Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!apple!julius.cs.uiuc.edu!zaphod.mps.ohio-state.edu!wuarchive!uunet!mcsun!ukc!edcastle!aipna!cstr!rick
From: rick@cstr.ed.ac.uk (Rick Innis)
Newsgroups: comp.windows.x
Subject: idraw compilation error solution
Message-ID: <RICK.90Sep11150652@brodie.cstr.ed.ac.uk>
Date: 11 Sep 90 15:06:52 GMT
Sender: news@aipna.ed.ac.uk
Distribution: comp
Organization: CSTR, University of Edinburgh
Lines: 59


To recap, the problem I had was that when compiling idraw, I got the message

 errhandler.o: Undefined symbol set_new_handler() referenced from text segment

Thanks to Olof Damberg (oldam@liu.math.se) and Harald Vogt (harald@cs.ruu.nl)
for the fixes.

First Olof:

    The missing function 'set_new_handler' is in _handler.c in
	iv/src/libInterViews. 
A look at iv/src/libInterViews/Imakefile gives a clue to the solution:
    ---
    #if defined(Use_g_plusplus)
    INITOBJS = \
            _ctorlist.o \
            _dtorlist.o \
            _exit.o \
            _finish.o \
            _handler.o \
            _init.o
    #endif
    ---

The macro Use_g_plusplus is not defined anywhere; nor is it mentioned in the
documentation, as far as I can see. I defined this in IVlocal.def and remade
the makefiles.

The other part of the fix came from Harald Vogt:

Line 33 of iv/src/bin/idraw/errhandler.c line 33 has the following declaration;
  extern NewHandler* set_new_handler(NewHandler*);
which should read
  extern "C" NewHandler* set_new_handler(NewHandler*);
to agree with the declaration of set_new_handler() in
iv/src/libInterView/_handler.c 

Finally, there is a bug in libInterViews which causes idraw to crash on
exiting; this is the patch for it:

*** X11-world.c.~1~     Wed Jan 31 20:28:51 1990
--- X11-world.c Wed Aug  1 08:54:09 1990
***************
*** 164,169 ****
--- 164,171 ----

  World::~World () {
      XCloseDisplay(rep->display());
+     // Bug fix from net, AD, 1-8-90
+     canvas->id = nil; // to prevent ~Canvas from accessing _world
      delete rep->_itable;
      delete rep;
  }

Thanks to both parties, as well as Per Abrahamsen and Ciaran McHale, who also
sent me suggestions as to the cause of the problem.

	--Rick.