Python

Strings of characters, immutable

#!/usr/bin/python

s = 'Hello World!';

#access element
print s[0]; #H

#slicing
print s[:2]; #He

#slice
sl = slice(0, 10, 2);
print s[sl]; #HloWr

#concatenation
print s+' ...';

#repeat
print s*2;

#in
if 'H' in s:
    print 'Contain H';

#raw string, suppresses actual meaning of escape characters
print r'raw string\n';

#!/usr/bin/python

s = 'Hello World!';

#capitalize, capitalizes first letter of string
print s.capitalize();

#center
print s.center(20, '*');

#count
print s.count('l');

#endswith
if s.endswith('!'):
    print 'end with ! ...';

#find
print s.find('or'); #7

#index, find a string and raise an exception if the string is not found
print s.index('or');

#join
c = '-';
print c.join(['a', 'b', 'c']);

#lower
print s.lower();

#replace
print s.replace('l', '-', 2);

#split
str = "Line1-abcdef, \nLine2-abc, \nLine4-abcd";
print str.split();

#strip
print '  Hello ... '.strip();

#upper
print s.upper(); #HELLO WORLD!

Pattern

^, beginning of a line
$, end of a line
., matches any single character except newline
[...], matches any single character in brackets
[^...], matches any single character not in brackets
*, matches 0 or more occurrences of preceding expression
+, matches 1 or more occurrence of preceding expression
?, matches 0 or 1 occurrence of preceding expression
re{n}, match exact n time
re{n, m}, matches at least n and at most m occurrences
a|b, matches either a or b
\w, match word characters
\W, match none word characters
<.*>, greedy repetition
<.*?>, none greedy repetition, matches the smallest number of repetitions

Flags

re.I, performs case-insensitive matching
re.M, makes $ match the end of a line (not just the end of the string) and makes ^ match the start of any line

#!/usr/bin/python

import re;

phone = "2004-959-559 # This is Phone Number"

#match
m = re.match(r'(\d+)-(\d+)-\d+.*', phone);
if m:
    print type(m);
    print m.group(); #2004-959-559 # This is Phone Number
    print m.group(1); # 2004
    print m.groups(); # ('2004', '959')

#search
s = re.search(r'\d+', phone);
if s:
    print type(s);
    print s.group(); #2004

#findall
a = re.findall(r'\d+', phone);
print a; #['2004', '959', '559']

#replace
r = re.sub(r'\d', '*', phone);
print r; ****-***-*** # This is Phone Number

a sequence of code points, immutable

Python keep characters as unicode in memory

String read from a file is byte string, need to be decoded to unicode

Output unicode to a file, need to encode to byte string

Print a unicode on console, it is encoded to byte string automatically by default encoding method

ord, convert a string to its unicode code point

unichr, convert an integer into a Unicode string

type 'str' represents byte string in Python 2, type 'unicode' represent unicode

#!/usr/bin/python
# -*- coding: utf-8 -*-

import sys;

# set up default encoding
reload(sys)  # Reload does the trick!
sys.setdefaultencoding('UTF8')
print sys.stdout.encoding;

# unicode to str
s = 'Café';
print type(s), len(s); #'str', 5
s = u'Café';
print type(s), len(s); #'unicode', 4
print type(s.encode('utf-8')); #'str'

# read string from a file and print it to screen
f = open('temp.txt', 'rb');
l = f.next(); #read a line and save it to byte string
l = l.decode('utf-8'); #decode str to unicode
print type(l); #'unicode'
print l; #陈, print encode unicode to str with utf-8
f.close();

# get unicode code point
c = ord(u'陈'); #38472
print unichr(38472); #陈, do not print code point, it is encoded to str

Regular open() function is not able to read chunks since one unicode character can be represented by several bytes, which cause problems when read in the last character

If we read the whole file and encode, it may require a large memory

import codecs

f = codecs.open('codecs.html', encoding='utf-8', mode='r') # read line and decode to unicode

#for line in f: % codecs creates a buffer to read lines

	#print repr(line)

outf = codecs.open('output.html', encoding='utf-8', mode='w+') # encode line and write to output file

for line in f:

	outf.write(line)

outf.close()

f.close()

Tutorialspoint

Regular Expression

Regular Expression HowTo