forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tim Williams" <william...@gmail.com>
Subject Re: resolving project locationmaps [was: Re: svn commit: r328847 - in /forrest/trunk: main/webapp/locationmap.xml main/webapp/sitemap.xmap site-author/content/xdocs/docs_0_80/locationmap.xml
Date Sat, 25 Feb 2006 13:35:23 GMT
On 2/18/06, Ross Gardler <rgardler@apache.org> wrote:
> Tim Williams wrote:
> > This breaks the locationmaps exists selector and now if the project
> > locationmap doesn't exist we get errors.
>
> By "errors" you mean in the log don't you? The actual site build does
> not fail does it, or at least it doesn't on any of the many sites I have
> using this since applying the patch in October.

Yeah, in the log, but I was noticing some weird stuff happening with
lm resolution in mounted lm's after that so I thought maybe that was
having some unintended side effects as well -- not sure at this point.

> I accept having the log error is misleading when it is not really an error.

I don't think it's misleading at all.  It's an error;)  We're trying
to generate a file that doesn't exist and that's not good.

>  > Maybe a better way to get
>  > what this is trying to do is to mount the dynamic locationmap from the
>  > project locationmap?
>
> -1
>
> This assumes that the dynamic locationmap is provided by the users site.
> This was added because some plugins, such as the Daisy plugin, need to
> generate the locationmap.
>
> We could expect users to know they need to add an entry into the
> locationmap, but I don't like expecting the user to have to do stuff to
> make things work. I'd rather look for a solution that means the select
> behaves as expected, i.e. if an optional locationmap is not present it
> simply ignores the fact.
>
> Ross

I wasn't expecting the users to do anything for plugins.  The plugin
locationmap would then mount the dynamically generated locationmap. 
So, Daisy plugin, which already has a static locationmap that gets
mounted, would simply mount a dynamically generated one.  I think this
is better than having these exceptions but I'm not convinced it's a
great solution either.

--tim

Mime
View raw message