On Tue, 7 Oct 2003, Jan Materne <[EMAIL PROTECTED]> wrote:
> Maybe the JDepend classes are not on that Gump installation?
Even then a manifest-only jar should be created.
Oops, looks as if I had accidently removed jdepend when I introduced
the optional-jar macro.
Stefan
---
On Thu, 11 Sep 2003, Davanum Srinivas <[EMAIL PROTECTED]> wrote:
> In case you are not yet tired of this :)
should I? 8-)
> - Am using the same ant jars ("Ant version 1.6alpha compiled on
> September 6 2003") picked up from nightly build area fails on nagoya
> (SunOS) and works on win2k.
Just
Stefan,
In case you are not yet tired of this :)
- Am using the same ant jars ("Ant version 1.6alpha compiled on September 6
2003") picked up from
nightly build area fails on nagoya (SunOS) and works on win2k.
- Checked in your suggested fix
(http://marc.theaimsgroup.com/?l=axis-dev&m=10632349
On Wed, 10 Sep 2003, Davanum Srinivas <[EMAIL PROTECTED]> wrote:
> Can you please explain why on WIN2K, both 1.5.X and CVS HEAD both
> work fine?
1.5.x works on both platforms as most bugs in Ant are cross-platform,
too.
Maybe you've picked up a CVS HEAD version after Peter Reilly has
committed
Stefan,
Can you please explain why on WIN2K, both 1.5.X and CVS HEAD both work fine?
(and we see the
failure only on nagoya?). If this was the problem, then we'd see failures on
both platforms.
Thanks,
-- dims
--- Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> Sorry for coming back contardicting
On Wed, 10 Sep 2003, peter reilly <[EMAIL PROTECTED]> wrote:
> Should I roll back the change?
Not for me, I think this fixes a bug (but may be worth a note in the
"could break your build" section).
Others?
Stefan
-
To unsubscr
On Wednesday 10 September 2003 15:40, Stefan Bodewig wrote:
> On Wed, 10 Sep 2003, peter reilly <[EMAIL PROTECTED]> wrote:
> > I did not really understand the original code,
>
> I think setProject has initially been package private so recreating
> the Property tasks from the new project has been th
On Wed, 10 Sep 2003, peter reilly <[EMAIL PROTECTED]> wrote:
> I did not really understand the original code,
I think setProject has initially been package private so recreating
the Property tasks from the new project has been the only option then.
> The code is a bit strange, it would have been
On Wednesday 10 September 2003 13:28, Stefan Bodewig wrote:
> On 10 Sep 2003, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> > (I'll soon commit a version that should make Axis happy for the next
> > - i.e. tomorrow's - Gump run).
>
> I could use some help here.
>
> The Axis build works if I back out
Sorry for coming back contardicting my previous mails again.
On 10 Sep 2003, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> I'd say the Axis build works with Ant 1.5.x by accident
It relies on a bug that has been fixed[1].
> (I'll soon commit a version that should make Axis happy for the next
> -
On 10 Sep 2003, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> (I'll soon commit a version that should make Axis happy for the next
> - i.e. tomorrow's - Gump run).
I could use some help here.
The Axis build works if I back out Peter's change for Ant.java's
revision 1.84 - and I don't have a good g
On 10 Sep 2003, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> Ant 1.5.x will use the last one to finally set the property
Not really true. Even 1.5.4 uses the first definition if you use
multiple tags in Ant. I'd say the Axis build works with
Ant 1.5.x by accident (I'll soon commit a version tha
On Mon, 08 Sep 2003, Steve Loughran <[EMAIL PROTECTED]> wrote:
> FYI: a recent change (few weeks back?) is breaking Axis in its
> iterative build (uses a descendent of ant-contrib's foreach)
I think I've tracked this down, finally.
target compileJunit in buildTest contains a task that under
the
On 10 Sep 2003, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> I'll try to get Gump into using 1.5.4 next, to see whether it really
> is a problem with a recent change in Ant.
At least it gets past the compilation - it is running the tests know
which is known to take ages 8-)
Next is a line by line
so far I haven't found anything too helpful while trying to debug the
situation, but ...
On Tue, 09 Sep 2003, Steve Loughran <[EMAIL PROTECTED]> wrote:
> Dims says:
>
> one more data point (nagoya, jdk13, ant-nightly):
> "ant -f buildSamples.xml compile" (FAILS - compiles same build.xml
> many t
On Tue, 09 Sep 2003, Steve Loughran <[EMAIL PROTECTED]> wrote:
>> OK, I've found a SOAPStruct.java in interop3/groupE.
>
>> Could it be that your compilation depends on the order of the files
>> in a fileset?
>
> Anything is possible, but I think maybe some other change w/
> classloading, vms, o
Stefan Bodewig wrote:
On Mon, 08 Sep 2003, Steve Loughran <[EMAIL PROTECTED]> wrote:
FYI: a recent change (few weeks back?) is breaking Axis in its
iterative build (uses a descendent of ant-contrib's foreach)
fails as it cannot find SOAPStruct and in fact a
find xml-axis -name \*SOAPStruct.class
On Mon, 08 Sep 2003, Steve Loughran <[EMAIL PROTECTED]> wrote:
> FYI: a recent change (few weeks back?) is breaking Axis in its
> iterative build (uses a descendent of ant-contrib's foreach)
fails as it cannot find SOAPStruct and in fact a
find xml-axis -name \*SOAPStruct.class
comes up empty
FYI: a recent change (few weeks back?) is breaking Axis in its iterative
build (uses a descendent of ant-contrib's foreach)
Davanum Srinivas wrote:
Tried both j2sdk-1.3.1_08 and j2sdk-1.4.1_03 on nagoya. Problem occurs on both
with latest ANT
nightly.
-- dims
--- Davanum Srinivas <[EMAIL PROTECT
19 matches
Mail list logo
