Regex to match possible names from a string

Question

I want to match possible names from a string. A name should be 2-4 words, each with 3 or more letters, all words capitalized. For example, given this list of strings:

Her name is Emily.
I work for Surya Soft.
I sent an email for Ery Wulandari.
Welcome to the Link Building Partner program!

I want a regex that returns:

None
Surya Soft
Ery Wulandari
Link Building Partner

currently here is my code:

data = [
   'Her name is Emily.', 
   'I work for Surya Soft.', 
   'I sent an email for Ery Wulandari.', 
   'Welcome to the Link Building Partner program!'
]

for line in data:
    print re.findall('(?:[A-Z][a-z0-9]{2,}\s+[A-Z][a-z0-9]{2,})', line)

It works for the first three lines, but it fail on the last line.

Don't you basically want any situation where there's a character sequence beginning with a capital letter? — mikebabcock
– mikebabcock, Commented Jun 6, 2013 at 4:13

flowfree · Accepted Answer · 2013-06-06 04:57:25Z

2

You can use grouping for repeating structure as given below:

compiled = re.compile('(?:(([A-Z][a-z0-9]{2,})\s*){2,})')
for line in data:
    match = compiled.search(line)
    if match:
       print match.group()
    else:
       print None

Output:

None
Surya Soft
Ery Wulandari
Link Building Partner

edited Jun 6, 2013 at 4:57

flowfree

16.5k12 gold badges55 silver badges79 bronze badges

answered Jun 6, 2013 at 4:30

ashokadhikari

1,2005 gold badges16 silver badges29 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Zangetsu Over a year ago

Yes this seems the better one; output is as expected

JBernardo · Accepted Answer · 2013-06-06 04:18:13Z

2

You can use:

re.findall(r'((?:[A-Z]\w{2,}\s*){2,4})', line)

It may add a trailing whitespace that can be trimmed with .strip()

answered Jun 6, 2013 at 4:18

JBernardo

33.6k13 gold badges92 silver badges120 bronze badges

Comments

Ashwini Chaudhary · Accepted Answer · 2013-06-06 04:46:15Z

1

Non-regex solution:

from string import punctuation as punc
def solve(strs):
   words = [[]]
   for i,x in enumerate(strs.split()):
      x = x.strip(punc)
      if x[0].isupper() and len(x)>2:
         if words[-1] and words[-1][-1][0] == i-1:
            words[-1].append((i,x))
         else:
            words.append([(i,x)])

   names = [" ".join(y[1] for y in x) for x in words if 2 <= len(x) <= 4]
   return ", ".join(names) if names else None


data = [
   'Her name is Emily.', 
   'I work for Surya Soft.', 
   'I sent an email for Ery Wulandari.', 
   'Welcome to the Link Building Partner abc Fooo Foo program!'
]
for x in data:
   print solve(x)

output:

None
Surya Soft
Ery Wulandari
Link Building Partner, Fooo Foo

edited Jun 6, 2013 at 4:46

answered Jun 6, 2013 at 4:21

Ashwini Chaudhary

252k60 gold badges478 silver badges519 bronze badges

2 Comments

JBernardo Over a year ago

"I want to match possible names from a string". Unlike findall, this can only match the last one...

Ashwini Chaudhary Over a year ago

@JBernardo missed that line, fixed the solution.

Adobe · Accepted Answer · 2013-06-06 04:32:54Z

0

for line in data:
    print re.findall("[A-Z][\w]+", line)

edited Jun 6, 2013 at 4:32

answered Jun 6, 2013 at 4:19

Adobe

13.5k12 gold badges94 silver badges132 bronze badges

Collectives™ on Stack Overflow

Regex to match possible names from a string

4 Answers 4

1 Comment

Comments

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

1 Comment

Comments

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related