Updating NEWS for 1.6.8 release.

Updating NEWS for 1.6.8 release.

[Rob Browning]

If you have time, could those of you who have committed changes to 1.6
(since 1.6.7) check out a current 1.6 tree, look over the output of

  cvs -qz5 diff -u -r release_1-6-7

and make sure that all of your changes have appropriate NEWS entries?

Thanks

-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4


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

Re: Updating NEWS for 1.6.8 release.

[Neil Jerram]

Rob Browning <rlb@defaultvalue.org> writes:

> If you have time, could those of you who have committed changes to 1.6
> (since 1.6.7) check out a current 1.6 tree, look over the output of
>
>   cvs -qz5 diff -u -r release_1-6-7
>
> and make sure that all of your changes have appropriate NEWS entries?

There are two bug fixes by me, but I don't think they merit NEWS
entries.  (Bug fixes don't normally, do they?)

      Neil



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

Re: Updating NEWS for 1.6.8 release.

[Rob Browning]

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

> There are two bug fixes by me, but I don't think they merit NEWS
> entries.  (Bug fixes don't normally, do they?)

The general rule I've heard is that NEWS should include "user visible
changes", but of course that leaves a lot of room for discretion.

I suppose I'd say that it depends on the bug fix.  If a bug was fixed
that people might have noticed and were having to work around, then a
perhaps a NEWS entry would be appropriate.

Of course I don't remotely consider myself an expert on the guidelines
regarding NEWS, so I'd be more than happy to hear relevant comments
and opinions.

-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4


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

Re: Updating NEWS for 1.6.8 release.

[Neil Jerram]

Rob Browning <rlb@defaultvalue.org> writes:

> The general rule I've heard is that NEWS should include "user visible
> changes", but of course that leaves a lot of room for discretion.
>
> I suppose I'd say that it depends on the bug fix.  If a bug was fixed
> that people might have noticed and were having to work around, then a
> perhaps a NEWS entry would be appropriate.

Thanks.  In this case, I'm pretty sure my bug fixes don't qualify as
NEWS.

        Neil



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

Re: Updating NEWS for 1.6.8 release.

[Kevin Ryde]

Neil Jerram <neil@ossau.uklinux.net> writes:
>
> Thanks.  In this case, I'm pretty sure my bug fixes don't qualify as
> NEWS.

What is the continuations versus stack bit you added some tests for?
Is it a fix for something?


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

Re: Updating NEWS for 1.6.8 release.

[Neil Jerram]

Kevin Ryde <user42@zip.com.au> writes:

> Neil Jerram <neil@ossau.uklinux.net> writes:
>>
>> Thanks.  In this case, I'm pretty sure my bug fixes don't qualify as
>> NEWS.
>
> What is the continuations versus stack bit you added some tests for?
> Is it a fix for something?

Yes, for `(call/cc make-stack)'.  make-stack is supposed to be able to
construct a stack from a continuation, a debug object, or from `#t',
which means the current evaluation stack.  But the make-stack code was
missing a lot of the relocations that are needed in the
continuation case.  The symptom was a segmentation fault.

I forget now exactly how I discovered this, though.  Is it worth me
trying to dig that up?

       Neil



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

Re: Updating NEWS for 1.6.8 release.

[Rob Browning]

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

> Yes, for `(call/cc make-stack)'.  make-stack is supposed to be able to
> construct a stack from a continuation, a debug object, or from `#t',
> which means the current evaluation stack.  But the make-stack code was
> missing a lot of the relocations that are needed in the
> continuation case.  The symptom was a segmentation fault.
>
> I forget now exactly how I discovered this, though.  Is it worth me
> trying to dig that up?

Not as far as I'm concerned, though would a trivial NEWS entry be
appropriate?  i.e. perhaps something like

 ** make-stack can now correctly construct a stack from a continuation.

if that's correct.

-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4


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

Re: Updating NEWS for 1.6.8 release.

[Neil Jerram]

Rob Browning <rlb@defaultvalue.org> writes:

> Not as far as I'm concerned, though would a trivial NEWS entry be
> appropriate?  i.e. perhaps something like
>
>  ** make-stack can now correctly construct a stack from a continuation.
>
> if that's correct.

Yes, that would be fine.

     Neil



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

Re: Updating NEWS for 1.6.8 release.

[Rob Browning]

Rob Browning <rlb@defaultvalue.org> writes:

> If you have time, could those of you who have committed changes to 1.6
> (since 1.6.7) check out a current 1.6 tree, look over the output of
>
>   cvs -qz5 diff -u -r release_1-6-7
>
> and make sure that all of your changes have appropriate NEWS entries?

I don't know that anyone noticed, but our previous entry for this
release was quite wrong.  Apparently we (I) had never added a new
section for 1.6.8 after the 1.6.7 release.  Here's the corrected NEWS
(also in CVS):

  Changes since Guile 1.6.7 (changes in 1.6.8):

  * Changes to the distribution

  ** A number of GCC 4.0 related build problems have been fixed.

  A number of changes have been made to accomodate GCC 4.0.

  * Changes to the stand-alone interpreter

  * Changes to Scheme functions and syntax

  ** The r5rs numerator and denominator functions have been added.

  These are now available in the default environment.

  * Changes to the C interface 

There are only two changes that the user would want to know about?

-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4


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

Re: Updating NEWS for 1.6.8 release.

[Kevin Ryde]

Rob Browning <rlb@defaultvalue.org> writes:
>
>   Changes since Guile 1.6.7 (changes in 1.6.8):

I find the way 1.6.7 is said first here pretty unclear.  I know it's
been that way for a long time, but I always have to think for which
version I'm reading about.  Maybe just,

	Guile 1.6.8 changes:

	...


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

Re: Updating NEWS for 1.6.8 release.

[Rob Browning]

Kevin Ryde <user42@zip.com.au> writes:

> I find the way 1.6.7 is said first here pretty unclear.  I know it's
> been that way for a long time, but I always have to think for which
> version I'm reading about.

I agree.  I thought the same thing myself today, but went ahead and
stuck with the convention.

> Maybe just,
>
> 	Guile 1.6.8 changes:

Alternately, we could use the Emacs approach:

  * Changes in Guile 1.6.8:

and I supppose if there's any advantage, we could leave the most
recent header as

  * Changes since Guile 1.6.7:

until time for the release, but I'm not sure that's helpful.

-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4


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

Re: Updating NEWS for 1.6.8 release.

[Kevin Ryde]

Rob Browning <rlb@defaultvalue.org> writes:
>
>   cvs -qz5 diff -u -r release_1-6-7

I added the bits I did, mostly fixes from bug reports.  I've tried to
keep it brief, none are too terrible, unless you actually wanted to
use something affected :).


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

Re: Updating NEWS for 1.6.8 release.

[Rob Browning]

Kevin Ryde <user42@zip.com.au> writes:

> Rob Browning <rlb@defaultvalue.org> writes:
>>
>>   cvs -qz5 diff -u -r release_1-6-7
>
> I added the bits I did, mostly fixes from bug reports.  I've tried to
> keep it brief, none are too terrible, unless you actually wanted to
> use something affected :).

Great, and thanks.
-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4


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