[ 
https://issues.apache.org/jira/browse/GIRAPH-413?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Nitay Joffe updated GIRAPH-413:
-------------------------------

    Description: 
Until Facebook publishes jars for it's open sourced version of Hadoop to some 
maven repositories, we can just include the built jar in our 
codebase directly. This removes the need for the whole -Dhadoop.jar.path 
whatnot and lowers the barrier for using this profile.

Also now the following works:

>> dev-support/for-each-profile.sh clean install

Rebuilds current codebase on every profile. Some profiles are still failing in 
formats because they don't work with the given hadoop version, but the script 
now functions.

I also removed the -Dhadoop=xx option for profile activation. Is there a reason 
for having this instead of just having everyone consistently use -Phadoop_xx?

https://reviews.apache.org/r/7924/

  was:
Until Facebook publishes jars for it's open sourced version of Hadoop to some 
maven repositories, we can just include the built jar in our 
codebase directly. This removes the need for the whole -Dhadoop.jar.path 
whatnot and lowers the barrier for using this profile.

Also now the following works:

>> dev-support/for-each-profile.sh clean install

Rebuilds current codebase on every profile. In fact this is how I tested the 
patch :).

I also removed the -Dhadoop=xx option for profile activation. Is there a reason 
for having this instead of just having everyone consistently use -Phadoop_xx?

https://reviews.apache.org/r/7924/

    
> Make building Facebook profile easier
> -------------------------------------
>
>                 Key: GIRAPH-413
>                 URL: https://issues.apache.org/jira/browse/GIRAPH-413
>             Project: Giraph
>          Issue Type: Improvement
>            Reporter: Nitay Joffe
>            Assignee: Nitay Joffe
>         Attachments: GIRAPH-413.patch, hadoop-0.20.1-dev-core.jar
>
>
> Until Facebook publishes jars for it's open sourced version of Hadoop to some 
> maven repositories, we can just include the built jar in our 
> codebase directly. This removes the need for the whole -Dhadoop.jar.path 
> whatnot and lowers the barrier for using this profile.
> Also now the following works:
> >> dev-support/for-each-profile.sh clean install
> Rebuilds current codebase on every profile. Some profiles are still failing 
> in formats because they don't work with the given hadoop version, but the 
> script now functions.
> I also removed the -Dhadoop=xx option for profile activation. Is there a 
> reason for having this instead of just having everyone consistently use 
> -Phadoop_xx?
> https://reviews.apache.org/r/7924/

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to