Add ArrayAgg aggregation function

[xiangfu0]

Add aggregation function: ArrayAgg.

• Usage: ArrayAgg(column, 'dataType' [, 'isDistinct'])
• Spports data BOOLEAN, INT, LONG, FLOAT(only in V1), DOUBLE, STRING, TIMESTAMP.
  E.g. ArrayAgg(intCol, 'INT') returns ARRAY<INT>

For implementation, separated the implementations per data type:

• Base abstract class is ArrayAggFunction,
• Each type has its own abstract class ArrayAggBase<type>Function,
• There are two implementations:
  • distinct agg: ArrayAggDistinct<type>Function,
  • non-distinct agg: ArrayAgg<type>Function

Note: Float type column is treated as Double in v2 engine, so FLOAT type is not supported.

Release Note

• Support ArrayAgg Aggregation function

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 40.45534% with 340 lines in your changes missing coverage. Please review.

Project coverage is 62.88%. Comparing base (226087f) to head (c9addb1).
Report is 2920 commits behind head on master.

Files with missing lines	Patch %	Lines
...gation/function/array/BaseArrayAggIntFunction.java	15.15%	26 Missing and 2 partials ⚠️
...ation/function/array/BaseArrayAggLongFunction.java	16.12%	24 Missing and 2 partials ⚠️
...ion/function/array/BaseArrayAggDoubleFunction.java	17.24%	22 Missing and 2 partials ⚠️
...tion/function/array/BaseArrayAggFloatFunction.java	17.24%	22 Missing and 2 partials ⚠️
...gregation/function/array/BaseArrayAggFunction.java	33.33%	23 Missing and 1 partial ⚠️
...ion/function/array/BaseArrayAggStringFunction.java	17.24%	22 Missing and 2 partials ⚠️
...function/array/ArrayAggDistinctDoubleFunction.java	33.33%	16 Missing ⚠️
.../function/array/ArrayAggDistinctFloatFunction.java	33.33%	16 Missing ⚠️
...function/array/ArrayAggDistinctStringFunction.java	30.43%	16 Missing ⚠️
...egation/function/array/ArrayAggDoubleFunction.java	33.33%	16 Missing ⚠️
... and 12 more

Additional details and impacted files

@@             Coverage Diff              @@
##             master   #11822      +/-   ##
============================================
- Coverage     63.02%   62.88%   -0.14%     
- Complexity     1140     1141       +1     
============================================
  Files          2351     2367      +16     
  Lines        127324   127888     +564     
  Branches      19611    19732     +121     
============================================
+ Hits          80244    80424     +180     
- Misses        41373    41733     +360     
- Partials       5707     5731      +24     

Flag	Coverage Δ
custom-integration1	<0.01% <0.00%> (-0.01%)	⬇️
integration	<0.01% <0.00%> (-0.01%)	⬇️
integration1	<0.01% <0.00%> (-0.01%)	⬇️
integration2	0.00% <0.00%> (ø)
java-11	62.83% <40.45%> (-0.14%)	⬇️
java-21	62.76% <40.45%> (-0.09%)	⬇️
skip-bytebuffers-false	62.87% <40.45%> (-0.14%)	⬇️
skip-bytebuffers-true	62.72% <40.10%> (-0.08%)	⬇️
temurin	62.88% <40.45%> (-0.14%)	⬇️
unittests	62.88% <40.45%> (-0.14%)	⬇️
unittests1	66.88% <40.45%> (-0.20%)	⬇️
unittests2	14.44% <0.00%> (-0.08%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[walterddr]

we should probably consider how to parse this. b/c the proper way to express distinct array agg is

SELECT key, ARRAY_AGG(DISTINCT col) 
FROM tbl 
GROUP BY key

given that our functions are not standard it is fine we do it now but good to note this

[walterddr]

also see: https://www.postgresql.org/docs/16/functions-aggregate.html#FUNCTIONS-AGGREGATE

[xiangfu0]

agreed, I was also thinking about this, however it's not parsed in v1 query engine.

[walterddr]

CC @Jackie-Jiang this is related to #11518

[gortiz]

This doesn't seem to be postgres compatible. Documentation says:

array_agg ( anynonarray ) → anyarray
Collects all the input values, including nulls, into an array.

And in fact that is how it behaves
Example:

postgres=# insert into A(a) values (1), (2), (3), (null), (5);
INSERT 0 5
postgres=# select array_agg(a) from A;
   array_agg    
----------------
 {1,2,3,NULL,5}
(1 row)