> (I am worried that the second way with aliases would not work well
> with charts?)
True, but these aliases cannot be used in other cases in charts, too.
They make only sense for unique, singular fields.
Actually, I don't use aliases very often, as the numeric relative
offsets look more clear to me. I use them typically if the referred
field is in a rather distant location.
> The only objection I have now is that specifying the radio buttons is
> not regular enough, i.e. I (as the application programmer) have to
> know up-front which button is the first one in the group and create it
> differently from the rest.
Well, the first field will often be different anyway, because it typically
will have a +Var or +E/R prefix. This is how I tested it here:
(default *G1 "value1a" *G2 "value2a")
(gui '(+Var +Radio) '*G1 NIL "value1a")
(gui '(+Radio) -1 "value1b")
(gui '(+Radio) -2 "value1c")
(gui '(+Var +Radio) '*G2 NIL "value2a")
(gui '(+Radio) -1 "value2b")
(gui '(+Radio) -2 "value2c")
so that I could check interactively
: (cons *G1 *G2)
-> ("value1a" . "value2a")
> It would be better if I could just call:
> (gui '(+Radio) 'group2 "value2c")
> Imagine the following scenario: An admin user can enter an enrolment
> In such scenario, I do not know up-front what is going to be in the
> form. I have to write a piece of code which finds out which radio
> button is the first one. I thing this piece of code should be part of
I'm not sure if I understood your intention. But searching for NIL
should clearly indicate which is the first one?
That's wise! In fact, I quite often use 'w3m' which doesn't know about
UNSUBSCRIBE: mailto:[EMAIL PROTECTED]