Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://github.com/pmd/pmd/issues/4861 below:

[java] UnusedPrivateMethod - false positive with static methods in core JDK classes · Issue #4861 · pmd/pmd · GitHub

• Pricing

Affects PMD Version: 7.0.0

Rule: UnusedPrivateMethod

Description:

The regression tester shows, that the rule finds the following allegedly unused private methods:

• https://github.com/openjdk/jdk/tree/jdk-11+28/src/java.base/share/classes/java/lang/Integer.java#L170 'toStringUTF16(int, int)'
• https://github.com/openjdk/jdk/tree/jdk-11+28/src/java.base/share/classes/java/lang/Integer.java#L338 'toUnsignedString0(int, int)'
• https://github.com/openjdk/jdk/tree/jdk-11+28/src/java.base/share/classes/java/lang/Integer.java#L389 'formatUnsignedIntUTF16(int, int, byte[], int, int)'
• https://github.com/openjdk/jdk/tree/jdk-11+28/src/java.base/share/classes/java/lang/Long.java#L154 'toStringUTF16(long, int)'
• https://github.com/openjdk/jdk/tree/jdk-11+28/src/java.base/share/classes/java/lang/Long.java#L241 'toUnsignedBigInteger(long)'
• https://github.com/openjdk/jdk/tree/jdk-11+28/src/java.base/share/classes/java/lang/Long.java#L417 'formatUnsignedLong0UTF16(long, int, byte[], int, int)'
• https://github.com/openjdk/jdk/tree/jdk-11+28/src/java.base/share/classes/java/lang/Object.java#L41 'registerNatives()'

Expected outcome:

All these private static methods are indeed in use. So, none of them should be reported.

Additional context:

• [java] UnusedPrivateMethod - fix false positive with lambdas #4841 (comment)
• [java] UnusedPrivateMethod - fix false positive with lambdas #4841 (comment)

  A very preliminar look into the FP on openjdk shows that the node is null not because inference failed, but because, as those classes (ie: java.lang.Integer) are part of the auxclasspath, they resolve to an ASM-backed symbol instead of an AST-backed symbol… and ASM-backed symbols can't reference the AST, so tryGetNode() returns null.

  3 possible ways around this would be:

  • Understand why ASM is prefered over AST-backed and see if it makes sense to switch over (probably there are performance issues at play, and ASM is objectively more reliable with things such as extending a class in a library).
  • Change the rule to compare matches by JMethodSig instead of by node (this is in practice similar to what this PR tried, but without reimplementing arg-matching manually).

DashwoodIce9, mihalyr-prospect, AlexisJehan and axthosarouris

You can’t perform that action at this time.

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4