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

Switch to Plain View
Hadoop, mail # general - why hadoop choose maven as its build tool?


+
smith jack 2012-05-16, 14:15
+
Roman Shaposhnik 2012-05-16, 14:31
+
Konstantin Boudnik 2012-05-17, 04:40
+
Steve Loughran 2012-05-17, 21:10
+
Keith Wiley 2012-05-17, 23:00
+
Jagane Sundar 2012-05-17, 21:23
+
Ted Dunning 2012-05-17, 22:15
+
Konstantin Boudnik 2012-05-17, 23:10
Copy link to this message
-
Re: why hadoop choose maven as its build tool?
Steve Loughran 2012-05-18, 00:48
On 17 May 2012 15:15, Ted Dunning <[EMAIL PROTECTED]> wrote:
> If you depend on released artifacts that you havent overridden then maven is deterministic.
>
> If you depend on snapshot artifacts then you get the latest snapshot which could change between builds.  But you asked for that.
true, but as trunk generates it's own snapshots the only way to be
100% sure that mvn isn't downloading the nightly build versions and
not your locally built stuff is to build with the -o setting. This
ensures that your local artifacts are the ones that get used, but also
mean that the moment you reach some state in maven that needs to
download 300 JARs for some plugin then you are in trouble.

But I digress.

Speaking of test reporting -what do you do to get good reports, where
reports mean: "human readable information needed to track down bugs
fast". I know I can get good stuff with Jenkins, and in theory could
bring up Jenkins nearby sharing git repositories and doing CI builds
of my own code, but I'd like something simpler such as the output of
"mvn test" to go into HTML files -with stack traces.I haven't got mvn
site to pull in the whole stack traces, which is what the IDE needs.