[portland] Meeting Tomorrow?

[kirby urner]

Yeah, hard to debug this way, I agree.

Note that if you're trying to do a sort within sort, you can do like
a database key (m, n), where m, n are your two fields.

def extract(row):
        """return (m,n) for sorting purposes"""
	return (row[2],row[0])

>>> data = [
('pia', 'field1', 'scurve', 3, 4, 5),
('zaza', 'field1', 'bellcurve', 9, 8, 8),
('kirby', 'field1', 'bellcurve', 1, 2, 3),
('divine', 'field1', 'trapezoid', 1, 2, 3)]

>>> sorted(data, key = extract)
[('kirby', 'field1', 'bellcurve', 1, 2, 3),
('zaza', 'field1', 'bellcurve', 9, 8, 8),
('pia', 'field1', 'scurve', 3, 4, 5),
('divine', 'field1', 'trapezoid', 1, 2, 3)]

i.e. I'm sorting on field0 (proper name) *within* field 2 (test type).

Kirby


On Mon, Mar 10, 2008 at 4:55 PM, Rich Shepard <rshepard at appl-ecosys.com> wrote:
> On Mon, 10 Mar 2008, kirby urner wrote:
>
>  > My personal advice, coder to coder, would be to implement a non-itertools
>  > solution first, then go to itertools later if you feel there's a
>  > significant advantage.
>
>  Kirby,
>
>    That's excellent advice. Unfortunately, I did not come up with working
>  code before Dylan suggested the itertools approach.
>
>    What I got was the first curve, then all the rest on a single set of axes.
>  I had fits trying to find how to loop through all the tupples in the list,
>  grouping each set by item[1].
>
>    I don't care which approach is used, but I'm missing the python experience
>  to get it working.
>
>  Thanks,
>
>
>
>  Rich
>
>  --
>  Richard B. Shepard, Ph.D.               |  Integrity            Credibility
>  Applied Ecosystem Services, Inc.        |            Innovation
>  <http://www.appl-ecosys.com>     Voice: 503-667-4517      Fax: 503-667-8863
>  _______________________________________________
>  Portland mailing list
>  Portland at python.org
>  http://mail.python.org/mailman/listinfo/portland
>