* Spark SQL => Honey SQL 2

* Fix stuff 

* Oracle Honey SQL 2

* Remove debugging function

* Fix concat()

* Oracle should use numbers for booleans when loading test data

* Remove driver methods deprecated in 43 and older

* Oops don't delete read-column-thunk

* Update Vertica to implement read-column-thunk instead of read-column

* Remove stray reference to syncable-schemas

---------

Co-authored-by: Ngoc Khuat <qn.khuat@gmail.com>

* Port `:vertica` driver to Honey SQL 2

* Fix 

* One last test fix 

* Support Honey SQL 2 for SQL QP

* Sort namespace

* Make Postgres/Redshift be Honey SQL 2-only

* Test fix 

* Fix busted

* Redshift test fixes 

* Sort namespaces

* Fix Postgres convert_timezone

* Fix linter config.

* Fix one last test failure

* Redshift test fix

* Test fixes and some extra validation etc.

* Update common.clj

* Update common.clj

* Update common.clj

* Enable Kondo `:warn-on-reflection` linter

* Enable warnings for metabase.api.setting-test

* Add warn-on-reflection to metabase.util.jvm

* Adding support for timezone awareness to mongo

* Remove unused ddl multimethod

* Remove mongo from the sql driver type test

* Fix tests and linters

* Add test for filtering datetimes by date in timezone

* Use test-data-with-timezones dataset as it should load in more dbs

* Exclude broken drivers, vertica especially returns no rows

* Address pr review, add timezone to isodate, include a full year of dates in the general timezone test

* Assert against each row

* Presto-jdbc year was not applying timezone

This has the same interface as `clojure.tools.logging` but supports CLJS
as well. It was formerly known as `metabase.shared.util.log`.

Added bans for print and friends. Fixed issues in kondo config where a macroexpansion put in a print by replacing with (constantly nil) as a no-op.

* Switch Audit App to Honey SQL 2

* Compile using Honey SQL 2

* Mark some more tests as ^:parallel

* Remove unused namespace

* Backport sql.qp/current-datetime-honeysql-form :mysql Honey SQL 2 support

* Add `hx/call` and `hx/raw`

* Update enterprise/backend/src/metabase_enterprise/audit_app/pages/common.clj

Co-authored-by: metamben <103100869+metamben@users.noreply.github.com>

* `#schema` => `#hawk/schema`

* Mark `metabase.async.streaming-response-test` as `^:mb/once`

---------

Co-authored-by: metamben <103100869+metamben@users.noreply.github.com>

* Bump athena driver to 2.0.35

* Include aws core for auth

Replaced, as needed, instances of str/lower|upper-case with the -en version. Also updated rules in our kondo config.

Bigquery was checking that we can set a project-id in the db details of
a bigquery database. Normally this comes from the json token:

```
export MB_BIGQUERY_CLOUD_SDK_TEST_SERVICE_ACCOUNT_JSON='{
  "type": "service_account",
  "project_id": "metabase-bigquery-driver", // <- ordinarily source of truth
  "private_key_id": "keykeykeykeykeykeykeykey",
  "private_key": "-----BEGIN PRIVATE KEY----- ...."
```

The db details normally connects to project-id
"metabase-bigquery-driver" and this test swaps out to
"bigquery-public-data" with a schema filter on chicago_taxi_trips.

to prove it works it selects from that dataset:

```
(mt/first-row
  (mt/run-mbql-query taxi_trips
    {:filter [:= [:field (mt/id :taxi_trips :unique_key) nil]
                 "67794e631648a002f88d4b7f3ab0bcb6a9ed306a"]}))
```

The test fails because this row seems to no longer be in the
results. The query isn't failing, it's just returning no rows (and
presumably because the row with this id from 2014 is no longer in the
dataset.)

I ran the same with with a limit 4, see it's working and grab one of
those rows and use the id as in the filter clause.

old: "67794e631648a002f88d4b7f3ab0bcb6a9ed306a" ("2014-01-12T00:45:00Z")
new: "ff0b96c0cada768361b1c9341e11905254644afb" ("2016-09-20T07:15:00Z")

* Simplify the calculation of mongo column types

* Remove TODO, since driver.common/class->base-type is being used

We can now filter out any ns tag for tests.

Use the vertica/vertica-ce:12.0.2-0 image.

Also point out that PKCS12 (.p12) keystores are now supported.

* move malli schema from metabase.util.schema to a new ns metabase.util.malli.schema

* revert name changes for plumatic schema. e.g: NonBlankStringPlumatic -> NonBlankString

* add clj-kondo rule to enforce consistent alias for metabase.util.malli.schema

* Switch to Honey SQL 2, Part 1

* Test fix

* Fix Kondo error

* Always return fields selected by $project in native MongoDB queries
* Keep _id at the position specified in $project if it's included there or at first position if it's not included

* Fix mongo version checking for now

* Add minor versions for comparison

* Handle nil case

* newline

* Replace _

* Use dbms-version semantic-version instead

* Compare only the major and minor version

* Fix mongo database-supports?

* Fix duplicate require

* Fix typo

* Remove expressions from version check

* Fix various mongo expressions

`trim`, `rtrim`, `ltrim` needed to be wrapped in `{"input" expr}` object
`replace` needed to be wrapped in `{"input" expr "find" ...
"replacement" ...}' object
`substring` needed to fill in the 3rd argument, optional in mbql but
required in mongo. Also to use a 1 based index

* Generalize semantic-version-gte

* Used synced dbms_version for testing feature support

* Expressions are only supported by mongo 4.2+

* Disable some tests

* Fix mongo division

Handle nulls with an upfront condition check.
Handle multiple divisors.

* Remove unused namespace

* Throw exeception if using replace on mongo < 4.4

* Skip test because of #27249

* Move minimum Mongo CI version to 4.2

* Fix sorting by expressions

$sort needs to come before $project otherwise we can only see the
projected fields, however expressions are only added in $project.

So now if a sort includes an expression, we will use $addFields to be
able to sort by that expression.

* Disable tests for expressions inside aggregations

To be addressed by #27275

* Handle aggregation nested in an expression

* Remove :truncation-start workaround

* Enable tests for expressions inside aggregations

* Fix datetime-math-tests

* Make sure dbms_version is included when fetching database for store

* Update doc for replace to indicate it should replace all occurrences

* Add mongo datetime-diff

* Fix database-supports?

* Fix order-by-test

* Handle embedded special aggregations (#27315)

* Handle embedded special aggregations
* Preserve aggregation options for nested aggregations
* Use top-level aggregation name as group name

* Disable nil punning on semantic version check (#27317)

* Remove nil punning from database-supports?

* Formatting

* Refactor based on Tamas' suggestions

* Change datetime-diff from case statement to multimethod

Co-authored-by: Case Nelson <case@metabase.com>
Co-authored-by: Tamás Benkő <tamas@metabase.com>
Co-authored-by: metamben <103100869+metamben@users.noreply.github.com>

* Fix mongo version checking for now

* Add minor versions for comparison

* Handle nil case

* newline

* Replace _

* Use dbms-version semantic-version instead

* Compare only the major and minor version

* Fix mongo database-supports?

* Fix duplicate require

* Fix typo

* Remove expressions from version check

* Fix various mongo expressions

`trim`, `rtrim`, `ltrim` needed to be wrapped in `{"input" expr}` object
`replace` needed to be wrapped in `{"input" expr "find" ...
"replacement" ...}' object
`substring` needed to fill in the 3rd argument, optional in mbql but
required in mongo. Also to use a 1 based index

* Generalize semantic-version-gte

* Used synced dbms_version for testing feature support

* Expressions are only supported by mongo 4.2+

* Disable some tests

* Fix mongo division

Handle nulls with an upfront condition check.
Handle multiple divisors.

* Remove unused namespace

* Throw exeception if using replace on mongo < 4.4

* Skip test because of #27249

* Move minimum Mongo CI version to 4.2

* Fix sorting by expressions

$sort needs to come before $project otherwise we can only see the
projected fields, however expressions are only added in $project.

So now if a sort includes an expression, we will use $addFields to be
able to sort by that expression.

* Disable tests for expressions inside aggregations

To be addressed by #27275

* Handle aggregation nested in an expression

* Remove :truncation-start workaround

* Enable tests for expressions inside aggregations

* Fix datetime-math-tests

* Make sure dbms_version is included when fetching database for store

* Update doc for replace to indicate it should replace all occurrences

* Fix order-by-test

* Handle embedded special aggregations (#27315)

* Handle embedded special aggregations
* Preserve aggregation options for nested aggregations
* Use top-level aggregation name as group name

* Disable nil punning on semantic version check (#27317)

* Optimize produced query for division when dividing by literals

* Rename var

* Use reduce to build division

* Fix reduction

* Clean up formatting and document some details

Co-authored-by: Callum Herries <hi@callumherries.com>
Co-authored-by: Tamás Benkő <tamas@metabase.com>
Co-authored-by: metamben <103100869+metamben@users.noreply.github.com>

* Add datetime-diff implementation for athena

* Don't run some tests for drivers not supporting timestamptz

* Fix for date types

* Use date-trunc helper

* Add comment about Athena's timestamp with time zone type

* Add datetime-diff-time-zones-athena-test

* Fix typo

* Fix typo

* Rewrite comment

* Tidying

* Add athena-datetime-diff-mixed-types-test

* Fix typo

* Adjust comments + formatting

* Adjust comments

* Format SQL query

* Align requires

* Add sqlite datetime-diff

* Refactor sqlite

* Exclude sqlite from datetime-diff-type-test

* Fix sqlite temporal fields losing type info on sql.qp->honeysql

* Leave the default impl for sql.qp/date

* Add oracle datetime-diff

* Remove duplicate method impls

* Remove unused require

* adds mu/defn, which behaves like s/defn but for malli schemas

- includes link to malli.io with your schema and type.
- making room for some namespaces

* require and use malli.destructure in mu/defn

* fix require alias

* clean ns

* fix linter errror

* fix linter error again

* Use gvenzl/oracle-xe docker to test Oracle
* Configure SSL connection for Oracle 21.3
* Fix secret handling

* Fix BigQuery parsing of `bigdecimal` results

* Avoid big changes to `ns` form

* Update tests