[Pythonmac-SIG] timeout of connection?

ken manheimer ken.manheimer at gmail.com
Thu Dec 17 21:24:01 CET 2009


On Thu, Dec 17, 2009 at 2:51 PM, Jon Christopher
<jon.christopher at gmail.com>wrote:

> Thanks, I'll try the fix when I get home.  However, I should reiterate
> that I'm not seeing a timeout  (which would cause an exception,
> right?).  It's just that I (eventually) issue an applescript command
> which seems to go nowhere.
>
> As I understand the id-rollover bug:
>
> 1) it will cause events which want a reply to eventually timeout.
> 2) which would trigger a python exception
>
> I'm not sure how that would be at play here since:
>
> 1) the command I'm sending doesn't (apparently) need a reply
> 2) I'm not getting exceptions.
>
> Did I missunderstand something?
>

i don't think you're misunderstanding anything, but the two sets of
circumstances are not mutually exclusive.  i don't know what the system
you're dealing with is doing with the command you're sending, in particular
(1) whether it's internally issuing a command that does need a reply, in the
course of its handling - that seems likely - and (2) whether it's
"protecting" against runtime disruption by swallowing any exceptions
encountered.  perhaps you know all the systems involved well enough to be
sure this isn't happening.  though it's often bad practice to swallow errors
without any signs, (2) is not unusual or unwarranted in executive style
programs, though they should issue some kind of warning, like in a log, that
can be missed, and too often nothing is issued at all.

that doesn't mean the problem is happening and being covered up, just that
it might be, and a little bit of "experimental computer science" is probably
worth the effort.
-- 
ken
http://myriadicity.net



> On Thu, Dec 17, 2009 at 11:10 AM, ken manheimer <ken.manheimer at gmail.com>
> wrote:
> > On Thu, Dec 17, 2009 at 1:26 PM, Brad Howes <howes at ll.mit.edu> wrote:
> >>
> >> On Dec 16, 2009, at 2:26 PM, Jon Christopher wrote:
> >>
> >> > Thanks for the reply, Brad.
> >> >
> >> > I'm not familiar with the issue you describe.  Can you provide a
> >> > pointer to some related info?  How would I test to see if my script is
> >> > affected?
> >>
> >> http://db.tidbits.com/article/10643
> >
> > and see these threads for the description of the problem in python
> appscript
> > and the fix:
> >
> > http://old.nabble.com/-ann--appscript-0.20.2-tt26413893.html
> > and
> > http://old.nabble.com/-ann--py-appscript-0.21.0-tt26634638.html
> >
> > note that i experienced the problem in leopard as well as snow leopard.
> it
> > does seem to be well explained by the return id problem, even though the
> > above article attributes it to only snow leopard.  plus, has' fix seems
> to
> > have completely resolved the problem that i was, until the fix, seeing
> > consistently.  has the update failed to settle the problem you're seeing?
> > --
> > ken
> > http://myriadicity.net
> >
> >
> >>
> >> --
> >> Brad Howes
> >> Group 42
> >> MIT Lincoln Laboratory • 244 Wood St. • Lexington, MA 02173
> >> Phone: 781.981.5292 • Fax: 781.981.3495 • Secretary: 781.981.7420
> >
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pythonmac-sig/attachments/20091217/5e73f0b7/attachment-0001.htm>


More information about the Pythonmac-SIG mailing list