Disqus - Latest Comments for exarkun

Re: Don't Write Python Scripts, Write Python Libraries Posted on Apr 03, 2014 by Jeff Knupp

exarkun — Thu, 03 Apr 2014 21:06:07 -0000

Yes!

Re: Declaring dependencies in Python

exarkun — Tue, 09 Apr 2013 11:43:39 -0000

What's the reason for this idiom?

try:
....import argparse
except ImportError:
....install_requires.append('argparse')

Re: Python's += Is Weird, Part II | A. Jesse Jiryu Davis

exarkun — Wed, 02 Jan 2013 16:57:17 -0000

Not sure what you mean here, but it doesn't sound right. Consider:

>>> x = []
>>> y = id(x)
>>> x += [0]
>>> z = id(x)
>>> y == z
True
>>>

That is to say, += does *not* create a new list.

Re: Python's += Is Weird | A. Jesse Jiryu Davis

exarkun — Tue, 01 Jan 2013 16:12:51 -0000

You left off my favorite bit of weirdness:

x = ([],)
try: x[0] += [3]
except TypeError: print "+= raised TypeError"
print "x =", x

If you're not already familiar with this case, see if you can predict the output before running it :)

Re: Calling All Twisted Daemon Experts

exarkun — Mon, 27 Feb 2012 11:15:29 -0000

Use twistd and write a plugin (better) or a tac file (still okay) to glue it to your application.

twistd provides daemonization as well as reactor selection, logging configuration, privilege shedding, debugger integration, profiling, and several other esoteric platform-specific runtime configuration options.

Re: Watch This

exarkun — Mon, 27 Feb 2012 10:22:35 -0000

Ah, interesting. According to Wikipedia, perhaps they are only the same genus (Brassica) as the rest. However (also according to Wikipedia), the origin appears to be lost. Perhaps (my speculation) it was a very early cultivar of Brassica oleracea. Or not: a little more searching reveals that cabbage and turnips have a different number of chromosomes (20 vs 18). I don't really know, but this sounds like a stretch for being the result of selective cultivation. Apparently turnips and cabbage hybridize by both giving all (!) of their chromosomes to the offspring. Plants are crazy. Thanks for the challenge: I now find my vocabulary even less up to the task of describing these things. :)

Re: Tornado Unittesting: Eventually Correct

exarkun — Sun, 18 Dec 2011 21:07:47 -0000

Consider exploring some prior art. For example, trial has had pretty good support for asynchronous testing for many years: <http: twistedmatrix.com="" documents="" current="" core="" howto="" trial.html="">. Trial's solution is better than your assertEventuallyEquals because tests complete *exactly* when they complete, instead of one poll interval afterwards. If you have 7000 unit tests, you don't want to spend 700 seconds idling just because your tests don't immediately notice that they're already over.

Re: wxPython: ANN: Namespace Diff Tool

exarkun — Thu, 10 Nov 2011 11:55:43 -0000

Hi Mike, I tried to run this on Ubuntu 10.04 which has Python 2.6 and wxPython 2.8.10, but NDT.py fails with this import error:

Traceback (most recent call last): File "NDT.py", line 273, in <module> import wx.lib.agw.pybusyinfo as PBIImportError: No module named pybusyinfo

I know you're just republishing the announcement, but I don't see anywhere further upstream where comments can be posted.

Any ideas about this issue?

Re: Twisted.web vs Tornado Performance Test

exarkun — Sun, 13 Sep 2009 13:16:54 -0000

I've investigated this a bit further and what I found suggests that this behavior may result from the platform's use of syn cookies. When I was initially unable to reproduce this, I was using Ubuntu Hardy, which disables syn cookies by default. After some experimenting on Ubuntu Jaunty, after a tip from Thomas Herve, I was able to reproduce the problem (although only at a much higher concurrency level - around 500). Jaunty enables syn cookies by default. Then, returning to Hardy and enabling syn cookies, I was able to produce the behavior there. I was also able to provoke this behavior from ab running against the Tornado server - which is perfectly sensible if this explanation is correct, since it's primarily do to something the platform TCP stack is doing, not something either Twisted or Tornado is doing.

I haven't dug into this so deeply that I'm entirely confident that this completely explains the behavior, but I think it's a pretty good start. A direction for further investigation would be to write a client like ab but which reports more identifying information about each connection so that the failing connections can be traced through the entire network.

One thing I'm not sure about is whether OS X uses syn cookies. Some web pages I've found with Google suggest that it does, but it would be nice to get the information from a more authoritative/reliable source.

Re: Twisted.web vs Tornado Performance Test

exarkun — Fri, 11 Sep 2009 13:07:22 -0000

Also, I thought this might only show up after many more connections (I only ran 10,000 - but you ran 100,000). However, I get the same results when running 100,000 total requests.

Re: Twisted.web vs Tornado Performance Test

exarkun — Fri, 11 Sep 2009 13:05:26 -0000

> What is noticeable is that after 50 concurrent connections, twisted.web’s webserver began to close the connection and not complete the test.

This is strange. I've run this benchmark against Twisted Web many times in the past, and written lots of apps based on Twisted web. I've never seen this problem before. Indeed, I downloaded your twisted.web.py and ran ab against it at various levels of concurrency. The tests all completed, up to the highest setting I tried, 1000:

$ ab -c 1000 -n 10000 http://localhost:8080/
...
Concurrency Level: 1000
Time taken for tests: 6.25909 seconds
Complete requests: 10000
Failed requests: 0
Write errors: 0

I wonder if you can investigate the cause of the closed connections you saw? Or suggest what other factors might be involved which cause the behavior in your environment which are missing from mine.

Re: clepy library now on pypi

exarkun — Tue, 20 Jan 2009 18:05:42 -0000

Yep. It should be pretty clear when you've got it working properly, since the source file/line number the warnings system displays will be useful instead of redundant. :)

Re: clepy library now on pypi

exarkun — Tue, 20 Jan 2009 17:22:09 -0000

Sure, but no promises you'll like it (it currently has the same problem as you pointed out for the pylons function, for example (though I wouldn't mind fixing that)) . ;)

http://twistedmatrix.com/tr...

Re: clepy library now on pypi

exarkun — Tue, 20 Jan 2009 16:57:00 -0000

I already have a library for deprecating things. :)

Re: clepy library now on pypi

exarkun — Tue, 20 Jan 2009 16:44:13 -0000

Wouldn't it be better if it defaulted to using a stacklevel which pointed the warning message at the real caller, rather than at the implementation of the decorator?