[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Savannah-hackers] Commits mailing lists
From: |
Camm Maguire |
Subject: |
[Savannah-hackers] Commits mailing lists |
Date: |
02 Jul 2004 10:19:01 -0400 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.2 |
Greetings!
Mike Thomas <address@hidden> writes:
> Thanks Camm.
>
> I must have accidentally deleted this email (who knows how many
> others) at work before seeing it - I'm overwhelmed by spam there, I
> think through carelessness with my address on public forums. The
> problem with my home address is that I often go for days without
> looking at it as in this case.
>
> Have you been told about when they will revive the commits mailing
> list if at all?
>
They have indicated to me that it is very difficult. No timeline is
set to my knowledge. Am forwarding this to those in charge to recheck
the status.
Take care,
> Cheers
>
> Mike Thomas
>
> Camm Maguire wrote:
> > Hi Mike! A few thoughts on your latest commits:
> > 1) I *very* much appreciate the care you are taking not to break
> > other
> > platforms with your use of #ifdefs around function declarations,
> > etc. This is working out very well in my opinion -- i.e. there
> > are few if any collisions.
> > 2) In this case, your observation about the handler_function_type is
> > correct on all platforms. The function types have all been
> > traditionally declared void/int (*)() for two reasons (that I can
> > see)
> > a) because early versions of the C compiler could not process
> > argument types in function variables
> > b) because in certain circumstances, the argument types are
> > unknown, i.e. can legitimately take on several forms
> > b) is not applicable here, so I've implemented your void (*)(int)
> > across the board. If you are ever in doubt and want to fire
> > off a note to me on
> > issues like this, we can begin to thin out the ifdefs.
> > 3) FEerror takes a second argument which indicates the number of
> > printf-like arguments needed to format the error string. In the
> > cases you've added to pathname.d, there are no format control
> > strings, so 0 is appropriate here, as far as I can tell. Harmless
> > the way you had it, just led to a compiler warning and uses up a
> > little extra stack space.
> > Take care,
>
>
>
>
--
Camm Maguire address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah
- [Savannah-hackers] Commits mailing lists,
Camm Maguire <=