forrest-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ross Gardler <>
Subject Re: xinclude
Date Wed, 06 Sep 2006 15:40:10 GMT
Jim Dixon wrote:
> On Wed, 6 Sep 2006, Jim Dixon wrote:
>>Yes. If I set
>>  forrest.validate.xdocs=false
>>in, then the 'forrest site' works.
>>That is, this is indeed a simple workaround, for which I thank you.
>>However, bear in mind that 'forrest run' did not fail.  Also, 'forrest
>>validate-xdocs' returns
>>implying that there is an inconsistency in the way in which validation
>>is applied.  For exactly the same configuration
>>  forrest run            succeeds
>>  forrest validate-xdocs succeeds
>>  forrest site           fails
>>Looks like a bug to me.
> Correction: 'forrest validate-xdocs' fails only if forrest.validate.xdocs
> is set to false.  If it is set to true, or just commented out, validation
> fails.

Yes, that is all correct. It's not a bug but a "feature". Validating 
when doing a forrest run is irrelevant since we can edit the docs after 
the initial ru and they could then be invalid.

> This remains uncomfortable, however.  There appears to be no way to use
> XInclude without introducing a new DTD.

That is correct. You can use a DTD that allows the xi:include stuff, or 
we could add it to the XDoc DTD. I think the reason we never added it 
was because we have always intended on moving the internal format over 
to a subset of XHTML2, we would add it at this point.

However, to date we have not done either.

Patches to our DTDs are welcome.


View raw message