Return-Path: Delivered-To: apmail-jakarta-ant-user-archive@apache.org Received: (qmail 62528 invoked from network); 13 Mar 2002 23:45:56 -0000 Received: from unknown (HELO nagoya.betaversion.org) (192.18.49.131) by daedalus.apache.org with SMTP; 13 Mar 2002 23:45:56 -0000 Received: (qmail 27928 invoked by uid 97); 13 Mar 2002 23:45:55 -0000 Delivered-To: qmlist-jakarta-archive-ant-user@jakarta.apache.org Received: (qmail 27902 invoked by uid 97); 13 Mar 2002 23:45:55 -0000 Mailing-List: contact ant-user-help@jakarta.apache.org; run by ezmlm Precedence: bulk List-Unsubscribe: List-Subscribe: List-Help: List-Post: List-Id: "Ant Users List" Reply-To: "Ant Users List" Delivered-To: mailing list ant-user@jakarta.apache.org Received: (qmail 27891 invoked from network); 13 Mar 2002 23:45:54 -0000 Message-ID: <057b01c1cae9$54dbb480$b81c570f@cv.hp.com> From: "Steve Loughran" To: "Ant Users List" References: Subject: Re: My experiences with doing a large project with ANT Date: Wed, 13 Mar 2002 15:46:41 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 X-MailScanner: Found to be clean X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N ----- Original Message ----- From: "Jim Jackl-Mochel" To: "Ant-User" Sent: Wednesday, March 13, 2002 12:26 Subject: My experiences with doing a large project with ANT > Introduction > ------------ > > Some people on this list have asked me to post details of how I got a > hierarchical ant based build system working for a project I am involved in. > This > is a simple attempt to get most of my experiences down in a way that might > be > useful to the ANT community at large. For brevity, I am intentionally > leaving > out most of the numerous false starts and blind paths I went down. I am > detailing out the general rationales and notable blind alleys whenever I see > a chance to save someone else a great deal of time. hey, this is good...stick it on a web page and we will point to it from the ant web pages > Developers can no longer do a build at any level they want. > Most have no problem with this because the speed is overall > much better. this is a hard one...20 minutes is still slow for immediate feedback. > > It does the job but there are small things within ANT that > required me to hack things in various ways. > > Logging of what happens is stil somewhat ungainly and ugly. > > We are somewhat at a loss when we want to choose whether or not > a failed test or a failed deployment constitutes a faiilure of the build. > Since we label a source tree based on a successful build I would love to > have (for any target) the ability to specify whether or not a failure of > this > target causes the build to stop. Sometimes I use conditions to probe for things working; you can use this to write rigorous tests that things worked (like a JAR file contains named classes and properties) > The fact that JAR files cannot be updated easily causes a world of > workarounds. you mean updated while loaded, or jar -u isnt working? > > The ability to pass in a bundle of configuartion info (a properties file?) > into the build command line would help we have some good news there from the nightly builds: -propertyfile load all properties from file with -D properties taking precedence -steve -- To unsubscribe, e-mail: For additional commands, e-mail: