参考

• Python的隐藏特性(StackOverflow)

list

for i, e in enumerate(array)
zeroes = [0] * 100
nans = ('nan',) * 6
spaces = 100 * " "

#想要生成一个长链表但是并不想把它立刻存在内存
for i in xrange(1000000000):

dict

dict轮训keys的速度比list快的多

print

# Python2.7
print >> sys.stderr, "this is an error message"

print >>  file("myfile", "w"), "hello world"

exception

class MyException:
    def __init__(self, value):
        self.value = value
    def __str__(self):
        return `self.value`

path

sys.path # 查看当前的search path

自省

• help() some? #查询说明文档
• __dict__所有属性键值对
• vars() 返回 __dict__ 对象属性

for property, value in vars(theObject).iteritems():
    print property, ": ", value

import inspect
>>> for attr, value in inspect.getmembers(obj):
...     print attr, value

inspect.getmembers 返回的是元组 (attrname, value) 的列表。而且是所有的属性, 包括 __class__ , __doc__ , __dict__ , __init__等特殊命名的属性和方法。而 vars() 只返回 __dict__

下划线

• 在解释器中：在这种情况下，"_"代表交互式解释器会话中上一条执行的语句的结果
• "_"作为临时性的名称使用 (var_name/function_name/class_name/file_name)
  • _name API中非公开的部分
  • __name 防止子类覆盖 解释器自动改写为 "_classname__name"
  • __name__ Python中特殊的方法名。通常，你将会覆写这些方法，并在里面实现你所需要的功能，以便Python调用它们

推导式

results = [(x, y)
           for x in range(10)
           for y in range(10)
           if x + y == 5
           if x > y]

magic method

__some__ 操作符重载 自省

magic method1

magic method2

断言

assert condition

仅用做防御型的编程，确保按约定进行设计

@装饰器

@property

      @property
      @deprecated("``residues_`` is deprecated")
    def birth(self):
        return self._birth

    @birth.setter
    def birth(self, value):
        self._birth = value

from abc import ABCMeta, abstractmethod
...
class ...

    @abstractmethod
    @classmethod 
    @raises(ValueError)
    def some_func():
        pass

dict default

val = dct.get(key, def_val)
sum[value] = sum.get(value, 0) + 1

val = dct.setdefault(key, def_val)
dct = defaultdict(list)

list 步长与反转

>>> a = [1, 2, 3, 4, 5]
>>> a[::2]
[1, 3, 5]
>>> a[::-1]
[5, 4, 3, 2, 1]

for else

else 语句块会在循环结束后执行，除非在循环块中执行 break

for i in (1, 3, 5):
...     if i % 2 == 0:
...         break
... else:
...     print "var i is always an odd"

tree

def tree(): return defaultdict(tree)

counter py2.7 np.unique

>>> from collections import Counter
>>> d = [1, 1, 1, 2, 2, 3, 1, 1]
>>> Counter(d)
Counter({1: 5, 2: 2, 3: 1})

>>> counter = Counter()
... for _ in range(10):
...     num = int(raw_input("Enter a number: "))
...     counter.update([num])
...
... for key, value in counter.iteritems():
...     print "You have entered {}, {} times!".format(key, value)

链式比较

1 < x < 5

Python解释器中的”_”

>>> range(4)
[0, 1, 2, 3]
>>> _
[0, 1, 2, 3]

_ 即Python解释器上一次返回的值

try/except/else

try:
  put_4000000000_volts_through_it(parrot)
except Voom:
  print "'E's pining!"
else:
  print "This parrot is no more!"
finally:
  end_sketch()

zen

import this

用切片来删除序列的某一段

>>> a = [1, 2, 3, 4, 5, 6, 7]
>>> a[1:4] = []
>>> a
[1, 5, 6, 7]
当然用 del a[1:4] 也是可以的

isinstance可以接收一个元组

>>> isinstance(1, (float, int))

显式的声明一个集合

>>> set([1,2,3])
# 在Python 2.7 之后可以这么写了:
>>> {1,2,3}

and or

 m_var = var or def_var

abc

from six import with_metaclass
from abc import ABCMeta, abstractmethod

class Parent(with_metaclass(ABCMeta, A, B)):

    @abstractmethod
    def my_method1(self, ...):

    @abstractmethod
    def my_method2(self, arg):
        ...some impl...

class Child(Parent):
    ...
    def my_method2(self, arg):
        super(Child, self).my_method2(arg)