On Thu, 5 Jan 2006, Allin Cottrell wrote:
On Thu, 5 Jan 2006, Sven Schreiber wrote:
> Yes indeed: Not using the local decimal separator
> (Preferences-General, uncheck checkbox, and restart) solves all my
> reported problems. (after replacing commas with periods where
> necessary, of course)
>
> So that's a workaround, and not too bad. In the medium term it
> would be good if gretl could use an argument separator which is
> not used as a decimal separator somewhere in the world
> (semicolon?). If that's feasible while maintaining some backward
> compatibility, I don't know.
Thanks for tracking down the problem. I'll see if I can come up
with something to solve this. Here's an initial thought. Right
now, all the spaces are squeezed out of "genr" formulas in an early
stage of processing, because white space carries no meaning in that
context. What if we made spaces following commas significant, when
the decimal separator is a comma? That is, in such locales you
could write, e.g., "(1,5, 2,3)" and have it read as "one-point-five,
two-point-three".
My humble opinion is that any decimal seprator other than "." is EVIL;
being in a country where the traditional decimal separator is "," and
"."
is reserved for thousands, I can tell you that the potential for confusion
is immense, to the point of making the CSV format nearly useless. If we
allow commas as separators in scripts depending on the locale setting,
portability becomes a nightmare.
Riccardo `Jack' Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
jack(a)dea.unian.it
http://www.econ.univpm.it/lucchetti