|
Description:
There're quite a few python ORM's. However, most are not easy to use. In Ruby on Rails's ActiveRecord ORM, you don't have to define schema, just specify the relationship like "belongs_to" and "has_many", and ORM do rest of the work, it's very easy to learn and easy to use.
This recipe provide a python ORM that behave like ActiveRecord.
Source: Text Source
import string, new, MySQLdb
from types import *
from MySQLdb.cursors import DictCursor
bag_belongs_to, bag_has_many = [],[]
def belongs_to(what): bag_belongs_to.append(what)
def has_many(what): bag_has_many.append(what)
class Mysqlwrapper:
def __init__(self,**kwds):
self.conn = MySQLdb.connect(cursorclass=DictCursor,**kwds)
self.cursor = self.conn.cursor()
self.escape = self.conn.escape_string
self.insert_id = self.conn.insert_id
self.commit = self.conn.commit
self.q = self.cursor.execute
def qone(self,query):
self.q(query)
return self.cursor.fetchone()
def qall(self,query):
self.q(query)
return self.cursor.fetchall()
class MetaRecord(type):
def __new__(cls, name, bases, dct):
global bag_belongs_to, bag_has_many
if name in globals(): return globals()[name]
else:
Record = type.__new__(cls, name, bases, dct)
for i in bag_belongs_to: Record.belongs_to(i)
for i in bag_has_many: Record.has_many(i)
bag_belongs_to = []
hag_has_many = []
return Record
class Storm(dict):
__metaclass__ = MetaRecord
__CONN = None
@classmethod
def belongs_to(cls, what):
def dah(self):
belong_cls = globals().get(what,None)
if not belong_cls:
belong_cls = type(what,(Storm,),{})
return belong_cls.selectone(self[what+'_id'])
setattr(cls,what,new.instancemethod(dah,None,cls))
@classmethod
def has_many(cls, what):
def dah(self):
hasmany_cls = globals().get(what,None)
if not hasmany_cls:
hasmany_cls = type(what,(Storm,),{})
dct={}
dct[string.lower(cls.__name__)+'_id']=self['id']
return hasmany_cls.select(**dct)
setattr(cls,what,new.instancemethod(dah,None,cls))
@classmethod
def conn(cls, **kwds):
if not cls.__CONN: cls.__CONN = Mysqlwrapper(**kwds)
@classmethod
def exe(cls,s):
if not cls.__CONN: raise "Database not connected"
return cls.__CONN.qall(s)
@classmethod
def insert(cls,**kwds):
vs = [[k,cls.__CONN.escape(str(kwds[k]))] for k in kwds]
if vs:
s = "insert into %s (%s) values ('%s')" % (
string.lower(cls.__name__), ','.join([v[0] for v in vs]),
"','".join([v[1] for v in vs]))
cls.__CONN.q(s)
cls.__CONN.commit()
return cls.__CONN.insert_id()
else: raise "nothing to insert"
@classmethod
def select(cls,*args, **kwds):
if len(args)==1 and (type(args[0])==IntType or type(args[0])==LongType):
q = "select * from %s where id='%s'"%(string.lower(cls.__name__),args[0])
where = "where id='%s'"%args[0]
else:
if args: s = ",".join(args)
else: s = "*"
if kwds:
c,limit,orderby = [],'',''
for k in kwds:
if k == 'limit': limit = "limit "+str(kwds[k])
elif k == 'order': orderby = "order by "+str(kwds[k])
else: c.append(k+"='"+str(kwds[k])+"'")
where = " and ".join(c)
if where: where = "where %s"%where
where = "%s %s %s"%(where,orderby,limit)
else: where = ""
q = " ".join(['select',s,'from',string.lower(cls.__name__),where])
r = cls.__CONN.qall(q)
list = []
for i in r:
list.append(cls(i))
list[-1].__dict__['where'] = where
return list
@classmethod
def selectone(cls,*args, **kwds):
r = cls.select(*args,**kwds)
if r: return r[0]
else: return {}
@classmethod
def update(cls,cond,**kwds):
if not cond or not kwds: raise "Update What?!"
if type(cond) == IntType: w = "id='%d'" % cond
else: w = cond
vs = [[k,cls.__CONN.escape(str(kwds[k]))] for k in kwds]
if vs:
s = "UPDATE %s SET %s WHERE %s" % ( string.lower(cls.__name__),
','.join(["%s='%s'"%(v[0],v[1]) for v in vs]), w)
cls.__CONN.q(s)
cls.__CONN.commit()
@classmethod
def delete(cls,id):
if type(id) == IntType:
cls.__CONN.q("delete from %s where id='%d'"%
(string.lower(cls.__name__),id))
cls.__CONN.commit()
else: raise "Only accept integer argument"
def __init__(self,dct={}):
if not self.__class__.__CONN: raise "Database not connected"
dict.__init__(self,dct)
self.__dict__['cur_table']= string.lower(self.__class__.__name__)
self.__dict__['where']= ''
self.__dict__['sql_buff']={}
def sql(self,sql): self.__class__.__CONN.q(sql)
def save(self):
s = ""
if self.where:
f = []
for v in self.sql_buff:
f.append("%s='%s'"%(v,self.sql_buff[v]))
s = "UPDATE %s SET %s %s" % (
self.cur_table, ','.join(f), self.where)
else:
f,i=[],[]
for v in self.sql_buff:
f.append(v)
i.append(self.sql_buff[v])
if f and i:
s = "INSERT INTO %s (%s) VALUES ('%s')" % (
self.cur_table, ','.join(f), "','".join(i))
if s:
self.__class__.__CONN.q(s)
self.__class__.__CONN.commit()
else: raise "nothing to insert"
def __setattr__(self,attr,value):
if attr in self.__dict__: self.__dict__[attr]=value
else:
v = self.__class__.__CONN.escape(str(value))
self.__dict__['sql_buff'][attr] = v
self[attr] = v
def __getattr__(self,attr):
if attr in self.__dict__: return self.__dict__[attr]
try: return self[attr]
except KeyError: pass
raise AttributeError
__all__ = ['Storm', 'belongs_to', 'has_many']
Below is a session screenshot of using this ORM(Storm):
-------------------------------------------------------------
wang@dapper-03:~/spark/lib$ mysql -u root
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 46 to server version: 5.0.22-Debian_0ubuntu6.06-log
Type 'help;' or '\h' for help. Type '\c' to clear the buffer.
mysql> create database teststorm;
Query OK, 1 row affected (0.00 sec)
mysql> use teststorm;
Database changed
mysql> create table author(id int auto_increment primary key,name varchar(50));
Query OK, 0 rows affected (0.06 sec)
mysql> create table book(id int auto_increment primary key,author_id int,title varchar(100));
Query OK, 0 rows affected (0.01 sec)
mysql> describe author;
+-------+-------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+-------+-------------+------+-----+---------+----------------+
| id | int(11) | NO | PRI | NULL | auto_increment |
| name | varchar(50) | YES | | NULL | |
+-------+-------------+------+-----+---------+----------------+
2 rows in set (0.00 sec)
mysql> describe book;
+-----------+--------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+-----------+--------------+------+-----+---------+----------------+
| id | int(11) | NO | PRI | NULL | auto_increment |
| author_id | int(11) | YES | | NULL | |
| title | varchar(100) | YES | | NULL | |
+-----------+--------------+------+-----+---------+----------------+
3 rows in set (0.00 sec)
mysql> Bye
wang@dapper-03:~/spark/lib$ python
Python 2.4.3 (
[GCC 4.0.3 (Ubuntu 4.0.3-1ubuntu5)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> from storm import *
>>> class Author(Storm):
... has_many('book')
...
>>> class Book(Storm):
... belongs_to('author')
...
>>> Storm.conn(user='root',db='teststorm')
>>> a = Author()
>>> a.name = 'Tolstoy'
>>> a.save()
>>> Author.insert(name='Charles Dickens')
0L
>>> a.name, a['name']
('Tolstoy', 'Tolstoy')
>>> o = Author.selectone(2)
>>> o
{'id': 2L, 'name': 'Charles Dickens'}
>>> o.id, o.name, o['id'], o['name']
(2L, 'Charles Dickens', 2L, 'Charles Dickens')
>>> b = Book()
>>> b.author_id = 1
>>> b.title = 'Anna Karenina'
>>> b.save()
>>> b.title = 'War and Peace'
>>> b.save()
>>> b.author_id = 2
>>> b.title = 'Great Expectations'
>>> b.save()
>>> Book.insert(author_id=2,title='A Tale of Two Cities')
0L
>>> Book.insert(author_id=2,title='David Copperfield')
0L
>>> all = Book.select()
>>> all
[{'author_id': 1L, 'id': 1L, 'title': 'Anna Karenina'}, {'author_id': 1L, 'id': 2L, 'title': 'War and Peace'},
{'author_id': 2L, 'id': 3L, 'title': 'Great Expectations'}, {'author_id': 2L, 'id': 4L, 'title':
'A Tale of Two Cities'}, {'author_id': 2L, 'id': 5L, 'title': 'David Copperfield'}]
>>> o = Book.selectone(4)
>>> a = o.author()
>>> a
{'id': 2L, 'name': 'Charles Dickens'}
>>> a = Author.selectone(name='Tolstoy')
>>> a
{'id': 1L, 'name': 'Tolstoy'}
>>> b = a.book()
>>> b
[{'author_id': 1L, 'id': 1L, 'title': 'Anna Karenina'}, {'author_id': 1L, 'id': 2L, 'title': 'War and Peace'}]
>>> b[0].title, b[1].title
('Anna Karenina', 'War and Peace')
>>>
wang@dapper-03:~/spark/lib$
Discussion:
You create class that has the same name as db table.
You can use classmethod insert() or create an instance to insert to db. You get objects of query result using select() and selectone() class methods.
You can define "belongs_to" and "has_many" one-to-many relationship during class definition. They automatically add instance method that return the object of associated table. For example, when we define "class Book(Storm): belongs_to('author')", it add a instance method author() that return the object of class "Author" (whois id = instance_book.author_id).
This metaprogramming technique is from one of commentor "brantley" in this discussion:
http://blog.ianbicking.org/more-on-python-metaprogramming.html
The ORM also has update(), delete() class methods.
This ORM is sufficient for use in small db's that has simple relationship. For complex database, I wouldn't use ORM anyway, most likely will use dbapi directly instead.
This ORM use Mysql, it can be modified to work with Sqlite and Postgres easily.
|
