The goal here is to not lose content, so the regex must not consume (match) any input. Rather it must match between the last character of the previous target input and the first character of the next target input. eg for 8-character substrings, we need to break the input up (ie match) at the places marked below:
a b c d e f g h i j k l m n o p q r s t u v w x y z ^ ^ ^
Ignore the spaces in the input which were required to show between character positions.