www.neuron.yale.edu

Posted: **Sun May 22, 2005 11:00 am**

On 3/1/2005 Jose Ambros-Ingerson <jose@kiubo.net> wrote:
I just discovered that
oc> a = +3
gives you a parse error.
Isn't this kind of strange?

Posted: **Sun May 22, 2005 11:01 am**

On 3/1/2005 Michael Hines <michael.hines@yale.edu> wrote:
Well, now that you mention it... What do you think of this:

Code: Select all

oc>---4
       -4
oc>----4
       4

All this is defined in the nrn/src/oc/parse.y where

Code: Select all

%left   '+' '-'      /* left associative, same precedence */
%left   '*' '/' '%'     /* left assoc., higher precedence */
%left   UNARYMINUS NOT
%right  '^'     /* exponentiation */
...
expr:   NUMBER
...
       | '-' expr %prec UNARYMINUS
               { TPD; $$ = $2; code(negate); PN;}

I suppose it would be possible to add a UNARYPLUS.
Then you could write things like

Code: Select all

---+-+++-4

I see from the above that

Code: Select all

!!!!4

also is valid.

www.neuron.yale.edu

+3 is not a number

+3 is not a number