Data Types
datetime
import datetime
classes:
# datetime.date
date(year, month, day)
# methods:
ctime(...)
...
# data descriptors:
day
month
year
# datetime.datetime(datetime.date)
datetime(year, month, day[, hour[, minute[, second[, microsecond[,tzinfo]]]]])
# methods:
datetime.datetime.strptime(string, format)
format_datetime = datetime.datetime.strptime('20160824161431', '%Y%m%d%H%M%S') # return: datetime.datetime(2016, 8, 24, 16, 14, 31)
format_datetime = datetime.datetime.strptime('24 August 2016 16:14:31', '%Y%m%d%H%M%S') # return: datetime.datetime(2016, 8, 24, 16, 14, 31)
datetime.datetime.strftime(format[, tuple])
string_datetime = format_datetime.strftime("%d %B %Y %H:%M:%S") # return: '24 August 2016 16:14:31'
datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S') # return: '2017-02-07 23:07:32'
str(datetime.datetime.now())
datetime.datetime.now().strftime('%Y%m%d%H%M%S') # 当前时间戳
# datetime.time
# datetime.timedelta
# datetime.tzinfo
data:
MAXYEAR = 9999
MINYEAR = 1
calendar
collections
import collections
classes:
# collections.Callable
# collections.Container
# collections.Counter
# collections.Hashable
# collections.ItemsView
# collections.Iterable
# collections.Iterator
# collections.deque
# 返回一个deque类型的实例，也就是双端队列
deque([iterable[, maxlen]])
# collections.defaultdict
# 返回一个defaultdict类型的实例，类似于dict.setdefault()
defaultdict(default_factory[, ...])
functions:
namedtuple(typename, field_names, verbose=False, rename=False)
# 返回一个tuple的子类, 将只能通过index访问的tuple变为可以通过name来访问．
NewTuple = namedtuple('NewTuple', ['elm1', 'elm2', 'elm3'])
nt = NewTuple(1, 2, 3)
nt.elm1 # 1
data:
heapq
bisect
array
sched
Queue
队列数据结构．
import Queue
classes:
# Queue.Queue
Queue(maxsize=0) # 创建一个大小为maxsize的queue对象
# methods:
empty(self) # 如果queue为空返回True
full(self) # 如果queue达到maxsize返回True
get(self, block=True, timeout=None) # 从队列中取出一个元素并返回,如果block=True,会一直阻塞,直到队列中有元素．
get_nowait(self)
join(self)
put(self, item, block=True, timeout=None) # 把item放到队列中,如果block=true,会一直阻塞,直到队列有空间存放item.
put_nowait(self, item)
qsize(self) # 返回queue大小，近似值
task_done(self)
weakref
UserDict
UserList
UserString
types
copy
copy和deepcopy都只拷贝对象的类型和数值，不拷贝对象的ID.也就是==运算为True, is(id())运算为False.
import copy
functions:
copy.copy(x) # 浅复制, 只拷贝父对象，不拷贝内部的子对象.
copy.deepcopy(x, memo=None, _nil=[]) # 深复制, 拷贝父对象和内部的子对象
pprint
repr
Data Persistence
pickle/cPickle/marshal提供对象的序列化操作．
gdb相关的模块anydbm/whichdb/dbm/gdbm/dumbdbm提供类似字典和文件的对象．
shelve集合了以上两者的功能．
pickle
pickle不支持unicode,只支持ascii.
cPickle
python3将cPickle和pickle统一合并为pickle.
cPickle是c开发的，速度比pickle快，但是不支持被继承．
import cPickle
functions:
Pickler(file, protocol=0)
Unpickler(file)
dump(obj, file, protocol=0) # 写入到一个文件
dumps(obj, protocol=0) # 写入到一个字符串
load(file) # 从文件读取
loads(string) # 从字符串读取
data:
HIGHEST_PROTOCOL = 2
marshal
shelve
import shelve
with shelve.open(f, flag='n', writeback=True) as f:
r/w: 不存在报错
c: read-write to new/existing，不存在创建
n: read-write to new. 总是创建新的
anydbm
whichdb
dbm
gdbm
dumbdbm
cope_reg
pickletools
sqlite
python访问数据库两种方式：
- ORM
- DB-API
ORM是对象-关系管理器，相关模块有SQLAlchemy, SQLObject.
DB-API参考PEP248/249定义了Database的API。
https://www.python.org/dev/peps/pep-0249/
import sqlite3
classes:
# sqlite3.Connection
# methods:
close(...)
commit(...)
cursor(...)
execute(...)
...
# sqlite3.Cursor
# methods:
close(...)
...
functions:
adapt(obj, protocol, alternate)
connect(database[, timeout, isolation_level, detect_types, factory])
...
data: