Fixes CLI hanging on invalid queries

[johnedquinn]

Relevant Issues

• Closes v0.13.1 CLI hangs on invalid queries #1230

Description

• There was a bug introduced in our implementation of the CLI's RunnableWriter and RunnablePipeline.
  • To break it down:
    1. The main thread was designed to send interrupt signals to the relevant RunnableWriter and RunnablePipeline which would result in the tear-down of the child thread. We would then re-create the child thread and instantiate a new runnable in its place to continue to allow user inputs.
    2. There was a bug when the writer or pipeline were to fail for reasons beyond thread interruption. Essentially, on invalid queries, the child thread would die (due to unhandled exceptions), and the main thread would not know of its passing. So, the CLI would "hang" -- since we wouldn't be able to pass data between the runnables via the blocking queues.
• This PR fixes this bug by handling invalid queries.

Reviewer Testing

• I personally recommend running the following queries:

-- Regular query
SELECT a, b
FROM << { 'a': 1, 'b': 2 } >> AS t;

-- Regular query that should fail during eval
SELECT a, b, c, d
FROM << { 'a': 1, 'b': 2 } >> AS t;

-- Regular query that should fail during parse
SELECT a;

-- This adds to the global bindings
!add_to_global_env { 'a': 1 }

-- This should work
a;

-- This should hang on compilation. Use CTRL-C to escape.
SELECT COUNT(*)
FROM
  ([1, 2, 3, 4]) as x1,
  ([1, 2, 3, 4]) as x2,
  ([1, 2, 3, 4]) as x3,
  ([1, 2, 3, 4]) as x4,
  ([1, 2, 3, 4]) as x5,
  ([1, 2, 3, 4]) as x6,
  ([1, 2, 3, 4]) as x7,
  ([1, 2, 3, 4]) as x8,
  ([1, 2, 3, 4]) as x9,
  ([1, 2, 3, 4]) as x10,
  ([1, 2, 3, 4]) as x11
GROUP BY x1._1;

-- This should essentially run for a very long time during evaluation. Use CTRL-C to cancel.
SELECT *
FROM
  ([1, 2, 3, 4]) as x1,
  ([1, 2, 3, 4]) as x2,
  ([1, 2, 3, 4]) as x3,
  ([1, 2, 3, 4]) as x4,
  ([1, 2, 3, 4]) as x5,
  ([1, 2, 3, 4]) as x6,
  ([1, 2, 3, 4]) as x7,
  ([1, 2, 3, 4]) as x8,
  ([1, 2, 3, 4]) as x9,
  ([1, 2, 3, 4]) as x10,
  ([1, 2, 3, 4]) as x11;

-- If you start up the shell with --pipeline EXPERIMENTAL, you can run this:
EXPLAIN SELECT 1 FROM <<0, 1, 2>>

Other Information

• Updated Unreleased Section in CHANGELOG: YES
• Any backward-incompatible changes? NO
• Any new external dependencies? NO
• Do your changes comply with the Contributing Guidelines
  and Code Style Guidelines? YES

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 (7e3625f)	5916d8b	+/-
% Passing	92.33%	92.33%	0.00%
✅ Passing	5372	5372	0
❌ Failing	446	446	0
🔶 Ignored	0	0	0
Total Tests	5818	5818	0

Number passing in both: 5372

Number failing in both: 446

Number passing in Base (7e3625f) but now fail: 0

Number failing in Base (7e3625f) but now pass: 0

[codecov-commenter]

Codecov Report

Attention: 72 lines in your changes are missing coverage. Please review.

Comparison is base (7e3625f) 71.57% compared to head (dff43e8) 71.60%.

Files	Patch %	Lines
...cli/src/main/kotlin/org/partiql/cli/shell/Shell.kt	0.00%	35 Missing ⚠️
...ain/kotlin/org/partiql/cli/shell/RunnableWriter.kt	0.00%	29 Missing ⚠️
...n/kotlin/org/partiql/cli/shell/RunnablePipeline.kt	0.00%	8 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #1274      +/-   ##
============================================
+ Coverage     71.57%   71.60%   +0.02%     
  Complexity     2451     2451              
============================================
  Files           229      229              
  Lines         17626    17620       -6     
  Branches       3249     3252       +3     
============================================
  Hits          12616    12616              
+ Misses         4001     3995       -6     
  Partials       1009     1009              

Flag	Coverage Δ
CLI	13.74% <0.00%> (+0.04%)	⬆️
EXAMPLES	80.28% <ø> (ø)
LANG	78.71% <ø> (ø)

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.

[yliuuuu]

I can verify the fix works.

I feel like the state management is a bit tricky to understand.

Maybe we can think about if there is a better way to model this later.