Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

sql: pull type resolution code in more specific packages #49484

Closed
rohany opened this issue May 23, 2020 · 0 comments · Fixed by #51621
Closed

sql: pull type resolution code in more specific packages #49484

rohany opened this issue May 23, 2020 · 0 comments · Fixed by #51621
Labels
C-cleanup Tech debt, refactors, loose ends, etc. Solution not expected to significantly change behavior.

Comments

@rohany
Copy link
Contributor

rohany commented May 23, 2020

Now that #49354 has gone in, the type resolution code (and better control around when descriptors have resolved types or not) should be pushed into more specific packages.

@rohany rohany added the C-cleanup Tech debt, refactors, loose ends, etc. Solution not expected to significantly change behavior. label May 23, 2020
@rohany rohany mentioned this issue May 27, 2020
35 tasks
rohany added a commit to rohany/cockroach that referenced this issue Jul 20, 2020
Fixes cockroachdb#49484.
Preparation for cockroachdb#51385.

Up until now, the `planner` was responsible for installing user defined
type metadata in tables that contained user defined types. This was
slightly messy and caused leakage of responsibility regarding when
descriptors had user defined types vs when they didn't. This commit
pushes that responsibility into the `descs.Collection`. It also paves
the way for work to avoid copying `ImmutableTableDescriptor`s that
contain user defined types every time that they need hydration.

Release note: None
rohany added a commit to rohany/cockroach that referenced this issue Jul 22, 2020
Fixes cockroachdb#49484.
Preparation for cockroachdb#51385.

Up until now, the planner was responsible for installing user defined
type metadata in tables that contained user defined types. This was
slightly messy and caused leakage of responsibility regarding when
descriptors had user defined types vs when they didn't. This commit
pushes that responsibility into the descs.Collection. It also paves
the way for work to avoid copying ImmutableTableDescriptors that
contain user defined types every time that they need hydration.

Release note: None
craig bot pushed a commit that referenced this issue Jul 23, 2020
51621: descs: push descriptor type hydration into the desc.Collection r=rohany a=rohany

Fixes #49484.
Preparation for #51385.

Up until now, the `planner` was responsible for installing user defined
type metadata in tables that contained user defined types. This was
slightly messy and caused leakage of responsibility regarding when
descriptors had user defined types vs when they didn't. This commit
pushes that responsibility into the `descs.Collection`. It also paves
the way for work to avoid copying `ImmutableTableDescriptor`s that
contain user defined types every time that they need hydration.

Release note: None

51828: sql,stats: fix flakes in TestCacheWait and TestQueryCache r=jordanlewis a=rytaft

**stats: fix flake in TestCacheWait**

This commit fixes a flake in `TestCacheWait`, which was introduced
by #51616. That PR changed the call to `InvalidateTableStats` in
`TestCacheWait` to `RefreshTableStats`, but that change should not have
been made. The purpose of the test is to test that the cache invalidation
and waiting mechanisms work correctly, and therefore it must call
`InvalidateTableStats`, not `RefreshTableStats`.

Fixes #51712

**sql: fix flake in TestQueryCache/group/statschange**

This commit fixes a flake in TestQueryCache/group/statschange,
which was introduced by #51616. That PR made updates to the stats
cache asynchronous, so we can no longer expect the query cache to
be invalidated immediately after a stats update. This commit fixes
the problem by introducing a retry mechanism into the test.

Fixes #51693

51837: sql: don't show the "hidden" column flag in EXPLAIN r=RaduBerinde a=RaduBerinde

The ResultColumns.Hidden flag is used internally; it does not carry
useful information for users and should not be visible in EXPLAIN.

Release note (sql change): EXPLAIN no longer shows the "hidden"
annotation for columns.

Co-authored-by: Rohan Yadav <[email protected]>
Co-authored-by: Rebecca Taft <[email protected]>
Co-authored-by: Radu Berinde <[email protected]>
@craig craig bot closed this as completed in eaeb7c1 Jul 23, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
C-cleanup Tech debt, refactors, loose ends, etc. Solution not expected to significantly change behavior.
Projects
None yet
Development

Successfully merging a pull request may close this issue.

1 participant