Thanks again, Leo.

I am pretty sure that this jar (db2jcc.jar for AIX DB2 V9) does not have 
any dependencies of its own other that the Java SE java.sql package. The 
main benefit I am looking for is to stop carrying it around in every 
project lib directory that needs to access the database.

I would like to support disconnected builds so that developers (like me) 
can work off-line when needed. My understanding is that if I do a 
connected build before having to go off-line my dependencies should be 
available in the Gradle cache. If I am correct then repository 
(un)availability should be manageable. Having developers set up a local 
Ivy repository on their laptops is also a possibility, but that seems 
redundant with the Gradle cache so I am not pursuing that unless I learn 
differently.

So I think that the options you are presenting are:

A repository management tool - This is a worthwhile approach, but it seems 
like overkill for addressing this one (for now) file.
A Gradle build for deploying to the repository - This is the direction I 
was headed when I got frustrated and posted here. Any details or 
documentation on doing this would be appreciated.
A manual deployment - This is a direction I am willing to take too but got 
similarly frustrated trying to figure this out. Again, any details or 
documentation would be appreciated.

-Andy



Leonard Axelsson <[email protected]> 
04/05/2011 11:37 AM
Please respond to
[email protected]


To
[email protected]
cc

Subject
Re: [gradle-user] Placing an arbitrary file into Ivy






You're welcome.

Check out Nexus from Sonatype, its an open source version of their 
Repository manager. I've worked a little with it and it seems okay. 
There's also Artifactory from the JFrog guys, but I havent got any 
experience with that one.

To be honest I'm not sure what the best practice is in your case. There's 
probably other people on the list with more experience with this kind of 
problem, maybe they can answer what the best approach is.

To me it seems like having to maintain someone else artifact in your local 
Repository would mean extra work without getting much benefit. With other 
dependencies you can just go in and configure your repository to cache the 
dependencies from Maven Central but in the case of a custom artifact that 
you do not control yourself you'll either have to set up a project to 
deploy to your repository or just deploy it manually when you need to 
update it. That might not seem like a big deal, but it will mean that 
you're software can't build without the repository available which 
wouldn't be the case with all your other dependencies (unless you're on a 
restricted network, then my point is kinda moot in the first place).

Finally, both options should work just fine. You know more about your use 
case than I do so you're much better equiped to figuring out the what the 
best approach is in your case.

Regards,
/Leo

Reply via email to