LucasEby opened a new issue, #24968:
URL: https://github.com/apache/pulsar/issues/24968

   ### Search before reporting
   
   - [x] I searched in the [issues](https://github.com/apache/pulsar/issues) 
and found nothing similar.
   
   
   ### Read release policy
   
   - [x] I understand that [unsupported 
versions](https://pulsar.apache.org/contribute/release-policy/#supported-versions)
 don't get bug fixes. I will attempt to reproduce the issue on a supported 
version of Pulsar client and Pulsar broker.
   
   
   ### User environment
   
   Broker version: 4.2.0-SNAPSHOT
   Broker Operating system and hardware type: Linux x86_64 GNU/Linux
   Broker Java version: 17.0.16
   
   ### Issue Description
   
   The below test would compares the generated documentation of 
CmdGenerateDocsTest#testGenerateDocs.  The test assumes a deterministic order 
of the generated documentation but the order of the listed help and name 
options was not guaranteed. Due to two sources of nondeterminism hidden in the 
test and production code, the options could be listed in a different order. 
Picocli's CommandLine code uses 
[Class#getDeclaredFields](https://docs.oracle.com/javase/8/docs/api/java/lang/Class.html#getDeclaredFields)
 internally (via reflection), which has a non-deterministic specification to 
retrieve the @Option annotations and orders these options accordingly. Even 
though picocli supports an order attribute on @Option annotations, 
[CommandLine](https://github.com/remkop/picocli/blob/main/src/main/java/picocli/CommandLine.java)
 never sorts by this order - it just maintains whatever order the reflection 
API gave it. CmdGenerateDocs#generateDocument then iterates over the order that 
the `args` meth
 od returns, with the assumption that the order is deterministic. 
Unfortunately, the ordering can change due to different environments producing 
the contents in different orders despite the logical contents being the same. 
Harmless re-ordering could flip the tests from pass to fail despite the data 
being semantically the same.
   - `org.apache.pulsar.docs.tools.CmdGenerateDocsTest.testGenerateDocs`
   
   I discovered the problem with the 
[NonDex](https://github.com/TestingResearchIllinois/NonDex) tool. NonDex 
systematically detects incorrect tests that rely on non-deterministic behaviors 
in Java APIs—like assuming order of name/value pairs in json files — by 
exploring all specification-allowed outcomes. It does this by instrumenting 
undetermined APIs and randomizing the returned order/behavior within what the 
spec allows. Failures it exposes reliably indicate flawed assumptions in order 
that were never promised and should be fixed.
   
   ### Error messages
   
   ```text
   [INFO] Running org.apache.pulsar.docs.tools.CmdGenerateDocsTest
   [ERROR] Tests run: 2, Failures: 1, Errors: 0, Skipped: 1, Time elapsed: 
3.273 s <<< FAILURE! -- in org.apache.pulsar.docs.tools.CmdGenerateDocsTest
   [ERROR] org.apache.pulsar.docs.tools.CmdGenerateDocsTest.testGenerateDocs -- 
Time elapsed: 0.007 s <<< FAILURE!
   java.lang.AssertionError: 
   expected [# test
   
   
   
   
   $ pulsar test options
   
   
   |Flag|Description|Default|
   |---|---|---|
   | `-h, --help` | Show this help message|false|
   | `-n, --name` | Name|null|
   
   ] but found [# test
   
   
   
   
   $ pulsar test options
   
   
   |Flag|Description|Default|
   |---|---|---|
   | `-n, --name` | Name|null|
   | `-h, --help` | Show this help message|false|
   
   ]
        at org.testng.Assert.fail(Assert.java:110)
        at org.testng.Assert.failNotEquals(Assert.java:1577)
        at org.testng.Assert.assertEqualsImpl(Assert.java:149)
        at org.testng.Assert.assertEquals(Assert.java:131)
        at org.testng.Assert.assertEquals(Assert.java:655)
        at org.testng.Assert.assertEquals(Assert.java:665)
        at 
org.apache.pulsar.docs.tools.CmdGenerateDocsTest.testGenerateDocs(CmdGenerateDocsTest.java:79)
        at 
java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at 
java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
        at 
java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
        at java.base/java.lang.reflect.Method.invoke(Method.java:569)
        at 
org.testng.internal.invokers.MethodInvocationHelper.invokeMethod(MethodInvocationHelper.java:139)
        at 
org.testng.internal.invokers.InvokeMethodRunnable.runOne(InvokeMethodRunnable.java:47)
        at 
org.testng.internal.invokers.InvokeMethodRunnable.call(InvokeMethodRunnable.java:76)
        at 
org.testng.internal.invokers.InvokeMethodRunnable.call(InvokeMethodRunnable.java:11)
        at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
        at 
java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
        at 
java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
        at java.base/java.lang.Thread.run(Thread.java:840)
   
   [INFO] 
   [INFO] Results:
   [INFO] 
   [ERROR] Failures: 
   [ERROR] org.apache.pulsar.docs.tools.CmdGenerateDocsTest.testGenerateDocs
   [INFO]   Run 1: PASS
   [ERROR]   Run 2: CmdGenerateDocsTest.testGenerateDocs:79 expected [# test
   
   
   
   
   $ pulsar test options
   
   
   |Flag|Description|Default|
   |---|---|---|
   | `-h, --help` | Show this help message|false|
   | `-n, --name` | Name|null|
   
   ] but found [# test
   
   
   
   
   $ pulsar test options
   
   
   |Flag|Description|Default|
   |---|---|---|
   | `-n, --name` | Name|null|
   | `-h, --help` | Show this help message|false|
   
   ]
   [INFO] 
   [INFO] 
   [ERROR] Tests run: 1, Failures: 1, Errors: 0, Skipped: 0
   ```
   
   ### Reproducing the issue
   
   You can replicate the problem with the 
[NonDex](https://github.com/TestingResearchIllinois/NonDex) tool command:
   ```
   mvn -pl pulsar-docs-tools 
-Dtest=org.apache.pulsar.docs.tools.CmdGenerateDocsTest#testGenerateDocs 
-DnondexRuns=10 -DreuseForks=false 
edu.illinois:nondex-maven-plugin:2.2.1:nondex -Dcheckstyle.skip=true 
-DnondexRunsWithoutShuffling=0 
   ```
   
   ### Additional information
   
   _No response_
   
   ### Are you willing to submit a PR?
   
   - [x] I'm willing to submit a PR!


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to