问题
在django的使用中，经常会出现大量数据的遍历操作，或者是对大量数据进行遍历迁移跟新，比如
for user
in User.objects.all():
user.A = user.B
User.B =
None
等种种情况。
在本地开发环境中QuerySet对象最初具有非常小的内存占用，随着业务量的增长QuerySet对象在我遍历它们时缓存每个model_instance，all()返回的QuerySet会越来越来，可能最终耗尽内存，被托管服务提供商杀死线程。
解决方法:
import copy
from decimal
import Decimal
class MemorySavingQuerysetIterator(object):
def __init__(self, queryset, max_obj_num=1000):
self._base_queryset = queryset
self._generator = self._setup()
self.max_obj_num = max_obj_num
def _setup(self):
for i
in xrange(
0, self._base_queryset.count(), self.max_obj_num):
smaller_queryset = copy.deepcopy(self._base_queryset
)[i:i + self.max_obj_num]
for obj
in smaller_queryset.iterator():
yield obj
def __iter__(self):
return self
def next(self):
return self._generator.next()
调用：
Users = User.objects.all()
for user
in MemorySavingQuerysetIterator(users,
100):
Pass
python mysql原生操作
import MySQLdb
class QuerySetIterator(object):
def __init__(self, cursor, query, max_num):
self.query = query
self.max_num = max_num
self._cursor = cursor
self._generator = self._setup()
def _setup(self):
for i
in xrange(
0,
90000000, self.max_num):
new_query =
"{query} limit {limit} offset {offset}".format(
query=self.query, limit=self.max_num, offset=i
)
self._cursor.execute(new_query)
result = self._cursor.fetchall()
if not result:
break
for obj
in result:
yield obj
def __iter__(self):
return self
def next(self):
return self._generator.next()
class TestModel(object):
db = MySQLdb.connect(
"localhost",
"root",
"123456",
"test")
cursor = db.cursor()
def __init__(self, tb_name, max_num=100):
self.tb_name = tb_name
self.max_num = max_num
self._query_sql_tpl =
"select * from {tb_name}".format(tb_name=tb_name)
def query_all(self, query_sql=None):
if not query_sql:
query_sql = self._query_sql_tpl
return QuerySet(self.cursor, query_sql, self.max_num)
test = TestModel(
'test')
result = test.query_all()
for obj
in result:
print obj
