I changed a few things to allow for better pep8 compatability.

Corrected travis image link

Fix for #71 - looks good to me.

Picked versions are stored for later printing/writing in the Installer
also when show-picked-versions is set to false but update-versions-file is set
to a file. Previously, show-picked-versions=false would also disable the versions
file updating.

I changed it to make it applicable whether the picked versions
are printed or not. 'This info has been written to' doesn't work
if this hasn't been printed on the previous lines :-)

`update-versions-file = true` has no effect when `show-picked-versions` is
false.

The old "hashbang" url redirected to https://travis-ci.org//buildout/buildout, with two slashes, which didn't work. Travis bug, but using the new https://travis-ci.org/buildout/buildout url ourselves is easier :-)

Prep for release.

Provoke a travis-ci run. :)

Fix for #64: __pycache__ dir inside distutils scripts metadata directory

Reinout distutils script encoding fix for #65

Should never ever happen, of course, but somewhere in the universe exists a
distutils script with just the `#!bin/python` line... Added a len() to guard
against that.

Added a note as suggested by Marius

The second line is now inspected for a source encoding

I used Marius' suggestion (with one additional fix). The issue is demonstrated
by the test in @77cd8b16. This commit fixes
the failing test.

Fixes #64

A directory hacked inside the distutils `EGG-INFO/scripts` metadata directory
(which can apparently happen in python 3.3 with a `__pycache__` directory)
breaks the current distutils script detection code.

Refs #64

Fixed filename of distutils-scripts on windows. Fixes #52

Adding missing *.test files to manifest, needed for manuel tests

Merging buildout-versions' functionality into buildout

Unlike the console-script entrypoint scripts, the distutils
scripts didn't get a -script.py appended to their name on
windows. This is now fixed.

As discussed on the mailinglist: better to have python version checking
as a separate buildout option instead of hidden in the [versions] part.
And... the python version checking itself needs to be more robust.
sys.version is too long, minor/major needs explicit checking.

Drop trove classifiers for Python 2.4 and 2.5

I guess this was why the pull request wasn't build.