^{Idea shamelessly stolen from caird and rak1507}

Shuffle a subset of a list of unique, positive integers with uniform randomness, given the indices of that subset. For example, given the list $[A, B, C, D, E, F, G, H]$ and the indices $[0, 3, 4, 5, 7]$ (0-indexed), you would extract the list $[A, D, E, F, H]$, shuffle it, and insert the shuffled elements back according to the list of indices. Some possible results of this are (elements that stayed in place are bolded) $$[H, \textbf{B}, \textbf{C}, D, A, E, \textbf{G}, F]$$ $$[A, \textbf{B}, \textbf{C}, D, E, F, \textbf{G}, H]$$ $$[F, \textbf{B}, \textbf{C}, E, A, H, \textbf{G}, D]$$.

Rules

• Every possible rearrangement of the list should have a non-zero chance of being chosen.
• You may use zero- or one-indexing, but please specify which you use.
• The indices in the input are guaranteed to be unique and valid.
• This is code golf, so least number of bytes wins.

Test cases

All of these examples use 0-indexing.

List
Indices
Possible output

[1,2,3,4,5,6,7]
[0,1,2,3,4,5,6]
[2,3,6,1,0,4,5], etc.

[1,2,3,4,5,6,7]
[0,2,4,6]
[3,2,7,4,5,6,1], etc.

[93,6,10,1,200,41,78,31,34,27]
[0,3,4,8,9]
[1,6,10,27,93,41,78,31,200,34], etc.

posted over 3 years ago

CC BY-SA 4.0

3y ago by General Sebast1an‭

user‭

1546 reputation 9 59 174 89