[GR-41420] Native Image FileSystem throws exception when resolving against an empty Path

[sbrannen]

Overview

As stated in #5020 (comment) and #5020 (comment), if you invoke fileSystem.getPath("") against the native image FileSystem and then invoke resolve(...) on that Path, you get the following exception.

java.lang.ArrayIndexOutOfBoundsException: Index -1 out of bounds for length 0
       com.oracle.svm.core.jdk.resources.NativeImageResourcePath.resolve(NativeImageResourcePath.java:302)
       com.oracle.svm.core.jdk.resources.NativeImageResourcePath.resolve(NativeImageResourcePath.java:327)

Whereas, a Path created by the standard OS FileSystem does not throw an exception in that scenario.

For example...

System.out.println("FileSystems.getDefault().getPath(\"\"): [%s]".formatted(FileSystems.getDefault().getPath("")));
System.out.println("FileSystems.getDefault().getPath(\"\").resolve(\"test\"): [%s]".formatted(FileSystems.getDefault().getPath("").resolve("test")));

... the above code prints the following on the JVM as well as within a native image.

FileSystems.getDefault().getPath(""): []
FileSystems.getDefault().getPath("").resolve("test"): [test]

Whereas, when executing the following within a native image...

URI root = new URI("resource:/");
FileSystem nativeImageFileSystem = FileSystems.newFileSystem(root, Map.of());

System.out.println("nativeImageFileSystem.getPath(\"\"): [%s]".formatted(nativeImageFileSystem.getPath("")));
System.out.println("nativeImageFileSystem.getPath(\"\").resolve(\"test\"): [%s]".formatted(nativeImageFileSystem.getPath("").resolve("test")));

... that prints:

nativeImageFileSystem.getPath(""): []

... and then throws the aforementioned ArrayIndexOutOfBoundsException.

Cause of the exception

In the following line of code...

graal/substratevm/src/com.oracle.svm.core/src/com/oracle/svm/core/jdk/resources/NativeImageResourcePath.java

Line 302 in 3950e04

if (p1.path[p1.path.length - 1] == '/') {

... if p1.path.length is 0, then you get the ArrayIndexOutOfBoundsException.

System Tested

• GraalVM: OpenJDK Runtime Environment GraalVM CE 22.2.0 (build 17.0.4+8-jvmci-22.2-b06)
• Mac OS: Darwin Kernel Version 21.6.0: Mon Aug 22 20:19:52 PDT 2022; root:xnu-8020.140.49~2/RELEASE_ARM64_T6000 arm64

[jovanstevanovic]

Fix is on a master.

[sbrannen]

Fix is on a master.

Great! 👍

@fniephaus and @vjovanov, is there any way to have related commits automatically linked to the corresponding GitHub issue?

For example, I would like to see the commit(s) associated with the fix, but I don't know how to find them (at least not easily).

It looks like this is the commit: 3816914

So it would be great if commit messages would include the related GitHub issue(s) (like #5081 or gh-5081).

[fniephaus]

@fniephaus and @vjovanov, is there any way to have related commits automatically linked to the corresponding GitHub issue?

Thanks for the feedback, @sbrannen. Typically, we add the internal tracking code (e.g., GR-41420) as a prefix to the GitHub issue and then there is going to be a merge with the fix matching that prefix (@jovanstevanovic is a relatively new team member and doesn't know all these details yet, I took care of it just now). I understand this situation is not ideal and GitHub provides different ways to connect issues, commits, and PRs. Although I'm glad that our CI system is smart enough to automate a lot of things for us behind the curtain, it's not yet smart enough to take advantages of those GitHub features but we're working on it. :)