forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Steven Noels <stev...@outerthought.org>
Subject Re: Forresbot WANTED
Date Fri, 29 Nov 2002 08:16:35 GMT
Jeff Turner wrote:

> On Thu, Nov 28, 2002 at 09:33:07PM +0100, Steven Noels wrote:
> 
>>Jeff Turner wrote:
>>
>>
>>>Typing 'forrestbot' in src/resources/forrestbot/samples should now
>>>generate the following sites from CVS:
>>>
>>> &forrest-site;
>>> &forrest-template;
>>> &xindice-site;
>>> &xml-site;
>>> &fop-site;
>>>
>>>Please remember to change the 'email' entity first :]
>>
>>Stupid idea to bypass the 'cvs-only' policy to update the websites 
>>hosted at daedalus:
>>
>>1) trigger forrestbot using a cron job (dead easy)
>>2) cvs import the generated files into a repository on cocoondev.org
>>3) nuke the existing content of the live website @ daedalus for 
>>volunteering projects
>>4) cvs checkout the sites stored in the cocoondev.org repository to the 
>>correct dirs on daedalus (e.g. /x2/xml.apache.org/forrest/)
>>
>>next run:
>>
>>5) rm -Rf the modules from the cocoondev.org repository (since I don't 
>>get repeat cvs imports working - but if Nicola says it _should_, this 
>>means I'm stupid)
>>6) forrestbot kicks in
>>7) cvs import generated files into cocoondev.org repository
>>
>>in the meantime, on daedalus:
>>
>>8) Sam's script kicks in, cvs updating now the updated files from the 
>>cocoondev.org repository
> 
> 
> Okay.. but why does it have to be a cocoondev.org repository?  Can't we
> just delete and then recreate the xml-site/targets/<project> directory?

Oh - no strong reason other than moving all cron- & shell 
manipulation-based stuff away from daedalus. You are right that my 
solution requires an extra step.

What do you think: should forrestbot run on cocoondev.org? Some ASCII-art:

  +---------+               +------------+                 +---------+
  | sources |    <---->     | forrestbot |      <----->    | results |
  +---------+     Pull      +------------+       Pull      +---------+

    icarus                     cocoondev                     icarus (cvs)


                                                                ^
                                                                |
                                                                |
                                                                v

                                                           +---------+
                                                           | website |
                                                           +---------+

                                                            daedalus


In your scenario, where should be run the forrestbot? And how will the 
sync be made? I understand the general feeling at 
infrastructure@apache.org is that daedalus would need to pull updates 
instead of updates being pushed (using scp).

</Steven>
-- 
Steven Noels                            http://outerthought.org/
Outerthought - Open Source, Java & XML Competence Support Center
Read my weblog at              http://radio.weblogs.com/0103539/
stevenn at outerthought.org                stevenn at apache.org


Mime
View raw message