build.xml changes and unversioned jar fles [Re: Build System]

[Aleksander Slominski]

Anthony Elder wrote:

Peter, I agree with you the current setup with the scripts isn't so good,
and it would be great if you could become more involved with WSIF and give
us a fresh perspective on things. One pain with the classpath script is
that it sets up the classpath relative to the wsif root,  so as soon as you
change out of the wsif directory it doesn't work. 

please do fix it but i do not know how to do this on windows ... however 
this has really nothing to do with changing build.xml to work directly 
with ANT as when you compile with ANT you always do it from project top 
level directory. this can only matter when you run java programs?

Easy to fix but no body
has yet. Alek, how about holding off for a while and letting Peter help
with this?
i have modified build.xml so it can work directly with ant and tested it 
with ANT 1.5.2.

for hassle free building and running i would still recommend using build 
and run scripts (build.sh,build.bat,run.sh,run.bat) but for developers 
that know exactly what they need they can use now ant directly and 
classpath script to set classpath to execute java programs (such s 
samples etc.)

Peter and Ant: please take a look and do any other modifications if 
necessary.

also could somebody fix names of JAR files that have no version number 
like saaj and jaxrpc, commons logging and discovery (preferably the 
person who put them into lib/ and knows what versions they are...)? that 
can bite us later when revised jaxrpc/saaj versions are available.

thanks,

alek


      ...ant

Anthony Elder
[EMAIL PROTECTED]
Web Services Development
IBM UK Laboratories,  Hursley Park
(+44) 01962 818320, x248320, MP208.
"Peter G. Lane" <[EMAIL PROTECTED]> on 20/03/2003 17:32:30

Please respond to [EMAIL PROTECTED]

To:    [EMAIL PROTECTED]
cc:
Subject:    Build System


I was just curious about the reasons behind having scripts on top of the
ant build file.  I
personally would much rather have this done away with in favor of pure ant,
but I don't know if
there's something I don't know...you know?
If it's a matter of man power, I would be more than happy to do this.

Peter



 



--
"Mr. Pauli, we in the audience are all agreed that your theory is crazy. 
What divides us is whether it is crazy enough to be true." Niels H. D. Bohr