Hi! Where I work we have a heavily restricted and extremely slow Internet access, so letting Ivy download the dependencies for us from the web is just not an option.
We download stuff at our houses, and then copy them in usb portable storage and bring them to work. The problem is that we did not use anything like Ivy, so everything was just copied in to the WEB-INF\lib folders.... Now we want to use Ivy to manage library dependencies better... So, we want to upload our WEB-INF\lib .jar files (for hibernate, seam, spring, commons, etc) in to a shared repository to a server inside our intranet... I guess that to do that we need to use the task ivy:publish... Am I right? But since we have hundreds of files, we would like to be able to do it in single shot, something like <ivy:publis-as-repository source-dir="WEB-INF\lib\"> And have Ivy read each of the MANIFEST.MF files inside each of the .jar files and "guess" the right repository structure by reading stuff like (from commons-beansutils.jar) Inside its MANIFEST.MF we can read: Implementation-Vendor: Apache Software Foundation Implementation-Version: 1.6 and generating a structure like shared-repository\apache\commons-beanutils\1.6\commons-beansutils.jar Is it possible to automate this with Ivy? or do I have to write my own ivy:publis-as-repository ant task? Thanks Regards, Francisco Jose Peredo -- View this message in context: http://www.nabble.com/private-repository-MANIFEST.MF-guessing-the-right-repository-structure-tp22687302p22687302.html Sent from the ivy-user mailing list archive at Nabble.com.
