Why site-packages?

Ian Kelly ian.g.kelly at gmail.com
Wed Jul 4 01:44:05 CEST 2012


On Tue, Jul 3, 2012 at 5:06 PM, Ian Kelly <ian.g.kelly at gmail.com> wrote:
> On Tue, Jul 3, 2012 at 4:38 PM, Dan Stromberg <drsalists at gmail.com> wrote:
>>> Right now this all seems highly speculative to me.  I think it might
>>> be informative, either to you or to me, to do an actual timing test.
>>> Why don't you try setting up two side-by-side installations of Python,
>>> one with all the site-packages cruft, and one trimmed down to only
>>> what you think should be in there, and see if you can measure a real
>>> difference in startup time?
>>
>> In the original stackoverflow thread I mentioned, there's a speed
>> comparison.
>
> That's just comparing using -S against not using -S, though.  As I
> understand it, you don't want to use -S; you still want to import site
> but just want to pare down your site-packages for a speedup.

I ran a quick test where I copied my Python 2.5 installation on
Windows and cleared out the site-packages folder and ran the test from
the stackoverflow thread.

With crufty site-packages: 0.18s
With empty site-packages: 0.16s
With -S (either installation): 0.10s

So there does appear to be some improvement, but only a fraction of
what I get by using -S.



More information about the Python-list mailing list