This is the mail archive of the mauve-discuss@sourceware.cygnus.com mailing list for the Mauve project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: SomeOne posted to me that the Process.exitValue() are ....


Tom Tromey wrote:

> >>>>> ">" == Uncle George <gatgul@voicenet.com> writes:
>
> >> I had originally though that the exit valus of the subprocess would
> >> be a composite of system & user exit codes. It appears ( right now
> >> ) on the face of it that the user exit code is only returned.
>
> What are the system and user exit codes?  I have no idea what you
> mean.

that would be something like ( as an example of a usr exit code )
exit(7);
an example of a system exit code would be
    char *p = 0xffffffffffffffff;
    *p = 1;                                        ( and now u get a
segfault/ ie  system exit code. )

>
>
> >> So what does this gotta do with this list ? i'd like to devel a
> >> test for process.exitValue(). I suppose this test will be somewhat
> >> 'unix' centric.
>
> >> what r my choices.
>
> This is probably hard to do in our framework.
> You could try writing a couple of shell scripts, and creating
> processes that run them.  Then check the exit status of the process
> objects.  (Or you could just use /bin/true and /bin/false, given that
> the scripts will probably be limited to Unix anyway.)
> You'd probably want to put a new tag to indicate that this test can
> only be run on certain systems ("POSIX" comes to mind).
>
> T

can I submit a 'c' pgm and have java exec a compile of the program ?
    exec("cc -g -o a a.c");
    process a = exec(" a 4");
    if( a.exitValue() != 4 ) inconsistent return codes ?

Its a little harder to 'cause' deliberately core-crashes, but a few can
be done
gat


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]