In Perl, unpack can be used to neatly split a string of fixed-length fields into its parts. What follows is an example of it being used to accomplish this:
# get a 5-byte string, skip 3, then grab 2 8-byte strings, then the rest
my ($leading, $s1, $s2, $trailing) = unpack("A5 x3 A8 A8 A*", $data);
Is there an equivalent to unpack in Clojure?
No, and there's no the corresponding pack.
I'd use RegExes. Although I can't really tell how exact of an alternative that is since I don't know Perl and how it handles multibyte characters in strings.
;; If you prefer named groups.
(let [data "xxxxx___yyyyyyyyzzzzzzzz1234awerasdf"
pattern #"(?<leading>.{5}).{3}(?<s1>.{8})(?<s2>.{8})(?<trailing>.*)"
matcher (re-matcher pattern data)]
(re-find matcher)
(into {}
(map (fn [[n idx]]
[(keyword n) (.group matcher ^long idx)]))
(.namedGroups matcher)))
;; If you prefer to rely on positions.
(let [data "xxxxx___yyyyyyyyzzzzzzzz1234awerasdf"
pattern #"(.{5}).{3}(.{8})(.{8})(.*)"
[_ leading s1 s2 trailing] (re-matches pattern data)]
{:leading leading
:s1 s1
:s2 s2
:trailing trailing})