unicodedata with chr() not the same between python 3.4 and 3.5

[Chris Angelico]

On Wed, Dec 23, 2015 at 2:27 AM, Vincent Davis <vincent at vincentdavis.net> wrote:
> I was expecting the code below to be the same between python3.4 and 3.5. I
> need a mapping between the integers and unicode that is consistant between
> 3.4 and 3.5
>
>>>>
> import unicodedata
>>>>
> u = ''.join(chr(i) for i in range(65536) if (unicodedata.category(chr(i))
> in ('Lu', 'Ll')))[945:965]

Not sure why you're slicing it like this, but it makes little
difference. The significant thing here is that the newer Pythons are
shipping newer Unicode data files, and some code points have changed
category.

rosuav at sikorsky:~$ python3.4
Python 3.4.2 (default, Oct  8 2014, 10:45:20)
[GCC 4.9.1] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import unicodedata
>>> unicodedata.unidata_version
'6.3.0'
>>>
rosuav at sikorsky:~$ python3.5
Python 3.5.0b1+ (default:7255af1a1c50+, May 26 2015, 00:39:06)
[GCC 4.9.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import unicodedata
>>> unicodedata.unidata_version
'7.0.0'
>>>
rosuav at sikorsky:~$ python3.6
Python 3.6.0a0 (default:6e114c4023f5, Dec 20 2015, 19:15:28)
[GCC 4.9.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import unicodedata
>>> unicodedata.unidata_version
'8.0.0'
>>>

Have a read here of what changed in those two major versions:

http://unicode.org/versions/Unicode7.0.0/
http://unicode.org/versions/Unicode8.0.0/

I'm not sure what the best way is to create the mapping you want, but
I would advise freezing it to a specific set of codepoints in your
source code, rather than depending on something external.

ChrisA