On Tue, Jun 15, 2010 at 8:26 AM, Nicolas M. Thiery
<nicolas.thi...@u-psud.fr> wrote:
> I have a question about passing down strings to functions of other
> systems. For (a stupid) example, if I want to concatenate two strings
> in GAP, I need to do:
>
> sage: gap.Concatenation("\"foo\"", "\"bar\"")
> foobar
Or
sage: gap.Concatenation('"foo"', '"bar"')
> Is there a reason that we can't just do the natural:
>
> sage: gap.Concatenation("foo", "bar")
> foobar
Currently, strings are passed directly down to the interfaces. So, if
you had two strings stored in the variables f and g in the GAP
session, doing
gap.Concatenation("f", "g")
would return the concatenation of those to string as opposed to "fg".
However, one could get a similar effect with
gap.Concatenation(gap.f, gap.g)
That only works for things where the underlying thing has a name which
is a valid Python identifier. You could instead want to do
gap.Concatenation(gap("f") gap("g"))
but then you might expect those to be strings too.
There is already a ticket for this at
http://trac.sagemath.org/sage_trac/ticket/2963 . The main reason it
isn't done is that it could break a lot of existing code, but we could
try to come up with a transition plan.
--Mike
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
