Lothar Scholz wrote: > Hello ES, > > E> The second and far more exciting change is RueDoc.normalize, > E> which takes the data structure generated by RDoc and rearranges > E> it into a much more sensible one. This makes it much easier to > E> do arbitrary things with the data (see Examples for a plain YAML > E> dump of a structure). I included a description of the generated > E> data structure at the bottom of this message. > > Did you work on the mny bugs in the ruby doc parser. Like skipping > inheritance classes, no differences between modules and classes, wrong > visibility declarations (not stored at all), infinite loops (for > rails) etc. ? No. These are specifically hacks to ease the pain until I or someone else comes up with a new, simple and workable documentation generator. I will eventually take that up, time permitting, if no-one else does. > Or is anybody working on it instead of frontend hacks ? Ryan Davis and Eric Hodel are tending to RDoc itself but I could not venture a guess what their position might be. E -- template<typename duck> void quack(duck& d) { d.quack(); }