Skip to main content
LeetCode 749, Hard. Topics: Array, Depth-First Search, Breadth-First Search, Matrix, Simulation. View on LeetCode. Generate this problem as a practice environment: tested reference solution, 16 parametrized pytest cases, and a playground notebook:

Problem

A virus is spreading rapidly, and your task is to quarantine the infected area by installing walls. The world is modeled as an m x n binary grid isInfected, where isInfected[i][j] == 0 represents uninfected cells, and isInfected[i][j] == 1 represents cells contaminated with the virus. A wall (and only one wall) can be installed between any two 4-directionally adjacent cells, on the shared boundary. Every night, the virus spreads to all neighboring cells in all four directions unless blocked by a wall. Resources are limited. Each day, you can install walls around only one region (i.e., the affected area (continuous block of infected cells) that threatens the most uninfected cells the following night). There will never be a tie. Return the number of walls used to quarantine all the infected regions. If the world will become fully infected, return the number of walls used.

Examples

Example 1
Example 2

Constraints

  • m == isInfected.length
  • n == isInfected[i].length
  • 1 <= m, n <= 50
  • isInfected[i][j] is either 0 or 1.
  • There is always a contiguous viral region throughout the described process that will infect strictly more uncontaminated squares in the next round.

Solution

Reference implementation from solution.py on GitHub, full suite in test_solution.py:

Complexity

Tags

Last modified on September 7, 2026