2.6 and sys.exit()

hetchkay hetchkay at gmail.com
Fri Nov 13 07:19:41 CET 2009


On Nov 13, 10:28 am, hetchkay <hetch... at gmail.com> wrote:
> On Nov 13, 9:50 am, John Yeung <gallium.arsen... at gmail.com> wrote:
>
>
>
> > On Nov 12, 11:22 pm, r <rt8... at gmail.com> wrote:
>
> > > On Nov 12, 10:07 pm, hetchkay <hetch... at gmail.com> wrote:
> > > > I have the following in exit.py:
> > > > import sys
> > > > sys.exit(0)
>
> > > > I now try 'python -i exit.py':
>
> > > > In 2.5, the script exits as I would expect.
>
> > > > In 2.6, the following error is printed:
>
> > > > Traceback (most recent call last):
> > > >   File "exit.py", line 2, in <module>
> > > >     sys.exit(0)
> > > > SystemExit: 0
>
> > > > I couldn't find anything related to this in "What's new in 2.6".
>
> > > Look here ;-)
> > >http://docs.python.org/library/exceptions.html#exceptions.SystemExit
>
> > How does that answer the OP's question?  Namely, how to make 2.6
> > behave like 2.5?  (Even saying "You can't make 2.6 behave like 2.5"
> > would have been a better answer.)
>
> > Failing that, how about something that explains why 2.6 behaves
> > differently than 2.5, and why one of them is better or more correct
> > than the other?
>
> > Personally, I think 2.6's is probably the more correct behavior.
> > Specifically, if the point of the -i command line option is to force
> > interactive mode after completion of the script (which in this case
> > completed with sys.exit), then it should go to interactive mode
> > regardless of whether the script terminates "normally" or not.  I
> > think 2.5's behavior of allowing interactive mode to be skipped is
> > against the spirit of -i.  Unless -i meant something different in 2.5.
>
> > Is there some kind of environment variable to set up to control this?
>
> > John
>
> I can understand the behavior from a '-i' point of view. My
> requirement is somewhat different. Consider a geometry tool that can
> be used to create objects, merge objects etc. I have python 'commands'
> for doing any of these operations and for saving the objects to a
> file. The user could write a file containing a set of commands (in
> python format) and load this file in the GUI. Optionally, one of the
> commands could be to exit in which case the GUI should shut down. I am
> using 'execfile' to execute the file, and this does not exit if the
> user has used sys.exit (or even if I expose a command called 'Exit'
> which calls sys.exit).
>
> May be I should not be using execfile but I am not sure what else I
> should use. The user-written file could contain loops and other
> constructs as well.
>
> -Krishnan

Hi,

I had been starting my application with -i option. I have removed this
now and made a few other changes and things work OK now.

Thanks to everyone who helped in this.

Regards,
Krishnan



More information about the Python-list mailing list