Guy_1463
0
I'm searching for UUIDs in blocks of text using a regex. Currently I'm relying on the assumption that all UUIDs will follow a patttern of 8-4-4-4-12 hexadecimal digits.
Can anyone think of a use case where this assumption would be invalid and would cause me to miss some UUIDs?
I agree that by definition your regex does not miss any UUID. However it may be useful to note that if you are searching especially for Microsoft's Globally Unique Identifiers (GUIDs), there are five equivalent string representations for a GUID:
"ca761232ed4211cebacd00aa0057b223"
"CA761232-ED42-11CE-BACD-00AA0057B223"
"{CA761232-ED42-11CE-BACD-00AA0057B223}"
"(CA761232-ED42-11CE-BACD-00AA0057B223)"
"{0xCA761232, 0xED42, 0x11CE, {0xBA, 0xCD, 0x00, 0xAA, 0x00, 0x57, 0xB2, 0x23}}"
By definition, a UUID is 32 hexadecimal digits, separated in 5 groups by hyphens, just as you have described. You shouldn't miss any with your regular expression.
Here is the working REGEX: https://www.regextester.com/99148
const regex = [0-9a-fA-F]{8}\-[0-9a-fA-F]{4}\-[0-9a-fA-F]{4}\-[0-9a-fA-F]{4}\-[0-9a-fA-F]{12}
For bash:
grep -E "[a-f0-9]{8}-[a-f0-9]{4}-4[a-f0-9]{3}-[89aAbB][a-f0-9]{3}-[a-f0-9]{12}"
For example:
$> echo "f2575e6a-9bce-49e7-ae7c-bff6b555bda4" | grep -E "[a-f0-9]{8}-[a-f0-9]{4}-4[a-f0-9]{3}-[89aAbB][a-f0-9]{3}-[a-f0-9]{12}"
f2575e6a-9bce-49e7-ae7c-bff6b555bda4
If using Posix regex (grep -E, MySQL, etc.), this may be easier to read & remember:
[[:xdigit:]]{8}(-[[:xdigit:]]{4}){3}-[[:xdigit:]]{12}
Edit: Perl & PCRE flavours also support Posix character classes so this'll work with them. For those, change the (…) to a non-capturing subgroup (?:…).
Wanted to give my contribution, as my regex cover all cases from OP and correctly group all relevant data on the group method (you don't need to post process the string to get each part of the uuid, this regex already get it for you)
([\d\w]{8})-?([\d\w]{4})-?([\d\w]{4})-?([\d\w]{4})-?([\d\w]{12})|[{0x]*([\d\w]{8})[0x, ]{4}([\d\w]{4})[0x, ]{4}([\d\w]{4})[0x, {]{5}([\d\w]{2})[0x, ]{4}([\d\w]{2})[0x, ]{4}([\d\w]{2})[0x, ]{4}([\d\w]{2})[0x, ]{4}([\d\w]{2})[0x, ]{4}([\d\w]{2})[0x, ]{4}([\d\w]{2})[0x, ]{4}([\d\w]{2})
Official uuid library uses following regex:
/^(?:[0-9a-f]{8}-[0-9a-f]{4}-[1-5][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}|00000000-0000-0000-0000-000000000000)$/i
See reference
Generalize one, where underscore is also neglected properly and only alphanumeric values are allowed with the pattern of 8-4-4-4-12.
^[^\W_]{8}(-[^\W_]{4}){4}[^\W_]{8}$
or
^[^\W_]{8}(-[^\W_]{4}){3}-[^\W_]{12}$
both give you the same result, but the last one is more readable. And I would like to recommend the website where one can learn as well as test the Regular expression properly: https://regexr.com/