Implement UDF Plan

[xinlifoobar]

Which issue does this PR close?

Closes #11242 and part of #11207

Rationale for this change

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

[jayzhan211]

IMO I don't think this is a good abstration, because extract and position are actually quite different concept. For the point of view to user-defined planner, they could have their own implementation that may not be UDF, so it does not make sense to place them to plan_udf for me.

I would prefer to have plan_extract and plan_position separately.

Btw, I suggest using plan_udaf in #11229, as we are customizing the query of aggregate function. Similarly, I think here should be the correct place for plan_udf

@xinlifoobar What do you think?

[jayzhan211]

it seems not to be extract specific anymore, we could name it args

[alamb]

I would prefer to have plan_extract and plan_position separately.

Here is a PR for planning position: #11243

It isn't clear to me what we should do with this PR now (should we close it?)

[alamb]

IMO I don't think this is a good abstration, because extract and position are actually quite different concept. For the point of view to user-defined planner, they could have their own implementation that may not be UDF, so it does not make sense to place them to plan_udf for me.

I would prefer to have plan_extract and plan_position separately.

I agree that even though the functions are somewhat repetitively named, it is better to be explicit in the function names rather than being implicit (via function name) as the API proposed in this PR would be

Sorry I didn't see the comments on #11207 (comment) on #11207 earlier

[jayzhan211]

I would prefer to have plan_extract and plan_position separately.

Here is a PR for planning position: #11243

It isn't clear to me what we should do with this PR now (should we close it?)

I think we can close #11263(?) And proceed #11243