@roopeshchander Hi! I saw vfmd and was interested in it, but it doesn't seem to define an AST intermediate form. Any plans for one?
1
@dsilverstone The AST format is not in the scope of the spec, but the implementation parses md to AST and renders the AST to HTML.
2
@roopeshchander I would suggest that the AST be formally part of the spec since for some of us, mdwn->AST is more useful than mdwn->HTML.
1
@dsilverstone What do you mean by formally specifying the AST? Something like pandoc's json? What is the motivation?
1
@roopeshchander Well, I'd like to have tools which can read mdwn, manipulate the AST and write mdwn back out for a start.
1
@dsilverstone You can do that by writing code that uses vfmd source. Why do you want the AST spec-ed out for that?
1
@dsilverstone Or maybe you mean that you'd like to know how vfmd's AST looks like in memory, so you can manipulate it?
1
@roopeshchander Mostly having it formally specced would make it easier to work with (e.g. to bind to another language like Lua)
1
@dsilverstone vfmd does define a swig interface for binding to other languages. 1/2
2
@dsilverstone If you intend to use the AST from memory (not from disk), you'll need to only know how vfmd stores it in memory, right? 2/2
1