python使用正则表达式文本替换

# encoding: UTF-8
import re

# 将正则表达式编译成Pattern对象
p = re.compile(r‘(?P<folder>(\w+/)*)(?P<filename>\w+\.png)‘)

# 使用Pattern匹配文本，获得匹配结果，无法匹配时将返回None
#match = pattern.match(‘<key>xxx/duobaojiemian_L/yangpizi.png</key>‘)

the_str = """<key>XXXX/duobaojiemian2222_L/duobaojiemian_L/yangpizi.png</key>  
  <key>yangpizi2.png</key>  
  <key>yangpizi3.png</key> """

for m in p.finditer(the_str):
  # 使用Match获得分组信息
  print m.groupdict()

print ‘-------------------------------‘

#f = lambda m: m.group().find(‘XXXX/‘) == -1 and ‘XXXX/‘+m.group() or m.group() 
def f(m):
  s = m.group()
  return s.find(‘XXXX/‘) == -1 and ‘XXXX/‘+s or s

def f2(m2):
  d = m2.groupdict()
  return d[‘folder‘]+‘the_‘+d[‘filename‘]

print p.sub(f2, the_str)