You are viewing a plain text version of this content. The canonical link for it is here.
Posted to log4cxx-user@logging.apache.org by Curt Arnold <ca...@apache.org> on 2005/05/10 16:13:41 UTC

Fwd: ant 1.6.3 breakage of cpptasks

FYI:

Begin forwarded message:

> From: Stefan Bodewig <bo...@apache.org>
> Date: May 10, 2005 2:26:31 AM CDT
> To: dev@ant.apache.org
> Subject: Re: ant 1.6.3 breakage of cpptasks
> Reply-To: Ant Developers List <de...@ant.apache.org>
>
>
> On Mon, 9 May 2005, Curt Arnold <ca...@apache.org> wrote:
>
>
>> I have had two reports in log4cxx of compile failures that seem to
>> be caused by changes in Ant 1.6.3,
>> http://marc.theaimsgroup.com/?l=log4cxx-user&m=111522597306665&w=2
>>
>
> Bug in DirectoryScanner that I'll look into.
>
>
>> and
>> http://marc.theaimsgroup.com/?l=log4cxx-user&m=111565145332704&w=2.
>>
>
> See <http://ant.apache.org/faq.html#delegating-classloader-1.6> the
> paragraph starting with "The wrapper skript".  This setup doesn't work
> in any 1.6.x version.  Don't place tasks on the CLASSPATH, use -lib
> instead.
>
>
>> Any theories why this was not caught by Gump?
>>
>
> The first one, because line 873 in DirectoryScanner.java says
>
>                 if ((myfile == null || !myfile.exists()) && ! 
> isCaseSensitive()) {
>
> and Gump currently only runs on Linux, which is case sensitive.  So
> this branch is never reached.
>
> The second one since Gump is the worst testbed you can imagine for
> classloader tests.  In order to make things work, Gump enforces a
> totally different classloader setup than "normal operations".
>
> Stefan
>



On May 10, 2005, at 2:37 AM, Stefan Bodewig wrote:
> On Tue, 10 May 2005, Stefan Bodewig <bo...@apache.org> wrote:
>
>> On Mon, 9 May 2005, Curt Arnold <ca...@apache.org> wrote:
>>
>>
>>> I have had two reports in log4cxx of compile failures that seem to
>>> be caused by changes in Ant 1.6.3,
>>> http://marc.theaimsgroup.com/?l=log4cxx-user&m=111522597306665&w=2
>>>
>>
>> Bug in DirectoryScanner that I'll look into.
>>
>
> fixed in CVS, will be fixed in 1.6.4.
>
> Stefan