PyTML / News: Recent posts

Well PyTML is still alive, and hasn't died at all. It is undergoing some intense reworking of the internals, and is becoming a quality application. It has grown with my abilities, and while I see constant improvement, I realize my earlier releases were kinda crap. But thats not the point, nor the focus. So many things have changed, and its continuing to just get better and better. Till then {oh, features to stay tuned for: Session Objects (done), Wide Array of Server Integrations (Partial [fcgi, cgi, mod_python]), reworking of templating system (done), optimizations of code, and possibly even some rewrites in C, or C++ (partial), data abstraction layer for allowing of relational mapping w/o using a standard platform (in process) }, okay, cya then.

There has been a lot of work on pytml lately, and it has developed a lot more into the web framework direction. After a summer of building a CMS in php, I learned a lot about modularity, and what is possible with good code. PyTML is not dead, and it is still being developed. The framework is now much more modular, and adheres much more closely to pythonic programming, along with the fact that the framework is being moved to a total MVC (model-view-controller). There has been almost a complete revamp of all the underlying code, which is leading to performance improvements, and overall feature enhancements. More than being just a templating system with a scripting engine, the system is growing much closer to a real framework, and after the conversion to total MVC, a new release will be out. Just you wait...

Some delays have been incurred as my time has been allocated to another project, in Orlando Florida. Fear not, my project is still continuing to grow, and features are still being added, revised, revamped, and replaced. Look forward to beta 3, as I know I am.

Beta 3 is almost out, as the listManager library is being perfected. This listManager is almost like a simple xml transformation block, but it also gives the ability to filter/modify/add/delete xml sections. You can use the xml file as an interactive data object( ie sql db). This means you can have dynamic interactive content w/o the use of a DB, and since all the files are cached, the only performance hit is with recaching the file after it is changed.

A few minor bug fixes, interfaced with mod_python or regular cgi, and new data object wrapper.

I recently tried to modify the system to be compatible with mod_python, and it resulted in unhappy events. Apparently objects persist across page calls, and with my system being as dynamic as it is, the module data is loaded into a varible and then processed from there. This opens up problems for the data not being flushed out, and multiple pages being written over, and everything goes haywire. If any mod_python programmers are out there I would ask for someone to look at the code, and see if they cannot come up with a way to allow this system to run with mod_python, and not freak out.

This is the initial release of PyTML after about a month and half of development. The framework came about as a tool created for personal and use, and for project use. It has finally reached a level of stability and maturity that warrented its release into the Open Source Community.