Re: [PATCH] GIterator
- From: Joel Becker <jlbec evilplan org>
- To: Havoc Pennington <hp redhat com>
- Cc: Daniel Elstner <daniel elstner gmx net>, James Henstridge <james daa com au>, gtk-devel-list <gtk-devel-list gnome org>, yosh gimp org
- Subject: Re: [PATCH] GIterator
- Date: Wed, 12 Jun 2002 09:50:22 +0100
On Tue, Jun 11, 2002 at 09:05:26PM -0400, Havoc Pennington wrote:
> We would end up sort of pseudo-deprecating all the foreach()
> functions, all functions that return a list, list->next style
> iteration, etc., and that's just too much churn in the "recommended"
> API set, even if the old API still works.
I don't think this is the case. Adding a type that makes my
life easier for my code (I hate hate hate _foreach() functions) and for
the code of others is a good thing.
In no way am I proposing "Replace all current API with GIterator
for 2.(x>2)". This isn't feasable. It only really makes sense for a
small set of the API (eg, gtk_container_iterate_children() returning a
GValueIterator provides more information than the GList of
gtk_container_get_children()). Even then, I haven't included such API
in my patches because I'm not sure it would be accepted.
However, new API can and should take advantage of such a thing.
We've discussed it already for GtkTreeSelection. A list is a list, and
an iterator is a different beast. Folks work with lists and iterators
in Python, Java, C++, and elsewhere without worrying. I don't think the
complexity would really increase for the user.
Joel
--
Life's Little Instruction Book #30
"Never buy a house without a fireplace."
http://www.jlbec.org/
jlbec evilplan org
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]