#####   ######  #####    ###    #   #  ###  #   #  ######
##  ##  ##      ##  ##  ## ##   #   #   #   #   #  ##
#####   ####    #####   #   #   #   #   #   #   #  ####
##  #   ##      ##      ## ##    # #    #    # #   ##
##   #  ######  ##       ###      #    ###    #    ######

$ curl repovive.com/roadmaps/data-structures/wavelet-trees/count-of-value-in-range

░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██████████████████████████████████████████████████████████████████████████████████

#####   ######  #####    ###    #   #  ###  #   #  ######
##  ##  ##      ##  ##  ## ##   #   #   #   #   #  ##
#####   ####    #####   #   #   #   #   #   #   #  ####
##  #   ##      ##      ## ##    # #    #    # #   ##
##   #  ######  ##       ###      #    ###    #    ######

$ curl repovive.com/roadmaps/data-structures/wavelet-trees/count-of-value-in-range

Repovive

Count of Value in Range - Wavelet Trees | Data Structures | Repovive

Repovive.

Data Structures19 sections · 729 units49h total

Open in Course

Count of Value in Range

Frequency query

Count occurrences of value $v$ in $A[l..r]$ :

function count(node, l, r, v)
 if l > r then
 return 0
 if node.lo = node.hi then
 return r - l + 1

 mid := (node.lo + node.hi) / 2
 if v <= mid then
 // v is in left subtree
 newL := rank0(node.bitvector, l)
 newR := rank0(node.bitvector, r + 1) - 1
 return count(node.left, newL, newR, v)
 else
 // v is in right subtree
 newL := rank1(node.bitvector, l)
 newR := rank1(node.bitvector, r + 1) - 1
 return count(node.right, newL, newR, v)

Follow the path to value $v$ , narrowing the range at each step.

Time: $O(\log \sigma)$ .