Eric Sammer wrote:
> Tim Johnson wrote:
> > It seems to me that it would improve performance to be able
> > to create a "pipe" directly to tpage, but is that possible?
> 
> One could create a lightweight persistent perl daemon that 
> listens on a fifo for parameters and a template name that, 
> in turn, builds the resulting pages. In other words, rewrite 
> tpage (which is super simple) as a persist daemon that 
> handles rebuild requests.
> 
> Now, you have a "page builder" daemon that can be used by any 
> application.

Interesting, as I am about to embark on a similar project. I plan to create
a service that accepts input from a few different "data sources" -- XML,
tab-delim text, DB table, etc. and provides the data to one or more
templates.

Users will be able to create "packages" that map a data source to templates,
and also specify an output location (dir, ftp location, etc).

I plan to provide a few different ways to trigger the output: Manual
triggering via a web interface listing the packages with a "Publish" button,
an option to poll the data source on a regular basis and trigger output upon
change, and via some sort of API through which you can request a rebuild of
a particular package.

The idea is to a TT-based publishing engine that Java and Python
applications can use for output.

- Mark


-- 
Mark Thomas                    [EMAIL PROTECTED] 
Internet Systems Architect     DigitalNet, Inc. 


_______________________________________________
templates mailing list
[EMAIL PROTECTED]
http://lists.template-toolkit.org/mailman/listinfo/templates

Reply via email to