[OT] Re: return level?

[Matthias Andree]

Nick Simicich <njs at scifi.squawk.com> writes:

> At 10:59 PM 2003-02-18 +0100, Matthias Andree wrote:
>
>>May I ask all of the interested parties to read waitpid(2) instead? The
>>original suspicion that 256 might be shifted by 8 bits is correct; the
>>application that misrepresents the bogofilter exit code as being 256
>>should have used the WEXITSTATUS macro (POSIX btw), but failed to do
>>so. Fix the caller.
>
> It is possible python does this to allow the caller to tell whether the
> command ended on a signal or normally in a single status value -- which
> is exactly what those macros do.  However, only one level of program can
> make that determination.  Unless there is a decision to do a "throw" or
> die equivalent if the command dies, keeping the status separate is the
> only way to allow the caller to tell "signal 1" from exit(1).

What Python library/function are you using to spawn bogofilter?

-- 
Matthias Andree