On 2005-09-06 13:20:35, Ian Clelland wrote:
On 9/6/05, Stefan Scholl sscholl@common-lisp.net wrote:
New release CL-WIKI 0.0.4 Changes:
- config file
I can tell already that I'm really going to get a lot of use out of the config file :) I really appreciate that it's just a plain lisp file as well, and not a text file with a custom parser.
Maybe at some time there will be a config editor for people who don't like Lisp. But at the moment I don't know of any easy format which can handle lists of key-value pairs like the one for *WIKI-PAGE-LIST*
I've tested trivial-configuration-parser and thought about implementing an INI file parser myself.
But at the end only Lisp can make me happy. :-)
Looking through the source, I can see that the other thing I'd like to be able to set in the config file is (html-mode) -- I'm trying to get a lot of my sites into XHTML, and it looks like that would be an easy way to do it with all of the wikis (well, that and an XML-ish template file) -- that's a wishlist feature though, not critical at all.
I've changed my mind about XHTML some months ago. The effort to do it right isn't worth the gain for the most time. I've seen so many sites and software that claim to use XHTML but it's still corrupt HTML with some other DTD and other style for empty elements and attributes.
And most of the browsers don't even support HTML 4 and CSS 2 in complete. The patch for wiki codes for external links I've received sets a class="external http" for http links. That's really only needed for CSS 2 unaware browsers like Internet Explorer (which can't cope with the right XHTML content-type, by the way). With Firefox you can use more sophisticated CSS selectors.
That's why I've sent Edi Weitz a patch for CL-WHO (released as 0.6.0) to let it output normal HTML.
And it's really easier to implement new wiki codes without thinking about 100 % right HTML output in combination with other codes. With XHTML I would feel really bad if CL-WIKI could produce invalid XHTML.
But it should be possible to implement some scheme to output XHTML in the future. Must check the Accept header to determine the right content-type and maybe implement a call to tidy to turn the output into valid XHTML. Just don't think about it at the moment. Gives only headaches.
Regards, Stefan