Question

When I use this code (adapted from Stephen Holiday code – thanks, Stephen for your code!):

#!/usr/bin/env python    
# -*- coding: utf-8 -*-
# -*- coding: utf-8 -*-

"""
USSSALoader.py
"""
import os
import re
#import urllib2
from zipfile import ZipFile
import csv
import pickle

def getNameList():
    namesDict=extractNamesDict()
    maleNames=list()
    femaleNames=list()
    for name in namesDict:
        counts=namesDict[name]
        tuple=(name,counts[0],counts[1])
        if counts[0]>counts[1]:
            maleNames.append(tuple)
        elif counts[1]>counts[0]:
            femaleNames.append(tuple)
    names=(maleNames,femaleNames)
    return names

def extractNamesDict():
    zf=ZipFile('names.zip', 'r')
    filenames=zf.namelist()

    names=dict()
    genderMap={'M':0,'F':1}

    for filename in filenames:
        file=zf.open(filename,'r')
        rows=csv.reader(file, delimiter=',')

        for row in rows:
            name=row[0].upper()
           # name=row[0].upper().encode('utf-8')
            gender=genderMap[row[1]]
            count=int(row[2])

            if not names.has_key(name):
                names[name]=[0,0]
            names[name][gender]=names[name][gender]+count

        file.close()
        print '\tImported %s'%filename
    return names

if __name__ == "__main__":
    getNameList()

I got this error:

  iterator = raw_query.Run(**kwargs)
  File "C:\Program Files (x86)\Google\google_appengine\google\appengine\api\datastore.py", line 1622, in Run
    itr = Iterator(self.GetBatcher(config=config))
  File "C:\Program Files (x86)\Google\google_appengine\google\appengine\api\datastore.py", line 1601, in GetBatcher
    return self.GetQuery().run(_GetConnection(), query_options)
  File "C:\Program Files (x86)\Google\google_appengine\google\appengine\api\datastore.py", line 1490, in GetQuery
    filter_predicate=self.GetFilterPredicate(),
  File "C:\Program Files (x86)\Google\google_appengine\google\appengine\api\datastore.py", line 1534, in GetFilterPredicate
    property_filters.append(datastore_query.make_filter(name, op, values))
  File "C:\Program Files (x86)\Google\google_appengine\google\appengine\datastore\datastore_query.py", line 107, in make_filter
    properties = datastore_types.ToPropertyPb(name, values)
  File "C:\Program Files (x86)\Google\google_appengine\google\appengine\api\datastore_types.py", line 1745, in ToPropertyPb
    pbvalue = pack_prop(name, v, pb.mutable_value())
  File "C:\Program Files (x86)\Google\google_appengine\google\appengine\api\datastore_types.py", line 1556, in PackString
    pbvalue.set_stringvalue(unicode(value).encode('utf-8'))
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe1 in position 1: ordinal not in range(128)

This happens when I have names with non-ASCII caracters (like “Chávez” or “Barañao”). I tried to fix this problem doing this:

     for row in rows:
           # name=row[0].upper()
            name=row[0].upper().encode('utf-8')
            gender=genderMap[row[1]]
            count=int(row[2])

But, then, I got this other error:

 File "C:\Users\CG\Desktop\Google Drive\Sci&Tech\projects\naivebayes\USSSALoader.py", line 17, in getNameList
    namesDict=extractNamesDict()
  File "C:\Users\CG\Desktop\Google Drive\Sci&Tech\projects\naivebayes\USSSALoader.py", line 43, in extractNamesDict
    name=row[0].upper().encode('utf-8')
UnicodeDecodeError: 'ascii' codec can't decode byte 0xed in position 3: ordinal not in range(128)

I also tried this:

def extractNamesDict():
    zf=ZipFile('names.zip', 'r', encode='utf-8')
    filenames=zf.namelist()

But ZipFile doesn’t have such argument.

So, how to fix that avoiding this UnicodeDecodeError for non-ASCII names?

I’m using this code with GAE.

Answer 1

It looks like your first traceback is AppEngine-related. Are you building a loader that will populate the datastore? If so, seeing the code that comprises the models and does the put‘ing would be helpful. I will probably be corrected by someone, but in order for that piece to work I believe you actually need to decode instead of encode (i.e. when you read the sheet prior to the put, convert the string to unicode by using decode('utf-8') or decode('latin1'), depending on your situation).

As far as your local code, I won’t pretend to know the deep internals of Unicode handling, but I’ve generally used decode() and encode() to handle these types of situations. I believe the correct encoding to use depends on the underlying text (meaning you’d need to know if it were encoded utf-8 or latin-1, etc.). Here is a quick test with your example:

>>> s = 'Chávez'
>>> type(s)
<type 'str'>
>>> u = s.decode('latin1')
>>> type(u)
<type 'unicode'>
>>> e = u.encode('latin1')
>>> print e
Chávez

In this case, I needed to use latin1 to decode the encoded string (I was using the terminal), but in your situation using utf-8 may very well work.