On Thu, 2004-04-15 at 07:37, Alex Levy wrote:
On Thu, 2004-04-15 at 06:47, David Reid wrote:
Not Concerns of a split:
* Backwards compatibility.
Just to clarify: you don't mean changing twisted.web to break backwards-compatibility, but instead mean that apps written for twisted.web will not always work with this new module. Correct?
I do mean that the new twisted-web will NOT include woven, and things like distrib and vhosts will work differently. It will be written primarily to simply the usage of Nevow but keep the two seperate enough so that Nevow is not a dependency of twisted-web and that other style rendering engines (I guess that is what Nevow has become, far more than a templating engine, because it has changed how Requests act and how children are located and how pages are renderered.)
I don't see a major problem with splitting code, as long as the old twisted.web and twisted.web.woven modules remain (so that people can continue to use applications that've already been written).
The old modules will remain for a period with deprecation warnings, but this shouldn't be the responsibility of twisted-web. The people currently interested in this split seem to have never used woven, and never want to, so having to support it shouldn't be a responsibility.
It's been pointed out before that twisted.web is an app server, not a protocol implementation, and thus there is justification for moving it outside of Twisted. What do the gurus think?
twisted.web includes and appserver but also httpclient, which apprently isn't very good anyway. The twisted-web will be focused on being a useable and perhaps even efficient standalone webserver of static files and that can be easily handle integration with Nevow based applications.
Right now I make $6.75US/hr and go to school 2 days a week, i have probably as much free time as anyone else and I really really want this stuff to work.
I'm graduating in a month, with no job, so I may have a hell of a lot of time to spend on this soon. :)
good, maybe i'll just wait for you to do this ;) Other notes. Part of the reason I'm pretty decided that t-web will NOT be built on Nevow but instead written to support Nevow is that I've seen twisted.web go from webwidgets, to domtemplate, to woven, and now to nevow, and I have absolutely no doubt that someone is going to come up with a newer, faster, better, stronger system in the future. -- David Reid