moin Kees,   
      
   Friday May 24 2013 13:20, you wrote to Roy Witt:   
      
    KE>  Lets hope that we can entice Ulrich to have a go, or whoever compiled   
    KE> the executables for OS/2.   
      
   rel 3.3.3   
      
   I've tried to compile with OS/2  gcc3   
   but didn't found any make file that work to the prepared environment   
   that I've used for the binkley-xe compile   :-P   
   to customize a makefile to get it running, I'm missing the experience   
   so no success to get an OS/2 executable out of the mn333src   
      
   I've tried to compile with DOS OpenWatcom 1.9   
   compile still works and produces an exe file.   
   starting testing with my production r56 makenl.ctl works (as expected :) ..)   
      
   ok, using the "special" Pvt 000.0.0.0.0 test ...   
   was replaced  by makenl_ng with Pvt,..,-Unpublished-,...   
   no error output, no warning message, nothing, replaced in quiet.   
   using loglevel 4 doesn't give any more info about this behaviour.   
      
   adding Allowunpub 1  and Minphone 1   
   still runs and corrects to Pvt,..,-Unpublished-,...   
      
   so far, I did the tests in a dos box under winxp ...   
   now moving over to OS/2 warp3 ....   :D   
      
   and it runs _with_ the faulty Pvt line in a segment file   
   w/o termination  =:)   
      
   makenl_ng rev 3.3.3 openwatcom 1.9 compiled   
   used under OS/2 warp3 in a DOS box   
   succeeded   
   Pvt,..,000.0.0.0.0,...   replaced with  Pvt,..,000.0.0.0.0,...   =:D   
      
   So I cannot reproduce the reported error for a rev 3.3.3 DOS executable in an   
   OS/2 DOS box, nor that there is an error under DOS :)   
      
   If someone can send me in a compiled OS/2 executable, I can start with a test.   
   Also helpful, a test scenario description or better control and segment files   
   to check, to test reproducing the reported errors ...   
      
   The Pvt,..,000.0.0.0.0,... was an easy testing ... by adding one more testline   
   in an existing segment file and using the still working makenl.ctl file.   
      
      
    KE>  As a last thought, there should not be a problem in using NLMAKE, if   
    KE> it can handle the envisioned changes to the nodelist.   
      
    KE> Kees   
      
   regards, uli   ;-)   
      
   ---   
    * Origin: AMBROSIA - Frankfurt/Main - Germany (2:244/1120)