Bernd Zeimetz wrote: > The camelcase plugin does not honour preformatted text tags, which > results in links being added at places where you would not expect one. > I think it's needed here to make the check for preformatted text tags > dependent on the type of page, as the way to define.
This is a more generic problem, it affects all forms of links, not just camelcase. http://ikiwiki.info/bugs/wiki_links_still_processed_inside_code_blocks/ CamelCase is probably the most annoying instance, since it's very easy for text in a code block to accidentially look like CamelCase. Also, unlike regular [[links]], which can be escaped to \[[link]] to avoid it being treated as a link, if you write \CamelCase, the '\' is not removed. (Maybe CamelCase should at least get support for escaped links?) One approach is to move linkification to happen after htmlization, so that the linkification code can check the generated html. But there's a chicken and egg problem -- ikiwiki needs to be able to scan pages for links very early, before it htmlizes pages. > For example: > > - html: > <pre> </pre> > <code> </code> > > - markdown: > ` ` > `` `` > > - creole: > <pre> </pre> > {{{ }}} And if this route is taken, every htmlize plugin will basically need to provide its own checks for whether a given link is in a code block. -- see shy jo
signature.asc
Description: Digital signature
