LeetCode 7 · Bit Manipulation · concept: math-and-geometry
Problem
Reverse the digits of a signed 32-bit integer. If reversing overflows the 32-bit range [−2³¹, 2³¹−1], return 0.
123 -> 321
-123 -> -321
120 -> 21
Approach — pop and push digits, checking overflow before it happens
Peel the last digit with % 10, append it to the reversed number with * 10 +. The real problem is overflow detection without a 64-bit type: check before each multiply-and-add whether it would exceed the bounds.
def reverse(x):
INT_MAX, INT_MIN = 2**31 - 1, -2**31
sign = -1 if x < 0 else 1
x = abs(x)
result = 0
while x:
digit = x % 10
x //= 10
# would result * 10 + digit overflow the positive bound?
if result > (INT_MAX - digit) // 10:
return 0
result = result * 10 + digit
return sign * resultTime O(log x) (digit count), space O(1).
The overflow check is the whole problem
The digit reversal is trivial; the interview point is catching overflow before it corrupts the value. Rearranging result * 10 + digit > INT_MAX into result > (INT_MAX - digit) // 10 tests the danger without actually computing the overflowing quantity — essential in fixed-width languages (Python wouldn’t overflow, but the check models the constraint). Grouped with bit manipulation because it lives and dies by integer-bound reasoning.
Key insight
Digit reversal is easy; the skill is pre-emptive overflow checking — test whether the next operation would exceed the bound and bail, rather than overflowing and detecting after. A discipline that applies to any bounded-integer arithmetic.
Related
- concepts: math-and-geometry, bit-manipulation
- relative: Reverse Bits (same pop/push, on bits)
- prev: Sum of Two Integers — end of the NeetCode 150 🎉
Software engineering
Infrastructure & security
Computer science
Hardware & engineering