Added %s support for labels.

[pelson]

Fixes #775.

I did consider only testsing using png, since the code being tested is backend agnostic. Any thoughts?

[mdboom]

Isn't this equivalent to str(s)?

[pelson]

Apparently not (I always thought it was until this evening):

>>> a = u"Hello\u2026"
>>> str(a)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
UnicodeEncodeError: 'ascii' codec can't encode character u'\u2026' in position 5: ordinal not in range(128)
>>> '%s' % a
u'Hello\u2026'

[pelson]

How about I add a unicode example to see what happens!

[efiring]

With the change below, it looks like set_label accepts virtually any python object! So, perhaps change the docstring to reflect that?

[pelson]

There is a tricky one! The text.py module, which does the same (and is where I stole the idea from) states:

ACCEPTS: string or anything printable with '%s' conversion.

How does that sound?

[efiring]

That works.

[mdboom]

I don't know if this ever made it into the code guidelines, but we've tried not to have unicode-encoded files like this, but instead use \u escape sequences for unicode characters. I know it's not as obvious -- but some developers are still using editors that don't handle unicode encodings correctly -- I'm not one of them ;)

[pelson]

I thought you might mention this. :-)
My unicode knowledge is weak at best, but I will try to figure out a suitable unicode string which can go in the label.

[efiring]

On 2012/08/20 3:35 AM, Michael Droettboom wrote:

I don't know if this ever made it into the code guidelines, but we've
tried not to have unicode-encoded files like this, but instead use |\u|
escape sequences for unicode characters. I know it's not as obvious --
but some developers are still using editors that don't handle unicode
encodings correctly -- I'm not one of them ;)

Thank you. I am one of them, maybe the only one.

This may help: http://www.utf8-chartable.de

One incredibly verbose but effective mechanism for including unicode
looks like this:

u'D\N{LATIN SMALL LETTER E WITH ACUTE}velopp\N{LATIN SMALL LETTER E WITH
ACUTE}s'

String substitution with %s can be used to make it shorter.

Or one can use this:
u'D\xe9velopp\xe9s'

again with the option of string substitution to make it a bit clearer.

Eric

[mdboom]

Other than my comment about using escape sequences, this seems good to go.

[pelson]

Thanks @efiring. That was very helpful - I definitely need to read up about best practices wrt to unicode handling.
This should now be good to merge.