Python正则表达式来查找包含确切单词的短语

Question

I have a list of strings and wish to find exact phases.

So far my code finds the month and year only, but the whole phase including “- Recorded” is needed, like “March 2016 - Recorded”.

How can it add on the “- Recorded” to the regex?

import re


texts = [

"Shawn Dookhit took annual leave in March 2016 - Recorded The report",
"Soondren Armon took medical leave in February 2017 - Recorded It was in",
"David Padachi took annual leave in May 2016 - Recorded It says",
"Jack Jagoo",
"Devendradutt Ramgolam took medical leave in August 2016 - Recorded Day back",
"Kate Dudhee",
"Vinaye Ramjuttun took annual leave in  - Recorded Answering"

]

regex = re.compile('(?P<month>[a-zA-Z]+)\s+(?P<year>\d{4})\s')     

for t in texts:
    try:
        m = regex.search(t)
        print m.group()
    except:
        print "keyword's not found"

Answer 1

You got 2 named groups here: month and year which takes month and year from your strings. To get - Recorded into recorded named group you can do this:

regex = re.compile('(?P<month>[a-zA-Z]+)\s+(?P<year>\d{4})\s(?P<recorded>- Recorded)')

Or if you can just add - Recorded to your regex without named group:

regex = re.compile('(?P<month>[a-zA-Z]+)\s+(?P<year>\d{4})\s- Recorded')

Or you can add named group other with hyphen and one capitalized word:

regex = re.compile('(?P<month>[a-zA-Z]+)\s+(?P<year>\d{4})\s(?P<other>- [A-Z][a-z]+)') 

I think first or third option is preferable because you already got named groups. Also i recommend you to use this web site http://pythex.org/ , it really helps to construct regex :).

Answer 2

Use a list comprehension with the corrected regex:

regex = re.compile('(?P<month>[a-zA-Z]+)\s+(?P<year>\d{4})\s* - Recorded')

matches = [match.groups() for text in texts for match in [regex.search(text)] if match]
print(matches)
# [('March', '2016'), ('February', '2017'), ('May', '2016'), ('August', '2016')]