Access DateTimeIndexed dataframe by timestamp

[chinux23]

Hello,

I am new to pandas and thanks for this great library!

I have a data frame like this:

Gold_2012.head()

                              open  high    low close   volume
date_time                   
2012-01-02 18:01:00  1571.0  1571.0  1569.1  1569.8  351
2012-01-02 18:02:00  1569.8  1570.0  1569.7  1569.8  54
2012-01-02 18:03:00  1570.0  1570.0  1569.1  1569.9  247
2012-01-02 18:04:00  1570.0  1570.0  1569.8  1569.9  55
2012-01-02 18:05:00  1569.8  1569.9  1568.5  1568.5  48

I am trying to access the first element of this dataframe. If I use loc function, everything works out:

Gold_2012.loc[Gold_2012.index[0]]


open      1571.0
high      1571.0
low       1569.1
close     1569.8
volume     351.0
Name: 2012-01-02 18:01:00-06:00, dtype: float64

But if I do something like this, an error is thrown. Is this expected?

Gold_2012[Gold_2012.index[0]]

---

KeyError Traceback (most recent call last)
in ()
----> 1 Gold_2012[Gold_2012.index[0]]

/Users/chen/Virtualenvs/python3Env/lib/python3.3/site-packages/pandas/core/frame.py in getitem(self, key)
1926 else:
1927 # get column
-> 1928 return self._get_item_cache(key)
1929
1930 def _getitem_slice(self, key):

/Users/chen/Virtualenvs/python3Env/lib/python3.3/site-packages/pandas/core/generic.py in _get_item_cache(self, item)
568 return cache[item]
569 except Exception:
--> 570 values = self._data.get(item)
571 res = self._box_item_values(item, values)
572 cache[item] = res

/Users/chen/Virtualenvs/python3Env/lib/python3.3/site-packages/pandas/core/internals.py in get(self, item)
1382
1383 def get(self, item):
-> 1384 _, block = self._find_block(item)
1385 return block.get(item)
1386

/Users/chen/Virtualenvs/python3Env/lib/python3.3/site-packages/pandas/core/internals.py in _find_block(self, item)
1524
1525 def _find_block(self, item):
-> 1526 self._check_have(item)
1527 for i, block in enumerate(self.blocks):
1528 if item in block:

/Users/chen/Virtualenvs/python3Env/lib/python3.3/site-packages/pandas/core/internals.py in _check_have(self, item)
1531 def _check_have(self, item):
1532 if item not in self.items:
-> 1533 raise KeyError('no item named %s' % com.pprint_thing(item))
1534
1535 def reindex_axis(self, new_axis, method=None, axis=0, copy=True):

KeyError: 'no item named 2012-01-02 18:01:00-06:00'

[jreback]

@chinux23 .loc is the correct method of access here, as df[...] is really columnar access; though we do provide row slicing if possible
e.g.

df['2012-01-02 18:01'] would have worked , however using a datetime object was not supported

see associated PR #3925 which implementes this

[jreback]

@chinux23 changing this interface makes things inconsistent, so use .loc

[jreback]

closing this a not a bug