I guess unicode variable names are restricted to letters, and that symbols and punctuation shoud be ignored (except _).

I have tested other dots (punctuation) they dont work. 

Only
http://www.fileformat.info/info/unicode/char/00b7/index.htm
oddly enough has worked so far.


$ python3.2
Python 3.2.3 (default, Sep 10 2012, 18:14:40) 
[GCC 4.6.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> foo⋅bar=42
  File "<stdin>", line 1
    foo⋅bar=42
            ^
SyntaxError: invalid character in identifier
>>> print(ord("foo⋅bar"[3]))
8901
>>> foo·bar = 42
>>> print(ord("foo·bar"[3]))
183

I have sampled randomly in the same block as MIDDLE DOT and it seems to behave correctly.

The rules for python identifiers are documented here:

  http://docs.python.org/dev/reference/lexical_analysis.html#identifiers

Are you saying that the behavior does not match the documentation?

http://www.fileformat.info/info/unicode/char/b7/index.htm

the unicode category is Po (Ponctuation). 

Empirically, it cannot start a variable name so according to the rules given in the lexical analyser it should be one of : Mn, Mc, Nd, Pc

Which is not the case Po not in [ Mn, Mc, Nd, Pc ].

Modulo my weak brain, it does not seem right.

The characters with the Other_ID_Continue property are also included, i.e.:

00B7          ; Other_ID_Continue # Po       MIDDLE DOT
0387          ; Other_ID_Continue # Po       GREEK ANO TELEIA
1369..1371    ; Other_ID_Continue # No   [9] ETHIOPIC DIGIT ONE..ETHIOPIC DIGIT NINE
19DA          ; Other_ID_Continue # No       NEW TAI LUE THAM DIGIT ONE

See http://unicode.org/Public/UNIDATA/PropList.txt