You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@forrest.apache.org by Ross Gardler <rg...@apache.org> on 2005/07/08 18:16:48 UTC

Removing docs duplication in SVN

Assuming that the locationmap merge is now complete and there aren't any 
other hidden surprises in there I think we should consider using this 
new feature to remove the duplicated docs in our SVN.

During the docs restructuring it became necessary to duplicate earlier 
versions of the docs in site_author. However, these files should never 
be changed in trunk, rather in the release branch where they already 
exist. Using the locationmap we can have Forrest retrieve the 
documentation for earlier versions from the relevant SVN branch.

There is a demo for how this is done in the 0.8 docs (I only just 
committed it so it won't be live yet, see the locationmap.xml in 
site-author)

This demo uses the viewSVN interface to retrieve the head version of the 
document. However, now I think about it there is no need to do this 
since SVN operates over http anyway, we can (I think) retrieve it 
directly from SVN. Nevertheless, the demo serves as a good example of 
how this could be done.

WDYT?

Ross

Re: Removing docs duplication in SVN

Posted by David Crossley <cr...@apache.org>.
Ross Gardler wrote:
> Assuming that the locationmap merge is now complete and there aren't any 
> other hidden surprises in there I think we should consider using this 
> new feature to remove the duplicated docs in our SVN.
> 
> During the docs restructuring it became necessary to duplicate earlier 
> versions of the docs in site_author. However, these files should never 
> be changed in trunk, rather in the release branch where they already 
> exist. Using the locationmap we can have Forrest retrieve the 
> documentation for earlier versions from the relevant SVN branch.
> 
> There is a demo for how this is done in the 0.8 docs (I only just 
> committed it so it won't be live yet, see the locationmap.xml in 
> site-author)
> 
> This demo uses the viewSVN interface to retrieve the head version of the 
> document. However, now I think about it there is no need to do this 
> since SVN operates over http anyway, we can (I think) retrieve it 
> directly from SVN. Nevertheless, the demo serves as a good example of 
> how this could be done.

I am going bush for the weekend so will comment on the rest later.

Using viewSVN will be very inefficient.
Would be better to go directly to the SVN server.
http://svn.apache.org/repos/asf/forrest/trunk/site-author/...

-David