Source encoding not taken into account by the GUI

[VsevolodGolovanov]

Say the source contains cyrillic symbols in UTF-8:

// провокация бага
public static SimpleDateFormat df = new SimpleDateFormat();

The platform encoding is the default Cp1251, so let's explicitly set all the possible encoding options in the pom:

<properties>
    <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
</properties>

<build>
    <plugins>
        <plugin>
            <groupId>org.codehaus.mojo</groupId>
            <artifactId>findbugs-maven-plugin</artifactId>
            <version>3.0.4-SNAPSHOT</version>
            <configuration>
                <jvmArgs>-Dfile.encoding=UTF-8</jvmArgs>
                <encoding>UTF-8</encoding>
                <sourceEncoding>UTF-8</sourceEncoding>
                <outputEncoding>UTF-8</outputEncoding>

Nope:

Let's check the GUI's JVM arguments with Visual VM:

It's got file.encoding (and others) set twice, and one of those is Cp1251. I suspect the duplicates come from clonevm=true.

mvn findbugs:gui log

Just to be sure that the source is right, let's force the encoding using the environment variable:

set JAVA_TOOL_OPTIONS=-Dfile.encoding=UTF-8

Now it works: