unofficial mirror of guile-devel@gnu.org 
 help / color / mirror / Atom feed
From: Marius Vollmer <mvo@zagadka.de>
Cc: guile-devel@gnu.org
Subject: Re: Re-addition of deprecated stuff to 1.7.
Date: 17 May 2003 19:22:23 +0200	[thread overview]
Message-ID: <87iss97ak0.fsf@zagadka.ping.de> (raw)
In-Reply-To: <m3d6j9d5ic.fsf@laruns.ossau.uklinux.net>

Neil Jerram <neil@ossau.uklinux.net> writes:

> >>>>> "Marius" == Marius Vollmer <marius.vollmer@uni-dortmund.de> writes:
> 
>     Marius> Hi, I have started to re-add most of the things that were
>     Marius> deprecated in 1.6 and removed in 1.7.  I hope everyone
>     Marius> agrees that this is OK.
> 
> Yes, but can it be done in a way that the back compatible bits are
> kept separate from the core?

Yes, mostly.  When too much need to change just to support a re-added
deprecated thing, I'm not re-adding it.  For example, vcells are not
coming back.

> A while ago you mentioned something like _GNU_SOURCE, which seems
> exactly right to me.
>
> For example, an application would just say
> 
> #define _GUILE_SOURCE 1.6
> #include <libguile.h>
> 
> And a 1.7.x/1.8.x libguile.h would have something at the end like
> 
> #if _GUILE_SOURCE = 1.6
> #include <guile/compat-1.6.h>
> #endif

Hmm, I don't think I would want to include version numbers explicitely
in this mechanism.  I would say that when some code wants to be
compiled against the Guile 1.6 API, it should be compiled against
Guile 1.6.

But with a version specific switch, we can have conflicting APIs.  But
for that, I'd say, we should have one switch per conflicting feature.
And that switch could be deprecated over time...

To illustrate, consider the change from a two-arg version of scm_eval
to the three-arg one.  When making the decision to make that change
(instead of just adding a new function), we can offer a switch that
activates the three-arg version:


   /* old code, using two-arg eval */

   #include <libguile.h>

   -------------------
   /* new code, using three-arg eval */

   #define GUILE_ENABLE_THREE_ARG_EVAL
   #include <libguile.h>

Later, we can deprecate the 'enable' variant and prefer a 'disable'
one:

   /* old code, using two-arg eval */
   #define GUILE_DISABLE_THREE_ARG_EVAL
   #include <libguile.h>

   -------------------
   /* new code, using three-arg eval */

   #include <libguile.h>

And still later, we can just remove GUILE_DISABLE_THREE_ARG_EVAL and
let the old code fail.

>     Marius> The idea is that we should backwards compatability for a
>     Marius> longer time when it is easy enough to do.  I am adding the
>     Marius> things in as clean a way as possible, so that they do not
>     Marius> get in the way of the mainline code.
> 
> It's possible that I may have described what you are already doing,
> then :-)

Maybe... I try to only modify the files deprecated.c, deprecated.h and
deprecated.scm.  The rest of the code remains mostly untouched.

-- 
GPG: D5D4E405 - 2F9B BCCC 8527 692A 04E3  331E FAF8 226A D5D4 E405


_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/guile-devel


  reply	other threads:[~2003-05-17 17:22 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-03-26 18:04 Re-addition of deprecated stuff to 1.7 Marius Vollmer
2003-04-26  8:31 ` Neil Jerram
2003-05-17 17:22   ` Marius Vollmer [this message]
2003-05-17 17:47     ` Rob Browning
2003-05-17 17:57       ` Marius Vollmer
2003-05-17 18:12     ` Bruce Korb
2003-06-01 18:27       ` Marius Vollmer
2003-06-01 19:56         ` Bruce Korb
2003-06-09 19:46           ` Marius Vollmer

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/guile/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87iss97ak0.fsf@zagadka.ping.de \
    --to=mvo@zagadka.de \
    --cc=guile-devel@gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).