Ah that makes sense.
I guess I was a bit too keen.
I should have just waited for the next stable release,
but at least now I know how to compile Factor.
So, in the meantime, to continue with my experiments I'll
just keep using alien-invoke.
Regards to all,
Anton.
> On 3/22/07, Anton Rolls <[EMA
On 3/22/07, Anton Rolls <[EMAIL PROTECTED]> wrote:
> Oh no..! more trouble.
That's the joy of running of the darcs version. It's not always in a
runnable state. You might like to try checking on irc occasionally
before pulling darcs to see if it is stable.
Chris.
--
http://www.bluishcoder.co.nz
Oh no..! more trouble.
The Factor I've compiled, while it opens the Workspace ui,
when I click anywhere in the ui, on any buttons, text-areas
or scrollers, it pops a window saying
"Error: the wait operation timed out. [Traceback][Help][Edit]"
and if I click in that window it just pops another one
Berlin,
> It looks like you are trying to get the zlib and freetype dlls
> that come with
> mingw. I would get the ones of the factor site or doug's site.
> Those worked
> fine for the compile that I was doing. Also, all of doug's instructions
> seemed to work.
Actually, I just copied t
I've just now pulled the latest changes from the repository,
recompiled repos/Factor/, and attempted bootstrapping with
f -i=factor.image
where factor.image is copied from
http://factorcode.org/images/latest/boot.image.x86
... and...
f -shell=tty
compile
save
Can you confirm you bootstrapped again today?
The errors in compile-all are not causing the problem. They are
known errors.
It looks like you should be able to save after the compile-all and
run the ui since FT_Init_FreeType did not show up in the errors list.
Problem solved?
Doug
On Mar
Cool, this looks like the way forward. More info:
f -shell=tty
scratchpad> recompile
Compiling 14 words...
Compile finished.
scratchpad> save
scratchpad> bye
f
Words calling ``alien-invoke'' cannot run in the interpreter. Compile
th
Anton,
Your previous message says you copied the files and recompiled only,
with no mention of bootstrap. Every time I have had this problem a
bootstrap has fixed it. What about doing f -shell=tty and trying
'recompile' or 'compile-all'? Then 'save' your image, of course.
Also, try thi
On Thu, 22 Mar 2007 02:06:09 +1100, Anton Rolls wrote
> Doug,
>
> No no! I still get the error, even after copying these DLLs.
> (see my prior message.)
>
> Must sleep now, until tomorrow.
>
> Anton.
>
> > Anton,
> >
> > You recompiled, but you should have bootstrapped again since it's all
>
Doug,
No no! I still get the error, even after copying these DLLs.
(see my prior message.)
Must sleep now, until tomorrow.
Anton.
> Anton,
>
> You recompiled, but you should have bootstrapped again since it's all
> on the Factor side. Bootstrap again with the two dlls in the
> * Addition
Anton,
You recompiled, but you should have bootstrapped again since it's all
on the Factor side. Bootstrap again with the two dlls in the
directory with your f.exe and it should work.
From the blog, the step is listed in the "ingredients" section at
the top and as a bullet:
* Additionall
On Wed, 21 Mar 2007 18:39:29 +1100, Anton Rolls wrote
> Here's a status update:
>
> Today, roughly following instructions from:
> http://code-factor.blogspot.com/2006/11/compiling-factor-on-windows.html
>
> I have:
> - installed MinGW and MSYS
> - compiled Factor successfully
> - started Factor w
Hi Slava,
A hexeditor view of freetype6.dll and zlib1.dll
reveals that their symbols are plain - these particular
libraries do not have mangled function names.
Anton.
-
Take Surveys. Earn Cash. Influence the Future of IT
Hi Slava,
Well, actually I had immediately thought of that and
copyied freetype6.dll and zlib1.dll to repos/Factor/
but it gives the same errors.
I've even tried again
make clean
make windows
to rebuild Factor.
I'm afraid I've done something wrong somewhere.
Any ideas ? I'm still
On 21-Mar-07, at 3:39 AM, Anton Rolls wrote:
> Here's a status update:
>
> Today, roughly following instructions from:
> http://code-factor.blogspot.com/2006/11/compiling-factor-on-
> windows.html
>
> I have:
> - installed MinGW and MSYS
> - compiled Factor successfully
> - started Factor with h
Here's a status update:
Today, roughly following instructions from:
http://code-factor.blogspot.com/2006/11/compiling-factor-on-windows.html
I have:
- installed MinGW and MSYS
- compiled Factor successfully
- started Factor with http://factorcode.org/images/latest/boot.image.x86
Factor booted u
Well that was quick!
I now have downloaded the latest darcs repository
and I'll try recompiling Factor tomorrow.
The library now looks like having a high chance of success :)
Thanks for your help.
Anton.
> Hi Anton,
>
> I implemented automatic stdcall name mangling in the darcs
> repository.
Hi Anton,
I implemented automatic stdcall name mangling in the darcs
repository. You can remove the '@...' stuff from the symbol names,
and use FUNCTION: again. Everything should just work. Unmangled
stdcall DLLs will continue to work too.
Good luck with your library. We can add it to the r
The FMOD interface module is working, and I have
placed it in libs/fmod, which looks to me like the right
location for it.
Thanks to you both it's coming along nicely.
Anton.
-
Take Surveys. Earn Cash. Influence the Future
Hi Slava,
> > USE: compiler
> > \ FSOUND_GetVersion compile
> If you put this stuff in a source file, the words will be compiled
> for you, you don't have to call 'compile' yourself.
Yes, I noticed this behaviour. Thanks for verifying, I will try it.
> > but when I move everything into a subd
On 19-Mar-07, at 6:47 AM, Anton Rolls wrote:
> USE: compiler
> \ FSOUND_GetVersion compile
> \ FSOUND_Init compile
> \ FSOUND_Close compile
> \ FSOUND_GetError compile
> \ FMUSIC_LoadSong compile
> \ FMUSIC_FreeSong compile
> \ FMUSIC_PlaySong compile
> \ FMUSIC_StopSong compile
If you put this
Here is a working demo, which plays a ScreamTracker3 module.
Hooray !
! Date: 19-Mar-2007
! Author: "Anton Rolls"
! Status: "working"
! Purpose: "Play a song using the FMOD library."
! ToDo:
! - check return codes for errors and act accordingly
! Notes: {
! }
USING: kernel alien prettyprint io n
This indeed works!
> FSOUND_Init should probably be:
>
> : FSOUND_Init "char" "fmod" "[EMAIL PROTECTED]" { "int" "int" "uint" }
> alien-invoke ;
>
> Chris.
Thankyou once again. Now to get it to do something...
Regards,
Anton.
Whoops!! Let me try that...
Anton.
> On 3/19/07, Anton Rolls <[EMAIL PROTECTED]> wrote:
> > : FSOUND_Init "char" "fmod" "[EMAIL PROTECTED]" { } alien-invoke ;
> FSOUND_Init should probably be:
>
> : FSOUND_Init "char" "fmod" "[EMAIL PROTECTED]" { "int" "int" "uint" }
> alien-invoke ;
>
> Chris
On 3/19/07, Anton Rolls <[EMAIL PROTECTED]> wrote:
> : FSOUND_Init "char" "fmod" "[EMAIL PROTECTED]" { } alien-invoke ;
This says FSOUND_Init takes no arguments.
But the following you pass it three:
> 32000 64 0 FSOUND_Init . ! Problem here with "Operating system signal 11"
FSOUND_Init should
Ok, the latest attempt looks as below.
All is ok until the second last line (call to FSOUND_Init).
I'm wondering if the order I've put the
arguments on the stack is wrong or if they
need to be cast to int, int, uint or something like that.
! Date: 19-Mar-2007
! Author: "Anton Rolls"
! Status: "no
OMG! That's it! It's always so simple, isn't it ?
That works for me.
Thanks a lot !
Anton.
> On 3/19/07, Anton Rolls <[EMAIL PROTECTED]> wrote:
> > "Anton/fmod-version.factor" run-file
> > Error
> > Assertion failed
> > assert-got3
> > assert-expect 2
>
> This error means the code in the
On 3/19/07, Anton Rolls <[EMAIL PROTECTED]> wrote:
> "Anton/fmod-version.factor" run-file
> Error
> Assertion failed
> assert-got3
> assert-expect 2
This error means the code in the file left a value on the stack. Code
inside something that is run-filed should have an overall stack effect
Aha! I should have known - I wrote an external library interface
creator for the Rebol language and I extracted mangled function
names from the DLL there. I am just using a hex editor to examine
the end of the DLL file manually.
I found that this works when I type it in the listener:
USING: kern
Hi,
I only recently discovered that on Windows, some stdcall libraries do
funny name mangling; symbols are not named as one would expect. So a
function 'Foo' is actually '[EMAIL PROTECTED]' in the DLL, where x is the
amount
of stack space, in bytes, its parameters take up. For Factor 0.89,
Hi, I'm new to Factor and I'm trying to
interface to FMOD.DLL.
Here is what I have so far, in a file named:
Anton/test-fmod.factor
--
USING: kernel alien ;
IN: fmod
"fmod" "Anton/fmod.dll" "stdcall" add-library ! "stdcall" or "cdecl"
LIBRARY: fmod
! DLL_API float F_API F
31 matches
Mail list logo
