Comments on: Decision to use Ivy is a no-brainer

by: The Path to Zen » Blog Archive » Ivy has lots of potential

Tue, 10 Apr 2007 02:16:07 +0000

[…] As Ivy has recently been taken under the Apache wings, and become an Apache Incubator project, I hold high hope on its codebase maturing rapidly and eventually graduates to be yet another great Apache tool. I believe as far as build management is concerned, Ivy offers a great opportunity to leverage and combine the two valuable assets the Java community has built over the years - the mature Ant toolset and the sophisticated ibiblio maven repositories. As Colin Sampaleanu pointed out, the decision to use Ivy is a no-brainer. […]

by: lf

Thu, 12 May 2005 07:56:59 +0000

dear sir:
I am very sorry to bother you , I am a student and now I want to write a paper about spring or J2EE but I only know little about them ,I first plan to write a paper about distributed transaction in spring ,but I found spring do well throught the integration of JTA ,so I was lost and does’t know which aspect I should write ,can you give me some constructive advice ? or give me some I interesting tile about spring that worth studying ,I know in spring you are very famous .your swift reply will be very much appreciated.Thank you very much !
Best regards.
lf

minvfeng@163.com

by: Matthieu Brouillard

Mon, 09 May 2005 12:16:28 +0000

Hi Graham,

yes you can do what your are describing with Ivy and you can more.
Ivy is very simple to use and also very powerfull and configurable to handle all the different configurations you can imagine.

Just take a look at the Ivy site : ivy.jayasoft.org, i think you will find some answers to your questions.

Do not hesitate to jump into Ivy world

by: Graham Lea

Thu, 05 May 2005 02:22:58 +0000

So, Ivy handles dependencies…
Let’s say I require commons-digester 1.4 (I’m making this up, BTW).
So I create a dependcy list containing just that.
Now, let’s say commons-digester 1.4 requires commons-logging 2.1.
I assume the purpose of Ivy is that it will also get commons-logging 2.1 for me, without me having to know that commons-digester needs it?

But what if I want different dependencies to what something describes for itself?
For example, what if I want commons-digester 1.4 to run against commons-logging 2.3?
Is it possible to do this kind of overriding with Ivy?

Thanks.

by: Marc Logemann

Mon, 02 May 2005 11:51:33 +0000

Yeah i followed some discussions in the ivy forum and you are correct with your solutions. I have some multi module projects with some dependecies of module A from module B, it seems that i can handle this too, because i saw that you have the same needs with Spring (Spring X module is depending on Spring-core etc.) So after all i will use it, but as it seems, it definitely takes some time to get into it when having more than just a flat project structure. I will follow your progress with the Spring project and ivy during the next weeks

by: Colin Sampaleanu

Sat, 30 Apr 2005 00:27:33 +0000

There are a couple of ways to handle what you describe. First of all, the normal pattern with ivy is to retrieve the dependencies to the lib dir, either right directly inside lib, or else to different subdirectories under lib which are named after the various configs (think compile-time vs test-time vs runtime, etc.). Then you just build classpaths in standard fashion for ant. So there’s nothing stopping you from also just having a
lib/static
dir which holds the jars which you are not going to retrieve though Ivy, and this directory gets checked in.

Now this is probably not the preferred form. As you say, it’d be better to have another repository somewhere with your non-standard jars, and have Ivy retrieve those from there. But this is not a big deal. A filesystem reposiroty is just a directory with the jars in a standard hierarchical format (the format is definable by you) underneath it. So if you want, if nothing else is going to be using those jars, you can have a ‘repository’ dir right inside your project. Then Ivy will get some jars from iBiblio, and some from that internal repository. What’s nice about this is that all dependencies, no matter where they come from, are listed in your ivy.xml file, and you get them shown in the reports, etc.

There is also nothing stopping you from structuring your build so that normally ivy doesn’t run at all, i.e. it’s a manual process to run ivy and pull stuff down into your lib dir. Then you check those in, and build normally as if the project wasn’t even using ivy. Now I wouldn’t really work like this myself, but it’s certainly viable.

by: Marc Logemann

Fri, 29 Apr 2005 23:37:25 +0000

I am following the progress of ivy for a few months now and i am quite thrilled but somehow i feel that i am losing something when putting my libs out of Subversion/CVS. Often i am working from different sites with different PCs. Right now its easy to just checkout my project and i get all the stuff i need. When using ivy, i assume i have to create my own public ivy repository, because i am not only using standard jars which can be possibly found in iBiblio but also more or less custom ones.

Perhaps i should check their maillist to find the answers that blow away my fears.

by: Jason Carreira

Fri, 29 Apr 2005 21:06:00 +0000

So I guess you’ve made up your mind now Nice write up.