Problem

Given a sentence s containing words separated by spaces, return the longest word. If several words tie for the longest length, return the one that appears first.

Input / Output

• Input: string s.
• Output: the longest word (empty string if s has no words).

Examples

Example 1 Input: s = "the quick brown fox jumped" Output: "jumped"

Example 2 Input: s = "a bb cc dd" Output: "bb" (first of the length-2 words)

Example 3 Input: s = " " Output: ""

Constraints

• 0 <= s.length <= 10^5

Approach

Scan the string once, tracking the start of the current word. When a space (or the end of the string) is reached, compare the finished word's length to the best so far and update only if it is strictly longer — that guarantees the first of equal-length words wins.

Handling punctuation is a common extension: strip non-letters before measuring, or treat any non-letter as a separator.

max(s.split(), key=len) is the one-liner; it also returns the first maximal element because max keeps the first maximum it sees, but you should be able to explain the manual scan.

Complexity

• Time: O(n)
• Space: O(1) extra (besides the returned word).

Reference solution (Python)

def largest_word(s: str) -> str:
    best_start = best_len = 0
    i, n = 0, len(s)
    while i < n:
        if s[i] == ' ':
            i += 1
            continue
        j = i
        while j < n and s[j] != ' ':
            j += 1
        if j - i > best_len:
            best_start, best_len = i, j - i
        i = j
    return s[best_start:best_start + best_len]

Test cases

Input	Expected
"the quick brown fox jumped"	"jumped"
"a bb cc dd"	"bb"
" "	""
"single"	"single"