forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ross Gardler <>
Subject Re: plugin for DITA
Date Sun, 05 Dec 2004 14:59:02 GMT
scott hutinger wrote:
> Ross Gardler wrote:
>> Dave Brondsema wrote:
>>> Ross Gardler wrote:
>>>> Dave Brondsema wrote:
>>>>> Ross Gardler wrote:
>>>>>> scott hutinger wrote:


>>>>>> I would *guess* you need to:
>>>>>> DITA --> HTML --> XDoc --> Forrest site
>>>>>>      /\        /\       /\
>>>>>>      1         2        3
>>>>>> 1 would be the DITA XSLT's
>>>>>> 2 would be the .xsl
>>>>>> 3 is the normal Forrest stuff
>>>>>> 1 & 2 would be your DITA input plugin.
>>>>> 2 is part of forrest core so the DITA input plugin should only do 
>>>>> 1.  Right?


>>> What I wanted to make clear is that the input plugin can produce an 
>>> html document and that is enough.  It does not have to do the 
>>> conversion to our current intermediate format (xdoc).


> If one were to completely control the output, then the transformation to 
> html would be all one would need to do.  But to follow forrest, one 
> could output to xdoc format.  Wouldn't it be better to transform to xdoc?

In the past, yes, but what you may not realise is that we are planning 
on moving the internal Forrest format to XHTML in the next release.

Some (most?) projects provide stylesheets that convert to HTML from 
their internal formats. These stylesheets will be quite complete in most 
cases (in a previous thread we were led to believe that this is the case 
with DITA, but I have not checked this out). If the DITA stylesheets are 
suitable for creating just the content part of a Forrest page (i.e. no 
navigation etc.) then it would make sense to use them as there would be 
much less maintenance for us to do on the plugin.

We already support HTML as a source format so there is little work needs 
to be done to use the output from DITA's documents. There will be some 
tweaking needed of this functionality in Forrest as it has not yet been 
used in a wide variety of formats, but it makes sense since once it is 
fully working we can support any input format that has stylesheets to 
convert to (X)HTML.

Doing it the other way, converting to XDoc, will only be useful for 
DITA. For much more on this topic see the discussion about supporting 
full Docbook (very long thread)

However, if you know something about the DITA stylesheets that we don't 
know you may be right, as I said this is the *generic* answer.


View raw message