grep regex匹配电子邮件地址

Question

I have a file test.txt which contains the following content:

BC@ABSC.CA
ABCabc+-._@mcmaster.io.ca
ABCabc+-._@school.image
ABCabc+-._@school3-computer.image
ABCabc+-._@school3-IT.image.tor.chrome.ca
ABCabc+-._@school3-IT.image.tor.chrome.canadannn
ABC123abc+-._@school3-IT.imageal.tor.chrome.canadannn
ABCabc+-._@school3-*IT.image.tor.chrome.ca
ABCabc+-._@school3-IT.image.tor.chrome.caskdlfj
ABCab*c+-._@school3-IT.image.tor.chrome.caABCabc

I then use

grep -E '^[A-Za-z0-9+._-]+@([a-zA-Z0-9-]+\.)+[a-zA-Z]{2,6}' test.txt

trying to match valid email. The key here is that the last subdomain has to be a sequence of 2 to 6 char.

So I am hopping to the get the following output:

BC@ABSC.CA
ABCabc+-._@mcmaster.io.ca
ABCabc+-._@school.image
ABCabc+-._@school3-computer.image
ABCabc+-._@school3-IT.image.tor.chrome.ca

But i also get the following even tho the length of the last domain succeed 6 chars.

ABCabc+-._@school3-IT.image.tor.chrome.canadannn
ABC123abc+-._@school3-IT.imageal.tor.chrome.canadannn
ABCabc+-._@school3-IT.image.tor.chrome.caskdlfj

How do i solve this problem?

Answer 1

The problem is that grep matches anything in a line. If you want the exact whole line, add the $ terminator at the end. Let's look at an example:

ABCabc+-._@school3-IT.image.tor.chrome.canadannn

1. ABCabc+-._ matches ^[A-Za-z0-9+._-]+
2. @ matches @
3. school3-IT.image.tor.chrome. matches ([a-zA-Z0-9-]+\\.)+ . As far as I know, all quantifiers are greedy in grep .
4. canada matches [a-zA-Z]{2,6}
5. nnn gets ignored

Without the $ , there just has to be some part of the line that matches, not necessarily the whole thing.

Answer 2

Add an endline anchor to your regex: $ :

grep -E '^[A-Za-z0-9+._-]+@([a-zA-Z0-9-]+\.)+[a-zA-Z]{2,6}$' test.txt

More about it: http://www.regular-expressions.info/anchors.html

Answer 3

You can fix your query by adding a $ at the end of your string.

grep -E '^[A-Za-z0-9+._-]+@([a-zA-Z0-9-]+\.)+[a-zA-Z]{2,6}$' test.txt

Here is a live demo: https://regex101.com/r/NtZJQ0/1