Hi Andrew,

Saturday May 25 2013 01:42, you wrote to me:

 AL>    Hello Ulrich!
 AL> 24 May 13 23:44, you wrote to Kees van Eeten:
 US>> For further testing I'm still await an OS/2 executable or a
 US>> working makefile script in my inbound  =:)

first, I've found the compile problem that is caused by
the filepointer around BatchFILE in main routine (makenl.c  line ~325)
see latest report under bug#17

What I could debug down is that the routine SearchMaxMSG() crashes
in the return sequence to the calling routine OpenMSGFile()

Whatever workaround I've tried to fix the return sequence
it crashes under OS/2 32-bit  :(

the problem may relate to the local int variable, that will be used
to return the result to the calling routine, where the result
will be assigned to a static int variable.
Don't know if there is a problem in the byte length of the used integer
variable under SearchMaxMSG() ?!?
Another problem that I've found, while declaring and defining
a 0 value to an integer variable in the SearchMaxMSG() routine was:

version a:
   int maxnum = 0;
   print maxnum results in =>  710691913

version b:
   int maxnum;
   maxnum = 0;
   print maxnum results in =>  0            (as expected)

so whatever memory block has been read under version a, it tells me
that something unusual is going on here with the integer type
of the used variables
At least this indicates the crash by an integer return of
the SearchMaxMSG() function  (is a 64 bit integer required here ?!?)

Someone with much more C/C++ experiences then me should dig into this code
section


 AL> Andrew

regards, uli   ;-)

---
 * Origin: AMBROSIA - Frankfurt/Main - Germany (2:244/1120)