* run release code tests separately

* remove extra install from frontend unit tests

* use the right dir

* fix dependency issues

[Fixes #45572]

* Consolidate query analysis code

* Tighten up more

* Missed a spot

* More encapsulation

* Remove blank line

* Remove default Count aggregation without triggering a query update

* Fix flaky test

"Admins can't create queries on Metabase Analytics models"

* Create typed dashboard reducers

* fix reference

* fix export

* Move more reducers

* move parameterValues

* fixup! move parameterValues

* rtkCreateAction -> createAction

* parameterized static questions

* sdk parameter values

* use parameters to construct parameter input

* add a unit test for passing the correct parameters

* fix parameter parsing

* move TEST_PARAM to top of file

* update readme docs

* address code review

* refactor: rework QueryModals to a functional component

* destructure props

* Pass missing type to clean card in url

* Add reproduction for #34517

* Add repro for #42949

* Add more repros for #42949

* Add repro for the frontend part of metabase#23857

* Avoid cy.visit

* add skip_graph to coll perm PUT

- also no longer ignores revision and groups
  so we don't print this warning:
  2024-07-11 16:17:32,293 WARN api.common :: Unexpected parameters at [:put "/api/collection/graph"]: [:revision :groups]

* namespace should be a keyword

* decode the keyword

* add a typehint to avoid reflection

* Adds test for skip_graph

* handle revision 0 

* Do not reset `table.columns` when some column is deleted

* Do not reset `table.columns` when some column is deleted

* Fix

* Fix

* Make API calls more obvious

Add FE repro for the visual glitch when working with models that have an expression with the same name as the column (#45503)

* Add FE repro for #35840

* Add FE repro for #35840

* do not allow to render empty SaveQuestionModal content

when question is already saved and QB rerenders still open SaveQuestionModal there is a moment when content is empty before modal is closed

* fixup! do not allow to render empty SaveQuestionModal content

* parallelize coll perm graph group lookup w/ core.async

* respond to code review

* add + uses ExecutorCompletionService map function

* rename test

* more tests

* no concurrency for cljs

* fix timeout + add more tests

* use claypoole

* cleanup

* add kondo hooks for claypoole

* cleanup

* Can't use libraries that don't have a license

Revert "add kondo hooks for claypoole"

This reverts commit 9d93b55b28d69267b65260f2b046420d67604361.

* ignore unused value

-- TODO: check back in on claypoole to see if the kondo config gets merged

* fix linter

* refactor and improve milestone check code

* whitespace

* fix cross-version flakes

* more flake fix

* fix slack message

* tired_face

* ref name is what I want already

* Fix fields in joins

* Fix deselecting the last join column

* Update frontend/src/metabase/query_builder/components/notebook/steps/JoinStep/JoinStep.unit.spec.tsx

Co-authored-by: Kamil Mielnik <kamil@kamilmielnik.com>

* Fix reference

---------

Co-authored-by: Kamil Mielnik <kamil@kamilmielnik.com>

This was a huge footgun.

We were accidentally limiting *ALL* queries run in our transducer to
1000 results. This included the query to fetch permissions, so when we
checked permissions we were checking a small percentage of your *actual*
permissions (e.g. 1000 / ~6300 for me on stats).

* Preserve type namespace of base-type in legacy-expression-for-expression-clause

Fixes #44584.

This comes in two parts.

**Rebuilt query logic**

I rewrote the query exec logic to use one `future` and one `a/go` block,
reporting results to a `promise`. This is easier to follow and doesn't leak
threads like the original implementation with two `future`s.

The single future is guaranteed to exit. It blocks on the BigQuery call.
That will either complete or throw an exception. Either the exception or
the BigQuery result is sent to the promise, and then the future is
complete. The only reference to that future is overwritten after the `promise`
resolves, so the `future`, its closure, and its thread don't persist longer than needed.

**Return a reducible result**

Previously the results were returned as a lazy sequence, but that brings
a lot of `seq` overhead, and is a risk of retaining the entire result
set in memory until the query is all done. Above all, it doesn't support
`reduced?` to explicitly drop any further results.

The `thunk` function style of `qp.reducible/reducible-rows` doesn't fit
neatly here, so I implemented the reducible logic directly on top of the
`Iterable` BigQuery results.

**Verifying**

This is a bit tricky. I've been testing it like this:

- Add a BigQuery database to my instance
- Create a saved question and note its ID (say, 123)
- Create an API key in (EE) Metabase and save it in an env var
- Install the `siege` load testing tool
- Install VisualVM and monitor my Clojure REPL process's thread count
- `siege --concurrent=15 --time=30s --header="x-api-key: $API_KEY" "http://localhost:3000/api/card/123/query POST"`

That will create a bunch of new threads while it's getting called, but
within a minute of the last queries they'll gradually drop out of the
thread pools and it will return to the baseline number of threads.