I too use ant for my builds. It does not compile every file every time. However, admittedly there have been times when compile errors occur which only go away upon a full rebuild. This doesn't happen often but does happen often enough that its annoying. Ron Denis E-mail : [EMAIL PROTECTED] Consultant Lucent Technologies Inc. > -----Original Message----- > From: Mark Bucciarelli [mailto:[EMAIL PROTECTED]] > Sent: Monday, March 19, 2001 1:49 PM > To: Paul Kinnucan; [EMAIL PROTECTED] > Cc: Stephane; Richard den Adel; JDE Mailing List > Subject: RE: PROPOSAL: New Java Build Feature > > > > -----Original Message----- > > From: Paul Kinnucan [mailto:[EMAIL PROTECTED]] > > Sent: Monday, March 19, 2001 1:23 PM > > To: [EMAIL PROTECTED] > > Cc: Stephane; Richard den Adel; JDE Mailing List > > Subject: Re: PROPOSAL: New Java Build Feature > {snip!} > > > >> In any event, my proposal can be combined with make or ant. All I > > >> am suggesting is that instead of trying to compile every class in > > >> your project, ant or make could compile only one: a compile > > >> master that references every class in your project. The Java > > >> compiler would then make sure that all files that needed to be > > >> compiled would be compiled. I think this approach, which I did > > >> not originate, is brilliant. > > I'm going to jump in here. I get the sense there's been a fair amount > of back and forth here on this proposal. I searched the archives > looking for how to use Ant with JDE and read five or six posts > regarding Paul's proposal, so perhaps there are others that have not > yet been posted there ... but anyway, here goes ... > > I'm using Ant 1.3 and the <javac> task only compiles the classes that > need recompiling. Perhaps they have over-ridden the behavior of > javac--if you're curious, you can take a look at the source for the > javac task on Apache's webCVS page. > > I have been using Ant in conjunction with JUnit 3.5 for the past two > weeks and I can't recommend it strongly enough. > > I am starting to think about my classes so they are "testable" which > means I think about the interface first and foremost. A side-effect > is that I try to minimize the interface so I don't have to write many > test methods. > > I'm starting to write my test code first, and then the production > code. When the production code passes the test, I'm done. I've > become much more aggressive in refactoring code that I have tests > written for; I am (fairly :) confident that if it passes the tests, > then I'm done. > > I would suggest using the functionality built into Ant for dependency > checking when building. Again, sorry if this rehashes stuff that has > already been covered in gory detail, but I just signed (back) on to > the list today. >
RE: PROPOSAL: New Java Build Feature
Denis, Ronald J (Ronald)** CTR ** Mon, 19 Mar 2001 11:32:51 -0800
- RE: PROPOSAL: New Java Build Feature Paul Kinnucan
- Re: PROPOSAL: New Java Build Feature Kevin A. Burton
- Re: PROPOSAL: New Java Build Feature Kevin A. Burton
- Re: PROPOSAL: New Java Build Featur... Paul Kinnucan
- RE: PROPOSAL: New Java Build Feature Schewe, Jon (MN65)
- RE: PROPOSAL: New Java Build Featur... Paul Kinnucan
- Re: PROPOSAL: New Java Build Feature Paul Kinnucan
- RE: PROPOSAL: New Java Build Feature Schewe, Jon (MN65)
- RE: PROPOSAL: New Java Build Feature Sandip Chitale
- RE: PROPOSAL: New Java Build Feature Mark Bucciarelli
- RE: PROPOSAL: New Java Build Feature Denis, Ronald J (Ronald)** CTR **
- RE: PROPOSAL: New Java Build Feature Schewe, Jon (MN65)
- RE: PROPOSAL: New Java Build Feature Paul Kinnucan
- RE: PROPOSAL: New Java Build Feature Jeff Rancier
- Re: PROPOSAL: New Java Build Feature eric
- Re: PROPOSAL: New Java Build Featur... Raffael Herzog
- RE: PROPOSAL: New Java Build Feature Denis, Ronald J (Ronald)** CTR **
- Re: PROPOSAL: New Java Build Feature Nic Ferrier
