[python-uk] Library for (undirected) graphs in Python?

René Dudfield renesd at gmail.com
Tue Jan 25 12:30:24 CET 2011


Like on everything Jonathan said... and as moloko said 'the time is now'...
for python 3000.

Very nice list.

cu

On Tue, Jan 25, 2011 at 11:25 AM, Jonathan Hartley <tartley at tartley.com>wrote:

> On 25/01/2011 02:08, Alec Battles wrote:
>
>> Now, maybe the solution is to use Python 2.6 instead. Before starting
>>> working on my project I knew nothing about Python, which is one of the
>>> reasons I chose it over, say, Java, and thought that the 3rd version is
>>> the
>>> way to go. Is it not?
>>>
>> afaik, the main difference is the assert statement. i'm sure there are
>> other differences, but as someone who rarely uses python that
>> 'deeply', you should be fine if you start off on python 2.
>> _______________________________________________
>> python-uk mailing list
>> python-uk at python.org
>> http://mail.python.org/mailman/listinfo/python-uk
>>
> Personally I think there's *heaps* of new stuff in Python 3.0, 3.1 and 3.2
> which really make the language nicer to use. Some of these have been
> backported to Python 2.7, where it could be done without breaking
> compatibility, but if you can use version 3, then you really should!
>
> Some new features of Python 3.0 that I care about:
> * Many things that used to return lists now return iterators. e.g.
> dict.keys(), map, filter, range, zip, etc. These can be used just the same
> in 95% of your code, and are much more memory efficient, especially when
> chaining them.
> * Set literals: {1, 2, 3} creates a set.
> * Division is now sane: 5/2 returns 2.5, instead of 2. Use operator '//'
> for the old behaviour.
> * extended iterable unpacking: stuff like this just works:  "a, b,
> *everything_else = my_list"
> * packages and modules in the standard library have been moved and renamed
> to be more consistent and comprehensible.
> * Ordering comparisons (<, >=, etc) are now sane - comparing different
> types will now in the general case raise a type error, unless they are
> specific pairs of types which make sense to compare (e.g. int to float)
> * Dict comprehensions: Mirroring list comprehensions, create dicts using
> "{k: v for k, v in stuff}"
> * no more confusion between int and long - everything is now an int (which
> behaves much like the old 'long' did)
> * no more confusion between old- and new-style classes, everything is now a
> new-style class
>
>    Jonathan
>
>
> --
> Jonathan Hartley      Made of meat.      http://tartley.com
> tartley at tartley.com   +44 7737 062 225   twitter/skype: tartley
>
>
> _______________________________________________
> python-uk mailing list
> python-uk at python.org
> http://mail.python.org/mailman/listinfo/python-uk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-uk/attachments/20110125/38ad281e/attachment.html>


More information about the python-uk mailing list