Generates internal impl IR classes; updates partiql-plan

[rchowell]

Tasks

• Upgrade to Kotlin 1.6 #1075
• Use sealed interfaces
• Generate a Rewriter (VisitorTransform)

Relevant Issues / Description

Related to #1054 we are splitting out Plan and AST intermediate representations into their own packages. This changes the generator to produce internal implementations and public interfaces which adhere to Explicit API.

You cannot directly instantiate an IR node, but you get four ways to do so

• Type Safe Builders
• Type Safe Builder via a DSL (with optional custom factory)
• Default Factory methods
• Custom Factory methods

The default implementation of a domain's abstract factory is a top-level object with the domain name (same as the PIG builder). In short,

// Before, direct constructor
val myNode = Rex.Id("a") 

// After, default factory method
val myNode = Plan.rexId("a") 

Additionally, each IR node has an _id: String field. The _id field is a simple unique identifier which can be used as a key, but nodes can be directly used in other data structures via hashCode and equals.

Other Information

• Updated Unreleased Section in CHANGELOG: [YES/NO]
  YES

• Any backward-incompatible changes? [YES/NO]
  YES. This is a breaking change. Customers would need to replace the direct constructors with the default factory method. Nothing has changed beyond the "constructor" location.

• Any new external dependencies? [YES/NO]
  NO

License Information

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[github-actions]

Conformance comparison report

	Base (adb32f7)	b9c3b1c	+/-
% Passing	92.47%	92.47%	0.00%
✅ Passing	5380	5380	0
❌ Failing	438	438	0
🔶 Ignored	0	0	0
Total Tests	5818	5818	0

Number passing in both: 5380

Number failing in both: 438

Number passing in Base (adb32f7) but now fail: 0

Number failing in Base (adb32f7) but now pass: 0

[codecov-commenter]

Codecov Report

Patch coverage: 57.47% and no project coverage change.

Comparison is base (adb32f7) 73.67% compared to head (f0bd70c) 73.67%.

Additional details and impacted files

@@            Coverage Diff            @@
##               main    #1109   +/-   ##
=========================================
  Coverage     73.67%   73.67%           
- Complexity     2404     2407    +3     
=========================================
  Files           248      248           
  Lines         17729    17741   +12     
  Branches       3230     3235    +5     
=========================================
+ Hits          13062    13071    +9     
  Misses         3676     3676           
- Partials        991      994    +3     

Flag	Coverage Δ
CLI	14.28% <ø> (ø)
EXAMPLES	80.28% <ø> (ø)
LANG	79.46% <57.47%> (-0.01%)	⬇️

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

Impacted Files	Coverage Δ
...rtiql/lang/planner/transforms/plan/RexConverter.kt	53.52% <53.84%> (ø)
.../partiql/lang/planner/transforms/plan/PlanTyper.kt	48.00% <61.53%> (ø)
...rtiql/lang/planner/transforms/plan/RelConverter.kt	66.98% <61.90%> (ø)
...n/org/partiql/lang/planner/transforms/AstToPlan.kt	88.00% <100.00%> (ø)

... and 3 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[johnedquinn]

Left small comments, but everything looks pretty good!

I did have one question about testing. Have you considered setting up a build task for generating a test IR? It could be useful to generate the test IR and then attempt to access some of the generated properties/methods as part of the unit tests -- just to show that it compiles and to make it easier for review.