Reputation: 1
Regular expression
Below is my text Data from which I have to find MI, mi, MumbaiIndians, MUMBAIINDIANS words through regex.
HATHWAY / MBBS : [PM1-PTY-PM / MUMBAIINDIANS0/2/2] || WICKET DOWN
HATHWAY / MBBS : [PM1-PTY-PM /MUMBAIINDIANS0/2/2] || WICKET DOWN
HATHWAY / MBBS : [PM1-PTY-PM / MI0/2/2] || WICKET DOWN
HATHWAY / MBBS : [PM1-PTY-PM /MI0/2/2] || WICKET DOWN
HATHWAY / MBBS : [PM1-PTY-PM /MI ] || WICKET DOWN
HATHWAY / MBBS : [PM1-PTY-PM / mi0/2/2] || WICKET DOWN
HATHWAY / MBBS : [PM1-PTY-PM / MumbaiINDIANS0/2/2] || WICKET DOWN
HATHWAY / MBBS : [PM1-PTY-PM /Me ] || WICKET DOWN
I wrote regex query for this as below
`/((\bMi[^0-9-!\s$%^&*()_+|~=`{}\[\]:\";'<>?,.\/]*)|(MI[^0-9-!\s$%^&*()_+|~=`{}\[\]:\";'<>?,.\/]*)|(\bMi[^0-9-!\s$%^&*()_+|~=`{}\[\]:\";'<>?,.\/]*)|(MumbaiINDIANS[^0-9-!\s$%^&*()_+|~=`{}\[\]:\";'<>?,.\/]*)|(\bMi[^0-9-!\s$%^&*()_+|~=`{}\[\]:\";'<>?,.\/]*)|(mi[^0-9-!\s$%^&*()_+|~=`{}\[\]:\";'<>?,.\/]*))/`
It is working but It is not efficient and looks very complex ... what I want is MI, mi, me, MumbaiIndians , MUMBAIINDIANS words to words to gets selected irrespective of their locations.
Please anyone tell me how to optimize this query.
Upvotes: 0
Views: 76
Answers (2)
Reputation: 206593
You could use the ig for insensitive, global with a positive lookbehind / lookahead:
/(?<=[\d \/])mumbaiindians|mi|me(?=[\d \/])/ig
Upvotes: 0
Reputation: 163632
Seems like you might shorten your pattern to (using a case insensitive match)
To prevent a partial match, you could use an assertion to make sure that it is not followed by a char a-zA-Z using a negative lookahead (?!
(?i)\b(?:mumbaiindians|mi)(?![A-Za-z])
Or you can assert one of the listed chars to the right using a positive lookahead (?=
(?i)\b(?:mumbaiindians|mi)(?=[0-9-!\s$%^&*()_+|~=`{}\[\]:\";'<>?,.\/])
Upvotes: 1