On Mon, Feb 04, 2002 at 10:04:14AM -0500, David B. Bitton wrote:
> Also, does
> anyone (Andy?) have any insight in regards to my earlier question in
> regards to not having theprocessed template file copied to the dest dir 
> after processing.  Tx :)

Nope, you stumped me there  :-)

I think someone already suggested it, but the approach I typically use 
is to have an index page which generates the individual data pages as it
gets processed. 

But if you don't want an index page, then it gets a little more tricky.
You could have a library component which generates the pages, but you 
still need to INCLUDE or PROCESS it from another page which is getting 
processed by ttree. 

The solution I'd like to see is for ttree (or its replacement) to be able
to define any number of different strategies for processing, copying, 
renaming templates, and so on.  For example a 'copy' policy to copy the
file only, a 'foobar' policy for processing a page with a particular set
of headers/footers, a 'wizbang' policy defining a different set of 
wrapper templates, a 'null' policy for processing a template but not 
generating any output (like you require), and so on.  Then, you could map 
templates to policies either by matching file regexen, looking at a META 
tag inside the template, and so on.  

At the moment, this is all kinda hard-coded into ttree and there's no easy
way to extend these kind of things with getting your hands quite dirty.


A



Reply via email to