用于从 AWS ARN 中提取区域名称的正则表达式

Question

I'm trying to build a RegEx that extracts the region (us-east-1) from the following AWS ARN:

arn:aws:secretsmanager:us-east-1:123456789012:secret:catsndogs-3HieNb

I've tried using ([^:]*) which creates groups, but I can't seem to grab the value of the 4th group, ie us-east-1.

Answer 1

(Since the OP did not specify a regex flavor, the following assumes PCRE; minor differences can arise if one changes the flavor; for instance, Golang uses $1 instead of \1 to refer to the first capturing group.)

^(?:[^:]+:){3}([^:]+).*

Regex101

• ^ is to start matching at the beginning of the line
• (?:[^:]+:){3} matches 3 repetitions of (?:[^:]+:) , which is a non-capturing group containing the regex [^:]+: , which matches a sequence of 1 or more non : followed by a : ;
• ([^:]+) matches and captures the 4th occurrence of that pattern
• .* matches all the rest of the line

If the substitution expression is \1 , all the line will be substituted by the 4th : -separated field.