Closed GoogleCodeExporter closed 9 years ago
Stops instantly for me. I'm on 0.8 and Ubuntu 10.04
I will try this on 9.10 tomorrow. It may be the new PyQt makes a difference.
Try pressing ctrl-c just once and waiting. See if the event is not being seen
or if it is taking a long time to be acted on.
Original comment by miss...@hotmail.com
on 9 Jun 2010 at 1:49
I tried this on Ubuntu 9.10 and it stops instantly for me there also.
There must be some difference between our systems. My suspicion is that it
is just processing speed.
Original comment by miss...@hotmail.com
on 9 Jun 2010 at 7:55
Interesting; I'm using an Intel Core 2 Duo P8600 2.40GHz on a Thinkpad X200
w/4GB of RAM; you?
Original comment by aresnick...@gmail.com
on 9 Jun 2010 at 9:28
> Intel Core 2 Duo P8600 2.40GHz on a Thinkpad X200 w/4GB of RAM
Oh. Never mind. That can't be it.
I was going on this comment from issue 1:
>> What speed is the machine you are using?
> 1.4GHz
Have you tried this:
>> Try pressing ctrl-c just once and waiting. See if the event is not
>> being seen or if it is taking a long time to be acted on.
Original comment by miss...@hotmail.com
on 9 Jun 2010 at 10:42
> Have you tried this:
>>> Try pressing ctrl-c just once and waiting. See if the event is not
>>> being seen or if it is taking a long time to be acted on.
Yep, just tried, no joy. Interestingly, it works with some calls to duopoly,
and not others. See attached for a screenshot.
It stops duopoly(4,8,2,-2) instantly, but duopoly(4,2,8,-2) took five ^C's, and
I waited maybe 30s between each? You can see the offset of the path in the
attached screenshot.
Original comment by aresnick...@gmail.com
on 13 Jun 2010 at 12:55
Attachments:
> took five ^C's, and I waited maybe 30s between each
Ok. Seems like maybe pynguin is never receiving the ctrl-c event at all.
I'm going to make it easier to turn on debugging and put in a logger right
there where the keyboard events are coming in...
> You can see the offset of the path in the attached screenshot
Are you saying that before it stops the path changes?
Or is it that when you start it up again it is no longer on the same path?
Original comment by miss...@hotmail.com
on 13 Jun 2010 at 7:45
Pull the latest for a better logging setup.
Run with:
pynguin -D
for a log of each keypress. (tail -f logfile.log to watch the log).
Or run with:
pynguin -D info
for just a message for each press of ctrl-c if the other is too noisy.
You should get a log message immediately (within half a second anyhow) when the
keys are pressed.
Original comment by miss...@hotmail.com
on 13 Jun 2010 at 7:03
I've just been running "python main.py" from pynguin--how should I be running
pynguin?
Original comment by aresnick...@gmail.com
on 14 Jun 2010 at 1:00
Er. Sorry.
normal:
python main.py
full debug log:
python main.py -D
less verbose log:
python main.py -D info
Original comment by miss...@hotmail.com
on 14 Jun 2010 at 1:08
Hm; where does the log appear?
Original comment by aresnick...@gmail.com
on 14 Jun 2010 at 1:13
> where does the log appear?
Depends on what is in conf.py
By default:
logfile.log
Original comment by miss...@hotmail.com
on 14 Jun 2010 at 1:23
Huh, when I run it like that, I don't get any logfile. My conf.py doesn't seem
to have a logging variable:
"""
version = 'pynguin-0.8'
uidir = 'data/ui'
bug_url = 'http://code.google.com/p/pynguin/issues/list'
"""
Sorry, am I missing something?
Original comment by aresnick...@gmail.com
on 15 Jun 2010 at 2:31
Make sure you are running the most recent development version.
Make sure that after you do
hg pull
you then follow up with
hg update
When I first started using mercurial I sometimes forgot that 2nd step.
If you are on the most recent version and your conf.py looks like that (you
would have had to make modifications to the file) the program should crash when
run with the debug switch (-D)
Original comment by miss...@hotmail.com
on 15 Jun 2010 at 7:54
Oh, doh! Thanks--
Here's a reproduction of the problem, with logging.
Original comment by aresnick...@gmail.com
on 15 Jun 2010 at 9:02
Attachments:
Well... just seeing the log file does not really help.
What do all of those "Key: ...." lines correspond to?
What happens when you press ctrl-c?
tail -f logfile.log
and watch the logfile as you are using the program (hopefully you have enough
screen real estate that you can have pynguin over top of the log file window
and still see the log as you are pressing ctrl-c).
Does "Ctrl-C pressed" show up as soon as you press the keys?
Or is there a delay?
Or do you sometimes press ctrl-c and it never shows up?
I find it is helpful to use "focus follows mouse" then you can mouse over the
log window, press enter a few times to make a gap in the display of the log
file, then mouse back to pynguin and press ctrl-c.
Original comment by miss...@hotmail.com
on 15 Jun 2010 at 11:08
> Well... just seeing the log file does not really help.
> What do all of those "Key: ...." lines correspond to?
Hitting backspace, moving up in the history, etc. Getting to run the duopoly
command
> What happens when you press ctrl-c?
In the logfile:
"""
Key: 67
Ctrl-C pressed
Thread running
"""
The penguin stops immediately, and now I have no problems with it getting
confused in between, so it's all set! That's great. . .thanks!
> Does "Ctrl-C pressed" show up as soon as you press the keys?
Yep!
> Or is there a delay?
Nope.
> Or do you sometimes press ctrl-c and it never shows up?
No issues with that.
Original comment by aresnick...@gmail.com
on 15 Jun 2010 at 11:38
>The penguin stops immediately, and now I have no problems with it
> getting confused in between, so it's all set!
Ok. Cool.
I did have one bug that only showed up when I was not logging (I call it the
Heisenbug) so I will leave this open for now. Please let me know if it causes
any more trouble.
If I don't hear anything for a while, then I'll close this one out.
Thanks!
Original comment by miss...@hotmail.com
on 16 Jun 2010 at 12:24
Closing. Open new bug if you have any trouble.
Original comment by miss...@hotmail.com
on 30 Oct 2010 at 7:10
Original issue reported on code.google.com by
aresnick...@gmail.com
on 8 Jun 2010 at 11:07Attachments: