Jigsaw test failures

[Sean Mullan]

On 1/3/11 5:53 PM, Mandy Chung wrote:
> On 01/03/11 13:01, Sean Mullan wrote:
>> There are a few jigsaw regression test failures:
>>
>> * org/openjdk/jigsaw/cli/ModuleFormatTest01.sh
>> * org/openjdk/jigsaw/cli/jmod-signed.sh
>> * org/openjdk/jigsaw/cli/signed-module.sh
>>
>> The last one is due to the cacerts file being empty. We have a fix coming for
>> that.
>>
>> However, the previous two appear to be the same failure and is caused when it
>> compares the bytes of one of the classes in an installed module with the bytes
>> in the module file (I think). This is the stack trace from the
>> ModuleFileFormatTest01.sh test:
>>
>> java.lang.RuntimeException: file compare failed 1
>> at ModuleFormatTest01.compare(ModuleFormatTest01.java:472)
>> at ModuleFormatTest01.compare(ModuleFormatTest01.java:407)
>> at ModuleFormatTest01.compare(ModuleFormatTest01.java:425)
>> at ModuleFormatTest01.compare(ModuleFormatTest01.java:448)
>> at ModuleFormatTest01.doTestSingleClassModule(ModuleFormatTest01.java:114)
>> at ModuleFormatTest01.testSingleClassModule(ModuleFormatTest01.java:93)
>> at ModuleFormatTest01.test(ModuleFormatTest01.java:301)
>> at ModuleFormatTest01.run(ModuleFormatTest01.java:54)
>> at ModuleFormatTest01.main(ModuleFormatTest01.java:49)
>> Exception in thread "main" java.lang.Exception: 1/2 tests failed
>> at ModuleFormatTest01.run(ModuleFormatTest01.java:62)
>> at ModuleFormatTest01.main(ModuleFormatTest01.java:49)
>> result: Failed. Execution failed: exit code 1
>>
>> Has anyone looked at this more closely?
>
>
> Not that I know of. Do you know when these tests started to fail?

No I don't know ...

--Sean