Rather than sorting based on a-z, is it possible to sort based on occurrence?
Example of desired task,
Input =
Example
Example
Hello
Hello
Input
Input
List
List
World
World
...
Desired output =
Example
Hello
Input
List
World
Example
Hello
Input
List
World
...
So in this scenario it gets the first occurrence of each line and puts them in order, then the second occurrence of each line and puts them in order until there's no more occurences
You can always use the decorate - sort - undecorate approach:
<input awk -v OFS=: '{print ++n[$0], $0}' |
sort -t: -k1,1n -k2 |
cut -d: -f2-
Where awk prepends each line with the line's occurrence as a number for sort to sort on as its first key (numeric order), and the original line as the second key (default locale collation order).
Here using : as the separator instead of awk's default SPC and sort's default transition from a non-blank to a blank as with that default, the SPC would be included in the second key which might (unlikely) have an impact on sorting.
FS variable of awk to specify how input fields are separated (OFS is the output field separator) or use the default of delimiting on blanks, and replace ++n[$0] with ++n[$1] for instance to report the occurrence number of the first field as opposed to the whole line.