python 类型与字符串

python

python是一种解释型、面向对象，动态数据类型的高级程序语言。

//精度高，执行时间长
from decimal import Decimal as D

类型

>>> type(a)
<type 'int'>
>>> a = "gh"
>>> type(a)
<type 'str'>
>>> a=[1,2,3,4]
>>> dir(a)
['__add__', '__class__', '__contains__', '__delattr__', '__delitem__', '__delslice__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getitem__', '__getslice__', '__gt__', '__hash__', '__iadd__', '__imul__', '__init__', '__iter__', '__le__', '__len__', '__lt__', '__mul__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__reversed__', '__rmul__', '__setattr__', '__setitem__', '__setslice__', '__sizeof__', '__str__', '__subclasshook__', 'append', 'count', 'extend', 'index', 'insert', 'pop', 'remove', 'reverse', 'sort']
>>> a.count(3)
1
>>> a = "dfgfht"
>>> a.count("d")
1
>>> a.__class__
<type 'str'>
>>> a.__doc__
"str(object='') -> string\n\nReturn a nice string representation of the object.\nIf the argument is a string, the return value is the same object."
>>> a = 3.4
>>> type(a)
<type 'float'>
>>> 0.3*3
0.8999999999999999
>>> print(0.3*3)
0.9
>>> 0.3/3
0.09999999999999999
>>> 10/3
3
>>> 10.0/3
3.3333333333333335
>>> from decimal import Decimal as D
>>> D(0.3)*D(0.9)
Decimal('0.2699999999999999966693309261')

耗时比较

$ python -mtimeit -s "from decimal import Decimal as D" "D('1.2')+D('3.4')"
10000 loops, best of 3: 28.5 usec per loop
$ python -mtimeit -s "from decimal import Decimal as D" "1.2+3.4"
100000000 loops, best of 3: 0.0172 usec per loop
$ python -mtimeit -s "from decimal import Decimal as D" "float('1.2')+float('3.4')"
1000000 loops, best of 3: 0.494 usec per loop

常用的库

math 和 random

import math
//π
math.pi
//平方根
math.sqrt()
//对数
math.log10()
//x^y
math.pow(x,y)
//!x
math.factorial(x)
import random
//产生一个[0,1]的浮点数
random.random()
//从列表里等概率随机选择一个
random.choice([1,2,3,4])
//a-b之间整数随机取一个
random.randint(a,b)
//返回a~b之间均匀分布的随机数，浮点数
random.uniform(a,b)
//高斯分布，（均值，方差）
random.gauss(mu,lamda)

>>> import math
>>> math.factorial(5)
120
>>> math.pi
3.141592653589793
>>> dir(math)
['__doc__', '__name__', '__package__', 'acos', 'acosh', 'asin', 'asinh', 'atan', 'atan2', 'atanh', 'ceil', 'copysign', 'cos', 'cosh', 'degrees', 'e', 'erf', 'erfc', 'exp', 'expm1', 'fabs', 'factorial', 'floor', 'fmod', 'frexp', 'fsum', 'gamma', 'hypot', 'isinf', 'isnan', 'ldexp', 'lgamma', 'log', 'log10', 'log1p', 'modf', 'pi', 'pow', 'radians', 'sin', 'sinh', 'sqrt', 'tan', 'tanh', 'trunc']
>>> help(math)
……

numpy 和 scipy

numpy

• 产生数组和矩阵，正态分布的随机数
• 矩阵运算
• 矩阵求逆，转置等操作

scipy

• 拟合、线性插值，样条插值
• 积分，微分
• 解非线性方程
• 滤波器设计

字符串

>>> s = "use python do something"
>>> s[1]
's'
>>> s[-1]
'g'
>>> s[1:3]
'se'
>>> s[1:6:2]  //[start,stop,step]
's y'
>>> s[1:]
'se python do something'
>>> s[:-1]
'use python do somethin'
>>> s[:]
'use python do something'

常用方法集合

>>> "let us" + s,s*2
('let ususe python do something', 'use python do somethinguse python do something')
>>> s.upper()
'USE PYTHON DO SOMETHING'
>>> s.find("pa")
-1
>>> s.replace("python","java")
'use java do something'
>>> print "%s like %s" %("we","python")
we like python
>>> a = s.split(" ")
>>> a
['use', 'python', 'do', 'something']
>>> b = ", ".join(a)
>>> b
'use, python, do, something'
>>> s[0:4]*5
'use use use use use '
>>> s = " life is short, we use python! "
>>> a = s.split(",")
>>> a[1]
' we use python! '
>>> a[1].strip()
'we use python!'

//转义
>>> s="C:\newpython"
>>> print s
C:
ewpython
>>> len(s)
11
>>> s = r"C:\newpython"
>>> print s
C:\newpython

re 模块

Regular expression

• re.match(p,text) //从text头开始找
• re.search(p,text)
• re.findall(p,text)
• re.split(p,text)
• re.sub(p,s,text) //替换，text——>s

编译选项

• I(ignorecase) 忽略大小写
• L(locale) 使预定字符类 \w \W \b \B \s \S 取决于当前区域设置
• M(multipline) 多行模式，改变“^”和“$”的行为
• S(dotall) 点任意模式，改变“.”行为
• U(unicode) 使预定字符类 \w \W \b \B \s \S \s \D 取决于当前区域设置取决于unicode定义的字符属性
• X(verbose) 详细模式，这个模式下正则表达式可以多行，忽略空白字符，并可以加注释

import re
text = 'c++ python2 python3 perl ruby lua java javascript php4 php5 c'
#match,search,findall,split,sub
re.match(r'java',text)      #none(第一个不是“java”)
re.search(r'java',text)     #<_sre.SRE_Match object at 0x00000000031046B0>
re.match(r'c++',text)       #error
re.match(r'c\+\+',text)     #<_sre.SRE_Match object at 0x00000000031941D0>
print re.findall(r'python',text)        #['python', 'python']
print re.split(r' perl ',text)          #['c++ python2 python3', 'ruby lua java javascript php4 php5 c']
print re.sub(r'ruby','fortran',text)    #c++ python2 python3 perl fortran lua java javascript php4 php5 c   
# ^   start
# $   end
# .   except \n
print re.findall(r'^c..',text)		#['c++']
print re.findall(r'c+',text)		#['c', 'c', 'c']
print re.findall(r'c\++',text)		#['c++']
print re.findall(r'c$',text)      #['c']
# +   1-inf
# *   0-inf
# ?   0-1, 
# []  or
# {}  repeat
# [^] not
print re.findall(r'p+',text)    
#['p', 'p', 'p', 'p', 'p', 'p', 'p', 'p']
print re.findall(r'p[a-zA-Z]+',text)  #{1,}
#['python', 'python', 'perl', 'pt', 'php', 'php']
print re.findall(r'p[a-zA-Z]*',text)  #{0,}
#['python', 'python', 'perl', 'pt', 'php', 'php']
print re.findall(r'p[a-zA-Z]?',text)  #{0,1}
#['py', 'py', 'pe', 'pt', 'ph', 'p', 'ph', 'p']
print re.findall(r'p[a-zA-Z0-9]{3,}',text)  #{,1}
#['python2', 'python3', 'perl', 'php4', 'php5']
print re.findall(r'c[a-zA-Z]*',text)  #{,inf}
#['c', 'cript', 'c']
print re.findall(r'c[^a-zA-Z]*',text)  #{,inf}
#['c++ ', 'c', 'c']
# |   or
print re.findall(r'[pj][a-zA-Z]+',text)  #{,inf}
#['python', 'python', 'perl', 'java', 'javascript', 'php', 'php']
# |重写上面的pattern
print re.findall(r'p[^0-9]+|j[a-zA-Z]+',text)   
#['python', 'python', 'perl ruby lua java javascript php', 'php']
print re.findall(r'p[^0-9 ]+|j[a-zA-Z]+',text) 
#['python', 'python', 'perl', 'java', 'javascript', 'php', 'php']
# \w  [a-zA-Z0-9_], \W
# \d  [0-9], \D
# \s  [ \t\n\r\f\v], \S
print re.findall(r'p\w+',text)          #['python2', 'python3', 'perl', 'pt', 'php4', 'php5']
print re.findall(r'p\w+\d',text)        #['python2', 'python3', 'php4', 'php5']
print re.findall(r'p\w+[0-9]',text)     #['python2', 'python3', 'php4', 'php5']
print re.findall(r'p\w{5,9}',text)      #['python2', 'python3']
# \b  word boundary
# \B  not \b
# \A  input start, ^
# \Z  input end, $
print re.findall(r'\bp[^0-9]',text)     #['py', 'py', 'pe', 'ph', 'ph']
print re.findall(r'p[^0-9]\b',text)     #['pt']
# *?  0~inf non-greedy
# +?  1~inf non-greedy
print re.findall(r'p[a-z]*',text)       #['python', 'python', 'perl', 'pt', 'php', 'php']
print re.findall(r'p[a-z]*?',text)      #['p', 'p', 'p', 'p', 'p', 'p', 'p', 'p']
print re.findall(r'p[a-z]+\b',text)     #['perl', 'pt']
print re.findall(r'p[a-z]+?\b',text)    #['perl', 'pt']
# ()  group
# (?P<name>pattern)
a=re.search(r'(p[a-zA-Z]+)([0-9])','python2',re.X)
print a.group(1)    #python
print a.group(2)    #2
a=re.search(r'(?P<name>p[a-zA-Z]+)(?P<version>[0-9])','python2')
print a.group('name')       #python
print a.group('version')    #2
print a.groupdict()         #{'version': '2', 'name': 'python'}
pattern = re.compile(r'(?P<name>p[a-zA-Z]+)(?P<version>[0-9])')
results = pattern.search('python2')
print results.groupdict()           #{'version': '2', 'name': 'python'}
results = pattern.search('python3')
print results.groupdict()           #{'version': '3', 'name': 'python'}
results = pattern.search('php4')
print results.groupdict()           #{'version': '4', 'name': 'php'}
#########################################
for t in text.split(' '):
    results = pattern.search(t)
    if results:
      print results.groupdict()
#{'version': '2', 'name': 'python'}
#{'version': '3', 'name': 'python'}
#{'version': '4', 'name': 'php'}
#{'version': '5', 'name': 'php'}
###########################################
a = re.compile(r"""\d +  # 整数部分
                   \.    # 小数点
                   \d *  # 小数部分
                """, re.X)
b = re.compile(r"\d+\.\d*")