Enable tests to run in parallel by default

[ExtReMLapin]

To be merged after #3428

[gemini-code-assist]

Code Review

This pull request significantly expands OpenCypher function support, adding geospatial, temporal, vector, and string functions along with GQL-conformant aliases. It also improves test parallelization by updating the TestHelper and implementing resource locks across the test suite. Feedback focuses on fixing a potential NumberFormatException for empty vector strings, ensuring split() preserves trailing empty strings per Cypher specifications, restoring robustness in point() coordinate parsing, and handling potential nulls or incorrect types in map-based distance calculations.

[gemini-code-assist]

The current parsing logic will throw a NumberFormatException if the input string represents an empty array []. When inner is an empty string, split(",") returns an array containing one empty string, which Float.parseFloat() cannot handle. Additionally, it doesn't handle cases with extra commas or whitespace between values gracefully.

Suggested change

	final String inner = trimmed.startsWith("[") && trimmed.endsWith("]") ? trimmed.substring(1, trimmed.length() - 1) : trimmed;
	final String[] parts = inner.split(",");
	final float[] result = new float[parts.length];
	final String inner = trimmed.startsWith("[") && trimmed.endsWith("]") ? trimmed.substring(1, trimmed.length() - 1).trim() : trimmed;
	if (inner.isEmpty())
	return new float[0];
	final String[] parts = inner.split(",");
	final float[] result = new float[parts.length];

[gemini-code-assist]

Java's String.split(regex) discards trailing empty strings by default. To comply with the Cypher specification, which expects trailing empty strings to be preserved (e.g., split('a,b,', ',') should return ['a', 'b', '']), you should use the two-argument version of split with a negative limit.

Suggested change

	return List.of(str.split(Pattern.quote(delimiter)));
	return List.of(str.split(Pattern.quote(delimiter), -1));

[gemini-code-assist]

This change reduces robustness compared to the previous implementation. The previous code used GeoUtils.getDoubleValue(), which safely handles both Number and String inputs. The direct cast to Number will throw a ClassCastException if the input is a string representation of a coordinate.

Suggested change

	final double lat = ((Number) args[0]).doubleValue();
	final double lon = ((Number) args[1]).doubleValue();
	final double lat = com.arcadedb.function.sql.geo.GeoUtils.getDoubleValue(args[0]);
	final double lon = com.arcadedb.function.sql.geo.GeoUtils.getDoubleValue(args[1]);

[gemini-code-assist]

This code is unsafe as it assumes the map values are non-null and of type Number. If a coordinate is null in the map (which is possible in Cypher), this will throw a NullPointerException. If it's a String, it will throw a ClassCastException. It is safer to use GeoUtils.getDoubleValue() after verifying the values are not null.

[codacy-production]

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes. Give us feedback}

[codecov]

Codecov Report

❌ Patch coverage is 76.26667% with 89 lines in your changes missing coverage. Please review.
✅ Project coverage is 56.53%. Comparing base (5c288d2) to head (b09a93f).
⚠️ Report is 3 commits behind head on main.

Files with missing lines	Patch %	Lines
...com/arcadedb/function/geo/CypherPointFunction.java	74.60%	6 Missing and 10 partials ⚠️
.../query/opencypher/executor/CypherTrimFunction.java	72.50%	5 Missing and 6 partials ⚠️
...rc/main/java/com/arcadedb/GlobalConfiguration.java	0.00%	9 Missing and 1 partial ⚠️
...dedb/function/geo/CypherPointDistanceFunction.java	72.22%	4 Missing and 6 partials ⚠️
...y/opencypher/executor/CypherSubstringFunction.java	66.66%	3 Missing and 3 partials ⚠️
...java/com/arcadedb/function/math/RoundFunction.java	64.28%	2 Missing and 3 partials ⚠️
...n/java/com/arcadedb/function/sql/geo/GeoUtils.java	44.44%	4 Missing and 1 partial ⚠️
.../com/arcadedb/function/text/SubstringFunction.java	0.00%	1 Missing and 2 partials ⚠️
...query/opencypher/executor/CypherSplitFunction.java	66.66%	2 Missing and 1 partial ⚠️
...edb/function/temporal/DateConstructorFunction.java	71.42%	2 Missing ⚠️
... and 9 more

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #3895      +/-   ##
============================================
- Coverage     64.71%   56.53%   -8.19%     
- Complexity        0     1572    +1572     
============================================
  Files          1581     1585       +4     
  Lines        117023   117508     +485     
  Branches      24858    24957      +99     
============================================
- Hits          75730    66431    -9299     
- Misses        30933    41743   +10810     
+ Partials      10360     9334    -1026     

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

[ExtReMLapin]

/gemini review

[ExtReMLapin]

This was 100% vibecoded and tested by qwen 27b

[gemini-code-assist]

Code Review

This pull request introduces parallel test execution by enabling JUnit 5 class-level concurrency and increasing Maven fork counts. Key changes include updates to TestHelper for capturing and restoring GlobalConfiguration snapshots, adding @ResourceLock annotations to various test classes, and refining configuration dumping logic. Review feedback identifies significant risks with concurrent class execution due to shared global state, recommending that classes run sequentially within forks to avoid flakiness. Additionally, suggestions were made to optimize the configuration restoration logic and cache configuration values to reduce overhead.

[gemini-code-assist]

Enabling concurrent execution of test classes within the same JVM is highly risky for this codebase. Many tests (via TestHelper) mutate the static GlobalConfiguration and other global states. While some tests have been annotated with @ResourceLock, any test that lacks this annotation will still run in parallel and potentially interfere with others, leading to race conditions and flaky results. Since parallelism is already provided by Maven forks (forkCount in pom.xml), it is much safer to run classes sequentially within each fork.

junit.jupiter.execution.parallel.mode.classes.default=same_thread

[gemini-code-assist]

This synchronization only prevents multiple threads from executing the checkActiveDatabases logic simultaneously. It does not protect against other concurrent tests (running in different threads within the same JVM) opening or closing databases while this check is in progress. In a parallel execution environment, this will likely lead to intermittent test failures when one test's cleanup phase overlaps with another test's setup phase. Disabling class-level parallelism in junit-platform.properties is the recommended fix.

[gemini-code-assist]

Restoring every configuration value after every test method is expensive, especially since setValue can trigger side effects like logging or configuration dumping. It is more efficient to only restore values that were actually modified during the test execution.

      final GlobalConfiguration config = e.getKey();
      final Object value = e.getValue();
      if (value != null && !value.equals(config.getValue()))
        config.setValue(value);

[gemini-code-assist]

Since captureConfigSnapshot is called before every test method, caching the GlobalConfiguration.values() array in a constant can avoid repeated array allocations.

Suggested change

	private static final Object ACTIVE_DB_LOCK = new Object();
	private Map<GlobalConfiguration, Object> configSnapshot;
	private static final Object ACTIVE_DB_LOCK = new Object();
	private static final GlobalConfiguration[] CONFIG_VALUES = GlobalConfiguration.values();
	private Map<GlobalConfiguration, Object> configSnapshot;

[ExtReMLapin]

I'm not a java guy so I have no idea what i'm doing