commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From sebb <seb...@gmail.com>
Subject Re: [lang] test failures
Date Mon, 05 Mar 2012 17:11:13 GMT
On 5 March 2012 16:53, Jörg Schaible <Joerg.Schaible@scalaris.com> wrote:
> sebb wrote:
>
>> On 5 March 2012 15:09, Benedikt Ritter <bene@systemoutprintln.de> wrote:
>>> Am 05.03.2012 16:03, schrieb Benedikt Ritter:
>>>
>>>> Am 04.03.2012 12:26, schrieb Jörg Schaible:
>>>>>
>>>>> Benedikt Ritter wrote:
>>>>>
>>>>>> Just out of curiosity ;) Are there plans to do anything about that?
I
>>>>>> guess it would be desirable if the component compiles on Java 7 as
>>>>>> well. I still don't know if things are broken when compiling with
Java
>>>>>> 7 or if the errors are just caused by API changes in Java 7. If the
>>>>>> latter, we could use JUnit's Theories runner and write an assumeTrue()
>>>>>> statement for those test cases, that checks what Java version the
test
>>>>>> is running with.
>>>>>
>>>>>
>>>>> I am quite sure I voted on lang3 and that means I also tested it with
>>>>> Java
>>>>> 7. Actually I can build lang3 HEAD with Java 7:
>>>>
>>>>
>>>> sorry, I'm not sure if I got you right. Are you saying that the problems
>>>> Gary was experiencing are not a result of Java 7 usage? Then we should
>>>> investigate further.
>>>>
>>>
>>> I just checked out HEAD and run mvn test with my setup:
>>>
>>> D:\Entwicklung\workspaces\commons\lang>mvn -version
>>> Apache Maven 3.0.3 (r1075438; 2011-02-28 18:31:09+0100)
>>> Maven home: D:\Entwicklung\maven\3.0.3
>>> Java version: 1.7.0_01, vendor: Oracle Corporation
>>> Java home: D:\Entwicklung\Java\jdk1.7.0_01\jre
>>> Default locale: de_DE, platform encoding: Cp1252
>>> OS name: "windows 7", version: "6.1", arch: "amd64", family: "windows"
>>>
>>> I got the same result as Jörg. Gary, maybe you should have another look
>>> at that issue?
>>>
>>> Regards,
>>> Benedikt
>>
>> The tests also fail in Gump.
>> I've tweaked the assertion to try and help a bit.
>>
>> Had a possibly similar situation in JMeter recently; only the Gump build
>> failed. Turned out that the tests were being run in a different order by
>> the Gump JVM; this revealed a test bug which was failing to clear up data
>> from a previous test.
>
>
> We might set the runOrder to "random" in the surefire plugin. This helps to
> ensure that tests will not have unwanted side effects.

Good idea; I'll add that.
I've just added some checks to see that the data is not present before
the test starts.

BTW the JMeter test used Maven; it was the JVM which permuted the test order.

> - Jörg
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Mime
View raw message