What about using the SVN properties on the file(s) concerned? In
similar cases we use "svn:eol-style=LF" on files we want to be
unix-friendly, as we do all our development on Windoze.

Hope that helps with your particular situation.

On Wed, May 21, 2008 at 11:49 AM, sebb <[EMAIL PROTECTED]> wrote:
> On 20/05/2008, Oleg Kalnichevski <[EMAIL PROTECTED]> wrote:
>> On Tue, 2008-05-20 at 21:03 +0100, sebb wrote:
>>  > On 20/05/2008, Oleg Kalnichevski <[EMAIL PROTECTED]> wrote:
>>  > > On Tue, 2008-05-20 at 01:23 +0100, sebb wrote:
>>  > >  > It's about time (indeed overdue) for another JMeter release, so I've
>>  > >  > created JMeter 2.3.1 RC1 in the directory:
>>  > >  >
>>  > >  > http://people.apache.org/~sebb/jmeter-2.3.2/dist
>>  > >  >
>>  > >  > Site/Docs are here:
>>  > >  > http://people.apache.org/~sebb/jmeter-2.3.2/docs
>>  > >  >
>>  > >  > Tag:
>>  > >  > http://svn.apache.org/repos/asf/jakarta/jmeter/tags/v2_3_2RC1
>>  > >  >
>>  > >  > Keys are here:
>>  > >  > http://svn.apache.org/repos/asf/jakarta/jmeter/trunk/KEYS.txt
>>  > >  > also
>>  > >  > http://www.apache.org/dist/jakarta/jmeter/KEYS
>>  > >  >
>>  > >  > All feedback (and votes!) welcome.
>>  > >  >
>>  > >  > [ ]+1 - the release candidate is OK
>>  > >  > [ ]-1 - there is a problem (please indicate what it is)
>>  > >  >
>>  > >  > The vote will remain open for at least 72 hours.
>>  > >  >
>>  > >  > Note: If the vote passes, the intention is to release the archive
>>  > >  > files and create the release tag from the RC1 tag.
>>  > >  >
>>  > >
>>  > >
>>  > > Hi Sebastian
>>  >
>>  > Thanks for trying it out.
>>  >
>>  > >  JMeter compiles and builds just fine. All tests pass for me, but the
>>  > >  'ant test' terminates with a build failure.
>>  > >
>>  > >  BUILD FAILED
>>  > >  /home/oleg/temp/jakarta-jmeter-2.3.2/build.xml:1552: Files are not
>>  > >  identical.
>>  > >
>>  > >  Any idea what may be wrong?
>>  >
>>  > One of the tests uses two threads - the ramp up time is supposed to
>>  > ensure that the first thread finishes before the second starts, but
>>  > sometimes this does not happen, and the samples occur in the wrong
>>  > order.
>>  >
>>  > In my experience that is fairly rare, but perhaps the ramp-up needs to
>>  > be increased.
>>  >
>>  > If the comparison fails, there should be some BatchTestLocal.csv/.xml
>>  > files in the bin directory; these should agree with the same name
>>  > files in bin/testfiles.
>>  >
>>  > If you want to run it again without rebuilding everything, try
>>  >
>>  > ant batchtest
>>  > or
>>  > ant batchtestserver
>>  >
>>
>>
>> The batch test seems to be failing consistently for me. I did not get a
>>  single successful run out of 6.
>>
>
> Just tested it myself on Ubuntu (VMWare) and I see the same problem.
>
> Turns out the cause is that the CSV comparison file in testfiles has
> got the wrong line endings - CRLF instead of CR. This is because the
> tgz files were created on Windows.
> I removed the extra CR characters and the test ran fine.
>
> If the files are checked out from SVN on Linux it should also work OK,
> because the files are eol=native.
>
> I guess one solution is to ensure that all the eol=native files are
> converted to the "normal" line-ending for the archive, i.e. CRLF for
> zip and LF for tgz, though that does not solve the problem for MacOS
> users.
>
> I could fix the test case by updating the Ant script, but perhaps it
> would be better to ensure that the correct line endings were generated
> in the first place...
>
> Not quite sure how to fix this in general - any bright ideas anyone?
>
>>
>>  > If the test often fails on your system then clearly the test needs
>>  > fixing - what OS/Java/hardware are you using?
>>  >
>>
>>
>> Single core Intel Centrino CPU, 2GB
>>
>>  Ubuntu 8.04
>>
>>  Linux ubuntu 2.6.24-16-generic #1 SMP Thu Apr 10 13:23:42 UTC 2008 i686
>>  GNU/Linux
>>
>>  Sun JDK 1.5.0
>>
>>  java version "1.5.0_11"
>>  Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_11-b03)
>>  Java HotSpot(TM) Client VM (build 1.5.0_11-b03, mixed mode)
>>
>>  I do not think it is a blocker, but I would be nice to fix it.
>>
>>
>>  Oleg
>>
>>
>>
>>  > >  Oleg
>>  > >
>>  > >  > Here's my:
>>  > >  >
>>  > >  > +1
>>  > >  >
>>  > >  > ---------------------------------------------------------------------
>>  > >  > To unsubscribe, e-mail: [EMAIL PROTECTED]
>>  > >  > For additional commands, e-mail: [EMAIL PROTECTED]
>>  > >  >
>>  > >
>>  > >
>>  > >  ---------------------------------------------------------------------
>>  > >  To unsubscribe, e-mail: [EMAIL PROTECTED]
>>  > >  For additional commands, e-mail: [EMAIL PROTECTED]
>>  > >
>>  > >
>>  >
>>  > ---------------------------------------------------------------------
>>  > To unsubscribe, e-mail: [EMAIL PROTECTED]
>>  > For additional commands, e-mail: [EMAIL PROTECTED]
>>  >
>>
>>
>>  ---------------------------------------------------------------------
>>  To unsubscribe, e-mail: [EMAIL PROTECTED]
>>  For additional commands, e-mail: [EMAIL PROTECTED]
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to