Get keyerror accessing row by index in 16.0 Pandas dataframe in Python 3.4

Why do I keep getting a key error?

[edit] Here is the data:

GEO,LAT,LON
AALBORG   DENMARK,57.0482206,9.9193939
AARHUS   DENMARK,56.1496278,10.2134046
ABBOTSFORD BC  CANADA,49.0519047,-122.3290473
ABEOKUTA   NIGERIA,7.161,3.348
ABERDEEN   SCOTLAND,57.1452452,-2.0913745

[end edit] Can't find row by index, but its clearly there:

geocache = pd.read_csv('geolog.csv',index_col=['GEO']) # index_col=['GEO']
geocache.head()

Shows

                      LAT           LON
GEO     
AALBORG DENMARK       57.048221     9.919394
AARHUS DENMARK        56.149628    10.213405
ABBOTSFORD BC CANADA  49.051905  -122.329047
ABEOKUTA NIGERIA       7.161000     3.348000
ABERDEEN SCOTLAND     57.145245    -2.091374

So then I test it:

x = 'AARHUS DENMARK'
print(x)
geocache[x]

And this is what I get:

AARHUS DENMARK

KeyError Traceback (most recent call last) in () 2 x = u'AARHUS DENMARK' 3 print(x) ----> 4 geocache[x]

C:\Users\g\Anaconda3\lib\site-packages\pandas\core\frame.py in __getitem__(self, key)
   1785             return self._getitem_multilevel(key)
   1786         else:
-> 1787             return self._getitem_column(key)
   1788 
   1789     def _getitem_column(self, key):

C:\Users\g\Anaconda3\lib\site-packages\pandas\core\frame.py in _getitem_column(self, key)
   1792         # get column
   1793         if self.columns.is_unique:
-> 1794             return self._get_item_cache(key)
   1795 
   1796         # duplicate columns & possible reduce dimensionaility

C:\Users\g\Anaconda3\lib\site-packages\pandas\core\generic.py in _get_item_cache(self, item)
   1077         res = cache.get(item)
   1078         if res is None:
-> 1079             values = self._data.get(item)
   1080             res = self._box_item_values(item, values)
   1081             cache[item] = res

C:\Users\g\Anaconda3\lib\site-packages\pandas\core\internals.py in get(self, item, fastpath)
   2841 
   2842             if not isnull(item):
-> 2843                 loc = self.items.get_loc(item)
   2844             else:
   2845                 indexer = np.arange(len(self.items))[isnull(self.items)]

C:\Users\g\Anaconda3\lib\site-packages\pandas\core\index.py in get_loc(self, key, method)
   1435         """
   1436         if method is None:
-> 1437             return self._engine.get_loc(_values_from_object(key))
   1438 
   1439         indexer = self.get_indexer([key], method=method)

pandas\index.pyx in pandas.index.IndexEngine.get_loc (pandas\index.c:3824)()

pandas\index.pyx in pandas.index.IndexEngine.get_loc (pandas\index.c:3704)()

pandas\hashtable.pyx in pandas.hashtable.PyObjectHashTable.get_item (pandas\hashtable.c:12349)()

pandas\hashtable.pyx in pandas.hashtable.PyObjectHashTable.get_item (pandas\hashtable.c:12300)()

KeyError: 'AARHUS DENMARK'

No extra spaces or non-visible chars, Tried putting r and u before the string assignment with no change in behavior.

Ok, what am I missing?

Solution

As you didn't pass a sep (separator) arg to read_csv the default is comma separated. As your csv contained spaces/tabs after the commas then these get treated as part of the data hence your index data contains embedded spaces.

So you need to pass additional params to read_csv:

pd.read_csv('geolog.csv',index_col=['GEO'], sep=',\s+', engine='python')

The sep arg means that it will look for commas with optional 1 or more spaces in front of the commas, we pass engine='python' as the c engine does not accept a regex for separators.