Support user defined function

[lianghanzhen]

I hereby agree to the terms of the CLA available at: https://databend.rs/policies/cla/

Summary

Support user defined function

• All UDFs will be stored in meta KV store like user stages.
• UDFs will be verified before stored which they must be: a) Not builtin scalar or aggregate functions; b) Params must match format @\d+ and starts with @0; c) Params much be continuous, such as @0, @1, @2; d) The order of params doesn't matter; e) Params are repeatable;
• UDFs will be transformed into the combination of builtin scalar and aggregate functions in analyze_expr phase;
• CREATE/DROP/SHOW function is supported;

Examples are available in stateless tests.

Changelog

• New Feature

Related Issues

Fixes #3440

Test Plan

Unit Tests

Stateless Tests

[databend-bot]

Thanks for the contribution!
I have applied any labels matching special text in your PR Changelog.

Please review the labels and make any necessary changes.

[databend-bot]

Thanks for the contribution!
I have applied any labels matching special text in your PR Changelog.

Please review the labels and make any necessary changes.

[vercel]

This pull request is being automatically deployed with Vercel (learn more).
To see the status of your deployment, click below or on the icon next to each commit.

🔍 Inspect: https://vercel.com/databend/databend/GQjLtYjJZivc7tVCNCN6AYzjiFym
✅ Preview: Canceled

[Deployment for 25e29b8 canceled]

[lianghanzhen]

@BohuTANG The stateless test of 05_0012_ddl_show_udf failed, but I couldn't figure out why.

Output diff:

@@ -1 +1 @@
-isnotempty     not(isnull(@0)) This is a description
+isnotempty     not(isnull(@0)) This is a description

[BohuTANG]

@BohuTANG The stateless test of 05_0012_ddl_show_udf failed, but I couldn't figure out why.

Output diff:

@@ -1 +1 @@	
-isnotempty     not(isnull(@0)) This is a description
+isnotempty     not(isnull(@0)) This is a description

isnotempty not(isnull(@0)) This is a description -- The result should be a tab not 4 blanks :)

[codecov-commenter]

Codecov Report

Merging #3655 (25e29b8) into main (4f30a6b) will increase coverage by 0%.
The diff coverage is 62%.

@@          Coverage Diff           @@
##            main   #3655    +/-   ##
======================================
  Coverage     60%     60%            
======================================
  Files        678     694    +16     
  Lines      36484   37116   +632     
======================================
+ Hits       21951   22393   +442     
- Misses     14533   14723   +190     

Impacted Files	Coverage Δ
common/planners/src/plan_node.rs	41% <0%> (-3%)	⬇️
common/planners/src/plan_rewriter.rs	43% <0%> (-1%)	⬇️
query/src/interpreters/interpreter_factory.rs	32% <0%> (-3%)	⬇️
query/src/sql/sql_statement.rs	31% <0%> (-2%)	⬇️
common/functions/src/udfs/udf_transformer.rs	24% <24%> (ø)
common/planners/src/plan_user_udf_show.rs	33% <33%> (ø)
query/src/users/user_api.rs	44% <33%> (-6%)	⬇️
common/sql/src/expr/expr_visitor.rs	63% <63%> (ø)
query/src/sql/statements/analyzer_expr.rs	68% <64%> (+2%)	⬆️
query/src/sql/statements/statement_create_udf.rs	66% <66%> (ø)
... and 22 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2be8994...25e29b8. Read the comment docs.

[sundy-li]

Awesome, really surprised me that this pr can be landed in such a swift speed.

[sundy-li]

That's much better now, no extra function call.

[databend-bot]

Wait for another reviewer approval

[leiysky]

It seems this crate only involves expr_visitor, what about rename it to common-ast? Since the name common-sql looks too common.

[lianghanzhen]

I don't have a strong opinion about the naming, I will rename it in other MR.

[BohuTANG]

Great job, let's merge!
Thank you @lianghanzhen !