Python：2.7.6 熊猫：0.14 SQLAlchemy：0.9.4
尝试使用Pandas to_sql函数将数据帧写入Oracle DB时，出现以下错误
TypeError: expecting string, unicode or buffer object
Dataframe是通过将2个单独的数据帧附加在一起创建的，df1和df2。 df1信息如下：
Int64Index: 20 entries, 0 to 19
Data columns (total 19 columns):
AID 20 non-null int64
CHARGE_OFF_AMT 20 non-null object
FEE_PAID 20 non-null object
GRADE 20 non-null object
INT_PMT 20 non-null object
LATE_FEE_PMT 20 non-null object
OUTSTANDING_AMT 20 non-null object
PERIOD 20 non-null object
PRIN_PMT 20 non-null object
PRODUCT 20 non-null int64
PURCHASES_AMT 20 non-null object
PURCHASES_COUNT 20 non-null int64
RECOVERY_PMT 20 non-null object
TERM 20 non-null int64
TOTAL_PMT 20 non-null object
WHOLE_FLAG 20 non-null object
ORDERS_AMT 16 non-null float64
ORDERS_COUNT 16 non-null float64
IN_FUNDING 16 non-null float64
dtypes: float64(3), int64(4), object(12)
df2信息如下：
Int64Index: 1 entries, 0 to 0
Data columns (total 6 columns):
AID 1 non-null int64
PERIOD 1 non-null object
DEPOSITS 1 non-null int64
WITHDRAWALS 1 non-null int64
BONUS 1 non-null int64
TOTAL_CASH 1 non-null object
dtypes: int64(4), object(2)
然后我将df2附加到df1，结果是final_d：
>>final_d = df1.append(df2)
>>final_d.info()
Int64Index: 21 entries, 0 to 0
Data columns (total 23 columns):
AID 21 non-null int64
BONUS 1 non-null float64
CHARGE_OFF_AMT 20 non-null object
DEPOSITS 1 non-null float64
FEE_PAID 20 non-null object
GRADE 20 non-null object
INT_PMT 20 non-null object
IN_FUNDING 16 non-null float64
LATE_FEE_PMT 20 non-null object
ORDERS_AMT 16 non-null float64
ORDERS_COUNT 16 non-null float64
OUTSTANDING_AMT 20 non-null object
PERIOD 21 non-null object
PRIN_PMT 20 non-null object
PRODUCT 20 non-null float64
PURCHASES_AMT 20 non-null object
PURCHASES_COUNT 20 non-null float64
RECOVERY_PMT 20 non-null object
TERM 20 non-null float64
TOTAL_CASH 1 non-null object
TOTAL_PMT 20 non-null object
WHOLE_FLAG 20 non-null object
WITHDRAWALS 1 non-null float64
dtypes: float64(9), int64(1), object(13)
最后，我试图将final_d写入oracle 11g数据库。
>>db = sqlalchemy.create_engine("oracle+cx_oracle://foo:bar@(DESCRIPTION = (LOAD_BALANCE=on) (FAILOVER=ON) \
(ADDRESS = (PROTOCOL = TCP)(HOST = host.address.com)(PORT = 1521)) (CONNECT_DATA = (SERVER = DEDICATED) \
(SERVICE_NAME = replica)))")
notes_cash = all_notes_info.append(all_cash_info)
>>final_d.to_sql('data_table', db, if_exists='append', index=True)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/pandas/core/generic.py", line 950, in to_sql
index_label=index_label)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/pandas/io/sql.py", line 440, in to_sql
index_label=index_label)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/pandas/io/sql.py", line 815, in to_sql
table.insert()
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/pandas/io/sql.py", line 584, in insert
self.pd_sql.execute(ins, data_list)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/pandas/io/sql.py", line 783, in execute
return self.engine.execute(*args, **kwargs)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/engine/base.py", line 1682, in execute
return connection.execute(statement, *multiparams, **params)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/engine/base.py", line 720, in execute
return meth(self, multiparams, params)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/sql/elements.py", line 317, in _execute_on_connection
return connection._execute_clauseelement(self, multiparams, params)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/engine/base.py", line 817, in _execute_clauseelement
compiled_sql, distilled_params
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/engine/base.py", line 947, in _execute_context
context)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/engine/base.py", line 1111, in _handle_dbapi_exception
util.reraise(*exc_info)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/engine/base.py", line 917, in _execute_context
context)
File "/Users/adsouza/anaconda/lib/python2.7/site-packages/sqlalchemy/dialects/oracle/cx_oracle.py", line 916, in do_executemany
cursor.executemany(statement, parameters)
TypeError: expecting string, unicode or buffer object
我能够使用这种方法成功地将df1和df2写入oracle但是无法编写附加的数据帧。
以下是最终df的示例：
AID BONUS CHARGE_OFF_AMT DEPOSITS FEE_PAID GRADE \
0 5819681 NaN 36600.565886430978 NaN 0 E
INT_PMT IN_FUNDING LATE_FEE_PMT ORDERS_AMT ... \
0 113099.457782700363 257875 76.07 1091450 ...
PRIN_PMT PRODUCT PURCHASES_AMT PURCHASES_COUNT RECOVERY_PMT \
0 290035.650142431736 1 574400 2767 0
TERM TOTAL_CASH TOTAL_PMT WHOLE_FLAG WITHDRAWALS
0 36 NaN 403211.177925132099 FRACTIONAL NaN