Capturing emails with regex in Python

Question

I will be gathering scattered emails from a larger CSV file. I am just now learning regex. I am trying to extract the emails from this example sentence. However, emails is populating with only the @ symbol and the letter immediately before that. Can you help me see what's going wrong?

import re

String = "'Jessica's email is [email protected], and Daniel's email is [email protected]. Edward's is [email protected], and his grandfather, Oscar's, is [email protected].'"

emails = re.findall(r'.[@]', String)
names = re.findall(r'[A-Z][a-z]*',String)

print(emails)
print(names)

Jean-François Fabre · Accepted Answer · 2018-03-16 15:52:43Z

your regex e-mail is not working at all: emails = re.findall(r'.[@]', String) matches anychar then @.

I would try a different approach: match the sentences and extract name,e-mails couples with the following empiric assumptions (if your text changes too much, that would break the logic)

all names are followed by 's" and is somewhere (using non-greedy .*? to match all that is in between
\w matches any alphanum char (or underscore), and only one dot for domain (else it matches the final dot of the sentence)

code:

import re

String = "'Jessica's email is [email protected], and Daniel's email is [email protected]. Edward's is [email protected], and his grandfather, Oscar's, is [email protected].'"

print(re.findall("(\w+)'s.*? is (\w+@\w+\.\w+)",String))

result:

[('Jessica', '[email protected]'), ('Daniel', '[email protected]'), ('Edward', '[email protected]'), ('Oscar', '[email protected]')]

converting to dict would even give you a dictionary name => address:

{'Oscar': '[email protected]', 'Jessica': '[email protected]', 'Daniel': '[email protected]', 'Edward': '[email protected]'}

The general case needs more chars (not sure I'm exhaustive):

String = "'Jessica's email is [email protected], and Daniel's email is [email protected]. Edward's is [email protected], and his grandfather, Oscar's, is [email protected].'"

print(re.findall("(\w+)'s.*? is ([\w\-.]+@[\w\-.]+\.[\w\-]+)",String))

result:

[('Jessica', '[email protected]'), ('Daniel', '[email protected]'), ('Edward', '[email protected]'), ('Oscar', '[email protected]')]

cs95 · Accepted Answer · 2018-03-16 17:00:30Z

2

1. Emails

    In [1382]: re.findall(r'\S+@\w+\.\w+', text)
    Out[1382]: 
    ['[email protected]',
     '[email protected]',
     '[email protected]',
     '[email protected]']

How it works: All emails are [email protected]. One thing to note is a bunch of characters surrounding @, and the singular .. So, we use \S to demarcate anything that is not a whitespace. And + is to search for 1 or more such characters. \w+\.\w+ is just a fancy way of saying search for a string that only has one . in it.

2. Names

    In [1375]: re.findall('[A-Z][\S]+(?=\')', text)
    Out[1375]: ['Jessica', 'Daniel', 'Edward', 'Oscar']

How it works: Any word starting with an upper case. The (?=\') is a lookahead. As you see, all names follow the pattern Name's. We want everything before the apostrophe. Hence, the lookahead, which is not captured.

Now, if you want to map names to emails by capturing them together with one massive regex, you can. Jean-François Fabre's answer is a good start. But I recommend getting the basics down par first.

edited Mar 16, 2018 at 17:00

answered Aug 4, 2017 at 5:02

cs95

406k106 gold badges744 silver badges797 bronze badges

8 Comments

Jean-François Fabre Over a year ago

[email protected].: last e-mail has a dot in the end, others get the commas as well :)

cs95 Over a year ago

@Jean-FrançoisFabre Last 24 hours. Yeah it's been a tough day -- tough crowd to please.

cs95 Over a year ago

@Jean-FrançoisFabre Honestly I start exercising my hammer when I need to blow off some steam -- which has been happening quite often actually.

Jean-François Fabre Over a year ago

470 and you're complaining? I admit a lot of users don't accept answers (or accept the lamest/least upvoted answer there is because they understand it)

Jean-François Fabre Over a year ago

I just realised that [\w_] is redundant since \w contains _ !

|

furoscame · Accepted Answer · 2017-08-04 05:29:41Z

1

You need to find anchors, patterns to match. An improved pattern could be:

import re

String = "'Jessica's email is [email protected], and Daniel's email is 
[email protected]. Edward's is [email protected], and his 
grandfather, Oscar's, is [email protected].'"

emails = re.findall(r'[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+', String)
names = re.findall(r'[A-Z][a-z]*', String)

print(emails) 
print(names)

\w+ is missing '-' which are allowed in email adresses.

edited Aug 4, 2017 at 5:29

answered Aug 4, 2017 at 5:08

furoscame

1491 silver badge7 bronze badges

Comments

Anonta · Accepted Answer · 2017-08-04 05:10:29Z

0

This is because you are not using the repeat operator. The below code uses the + operator which means the characters / sub patterns just before it can repeat 1 to many times.

s = '''Jessica's email is [email protected], and Daniel's email is [email protected]. Edward's is [email protected], and his grandfather, Oscar's, is [email protected].'''

p = r'[a-z0-9]+@[a-z]+\.[a-z]+'
ans = re.findall(p, s)

print(ans)

answered Aug 4, 2017 at 5:10

Anonta

2,5402 gold badges17 silver badges25 bronze badges

1 Comment

EwokHugz Over a year ago

Thanks m8. Like I said I am new to this regex.

Collectives™ on Stack Overflow

Capturing emails with regex in Python

4 Answers 4

Comments

8 Comments

Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

8 Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related