feat: Support right expression

[Shekharrajak]

Closes #3182

Rationale for this change

Enable native execution for Spark's RIGHT(str, len) function to improve query performance by avoiding JVM fallback.

What changes are included in this PR?

• Added CometRight serializer that handles edge cases:
  • len <= 0: Returns empty string literal
  • len > 0: Transforms to Substring(str, -len, len) protobuf
• Registered Right expression in QueryPlanSerde string expressions map
• No protobuf or Rust changes required (reuses existing Substring implementation)

How are these changes tested?

Added 4 test suites in CometExpressionSuite:

• Basic functionality (various lengths: 0, -1, positive, exceeds length)
• Unicode character handling (emoji, multi-byte chars)
• Equivalence verification with SUBSTRING(str, -len, len)
• Dictionary encoding preservation

[andygrove]

@Shekharrajak can you run make to update the generated configs.md

[andygrove]

Review Feedback

Critical Issue: NULL handling bug when len ≤ 0

The implementation has a Spark compatibility issue in strings.scala:

if (lenInt <= 0) {
  val literalBuilder = LiteralOuterClass.Literal.newBuilder()
  literalBuilder.setIsNull(false)
  literalBuilder.setStringVal("")
  Some(ExprOuterClass.Expr.newBuilder().setLiteral(literalBuilder).build())
}

This returns empty string for len ≤ 0 without checking if str is NULL.

Per Spark's implementation (Right.replacement):

If(IsNull(str), Literal.create(null, dataType),      // Check NULL first!
  If(LessThanOrEqual(len, Literal(0)),
    Literal(UTF8String.EMPTY_UTF8),
    Substring(str, UnaryMinus(len), len)))

Spark checks IsNull(str) before checking len <= 0. So:

• RIGHT(NULL, 0) should return NULL (Spark behavior)
• RIGHT(NULL, 0) returns empty string (this PR)

Suggested Fix

Either:

1. Evaluate the str expression and return NULL when str is NULL (even when len ≤ 0), or
2. Mark this edge case as Incompatible

Missing Test Case

Please add a test for SELECT RIGHT(CAST(NULL AS STRING), 0) to verify NULL handling with non-positive length.

---

This review was generated with AI assistance.

[Shekharrajak]

Updated the PR, please trigger the workflow

[codecov-commenter]

Codecov Report

❌ Patch coverage is 7.69231% with 24 lines in your changes missing coverage. Please review.
✅ Project coverage is 59.87%. Comparing base (f09f8af) to head (4d37921).
⚠️ Report is 932 commits behind head on main.

Files with missing lines	Patch %	Lines
...rc/main/scala/org/apache/comet/serde/strings.scala	4.00%	24 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #3207      +/-   ##
============================================
+ Coverage     56.12%   59.87%   +3.74%     
- Complexity      976     1473     +497     
============================================
  Files           119      175      +56     
  Lines         11743    16193    +4450     
  Branches       2251     2687     +436     
============================================
+ Hits           6591     9695    +3104     
- Misses         4012     5150    +1138     
- Partials       1140     1348     +208     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Shekharrajak]

checks looking good. Please have a look.

[andygrove]

LGTM. Thanks @Shekharrajak