On Thu, Aug 2, 2012 at 12:50 PM, Matthias Friedrich <[email protected]> wrote:
> Hi,
>
> see below.
>
> On Wednesday, 2012-08-01, Josh Wills wrote:
>> On Wed, Aug 1, 2012 at 12:16 PM, Patrick Hunt <[email protected]> wrote:
> [...]
>>> 1) when do you think you'll be ready for a release?
>
>> Among issues outstanding, I think we need to fix these prior to release:
>
>> https://issues.apache.org/jira/browse/CRUNCH-25 (Conform the site and
>> build artifacts to Apache rules)
>> https://issues.apache.org/jira/browse/CRUNCH-4 (Add Hadoop 1.0/2.0
>> profiles and remove CDH dependencies)
>> https://issues.apache.org/jira/browse/CRUNCH-16 (Fix dependency versions)
>
>> CRUNCH-25 has a patch and I just missed it, so we need to update it to
>> work with the current head. CRUNCH-4 is currently blocked on the
>> release of HBase 0.94.1, which will be the first HBase release that
>> will work with Hadoop 2.0. I don't have a feel for how much work
>> CRUNCH-16 depends on CRUNCH-4 in my mind, and am more ambivalent about
>> blocking the release on it since I don't know how much work it
>> involves.
> [...]
>
> I'll have a look at CRUNCH-16 if someone tells me which Hadoop version
> we're going to use. There's also the issue of fixing Scrunch's
> namespace.

It will be Hadoop 1.0.3 and Hadoop 2.0.0-alpha.

>
> Apart from these issues we need:
>
>   * Work through Apache and Incubator release guidelines
>   * Write a release guide for Crunch
>   * Set up a distribution directory with a KEYS file
>   * Decide on a branching model and document it
>   * Configure the POM to
>     * create a source distribution
>     * make release via maven-release-plugin possible
>     * make deployment to Nexus possible
>   * Nominate a release manager
>
> Being able to publish the website is probably also a good idea (see
> INFRA-5092).

+1.

>
> I think writing an Apache-compliant release guide will be the most
> time consuming task.
>
> Regards,
>   Matthias



-- 
Director of Data Science
Cloudera
Twitter: @josh_wills

Reply via email to