Uwe Brauer <o...@mat.ucm.es> writes: >> Note that you can instead use Elisp formulas. See 3.5.3 Emacs Lisp forms >> as formulas. It is more flexible. > > Can you give me an example?
#+begin_src emacs-lisp (defun yant/convert-cd (x) "Replace \",\" in X string with \".\" and return number." (string-to-number (replace-regexp-in-string "," "." x))) (defun yant/convert-dc (x) "Replace \".\" in X number with \",\" and return string." (replace-regexp-in-string "\\." "," (format "%s" x))) #+end_src | 3,5 | 4,2 | 7,7 | #+TBLFM: $3='(yant/convert-dc (apply '+ (mapcar #'yant/convert-cd (list $1 $2)))) Yes, it is awkward. Yet possible. >> Alternatively, we may implement the comma separators as a built-in >> feature. Org table formulas already support notations like > >> | 3,5 | 4,2 | 7 | > >> #+TBLFM: $3=$1+$2;N > > Not sure I understand this > > | 3,5 | 4,2 | 7 | (7, 7) | (7, 7) | > | | | | | | > #+TBLFM: $3=$1+$2;N::$4=$1+$2;E::$5=$1+$2;L ;N means "convert to number using string-to-number". (string-to-number "3,5") ;; => 3 why? read the docstring (it ignores non-recognized trailing chars ",5") L makes no sense in calc formulas (it has no effect). E refers to dealing with empty field and has no effect here. Best, Ihor
