On Wed, Aug 28, 2013 at 12:54 PM, Tiziano Zito opossumn...@gmail.comwrote:
UTF-8 characters are not very pleasant to work with, as only a fraction
of the world knows how to type them (that fraction depends on the
corresponding character).
I notice that they have creeped in a bunch of
FWIW, there are quite a few more files with non ASCII character:
[vagrant@localhost scikit-learn]$ find sklearn/ -name *.py -exec grep
--color='auto' -H -P -n [\x80-\xFF] '{}' \;
sklearn/naive_bayes.py:428:C.D. Manning, P. Raghavan and H. Sch��tze
(2008). Introduction to
2013/8/28 David Cournapeau courn...@gmail.com:
FWIW, there are quite a few more files with non ASCII character:
To be fair, I have to say that probably all of these are my doing.
--
Lars Buitinck
Scientific programmer, ILPS
University of Amsterdam
On Mon, Aug 26, 2013 at 2:17 PM, Lars Buitinck l.j.buiti...@uva.nl wrote:
2013/8/26 Gael Varoquaux gael.varoqu...@normalesup.org:
I just removed a non-ASCII character from truncated_svd.py, because it
was crashing a certain version of IPython when displaying the help.
UTF-8 characters
I'll have to side slightly against Lars on this one.
I agree with Lars that any software that doesn't support these is broken,
that Unicode looks better than other ad-hoc formatting.
If the software works, often the fonts won't. Personally if I'd need to
see the source and find characters
UTF-8 characters are not very pleasant to work with, as only a fraction
of the world knows how to type them (that fraction depends on the
corresponding character).
I notice that they have creeped in a bunch of our files. I'll try to
remove them (if I find time). I believe that we should
2013/8/28 Gael Varoquaux gael.varoqu...@normalesup.org:
Grmbl... I see your point. We can change back to all-ASCII if it's
important. It just feels like stepping back in time :)
Vlad is summarizing my point of view very well.
Let's us be ahead of our time with regards to algorithms and
On Wed, Aug 28, 2013 at 06:26:13PM +0200, Lars Buitinck wrote:
I just pushed a patch that should change all the docstrings to ASCII
only. I did not touch the narrative docs, nor the comments, so
contributors' names can still be spelt without ASCII workarounds.
Thanks heaps!
Oh, and I
On 08/28/2013 02:22 PM, Olivier Grisel wrote:
2013/8/28 Tiziano Zito opossumn...@gmail.com:
UTF-8 characters are not very pleasant to work with, as only a fraction
of the world knows how to type them (that fraction depends on the
corresponding character).
I notice that they have creeped in a
On 08/26/2013 03:17 PM, Lars Buitinck wrote:
That's probably my doing, because I strongly prefer Unicode to ASCII
art when reading formulas. In this case, it's just a hyphen so it's
not much of a problem, but I think the restriction to ASCII is a
throwback to the 1980s. My window system,
10 matches
Mail list logo
