Turn on expand identifier in SqlValidator config

[xiangfu0]

Fix for #11447
For sql:

SELECT toBase64(toUtf8(AirlineID)) as encoded
FROM mytable
GROUP BY toBase64(toUtf8(AirlineID))
LIMIT 10

The issue is that calcite expand group by toBase64(toUtf8(AirlineID)) to toBase64(toUtf8(CAST(mytable.AirlineID AS VARCHAR CHARACTER SET ISO-8859-1))) but not the expression in SELECT.

Enabling identifier expansion is useful in situations where you want to be explicit about the source of each column or table. It can help in reducing ambiguity and making the query more self-describing.

[codecov-commenter]

Codecov Report

Merging #11457 (84c3221) into master (399f033) will increase coverage by 0.01%.
Report is 4 commits behind head on master.
The diff coverage is 100.00%.

@@             Coverage Diff              @@
##             master   #11457      +/-   ##
============================================
+ Coverage     62.99%   63.00%   +0.01%     
- Complexity     1094     1098       +4     
============================================
  Files          2302     2302              
  Lines        124025   124041      +16     
  Branches      18901    18903       +2     
============================================
+ Hits          78126    78156      +30     
+ Misses        40351    40333      -18     
- Partials       5548     5552       +4     

Flag	Coverage Δ
integration	<0.01% <0.00%> (ø)
integration1	<0.01% <0.00%> (ø)
integration2	0.00% <0.00%> (ø)
java-11	62.95% <100.00%> (-0.01%)	⬇️
java-17	62.87% <100.00%> (+0.02%)	⬆️
java-20	62.84% <100.00%> (-0.01%)	⬇️
temurin	63.00% <100.00%> (+0.01%)	⬆️
unittests	63.00% <100.00%> (+0.01%)	⬆️
unittests1	67.55% <100.00%> (+0.02%)	⬆️
unittests2	14.47% <0.00%> (-0.01%)	⬇️

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

Files Changed	Coverage Δ
...ava/org/apache/pinot/query/validate/Validator.java	80.00% <100.00%> (+0.68%)	⬆️

... and 18 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[gortiz]

This PR seems to produce a regression where select * from table returns the $rowId, $hostname, etc that should not be returned

[walterddr]

i was wondering if there's a better way to handle it. expanding both into a CAST seems a bit overkill b/c we might not really want to perform the cast operation, is there any rule in Calcite during parsing time to convert the group-by into just an index reference (e.g. SELECT toBase64(toUtf8(AirlineID)) as encoded FROM mytable GROUP BY encoded or ... GROUP BY 1 )

[walterddr]

there's a problem with this.

SELECT toBase64(toUtf8(AirlineID)) AS AirlineID
FROM mytable
GROUP BY toBase64(toUtf8(AirlineID))
ORDER BY toBase64(toUtf8(AirlineID)) DESC 
LIMIT 10

will failed b/c of the reference to AirlineID is both in the function toBase64 and AS aliasing
did this query ever succeed before the expand identifier?