Skip to content

False Positive: CloseSql.ql reports a ResultSet that is already scoped by try-with-resources on its parent Statement. #21533

@Carlson-JLQ

Description

Version
codeql 2.24.3

Checker

  • Checker id: Likely Bugs/Resource Leaks/CloseSql.ql
  • Checker description: This checker detects SQL resource objects (Connection, Statement, ResultSet) that are initialized locally and not guaranteed to be closed on method exit.

Description of the false positive

This case should stay out of the result set. The ResultSet is created from a Statement that is itself managed by try-with-resources, together with the owning Connection. Once the try block exits, both parent resources are closed, and the ResultSet lifetime is covered by that ownership chain.

The alias on the Statement does not change that. It is still the same tracked resource.

Affected test cases

NegCase7_Var3.java

stmtAlias is just another reference to originalStmt. The query appears to lose that alias relationship and treats rs as if it were detached from the try-with-resources scope.

// A locally initialized SQL ResultSet obtained via a method call on a Statement variable, where Statement is a transitive parent closeable resource, should not be flagged as unclosed.
package scensct.var.neg;

import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.Statement;
import java.sql.ResultSet;

public class NegCase7_Var3 {
    public void test() throws Exception {
        try (Connection conn = DriverManager.getConnection("jdbc:test");
             Statement originalStmt = conn.createStatement()) {
            // Alias the Statement
            Statement stmtAlias = originalStmt;
            // Always-true condition
            if (conn != null) {
                // Locally initialized ResultSet via method call on aliased Statement
                ResultSet rs = stmtAlias.executeQuery("SELECT 1"); // [REPORTED LINE]
            }
        }
    }
}

Cause analysis

Likely Bugs/Resource Leaks/CloseSql.ql is supposed to report SQL resources that are not guaranteed to be closed. That is not what this sample is doing.

The only reason this gets reported is that the Statement is referenced through stmtAlias instead of its original local name. If the query stops recognizing parent-resource ownership after a trivial alias, it will produce noise in ordinary JDBC code where objects are routinely passed through locals before use.

This is an ownership-tracking false positive, not a real leak.

Metadata

Metadata

Assignees

No one assigned

    Labels

    questionFurther information is requested

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions