_schema.columns matches pg_attribute.attnum after a column is dropped. 40511: exec: fix explain(vec) for queries with subqueries r=jordanlewis a=jordanlewis Also add logic tests that show the explain(vec) plans for all of the tpch queries. Closes #40484. Release note: None Co-authored-by: Radu Berinde <radu@cockroachlabs.com> Co-authored-by: Andrei Matei <andrei@cockroachlabs.com> Co-authored-by: Arul Ajmani <arula@cockroachlabs.com> Co-authored-by: Jordan Lewis <jordanthelewis@gmail.com>
Build failed (retrying...) |
craig bot
pushed a commit
that referenced
this pull request
Sep 5, 2019
40485: sql: Fix a bug with ordinal_position in information_schema.columns r=arulajmani a=arulajmani When a column other than the last is dropped, ordinal_position in information_schema.columns virtual table no longer matches attnum from the pg_attribute table. This PR fixes this issue. Fixes #39787 Release note (bug fix): ordinal_position in information_schema.columns matches pg_attribute.attnum after a column is dropped. 40511: exec: fix explain(vec) for queries with subqueries r=jordanlewis a=jordanlewis Also add logic tests that show the explain(vec) plans for all of the tpch queries. Closes #40484. Release note: None 40516: sql: fix table lookup for drop index r=pbardea a=pbardea Previously, when searching for the table relevant to a particular index when dropping the index, we would fetch all object names and require that all those tables exist. However, if a table was deleted in the same transaction that table name would not be resolvable and we would error. We already had a check to see if the table being looked up was nil, but this check would not be used because the `required` flag was set to true. This PR just sets the flag to false, and looks at moves on to the next table if one of them no longer is resolvable. Addresses #38768. Release note (bug fix): Fix faulty error when trying to delete a table and an unrelated index in the same transaction. Co-authored-by: Arul Ajmani <arula@cockroachlabs.com> Co-authored-by: Jordan Lewis <jordanthelewis@gmail.com> Co-authored-by: Paul Bardea <pbardea@gmail.com>
Build succeeded |
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.
Also add logic tests that show the explain(vec) plans for all of the
tpch queries.
Closes #40484.
Release note: None