《Python自然语言处理》答案第三章

s='colorless'
s=s[:s.index('r')]+'u'+s[s.index('r'):]

s[:s.index('-')]

p=r'[a-zA-Z]+'
nltk.re_show(p,'123asd456')
nltk.re_show(p,'123asd456asd')
p='[A-Z][a-z]*'
nltk.re_show(p,'123asd456asd')
nltk.re_show(p,'Aadsds123asd456asd')
p='p[aeiou]{,2}t'
nltk.re_show(p,'paat'')
nltk.re_show(p,'paat')
nltk.re_show(p,'padst')
nltk.re_show(p,'padsst')
p='\d+(\.\d+)?'
nltk.re_show(p,'2312.12345dsa')

a.
pattern = r'''(?x)  # set flag to allow verbose regexps
[][.,;"'?():-_`]        #  these are separate tokens
'''
nltk.regexp_tokenize(text, pattern)

b.
pattern =r'''(?x) # set flag to allow verbose regexps
([A-Z]\.)+ # abbreviations, e.g. U.S.A.
| [A-Z][a-z]*\s[A-Z][a-z]* # words with optional internal
| \$?\d+(\.\d+)?%? # currency and percentages, e.g. $12.40, 82%
| \d+-\d+-\d+
'''

S.split(sep=None, maxsplit=-1) -> list of strings

Return a list of the words in S, using sep as the
delimiter string.  If maxsplit is given, at most maxsplit
splits are done. If sep is not specified or is None, any
whitespace string is a separator and empty strings are
removed from the result.

sorted([w for w in text if w.lower().startswith('wh')])

result=[]
text=['a 10','b 20','c 30']
for line in text:
     ...:     w,x=tuple(line.split())
     ...:     result.append((w,x))

def unknown(url):
    unknown('http://www.gutenberg.org/files/11/11-h/11-h.htm')
    resp=urllib.request.urlopen('http://www.gutenberg.org/files/11/11-h/11-h.htm')
    raw=resp.read().decode('utf-8')
    words=nltk.word_tokenize(raw)
    unknown=[w for w in words if w not in wn.words()]

p1=r'e'
p2=r'i'
p3='o'
p4=r'[.]'
p5=r'ate'
p6=r'^s'
p7=r's'
p8=r'1'

def f(s):
    s=re.sub(p1,'3',s)
    s=re.sub(p2,'1',s)
    s=re.sub(p3,'0',s)
    s=re.sub(p4,'5w33t!')
    s=re.sub(p5,'8',s)
    s=re.sub(p6,'$',s)
    s=re.sub(p7,'5',s)
    s=re.sub(p8,'|',s)

saying=['After', 'all', 'is', 'said', 'and', 'done', ',', 'more', 'is', 'said', 'than', 'done', '.']
lengths=[]
for w in saying:
    lengths.append(w)

lengths=[w for w in saying]

silly='newly formed bland ideas are inexpressible in an infuriating way'
bland=silly.split()
from functools import reduce
s=reduce(lambda x,y:x+y,[w[1] for w in bland])
' '.join(bland)
sorted(bland)