[ 
https://issues.apache.org/jira/browse/CLIMATE-836?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15398428#comment-15398428
 ] 

Alex Goodman edited comment on CLIMATE-836 at 7/28/16 11:38 PM:
----------------------------------------------------------------

Hi [~lewismc],

I personally think it would be better to have it set up this way rather than 
using a channel with my own name of course, but do you know the logistics of 
setting up a channel with apache affiliation? I think it might be something 
other python apache projects would be very interested in as well, but then you 
also have to address the question of access (ie, share the channel password to 
committers for the project or use a shared access token, see the install 
documentation on the wiki for more info).

https://cwiki.apache.org/confluence/display/CLIMATE/Installing+OCW+using+the+conda+package+manager#InstallingOCWusingthecondapackagemanager-Maintainingthecondapackages(foradvancedusers))

As for documentation on maintaining the packages, I have already written 
information in the bottom half of the wiki page. It might be worthwhile for 
either you or one of your summer students ([~omkar20895] [~jarifibrahim]) to go 
through the process of at least building the packages from the recipe files. It 
should be noted that a few of the dependencies (most notably pyesgf) are not 
located in the main anaconda channels. Every dependency in a conda recipe must 
also have a conda package, so keeping those dependencies up to date will be our 
responsibility too if needed. This isn't a very difficult thing in practice, 
but just another thing to be aware of too. 

Let me know if you have any other questions.


was (Author: agoodman):
Hi [~lewismc],

I personally think it would be better to have it set up this way rather than 
using a channel with my own name of course, but do you know the logistics of 
setting up a channel with apache affiliation? I think it might be something 
other python apache projects would be very interested in as well, but then you 
also have to address the question of access (ie, share the channel password to 
committers for the project or use a shared access token, see the install 
documentation on the wiki for more 
info).[https://cwiki.apache.org/confluence/display/CLIMATE/Installing+OCW+using+the+conda+package+manager#InstallingOCWusingthecondapackagemanager-Maintainingthecondapackages(foradvancedusers))

As for documentation on maintaining the packages, I have already written 
information in the bottom half of the wiki page. It might be worthwhile for 
either you or one of your summer students ([~omkar20895] [~jarifibrahim]) to go 
through the process of at least building the packages from the recipe files. It 
should be noted that a few of the dependencies (most notably pyesgf) are not 
located in the main anaconda channels. Every dependency in a conda recipe must 
also have a conda package, so keeping those dependencies up to date will be our 
responsibility too if needed. This isn't a very difficult thing in practice, 
but just another thing to be aware of too. 

Let me know if you have any other questions.

> Transition conda packaging to Apache and add to release documentation
> ---------------------------------------------------------------------
>
>                 Key: CLIMATE-836
>                 URL: https://issues.apache.org/jira/browse/CLIMATE-836
>             Project: Apache Open Climate Workbench
>          Issue Type: Bug
>          Components: conda, documentation
>            Reporter: Lewis John McGibbney
>             Fix For: 1.2.0
>
>
> Hi [~agoodman], what are your thoughts on transitioning the packaging over to 
> Apache such that installation would be
> {code}
> conda install -c apache ocw
> {code}
> ???
> Also, regardless, we need to add maintenance and updating the conda recipe to 
> the release management documentation.
> BTW, I've just performed an install on a clean system and it all works very 
> well. Good work!



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to