Pandas cx_Oracle使用方法

正确安装好cx_oracle之后，要使用它来连接到oracle数据库进行操作，具体应该分3步走：

第一步：导入cx_Oracle，建立连接

>>>  import cx_Oracle      #导入模块
>>>  db = cx_Oracle.connect('hr', 'hrpwd', 'localhost:1521/XE') 建立连接，3个参数分开写

特别注意：这里的 'localhost:1521/XE'可以是你oracle net manager配置的链接名，如oracl，我就是在这里耽搁了半天

>>>  db1 = cx_Oracle.connect('hr/hrpwd@localhost:1521/XE') 建立连接，3个参数连写

>>>  dsn_tns = cx_Oracle.makedsn('localhost', 1521, 'XE')

>>>  print dsn_tns
(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=localhost)(PORT=1521)))
(CONNECT_DATA=(SID=XE)))
>>>  db2 = cx_Oracle.connect('hr', 'hrpwd', dsn_tns)

>>> <strong style="line-height: normal;">print db.version</strong><br style="line-height: normal;" />
10.2.0.1.0<br style="line-height: normal;" />
>>> <strong style="line-height: normal;">versioning = db.version.split('.')</strong><br style="line-height: normal;" />
>>> <strong style="line-height: normal;">print versioning</strong><br style="line-height: normal;" />
['10', '2', '0', '1', '0']<br style="line-height: normal;" />
>>> <strong style="line-height: normal;">if versioning[0]=='10':</strong><br style="line-height: normal;" />
...       <strong style="line-height: normal;">print "Running 10g"</strong><br style="line-height: normal;" />
... <strong style="line-height: normal;">elif versioning[0]=='9':</strong><br style="line-height: normal;" />
...      <strong style="line-height: normal;">print "Running 9i"</strong><br style="line-height: normal;" />
...<br style="line-height: normal;" />
Running 10g<br style="line-height: normal;" />
>>> <strong style="line-height: normal;">print db.dsn</strong><br style="line-height: normal;" />
localhost:1521/XE

第二步：建立Cursor光标

>>>cursor = db.cursor() 建立一个cursor

之后，我们可以调用这个cursor.execute（‘SQL‘） 来执行SQL语句。比如：

>>>cursor.execute(‘select * from tabs’)

执行完毕以后，可以调用cursor.fetchall()一次取完所有结果，或者cursor.fetchone()一次取一行结果

>>>row=cursor.fetchall()

>>>for x in row:

      For y in x:

         Print y,

       Print

这样就可以按照表格的形式打印取得的结果了！

在从oracle取出数据的时候，考虑到它的数据类型了吗？下面就是数据类型的对应表

Datatypes

During the fetch stage, basic Oracle data types get mapped into their Python equivalents. cx_Oracle maintains a separate set of data types that helps in this transition. The Oracle - cx_Oracle - Python mappings are:

Oracle	cx_Oracle	Python
VARCHAR2 NVARCHAR2 LONG	cx_Oracle.STRING	str
CHAR	cx_Oracle.FIXED_CHAR
NUMBER	cx_Oracle.NUMBER	int
FLOAT		float
DATE	cx_Oracle.DATETIME	datetime.datetime
TIMESTAMP	cx_Oracle.TIMESTAMP
CLOB	cx_Oracle.CLOB	cx_Oracle.LOB
BLOB	cx_Oracle.BLOB

带参数的查询：

>>> named_params = {'dept_id':50, 'sal':1000}
>>> query1 = cursor.execute('SELECT * FROM employees 
WHERE department_id=:dept_id AND salary>:sal', named_params)
>>> query2 = cursor.execute('SELECT * FROM employees 
WHERE department_id=:dept_id AND salary>:sal', dept_id=50, sal=1000)

这种是名字参数，还可以按位置参数：

r1 = cursor.execute('SELECT * FROM locations 
WHERE country_id=:1 AND city=:2', ('US', 'Seattle'))

注意：

当只有一次参数的时候，也要把它写成元组的形式，比如

Cursor.execute(‘select name from user where id=:1’,(login_Id,))

千万要注意，login_id后面还带有一个逗号，如果没有逗号，他其实就是一个数据对象，但是当他后面有个逗号的时候，他就变成了元组的一个数据项，千万要记住啊，我就是在这里徘徊了很久。！

Cursor. Prepare的用法，

这个方法就是在prepare之后，你再去execute的时候，就不用写上sql语句参数了

>>> cursor.prepare('SELECT * FROM jobs WHERE min_salary>:min')

>>> r = cursor.execute(None, {'min':1000}) #注意，第一个参数是None,

一次执行多条sql语句

Large insert operations don't require many separate inserts because Python fully supports inserting many rows at once with the cx_Oracle.Cursor.executemany method. Limiting the number of execute operations improves program performance a lot and should be the first thing to think about when writing applications heavy on INSERTs.

Let's create a table for a Python module list, this time directly from Python. You will drop it later.

>>> create_table = """<br style="line-height: normal;" />
CREATE TABLE python_modules (<br style="line-height: normal;" />
  module_name VARCHAR2(50) NOT NULL,<br style="line-height: normal;" />
  file_path VARCHAR2(300) NOT NULL<br style="line-height: normal;" />
)<br style="line-height: normal;" />
"""<br style="line-height: normal;" />
>>> from sys import modules<br style="line-height: normal;" />
>>> cursor.execute(create_table)<br style="line-height: normal;" />
>>> M = []<br style="line-height: normal;" />
>>> for m_name, m_info in modules.items():<br style="line-height: normal;" />
...     try:<br style="line-height: normal;" />
...       M.append((m_name, m_info.__file__))<br style="line-height: normal;" />
...     except AttributeError:<br style="line-height: normal;" />
...       pass<br style="line-height: normal;" />
...<br style="line-height: normal;" />
>>> len(M)<br style="line-height: normal;" />
76<br style="line-height: normal;" />
>>> cursor.prepare("INSERT INTO python_modules(module_name, file_path) VALUES (:1, :2)")<br style="line-height: normal;" />
>>> cursor.executemany(None, M)<br style="line-height: normal;" />
>>> db.commit()<br style="line-height: normal;" />
>>> r = cursor.execute("SELECT COUNT(*) FROM python_modules")<br style="line-height: normal;" />
>>> print cursor.fetchone()<br style="line-height: normal;" />
(76,)<br style="line-height: normal;" />
>>> cursor.execute("DROP TABLE python_modules PURGE")