ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From James Fuller <jim.ful...@ruminate.co.uk>
Subject Re: FYI: document the code
Date Tue, 02 Aug 2005 10:49:17 GMT
Kev Jackson wrote:

> James Fuller wrote:
>
>> Jan.Materne@rzf.fin-nrw.de wrote:
>>
>>  
>>
>>> You know that we (for several months :) are discussing about
>>> autogenerating the manual.
>>> I think we should have an eye on one of the Apache Google Summer of
>>> Code
>>> projects - RefDoc.
>>> Maybe we could reuse parts or the whole of it.
>>>
>>> See
>>> http://marc.theaimsgroup.com/?l=forrest-dev&m=112289337320271&w=2
>>> http://svn.apache.org/repos/asf/cocoon/gsoc/rgraham/refdoc/
>>>
>>>
>>>   
>>
>> +1
>>  
>>
> definitely +1
>
>> my only worry would be any dep on something as big as cocoon.... (though
>> current experience using native xml db eXist has been quite good, which
>> also uses cocoon)
>>
>>  
>>
> [OT]
> What's the database like?  I've never used a native XML format db
> (although I heard that DB2 has some pretty good native XML storage
> abilities)

eXist is great, also has a suite of Ant tasks (a bit flaky though
getting better).

>  :)

I think the approach needs to take into account several potential
upstream routes; xdoclet definately being one of them.

so the process could be;

Step 1: aggregate source

[source (xhtml)
source (docbook)
source (xdoclet)
source (rss/atom)
source (whatever xml)
....]

we could create a pluggable arch and just do those that we are
interested in...and leave it to others to implement if they want.

Step 2: generate canonical Ant Doc format which should render something
in the browser so we dont have to go another transform step

Step 3: XSLT generates final format

what u think?

--jim


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org
For additional commands, e-mail: dev-help@ant.apache.org


Mime
View raw message