What about RSXNT instead of cygwin32???

What about RSXNT instead of cygwin32???

[Bernd Schilpp]

Hi everybody,

Does anybody know about the RSXNT-Compiler ?
I tried to get my stuff compiled with cygwin32 - but I have some
problems, because the libaries are not threadsave, the socket-problems
and so on.
And there are always new problems reported in the mailinglist - you know
it.

Now I found
RSXNT-Compiler( ftp://ftp.uni-bielefeld.de/pub/systems/msdos/misc/
or http://www.sjc.ox.ac.uk/users/williams/ ).
The list of bugs and missing features is very short. The Libaries are
threadsave and it seems to work quite good.
Does anybody know the big difference to cygwin32 or is anybody
experienced with RSXNT ? Are any problems known ?
Please let me know your opinion.

thanks in advance,
Bernd


-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".

Re: What about RSXNT instead of cygwin32???

[Peter Dalgaard BSA]

Bernd Schilpp <Bernd.Schilpp@brokat.de> writes:

> Now I found
> RSXNT-Compiler( ftp://ftp.uni-bielefeld.de/pub/systems/msdos/misc/
> or http://www.sjc.ox.ac.uk/users/williams/ ).
> The list of bugs and missing features is very short. The Libaries are
> threadsave and it seems to work quite good.
> Does anybody know the big difference to cygwin32 or is anybody
> experienced with RSXNT ? Are any problems known ?
> Please let me know your opinion.

Yes, I've tried that and it certainly seems to be a class act,
although not as ambitious as cygwin32. The most obvious difference is
that there are no Unix utilities beyond 'make' coming with it. In
particular, there's no shell, so configure scripts can't work. In
contrast, cygwin32 is trying to reach a state where you can take Unix
sources and just type "./configure; make". 

On the other hand, RSXNT includes a resource compiler and working
samples of several native win32 applications, which makes it
attractive for purely windows-based programs. For people coming from a
Unix background, I think it can quite easily beat all of the
commercial Win32 C compilers.

Also, even though RSXNT implements fork() and some other Unix calls, I
bet that there are things that you can't do (not that I've checked,
but e.g. it took a while for cygwin32 to figure out how to do select()
on a mixture of files, pipes and sockets, so I wouldn't expect to find
it in RSXNT).

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".

Re: What about RSXNT instead of cygwin32???

[Bernd Schilpp]

Peter Dalgaard BSA wrote:

> Yes, I've tried that and it certainly seems to be a class act,
> although not as ambitious as cygwin32. The most obvious difference is
> that there are no Unix utilities beyond 'make' coming with it. In
> particular, there's no shell, so configure scripts can't work. In
> contrast, cygwin32 is trying to reach a state where you can take Unix
> sources and just type "./configure; make".

Thank you for writing of your experience with RSXNT.
I agree with you, that it is very convenient, if you have a
development-environment - like
cygwin32 offers - where you can act just like in a unix-environment.
But my Problem is, that threads are very important for the purpose of my
Programs - so I need threadsave libaries. Does anybody know, when cygwin32
is going to have threadsave libaries ?

--
Greetings,
Bernd


-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".

Re: What about RSXNT instead of cygwin32???

[Weiqi Gao]

Bernd Schilpp wrote:
> 
> Hi everybody,
> 
> Does anybody know about the RSXNT-Compiler ?
> I tried to get my stuff compiled with cygwin32 - but I have some
> problems, because the libaries are not threadsave, the socket-problems
> and so on.
> And there are always new problems reported in the mailinglist - you know
> it.
> 
> Now I found
> RSXNT-Compiler( ftp://ftp.uni-bielefeld.de/pub/systems/msdos/misc/
> or http://www.sjc.ox.ac.uk/users/williams/ ).
> The list of bugs and missing features is very short. The Libaries are
> threadsave and it seems to work quite good.
> Does anybody know the big difference to cygwin32 or is anybody
> experienced with RSXNT ? Are any problems known ?
> Please let me know your opinion.

RSXNTDJ v.1.31 and DJGPP v2.01 (not exactly RSXNT, I know) worked really
well on Windows 95 machines as a Win32 compiler.  You won't e able to
port UNIX code quite easily as you could with GNU-Win32 though.

--
Weiqi Gao
weiqigao@a.crl.com
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".

Re: What about RSXNT instead of cygwin32???

[Anthony Williams]

Bernd,

I am glad that you found RSXNT on my web page! I reckon it works really
well, and I have used it to produce several Win32 programs without hassles
- GUI programs, and command line apps, and DLLs, with both C and C++.
However, I don't know how compatible it is with CygWin32 (~100%, since the
libraries are standard?) and it doesn't contain all the utilities that
come with cygwin (not that I've ever found need for them) It's also good
for those who use compilers and assembles (such as older Borland ones)
that generate OMF object files, since it provides a conversion utility
to/from the GCC object format and OMF format. 

Anthony

On Mon, 13 Oct 1997, Bernd Schilpp wrote:

> Hi everybody,
> 
> Does anybody know about the RSXNT-Compiler ?
> I tried to get my stuff compiled with cygwin32 - but I have some
> problems, because the libaries are not threadsave, the socket-problems
> and so on.
> And there are always new problems reported in the mailinglist - you know
> it.
> 
> Now I found
> RSXNT-Compiler( ftp://ftp.uni-bielefeld.de/pub/systems/msdos/misc/
> or http://www.sjc.ox.ac.uk/users/williams/ ).
> The list of bugs and missing features is very short. The Libaries are
> threadsave and it seems to work quite good.
> Does anybody know the big difference to cygwin32 or is anybody
> experienced with RSXNT ? Are any problems known ?
> Please let me know your opinion.
> 
> thanks in advance,
> Bernd
> 
> 
> -
> For help on using this list (especially unsubscribing), send a message to
> "gnu-win32-request@cygnus.com" with one line of text: "help".
> 
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".

Re: What about RSXNT instead of cygwin32???

[Geoffrey Noer]

> Yes, I've tried that and it certainly seems to be a class act,
> although not as ambitious as cygwin32.

From what I have heard, RSXNT sounds like it does significantly
less but isn't beta quality.  I've also heard that nobody is actively
maintaining RSXNT -- anyone know differently?

I hope Cygwin32 will be a class act some day.  :-)

> On the other hand, RSXNT includes a resource compiler

BTW, future GNU-Win32 releases will include one.

-- 
Geoffrey Noer
noer@cygnus.com
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".

Re: What about RSXNT instead of cygwin32???

[Peter Dalgaard BSA]

Geoffrey Noer <noer@cygnus.com> writes:

> 
> > Yes, I've tried that and it certainly seems to be a class act,
> > although not as ambitious as cygwin32.
> 
> From what I have heard, RSXNT sounds like it does significantly
> less but isn't beta quality.  I've also heard that nobody is actively
> maintaining RSXNT -- anyone know differently?

As far as I can tell, the author is still around, but hasn't posted
anything on the net since the summer holidays. The current version is
from March 97, so it's not exactly rapidly changing, but it's probably
a bit early to declare it dead.

What you may have heard is what I've also seen on dejanews, namely
that Rainer Schnitker at some point abandoned RSXWIN *in favor of*
RSXNT.

BTW. It's probably fair to mention that also RSXNT has strings
attached when it comes to commercial development.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".

Re: What about RSXNT instead of cygwin32???

[Bernd Schilpp]

Peter Dalgaard BSA wrote:

> Geoffrey Noer <noer@cygnus.com> writes:
>
> >
> > > Yes, I've tried that and it certainly seems to be a class act,
> > > although not as ambitious as cygwin32.
> >
> > From what I have heard, RSXNT sounds like it does significantly
> > less but isn't beta quality.  I've also heard that nobody is actively
> > maintaining RSXNT -- anyone know differently?
>
> As far as I can tell, the author is still around, but hasn't posted
> anything on the net since the summer holidays. The current version is
> from March 97, so it's not exactly rapidly changing, but it's probably
> a bit early to declare it dead.

I know from Rainer Schnitker that there will be a new version (1.4) of
RSXNT with some improvements. But 1.4 will not come this October.



-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".