These are known bugs to be fixed or need to be tested:
*=something that QLIB does not support (is not a bug)
-=a bug in QLIB that I acknowledge

 * I can not seem to get debug info inserted into the EXE, even with the
   /z option.  Something to do with compilers.

 * QLIB does not support O_TEXT (or "t" with fopen()).  This means that
   you must always use \r\n because the \n will not be translated
   to \r\n for you.  I have no idea how BC (or WC) support O_TEXT.
   You will get a Warning if you try and use O_TEXT mode and QLIB will
   use O_BINARY instead.  _fmode by default is set to O_TEXT since this
   is the standard so if you do not provide O_BINARY then O_TEXT will be
   assumed but then QLIB will switch to O_BINARY with the Warning message.

 * win95_close_enable() does not work. I think it's because one of the
   RMODE sregs is not setup correctly to call this func.  Any ideas?

 - Ctrl+Break checking now works some what better (signal = SIGINT)
     DOS/4GW : works only with Ctrl+C (Ctrl+Break will terminate immediately)
     PMODE/W : works only with Ctrl+C (Ctrl+Break will terminate immediately)
     WDOSX : works with both
   NOTE : '^C' is printed by DOS before signal() gains control.
   So I suggest you use WDOSX if you plan on trapping Ctrl+Break properly.
   I need more info on how Ctrl+C and Ctrl+Break works to finish
   these exception handlers.
   Either way Ctrl+Break (and Ctrl+C) work pretty good now, but not perfect
   yet.
