Re: [ctwm] Evaluating our requirements

[Richard Levitte]

In message 20120109042459.gs72...@over-yonder.net on Sun, 8 Jan 2012 22:24:59 
-0600, Matthew D. Fuller fulle...@over-yonder.net said:

fullermd On Sun, Jan 08, 2012 at 10:45:32PM -0500 I heard the voice of
fullermd Stefan Monnier, and lo! it spake thus:
fullermd   fair conformance to C89.  What systems do we care about that don't
fullermd   have reasonably competent C99 support?
fullermd  
fullermd  C99 support is unclear: e.g., AFAIK, gcc doesn't fully support
fullermd  C99, tho it has supported many parts of it for quite a while.
fullermd 
fullermd Well, hence reasonably competent, rather than complete :)  Total
fullermd support is fairly uncommon in any mainstream compilers.

I have zero problems with the thought of upgrading to a more modern
standard.  Also, if it is important to keep support for uncommong
things (such as a 16-bit int), it's not really difficult to have a set
of support macros that do things differently based on, say, sizeof(int).

I foresee no real C99 problems on platforms like the ones mentioned
earlier in this thread (VMS, ...)

fullermd In contrast, a lack of that fairly common subset would be
fullermd more expected in obsolescent systems (AIXV3, say).
fullermd Clarifying how much real pain drawing the line in various
fullermd places causes actual users is what I want to draw out here.

Another view is to simply start working on it, and fix the introduced
limitations of someone screams...

Personally, I have an additional desire, and it's to check against
things like ICCCM and make additions to comply.

Those two things could very well be goals for version 4.0

Cheers,
Richard

-- 
Richard Levitte rich...@levitte.org
http://richard.levitte.org/

Life is a tremendous celebration - and I'm invited!
-- from a friend's blog, translated from Swedish

Re: [ctwm] Evaluating our requirements

[Stefan Monnier]

 fair conformance to C89.  What systems do we care about that don't
 have reasonably competent C99 support?

C99 support is unclear: e.g., AFAIK, gcc doesn't fully support C99,
tho it has supported many parts of it for quite a while.


Stefan

Re: [ctwm] Evaluating our requirements

[Matthew D. Fuller]

On Sun, Jan 08, 2012 at 10:45:32PM -0500 I heard the voice of
Stefan Monnier, and lo! it spake thus:
  fair conformance to C89.  What systems do we care about that don't
  have reasonably competent C99 support?
 
 C99 support is unclear: e.g., AFAIK, gcc doesn't fully support
 C99, tho it has supported many parts of it for quite a while.

Well, hence reasonably competent, rather than complete :)  Total
support is fairly uncommon in any mainstream compilers.

Large subsets are common across most current compilers (VS being the
major exception, to my knowledge).  And most of the things that strike
me as useful in ctwm would fit within that fairly well.  The
additional integral types (particularly fixed size) are topping my
mind right now, but things like block scoped declarations would be
useful too.  OTOH, if somebody has a good reason to use complex types
in ctwm, I want to see how (in much the same way that I want to see
how you blow up a continent with pop tarts, pipe cleaners, and excess
dryer lint, anyway).

In contrast, a lack of that fairly common subset would be more
expected in obsolescent systems (AIXV3, say).  Clarifying how much
real pain drawing the line in various places causes actual users is
what I want to draw out here.


-- 
Matthew Fuller (MF4839)   |  fulle...@over-yonder.net
Systems/Network Administrator |  http://www.over-yonder.net/~fullermd/
   On the Internet, nobody can hear you scream.