Question

我有像

这样的关键字

cat="AUTHORISATION,FORTHCOMING BOARD MEETINGS,PREVIOUS BOARD MEETINGS,BOARD MEETINGS,BOARD MEETING,MINUTES,BOARD PAPERS,AGENDA,COMMUNITY PROFILES,FORTHCOMING GOVERNOR MEETINGS,PREVIOUS GOVERNOR MEETINGS,GOVERNOR MEETINGS,GOVERNOR MEETING,GOVERNOR,COUNCIL OF GOVERNORS,GOVERNING BODY MEETINGS,COMPARISON,APC SUMMARY OF DECISIONS"

我有一些这样的预处理

cat_list=cat.split(',')
cat_list=filter(None, cat_list)
cat_list=[s.strip() for s in cat_list]
cat_list=[re.sub('\r\n' , ' ', s) for s in cat_list]
cat_list=[re.sub(r'([^\s])\s([^\s])', r'\1+(.)+\2',x) for x in cat_list]
cat_list=[re.sub(r'([a-z][a-z]+)', r'(\1)',a,flags=re.I) for a in cat_list]
regexes_cat=[re.compile((r'(?:%s)' % '|'.join(cat_list)),re.IGNORECASE),]

在列表中给出了re.compile表达式，供我执行re.search 所以处理后的最终正则表达式模式看起来像这样

(?:(AUTHORISATION)|(FORTHCOMING)+(.)+(BOARD)+(.)+(MEETINGS)|(PREVIOUS)+(.)+(BOARD)+(.)+(MEETINGS)|(BOARD)+(.)+(MEETINGS)|(BOARD)+(.)+(MEETING)|(MINUTES)|(BOARD)+(.)+(PAPERS)|(AGENDA)|(COMMUNITY)+(.)+(PROFILES)|(FORTHCOMING)+(.)+(GOVERNOR)+(.)+(MEETINGS)|(PREVIOUS)+(.)+(GOVERNOR)+(.)+(MEETINGS)|(GOVERNOR)+(.)+(MEETINGS)|(GOVERNOR)+(.)+(MEETING)|(GOVERNOR)|(COUNCIL)+(.)+(OF)+(.)+(GOVERNORS)|(GOVERNING)+(.)+(BODY)+(.)+(MEETINGS)|(COMPARISON)|(APC)+(.)+(SUMMARY)+(.)+(OF)+(.)+(DECISIONS))

但如果我打印组（0）

，我会得到这样的结果

GOVERNORS-MEETINGS.ASP?P=GOVERNORS%27.COUNCIL.MEETINGS

所以我搜索并发现我必须使用？使其非贪婪，但我无法获得所需的输出应该是

GOVERNORS-MEETINGS

我正在针对网页上显示的网址和文字执行re.search

http://www.qehkl.nhs.uk/governors-meetings.asp?p=governors%27.council.meetings&s=main&ss=becoming.a.foundation.trust

Answer 1

我建议的解决方案基于以下假设：

正则表达式匹配应发生在路径的最后一个子部分中（即在文件部分中，在任何最终查询字符串之前）
查询字符串是可选的

因此，解决方案是首先使用urlparse解析URL以仅获取运行正则表达式的字符串，并忘记外观。而不是(.)+，只需使用惰性(.*?)来匹配任何0 +字符尽可能少：

import re
from urlparse import urlparse

cat="AUTHORISATION,FORTHCOMING BOARD MEETINGS,PREVIOUS BOARD MEETINGS,BOARD MEETINGS,BOARD MEETING,MINUTES,BOARD PAPERS,AGENDA,COMMUNITY PROFILES,FORTHCOMING GOVERNOR MEETINGS,PREVIOUS GOVERNOR MEETINGS,GOVERNOR MEETINGS,GOVERNOR MEETING,GOVERNOR,COUNCIL OF GOVERNORS,GOVERNING BODY MEETINGS,COMPARISON,APC SUMMARY OF DECISIONS"
cat_list=cat.split(',')
cat_list=filter(None, cat_list)
cat_list=[s.strip() for s in cat_list]
cat_list=[re.sub('\r\n' , ' ', s) for s in cat_list]
cat_list=[re.sub(r'([^\s])\s([^\s])', r'\1(.*?)\2',x) for x in cat_list] # Allow anything in between the keywords, but as few as possible
cat_list=[re.sub(r'([a-z][a-z]+)', r'(\1)', a, flags=re.I) for a in cat_list]
regex_cat=re.compile(r"(?:{})".format('|'.join(cat_list)),re.IGNORECASE)
#print(regex_cat.pattern)
urls = "GOVERNORS/GOVERNORS-MEETINGS.ASP?P=GOVERNORS%27.COUNCIL.MEETINGS "
o = urlparse(urls)                       # Parse the URL
last_subpart = o.path.split('/').pop()   # Get the last subpart
m = regex_cat.search(last_subpart)       # Run the regex search
if m:                                    # If there is a match...
    print(m.group())                     # Print or do anything with the value

请参阅Python demo

Answer 2

请尝试以下代码 -

cat_list=cat.split(',')
cat_list=filter(None, cat_list)
cat_list=[s.strip() for s in cat_list]
cat_list=[re.sub('\r\n' , ' ', s) for s in cat_list]

#Till now all same, following statements have changes
cat_list=[re.sub(r'([^\s])\s([^\s])', r'\1+.+?\2',x) for x in cat_list]
cat_list=['(%s)'%re.sub(r'([a-z]+)', r'(\1)',a,flags=re.I) for a in cat_list]
regexes_cat=[re.compile((r'(?:%s)' % '|'.join(cat_list)),re.IGNORECASE),]

这是工作demo。

匹配来自re.compile的单个关键字，其中包含关键字列表

2 个答案: