Finding firstname.lastname.number pattern in email

[Steph Cheri]

My data contains observations that have [email protected]. I have variables firstname and lastname

What I want to do is tag observations that have firstname followed by lastname followed by some numbers before the "@" sign in their emailid. How do I go about tagging them? Also the number could be multiple digits

[George Ford]

split email , parse("." "@")

edit: that won't work if no number.

[George Ford]

probably a cleaner expression, but I don't use ustrregexm much.

asking whether a number appears before @.

g hasnum = ustrregexm(substr(var1,strpos(var1,"@")-1,1), "([0-9])")

[Steph Cheri]

probably a cleaner expression, but I don't use ustrregexm much.

asking whether a number appears before @.

g hasnum = ustrregexm(substr(var1,strpos(var1,"@")-1,1), "([0-9])")

this has helped me narrow down the observations with a number before the "@". And after using split email, p("@") the firstname.lasntname.number gets split out. Now my aim is tag these observatiosn based on the firstname and lastname variable i have. so something like this

firstname	lastname	email	emailbefore@	hasnum	tag
JOHN	DOE	[email protected]	JOHNDOE	0	0
JANE	DOE	[email protected]	JANEDOE122	1	1
JEFF	DOE	[email protected]	JEFF1DOE32	1	0

so tag = 1 because emailbefore@ == firstname + lastname + "somecombinationofnumber"

so i would ideally strip the numbers off of emailbefore@ and then check if the A-Z characters match the stripped emailbefore@ variable. BUT the problem is with the observation in row 3. That would get tagged as 1 but I want it to be tagged as 0

Thanks again George

[Clyde Schechter]

I think you want

gen byte wanted = ustrregexm(email, "^[A-Z]+\d+@")

[daniel klein]

generate byte tag = ustrregexm(email, "^"+firstname+lastname+"\d+@")

[George Ford]

#5 and #6 seem to work.