Andrew John Hughes wrote: > Does anyone know the current status of gjdoc? It would be nice for it to be > able to handle the generics branch at some point. Last I heard, the best > approach to this would be to fix the parsing process to unify the current > separate systems of parsing expressions and then manually parsing the > higher-level constructs. Of course it would be great if gjdoc could handle generics (and enums, static imports, etc.) Unfortunately, the current parser code is really messy. It's a hand-written scanner/parser hybrid thing and, I believe, nearly unmaintainable in its current form. What I'd really like to do is completely replace the parser with a new one written from scratch (preferrably, a generated one.) In addition, the parsing and type resolution steps should be cleanly separated - they are somewhat intermingled at the time. This, however, amounts to a rewrite of large parts of gjdoc's core. I've already started working on this back in 2005, but I regret to say that I won't get around to finishing it in the foreseeable future. With a bit of luck I can spend some more time on it in April, but at this point it's hard for me to tell when the new parsing framework could be mature enough to replace the current one. If you find an easy way to extend the current parser to handle the 1.5 constructs, that would be the only viable short-term solution. But I fear there might be no such easy way :( Julian
