2014-02-16 10:06 GMT+01:00 Michael Brand <michael.ch.brand@gmail.com>:

> > - I would like to have only the last element of the range filled.
>
> I always thought that this would not be possible with reasonable
> effort. Your question made me think again and it is possible, now I
> can change my own use cases with sub-total :-) . See docstring and
> take the procedure with vlen etc. from the TBLFM of the new ERT
> test-org-table/sub-total here:
>
> http://orgmode.org/w/org-mode.git?p=org-mode.git;a=blob;f=testing/lisp/test-org-table.el
>
> Or read the docstring there and just take this copy that I used to
> build the ERT:
> |-------+-------+-------|
> | Item  |  Item |  Sub- |
> | name  | value | total |
> |-------+-------+-------|
> | a1    |   4.1 |       |
> | a2    |   8.2 |       |
> | a3    |       |  12.3 |
> |-------+-------+-------|
> | b1    |  16.0 |  16.0 |
> |-------+-------+-------|
> | c1    |    32 |       |
> | c2    |    64 |  96.0 |
> |-------+-------+-------|
> | Total | 124.3 |       |
> |-------+-------+-------|
> #+TBLFM: @>$2 = vsum(@II..@>>) :: $3 = if(vlen(@-I$2..@0$2) ==
> vlen(@-I$2..@+I$2), vsum(@-I$2..@+I$2) +.0, string("")); EN f-1 ::
> @>$3 = string("")
>

I made:
    #+TBLFM: @>$2 = vsum(@<<..@>>) :: @<<$3..@>>$3 = if(vlen(@-I$2..@0$2)
== vlen(@-I$2..@+I$2), vsum(@-I$2..@+I$2), string(""))

-- 
Cecil Westerhof