Support non-ASCII characters in Sage sources

[robertwb]

This involves at least fixing the documentation build process and trac to support utf-8. Possibly other components as well.

Discussion in  August 2009.

More recent discussion  here.

[mpatel]

Script to prepend coding to .py(x) files. Not a patch.

[mpatel]

From my brief experience with Unicode in SageNB sources (#7249 adds them to doctests), we may just need to

• Prepend # -*- coding: utf-8 -*- to every .py file. I assume we should do this for .pyx files, too. I've attached a script that can do this, although I'm sure there are more succinct ways.

• Use unicode strings for docstrings that contain non-ASCII Unicode characters. For example,

  def f(n):
      u"""
      Transmogrifies ``n``, heinously. ☺
      """
      return transmogrify(n, algorithm='heinous')

Note: At #8000, Minh suggested polling sage-devel about allowing non-ASCII characters in Sage library code. I'll try to do this soon.