python正则替换中的unicode问题
17-19，结论：对于unicode字符串，要⽤unicode正则字符串查，unicode正则字符串的构成有两个要件，⼀是字符串本⾝由unicode字符构成，⼆是⽤'u'引导（python2）；
20-23，结论：对于str字符串，要⽤⼆进制正则字符串查，⽤'r'引导，如果⽤'u'引导，因为并不包含unicode字符，其实际上不起作⽤，相当于'r'；
In [17]: %paste
import re
html = u"abcd 32⼈喜欢 efgh"
html = re.sub(ur'(\d+)\u4eba\u559c\u6b22', r'<div class="like">\1</div>', html)
print html
## -- End pasted text --
abcd <div class="like">32</div> efgh
In [18]: %paste
import re
html = u"abcd 32⼈喜欢 efgh"
html = re.sub(r'(\d+)\u4eba\u559c\u6b22', r'<div class="like">\1</div>', html)
print html
## -- End pasted text --
abcd 32⼈喜欢 efgh
In [19]: %paste
import re
html = u"abcd 32⼈喜欢 efgh"
html = re.sub(r'(\d+)\xe4\xba\xba\xe5\x96\x9c\xe6\xac\xa2', r'<div class="like">\1</div>', html)
print html
## -- End pasted text --
abcd 32⼈喜欢 efgh
In [20]: %paste
import re
html = "abcd 32⼈喜欢 efgh"
html = re.sub(r'(\d+)\u4eba\u559c\u6b22', r'<div class="like">\1</div>', html)
print html
## -- End pasted text --unicode所有字符
abcd 32浜哄枩娆?efgh
In [21]: %paste
import re
html = "abcd 32⼈喜欢 efgh"
html = re.sub(ur'(\d+)\u4eba\u559c\u6b22', r'<div class="like">\1</div>', html)
print html
## -- End pasted text --
abcd 32浜哄枩娆?efgh
In [22]: %paste
import re
html = "abcd 32⼈喜欢 efgh"
html = re.sub(r'(\d+)\xe4\xba\xba\xe5\x96\x9c\xe6\xac\xa2', r'<div class="like">\1</div>', html)
print html
## -- End pasted text --
abcd <div class="like">32</div> efgh
In [23]: %paste
import re
html = "abcd 32⼈喜欢 efgh"
html = re.sub(ur'(\d+)\xe4\xba\xba\xe5\x96\x9c\xe6\xac\xa2', r'<div class="like">\1</div>', html)
print html
## -- End pasted text --
abcd <div class="like">32</div> efgh