On 6/24/16 Jun 24 -12:31 PM, rcn wrote: > Hi all, > > Will wild-module support ever be fully and "officially" supported? > > Currently there's what appears to be a sample implementation as part of > asdf/contrib but, if I recall correctly, asdf/contrib is bound to > disappear eventually (actually, the documentation says this is an empty > package). However, it's not clear how to use it, it's not documented and > it's not clear if it will ever be a part of asdf or not. > > In my mind, this is an useful feature. Not for small systems (which, > sadly, are the majority of Common Lisp open source projects), but for > large systems where new files may be added to a directory to incorporate > additional functionalities (a plug-in kind of thing). > > In these scenarios, wild-module definitions would ease development and > system maintenance greatly. As far as I know, asdf is perfectly capable > of implementing it and it looks to me like an implementation-agnostic > feature, so I see no reason not to include as a core component.
AFAICT, wild-modules are completely undocumented and, since I wasn't involved in their development, I don't understand what they are for, or what they are supposed to do. There are neither docstrings nor comments. I also note that the code in contrib is more than 3 years old, so likely doesn't even work. So I am not in a position to answer your question (except by saying "WTF is a wild module?"). In order to move towards integration, I would like to see some kind of specification to describe the intended function of WILD-MODULE, and how we intend to meet the specification. Then we could move forward with getting the code integrated, and some appropriate documentation into the manual. A test suite would also be required before we could "ship" WILD-MODULEs. Best, r
