>>> ... actually we do. So, maybe docstring can still be improved further. >> >> No to be a devils advoce, but I am not sure you do. If we look at: > > I think Ihor is referring to the doc string, it isn’t quite consistent > with the manual: > > org-capture-templates: >> %(sexp) Evaluate elisp (sexp) and replace it with the results. > > manual: >> ‘%(EXP)’ Evaluate Elisp expression EXP and replace it with the result. > > Best,
Ah, Sorry, I appologize, I missed it was the doc string. I was just focused on the manual. Anyway, doc string version is equally problematic for the very same reason. It just takes it from the other end. I think the problem here seems to be what we understand as a symbolic expression usually shortened as sexp or s-exp. 1, "2", eval, (+ 2 3), foo, (foo) <- these are all valid symbolic expressions. As a curiosa, I just looked up in Elisp manual what they say about symbolic expressions, but they don't spell it out. There is only a brief note in intro about Eval, saying they don't use the terminology in this manual. Anyway, lets not make this a lesson about lisp, that wasn't my intent. The point of the discussion is that we can't use variables, only functions. That's perhaps worth capturing in the documentation, more explicitly, if it is not of the broader interest to hack the code to fix the limitation.
