Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #98029 > unrolled thread

Python 2 vs Python 3 for teaching

Back to article view | Back to comp.lang.python

Contents

  Python 2 vs Python 3 for teaching Chris Angelico <rosuav@gmail.com> - 2015-11-01 20:43 +1100
    Re: Python 2 vs Python 3 for teaching beliavsky@aol.com - 2015-11-02 05:15 -0800
      Re: Python 2 vs Python 3 for teaching Chris Angelico <rosuav@gmail.com> - 2015-11-03 01:10 +1100

#98029 — Python 2 vs Python 3 for teaching

I'm proud to say that a Python tutoring company has just converted its
course over from teaching Python 2.7 to teaching 3.x. For the
naysayers out there, it actually wasn't much of a transition; putting
parentheses around all print calls, plus changing the way virtual
environments get created, pretty much covered it. The difference
between well-written 2.7 code and well-written 3.4 code is really not
huge.

Interestingly, the bytes/unicode distinction wasn't much of an issue.
Under 2.7, a lot of functions return Unicode strings, and their reprs
show a u prefix, which disappears under Py3. In both cases, a Unicode
string returned from a library will compare equal to a simple
double-quoted string:

>>> import json
>>> json.loads('["Hello", "World"]')
[u'Hello', u'World']
>>> _[0] == "Hello"
True

The bulk of the changes were actually just changing displayed output
to match a change to some object's repr (eg "<type 'bool'>" becomes
"<class 'bool'>", and "set([1])" becomes "{1}"), or the exact text of
an exception (the TypeError from evaluating None[0] looks different,
but it's still a TypeError).

Who out there is currently teaching/tutoring/training using Python 2?
Push to the common subset (parenthesized single string prints, never
assuming int/int yields int, etc), with a view to migration - it's
easier than you might think!

(This isn't meant to be an ad for a specific company, so much as a
general recommendation to push to Py3, but they deserve a bit of a
shout-out anyway. The company is Thinkful, www.thinkful.com.)

ChrisA

[toc] | [next] | [standalone]

#98096

I think Python 2.x is still used more than Python 3.x in scientific computing. The Python books I have in this area, such as "Python for Finance: Analyze Big Financial Data" and "Python for Data Analysis", still use Python 2.x . An aspiring computational scientist, data scientist, or financial quant may still be better off learning Python 2.x but using print(x) rather than print x and doing other things to future-proof his code.

[toc] | [prev] | [next] | [standalone]

#98100

On Tue, Nov 3, 2015 at 12:15 AM, beliavsky--- via Python-list
<python-list@python.org> wrote:
> I think Python 2.x is still used more than Python 3.x in scientific computing. The Python books I have in this area, such as "Python for Finance: Analyze Big Financial Data" and "Python for Data Analysis", still use Python 2.x . An aspiring computational scientist, data scientist, or financial quant may still be better off learning Python 2.x but using print(x) rather than print x and doing other things to future-proof his code.
>

That doesn't mean that Python 3 *can't* be used. Far as I know, all
the key libraries (numpy, pandas, statsmodels, scipy) are available
for Python 3 as well. Recommending the use of Python 2 simply because
all the books you have teach Python 2 is a purely circular argument.

But yes. If you're going to use Py2, aim for the common subset. Good
Py2 code is a lot more similar to good Py3 code than an enumeration of
language-level differences would suggest.

ChrisA

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web