Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hadoop >> mail # dev >> Developing cross-component patches post-split

Copy link to this message
Re: Developing cross-component patches post-split
Giridharan Kesavan wrote:
> I agree with you on this;
> We have some common places on people server
> people.apache.org/repo and people.apache.org/repository but I've not seen any ivy artifacts being published there.

Writing things there causes them to be published to repo{1,2}.maven.org.


For example, I copied the Avro release files to:


and 24 hours later they are at:


where they can be used as Maven and Ivy dependencies and found by Maven
search engines:


FWIW, I used Ivy's makepom Ant task and Ant's checksum Task to create
these files, then ran gpg manaually to sign them, as with normal Apache
releases, then used scp to post them to people.apache.org.  I created
the maven-metadata.xml files by hand, although I'm not sure they're
required.  YMMV.

I have not used the snapshot repositories, but they should behave similarly.