Re: HTML_FlexyFramework
| From: | Alan Knowles | Date: | Thu, 26 Aug 2004 02:27:19 +0000 |
| Subject: | Re: HTML_FlexyFramework | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-32951@lists.php.net to get a copy of this message | ||
DG wrote:
Hello Alan, I would like to know if you plan to import your HTML_FlexyFramework in pear ?Thats the $1,000,000 question. I have learned to be causious about contributing too much (I have enough trouble keeping up with bug reports and features on my existing packages) - It would be nice to propose it, (hint to paul Yawp may be better as HTML_YawpFramework), but see the notes below about refactoring..
Are you going to maintain it in the future ?Yes, It is being mainted as I have over 10 major sites depending (both mine and clients).
Or is it deprecated?Not yet, but there are a number of reasons I want to refactor it, enough to break old applications. - Some of the current issues that have grown out of It's history * configuration - I'd rather pull the majority of the configuration stuff out and put it in the index.php that calls the constructor. * start() / getAuth() returning redirects was a bad idea.. Framework::run('page/name') is far more intuative.. * caching support needs removing/seperating.. - it adds to much cruft to the core, that's rarely used. * url reading needs simplifying (I used the wrong _SERVER vars) Obviously that will break old applications - so it may end up as FlexyFramework2, I'm hoping that the changes to support the new version will not be too problematic, as I will probably have to modify my old code to support it. Mind you - this is all vapourware, so dont expect a timeframe on it...
I'm looking for a light and highly modular framework for my apps, and it looks very fine (I already heavily use flexy and db_DataObjects).If it works, and it's easy to understand - go for it.. - I suspect the config stuff in it at the moment may be the death of that.. (which is why I want to remove it..) Regards Alan
But I don't want to go in a dead end way... Thanks, Didier