[java] Fix AvoidUsingOctalValues false-positive

[Gold856]

Describe the PR

ASTNumericLiteral did not correctly determine the base when a number ended with a non-number (L, F, D, .) This PR checks if the number ends with one of those characters, and correctly returns a base of 10.

Related issues

• Fixes [java] AvoidUsingOctalValues triggers on non-octal double literals with a leading 0 #5007

Ready?

• Added unit tests for fixed bug/feature
• Passing all unit tests
• Complete build ./mvnw clean verify passes (checked automatically by github actions)
• Added (in-code) documentation (if needed)

[jsotuyod]

Thanks for the PR! We greatly appreciate the contribution.

I've added some comments to improve the code. We should also add a new test case for the AvoidUsingOctalValues rule. You can do so adding a new <test-code> node in AvoidUsingOctalValues.xml

[Gold856]

I added some test cases and fixed the logic; the code now checks for a period in the literal along with a float or double suffix at the end (after the hex and binary check), since those do cause the base to change to base 10.

[pmd-test]

	1 Message
📖	Compared to master: This changeset changes 0 violations, introduces 8 new violations, 0 new errors and 0 new configuration errors, removes 2 violations, 0 errors and 0 configuration errors. Download full report as build artifact
✅	Compared to master: This changeset changes 0 violations, introduces 8 new violations, 0 new errors and 0 new configuration errors, removes 2 violations, 0 errors and 0 configuration errors. Download full report as build artifact
✅	Compared to master: This changeset changes 0 violations, introduces 8 new violations, 0 new errors and 0 new configuration errors, removes 2 violations, 0 errors and 0 configuration errors. Download full report as build artifact
✅	Compared to master: This changeset changes 0 violations, introduces 63 new violations, 0 new errors and 0 new configuration errors, removes 2 violations, 0 errors and 0 configuration errors. Download full report as build artifact

Generated by 🚫 Danger

[jsotuyod]

Let's see what the regression tester shows, but it seems better.

I'd probably just add a test case before merging on the case 0x1.0000_ffff_eeeep+2f for the rule UseUnderscoresInNumericLiterals, as this was detected by the regression tester but not our own rules, so we are obviously missing this scenario there.

But other than that, if @pmd-test doesn't show anything weird, this is good to go.

[Gold856]

Added the float literal test case. Also added code to check for exponents (those can also change an octal to decimal) with a test case.

[jsotuyod]

This looks great! Thanks!

[oowekyala]

Thank you @Gold856!