>
>(c) include force in 0.20, default it to "false", but
>have a message in the release notes warning about the
>change.
>
>Technically, even if most users forgot to read the
>warning in (c) they would still be OK, as they would
>be happy if no time was spent with unnecessary
>regnerations.  However, due to internal links to other
>files that might have changed, etc.--which force does
>not capture, either here or in <xslt>--there could be
>a difference in functionality.

In those cases, the Ant <dependset> task can be used to capture those dependencies.  
(I believe that is the best way to handle this in Ant.)

>Thoughts?

Sorry for not responding sooner - I've been burning the midnight oil for a client this 
week and am behind on my email.

Although the patch I submitted was for (b), my preference would be for (c) for the 
reasons you state (conformance with <xslt> task and the most sensible long-term 
solution)

I would really like to see this in the 0.20 branch because I am using Ant + <xslt> + 
<fop> for a DocBook publishing toolchain which I would like to post to SourceForge or 
perhaps someday an Apache sub-project.  The idea is that Ant buildfiles can be used to 
build a (potentially large) set of documents with common components.  The Ant build 
would only rebuild the necessary pieces.  Since <xslt> and <fop> processing of large 
documents is measured in minutes eliminating unnecessary processing is a huge win.

The 1.0 release is at best several months away (right?) - so choice (c) provides a 
valuable feature to fop users in the meantime.  I am willing to help out by testing or 
responding to questions on the lists to help make (c) a reality.

Thanks, again, to the FOP group for this powerful tool.

Regards,

Sean


-- 
---------------------------------------------------------------------------
M. Sean Gilligan                    : mailto:[EMAIL PROTECTED]
Catalla Systems, Inc.               : 831-439-9568 x11
---------------------------------------------------------------------------

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]

Reply via email to