[ANNOUNCE] FrobTADS 0.3 released

[Guest]

Archived from groups: rec.games.int-fiction,rec.arts.int-fiction (More info?)

Version 0.3 of FrobTADS, a console-mode "portable port" of TADS2/TADS3,
has been released. Point your browser at:

http://www.tads.org/frobtads.htm

where you can download the interpreter as well as the authoring tools
for TADS 2 and TADS 3.

So far, the package has been tested and is known to work under Linux,
various flavors of BSD, Solaris, Mac OS X, BeOS, and Microsoft Windows.

The new version is in sync with the newest TADS release (2.5.9/3.0.9).
Other changes:


- Statusline now looks good even when colors are disabled or not
available.

- Fixed compilation problem on systems that don't support the SIGWINCH
signal ("unknown identifier: SIGWINCH"), like MS Windows with MinGW
and/or Dev-C++.

- Fixed the fix of the compilation problem with the <wchar.h>
functions (wcslen and wcscpy); this time for good, I hope.
Compilation in BeOS should hopefully work "out-of-the-box" now.

- Hopefully improved terminal resize handling with old/broken curses
libraries.


Questions, comments and bug-reports, should be sent to:

realnc at users dot sourceforge dot net

 

[Guest]

Archived from groups: rec.games.int-fiction,rec.arts.int-fiction (More info?)

Nikos Chantziaras wrote:
> Version 0.3 of FrobTADS, a console-mode "portable port" of TADS2/TADS3,
> has been released. Point your browser at:
>
> http://www.tads.org/frobtads.htm
>
> where you can download the interpreter as well as the authoring tools
> for TADS 2 and TADS 3.

The TADS 2 compiler package is corrupted and results in CRC errors when
unpacking (possibly a transfer error). Until the package is re-uploaded
to the official site, you can grab it here (direct link to the file):

http://members.lycos.co.uk/realnc/t2comp.tar.gz

The interpreter and TADS 3 compiler packages are not affected.

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Nikos Chantziaras <realnc@arcor.de> wrote:

>Version 0.3 of FrobTADS, a console-mode "portable port" of TADS2/TADS3,
>has been released.
....
> - Fixed the fix of the compilation problem with the <wchar.h>
> functions (wcslen and wcscpy); this time for good, I hope.
> Compilation in BeOS should hopefully work "out-of-the-box" now.

It does, thank you. (Although it still says in the INSTALL file
that nobody has tried to compile it "in things like BeOS", etc.

--
Sophie Frühling

"El arte no viste pantalones." --Rubén Darío

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Sophie Fruehling wrote:
> Nikos Chantziaras <realnc@arcor.de> wrote:
>
>> Version 0.3 of FrobTADS, a console-mode "portable port" of TADS2/TADS3,
>> has been released.
> ....
>> - Fixed the fix of the compilation problem with the <wchar.h>
>> functions (wcslen and wcscpy); this time for good, I hope.
>> Compilation in BeOS should hopefully work "out-of-the-box" now.
>
> It does, thank you. (Although it still says in the INSTALL file
> that nobody has tried to compile it "in things like BeOS", etc.

Oops... I forgot to change that.

I hope the above sentence -- "_things_ like BeOS" -- didn't sound
offensive or anything... It's just that I never used that system. I
even never saw a screenshot of what BeOS looks like.

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Somewhat related:

I was able to compile the entire kit and kaboodle on Windows through
Cygwin, using gcc 3.4.4. frob.exe works great - but t3make dies with a
segmentation fault whenever I try to use it to compile the sample (or Eric
Eve's Heidi version, for the record). It appears to get to the end, but
then conks out:

link -> sample.t3p
preinit -> sample.t3
/bin/sh: line 1: 1408 Segmentation fault (core dumped)
"/c/frobtads-0.3/t3
make" -f sample.t3m -a -FL "/c/frobtads-0.3/t3compiler/tads3/lib" -FI
"/c/frobta
ds-0.3/t3compiler/tads3/include" -I
"/c/frobtads-0.3/t3compiler/tads3/samples" -
Fs "/c/frobtads-0.3/t3compiler/tads3/samples" -Fy
"/c/frobtads-0.3/samples/obj"
-Fo "/c/frobtads-0.3/samples/obj"
make: *** [sample] Error 139

Anyone else had any luck here?

best,
james

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Nikos Chantziaras <realnc@arcor.de> wrote:

>I hope the above sentence -- "_things_ like BeOS" -- didn't sound
>offensive or anything...

Ah, no, especially not now that you are going to change it and
aren't going to mention Beos in that sentence any more.

>It's just that I never used that system. I
>even never saw a screenshot of what BeOS looks like.

I guess it's not particularly exciting, but here's an example:
http://www.generalcoffee.com/futureboy/images/sshot_beos1.jpg

(And it's on-topic too, yay!)

--
Sophie Frühling

"El arte no viste pantalones." --Rubén Darío

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

james cunningham wrote:
> Somewhat related:
>
> I was able to compile the entire kit and kaboodle on Windows through
> Cygwin, using gcc 3.4.4. frob.exe works great - but t3make dies with a
> segmentation fault whenever I try to use it to compile the sample (or
> Eric Eve's Heidi version, for the record). It appears to get to the
> end, but then conks out:
>
> link -> sample.t3p
> preinit -> sample.t3
> /bin/sh: line 1: 1408 Segmentation fault (core dumped)

That's the (by now) infamous "Strict Aliasing Bug(TM)". It's not
Cygwin's fault; same thing happens with MinGW, and in general any other
GCC compiler (even under Linux). This is mentioned in the INSTALL file.
In general, when using GCC to compile, one should configure like this:

configure CXXFLAGS="[other flags] -fno-strict-aliasing"

The problem is not FrobTADS-specific. It seems G++ produces buggy
machine code with compiling the T3VM. I don't think that this will get
"fixed" in the basecode, as no one is actually sure what's causing the
problem. If some part of the T3VM violates C++ aliasing rules, then the
problem is with the T3VM, not GCC. No one can tell if this is the case
though.

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

On Sun, 11 Sep 2005 16:26:04 -0400, Nikos Chantziaras <realnc@arcor.de>
wrote:

> That's the (by now) infamous "Strict Aliasing Bug(TM)". It's not
> Cygwin's fault; same thing happens with MinGW, and in general any other
> GCC compiler (even under Linux). This is mentioned in the INSTALL file.
> In general, when using GCC to compile, one should configure like this:
>
> configure CXXFLAGS="[other flags] -fno-strict-aliasing"
>

That worked great, thanks. Apologies for not thoroughly reading the
documentation before bothering you about it.

best,
james

 

[Guest]

Archived from groups: rec.games.int-fiction,rec.arts.int-fiction (More info?)

Nikos Chantziaras wrote:
> Nikos Chantziaras wrote:
>> Version 0.3 of FrobTADS, a console-mode "portable port" of
>> TADS2/TADS3, has been released. Point your browser at:
>>
>> http://www.tads.org/frobtads.htm
>
> The TADS 2 compiler package is corrupted and results in CRC errors when
> unpacking (possibly a transfer error).

The package has been re-uploaded to http://www.tads.org/frobtads.htm.
Hopefully, Murphy's Law of Bogus TADS Uploads doesn't apply to
*re*-uploads...

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

"Sophie Fruehling" <sfruehling@LOVELY-SPAM.aon.at> wrote in message
news:v019i117ppoio3e0ovgrbmb0ik7q90ff21@4ax.com...
> Nikos Chantziaras <realnc@arcor.de> wrote:
>
>>I hope the above sentence -- "_things_ like BeOS" -- didn't sound
>>offensive or anything...
>
> Ah, no, especially not now that you are going to change it and
> aren't going to mention Beos in that sentence any more.
>
>>It's just that I never used that system. I
>>even never saw a screenshot of what BeOS looks like.
>
> I guess it's not particularly exciting, but here's an example:
> http://www.generalcoffee.com/futureboy/images/sshot_beos1.jpg
>
> (And it's on-topic too, yay!)

Looks good. Now, if I'd get BeOS to install on my computer, but no... seems
the installer requires I use a mouse in the beginning - but has turned off
my USB mouse.