www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.bugs - The new std.boxer seems to be broken. :(

reply David L. Davis <SpottedTiger yahoo.com> writes:
Following the exact example as was posted in the D forum recently, the new
std.boxer.d is giving the below link error when the unbox functions are being
used. Is this a bug, or have I done something wrong in the code?

// WinXP SP2, dmd v0.125





















Output:
--------
C:\dmd>dmd boxed.d
C:\dmd\bin\..\..\dm\bin\link.exe boxed,,,user32+kernel32/noi;
OPTLINK (R) for Win32  Release 7.50B1
Copyright (C) Digital Mars 1989 - 2001  All Rights Reserved

boxed.obj(boxed)
Error 42: Symbol Undefined _assert_3std5boxer
--- errorlevel 1

C:\dmd>

David L.

-------------------------------------------------------------------
"Dare to reach for the Stars...Dare to Dream, Build, and Achieve!"
-------------------------------------------------------------------

MKoD: http://spottedtiger.tripod.com/D_Language/D_Main_XP.html
May 21 2005
next sibling parent reply Geert <gertje gertje.org> writes:
You should compile it in release mode and without unittest...
I'm not quite sure why it doesn't compile (and I really would like to 
know) if you try to compile it with unittest on and not in release mode.


David L. Davis wrote:

 Following the exact example as was posted in the D forum recently, the new
 std.boxer.d is giving the below link error when the unbox functions are being
 used. Is this a bug, or have I done something wrong in the code?
 
 // WinXP SP2, dmd v0.125




















 
 Output:
 --------
 C:\dmd>dmd boxed.d
 C:\dmd\bin\..\..\dm\bin\link.exe boxed,,,user32+kernel32/noi;
 OPTLINK (R) for Win32  Release 7.50B1
 Copyright (C) Digital Mars 1989 - 2001  All Rights Reserved
 
 boxed.obj(boxed)
 Error 42: Symbol Undefined _assert_3std5boxer
 --- errorlevel 1
 
 C:\dmd>
 
 David L.
 
 -------------------------------------------------------------------
 "Dare to reach for the Stars...Dare to Dream, Build, and Achieve!"
 -------------------------------------------------------------------
 
 MKoD: http://spottedtiger.tripod.com/D_Language/D_Main_XP.html
May 22 2005
parent David L. Davis <SpottedTiger yahoo.com> writes:
In article <d6pmnb$1m94$1 digitaldaemon.com>, Geert says...

You should compile it in release mode and without unittest...
I'm not quite sure why it doesn't compile (and I really would like to 
know) if you try to compile it with unittest on and not in release mode.


Geert, thank you for your reply, "-release" does allow the code to link and run.


David L.

-------------------------------------------------------------------
"Dare to reach for the Stars...Dare to Dream, Build, and Achieve!"
-------------------------------------------------------------------

MKoD: http://spottedtiger.tripod.com/D_Language/D_Main_XP.html
May 22 2005
prev sibling parent reply Burton Radons <burton-radons smocky.com> writes:
David L. Davis wrote:


 Following the exact example as was posted in the D forum recently, the new
 std.boxer.d is giving the below link error when the unbox functions are being
 used. Is this a bug, or have I done something wrong in the code?


Wow, this is a flash-back, I haven't seen that link error in years. 
This is caused because it's the first template in Phobos designed to be 
used by a client, so it's just exposed a code generation problem in 
templates when the -release flags are mismatched (Phobos is now compiled 
with -release).

Here's an example:

    [module f - compiled with -release]
    int i; // to keep the assert from being constfolded

    template T ()
    {
        void T ()
        {
            assert (i);
        }
    }

    [module g - compiled without -release]
    import f;

    void main ()
    {
        T! () ();
    }
May 22 2005
parent Burton Radons <burton-radons smocky.com> writes:
Burton Radons wrote:


 David L. Davis wrote:
 

 Following the exact example as was posted in the D forum recently, the 
 new
 std.boxer.d is giving the below link error when the unbox functions 
 are being
 used. Is this a bug, or have I done something wrong in the code?


 
 
 Wow, this is a flash-back, I haven't seen that link error in years. This 
 is caused because it's the first template in Phobos designed to be used 
 by a client, so it's just exposed a code generation problem in templates 
 when the -release flags are mismatched (Phobos is now compiled with 
 -release).


Err, rather "the first template in Phobos designed to be used by a 
client that asserts".  Forgot about stdarg.
May 22 2005