I am very tired about all this unicode problems. I need to build graph with object nodes. And I need do show this graph. I have problems with visualization unicode strings. So... To make object as graph node, I need to override equal and hash methods
class VkUser:
def __init__(self,uid=None,f_name = None,l_name = None,json=None):
if(json==None):
self.uid=uid
self.l_name=l_name
self.f_name=f_name
else:
self.uid = json['uid']
self.f_name = json['first_name']
self.l_name = json['last_name']
def __eq__(self, other):
if isinstance(other, VkUser):
return (self.uid == other.uid)
return NotImplemented
def __ne__(self, other):
result = self.__eq__(other)
if result is NotImplemented:
return result
return not result
def __hash__(self):
return hash(self.uid)
But, if I want to not Pointers as nodes of the graph, i need to override str
def __str__(self):
return '%s %s'%(self.f_name,self.l_name)
It works fine, while I have just english letters.
But I have Json input with unicode russian leters, and I need to show them on graph. Somethisg like
Me = VkUser(111,u'\u0410\u043b\u0435\u043a\u0441\u0430\u043d\u0434\u0440',u'\u0410\u043b\u0435\u043a\u0441\u0430\u043d\u0434\u0440')
And now I have got en error
label=str(label) # this will cause "1" and 1 to be labeled the same
UnicodeEncodeError: 'ascii' codec can't encode characters in position 0-8: ordinal not in range(128)
I tried different variants in str
def str(self): return u'%s %s'%(self.f_name,self.l_name) Same error
def __str__(self):
res = u'%s %s'%(self.f_name,self.l_name)
return res.encode('utf-8')
...
ValueError: matplotlib display text must have all code points < 128 or use Unicode strings
Help me please, I am rely tired from this.