Re: Recommended way to handle deprecation?
- From: Andreas Rottmann <a rottmann gmx at>
- To: James Henstridge <james daa com au>
- Cc: language-bindings gnome org
- Subject: Re: Recommended way to handle deprecation?
- Date: Wed, 31 Mar 2004 18:23:53 +0200
James Henstridge <james daa com au> writes:
> On 31/03/04 22:50, Andreas Rottmann wrote:
>
>>Hi!
>>
>>I wonder what the recommended way to handle deprecated API: to wrap or
>>not to wrap, that's the question here :). How do other bindings handle
>>this and is there a "recommended way"?
>>
> If you remove deprecated interfaces from your binding, then you will
> break programs written against old versions of your binding. So while
> I can write a C program today and know that it will work with all
> future GTK 2.x releases, if I write it against a language binding that
> doesn't wrap deprecated interfaces, it might break when the next 2.x
> release comes out.
>
> The best option would be to wrap the function, and emit a deprecation
> warning if possible (I know Python and Java can do this, and GCC has
> an extension to do this for C/C++).
>
Guile does this for its own API, so it should be possible to here
too. Thanks for the advice; So I think the rules can be formulated
like this:
1) If you start a new binding of some lib, don't wrap deprectated stuff.
2) If you underlying stuff becomes deprecated, provide it, along with
a deprecation message, if possible.
Maybe this should be noted in the Binding Rules somewhere.
Andy
--
Andreas Rottmann | Rotty ICQ | 118634484 ICQ | a rottmann gmx at
http://yi.org/rotty | GnuPG Key: http://yi.org/rotty/gpg.asc
Fingerprint | DFB4 4EB4 78A4 5EEE 6219 F228 F92F CFC5 01FD 5B62
A. Because it breaks the logical sequence of discussion
Q. Why is top posting bad?
[
Date Prev][
Date Next] [
Thread Prev][Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]