Peter,

On 2005 Aug 16 , at 18.43, Peter W. Draper wrote:

> OK, I've implemented three new macros:
>
>     AC_FC_HAVE_BOZ, to check if standard f95 integer BOZ constants are
>                     supported
>     AC_FC_HAVE_TYPELESS_BOZ, to check if these are extended to  
> allow for
>                              typeless assignments
>     AC_FC_HAVE_OLD_TYPELESS_BOZ, to check if the current usage is
>                                  supported (that's typeless 'xxxx'X  
> style)
>
> and a
>
>     AC_FC_HAVE_PERCENTLOC, to check if %LOC is supported (the fallback
>                            being to use plain LOC)

Sounds good.

> I'll commit these to the HEAD branch if no one objects.
>
> Norman, I've not documented these as I cannot work what's going on  
> with
> the autoconf.texi file. The log suggests that things like
> AC_FC_HAVE_PERCENTVAL should be documented on HEAD (merged from your
> branch?), but they do not seem to be present... Any ideas what's up?

I last merged from that branch at the end of March, making fortran.m4  
1.17.

I appear to have added AC_FC_HAVE_PERCENTVAL to the HEAD after that,  
in revisions 1.18 and 1.19, and then immediately hand-applied the  
change to the dev-nxg-20040116-add-fpp-support branch in revisions  
1.1.1.1.2.22 and .23.

I did it this way rather than merging the branch, because Toby was at  
that point still working on the branch, and it wasn't mergeable to  
the HEAD without major breakage.  Toby has recently committed a batch  
of changes to the branch, which are believed to leave the branch  
functional.  The plan is this:

1. I get this branch to a potentially submittable state -- fixing  
documentation and making sure no embarrasing regression tests fail;  
we call this a beta.

2. I make an internal release of that, by merging it to the HEAD and  
by creating a distribution tarball, and one of us then checks whether  
this breaks the HEAD.  At the same time, one of Toby's colleagues  
will check whether this distribution still works with their code.  In  
both cases, this will be best done by someone who doesn't know the  
internals, but is working from the documentation alone.  We iterate  
on the branch.

3. When everything appears to be working, we can submit the patch  
upstream, and Toby and I iterate on that list.

So given that you're using the BOZ macros at present, yes, can you  
commit them to the HEAD, but also hand-apply the changes to the  
branch.  The doc/autoconf.texi on the branch should have the  
documentation for the other macros we've added.

How's that?

Norman


-- 
----------------------------------------------------------------------
Norman Gray  /  http://www.astro.gla.ac.uk/users/norman/
Physics & Astronomy, Glasgow University, UK