Design Video Sharing Platform - Practice Coding Problems

Design Video Sharing Platform - Problem

Design a comprehensive video sharing platform similar to YouTube or TikTok that supports core functionalities like uploading, deleting, watching, and rating videos.

Each video is represented as a string of digits, where the i-th digit represents the content at minute i. For example, video "12345" has content '1' at minute 0, '2' at minute 1, and so on.

Key Features:

Smart ID Management: Videos get the smallest available ID starting from 0. Deleted video IDs can be reused.
Video Streaming: Users can watch specific time ranges of videos
Engagement Tracking: Track views, likes, and dislikes for each video
Content Management: Upload and remove videos dynamically

Implement the VideoSharingPlatform class with methods for upload, remove, watch, like, dislike, and analytics retrieval. The platform should efficiently handle video ID recycling and maintain accurate engagement metrics.

Input & Output

example_1.py — Basic Operations

$ Input: ["VideoSharingPlatform", "upload", "upload", "remove", "upload", "watch", "like", "dislike", "getLikesAndDislikes", "getViews"] [[], ["123"], ["456"], [0], ["789"], [1, 0, 1], [1], [1], [1], [1]]

› Output: [null, 0, 1, null, 0, "45", null, null, [1, 1], 1]

💡 Note: Upload "123" gets ID 0, upload "456" gets ID 1, remove video 0, upload "789" reuses ID 0, watch video 1 from minute 0-1 returns "45", like and dislike video 1, then get stats

example_2.py — Edge Cases

$ Input: ["VideoSharingPlatform", "upload", "watch", "watch", "watch"] [[], ["12345"], [0, 0, 2], [0, 3, 10], [0, 10, 15]]

› Output: [null, 0, "123", "45", ""]

💡 Note: Upload video "12345", watch minutes 0-2 returns "123", watch minutes 3-10 returns "45" (clamped to video length), watch beyond video length returns empty string

example_3.py — Invalid Operations

$ Input: ["VideoSharingPlatform", "watch", "like", "getLikesAndDislikes", "getViews"] [[], [999, 0, 1], [999], [999], [999]]

› Output: [null, "-1", null, [-1], -1]

💡 Note: Operations on non-existent video ID 999 return appropriate error values: "-1" for watch, [-1] for stats, -1 for views

Visualization

Tap to expand

Understanding the Visualization

Smart ID Assignment

Use min-heap to always assign the smallest available ID

Fast Video Access

Hash map provides instant access to any video by ID

Engagement Tracking

Separate stats tracking for views, likes, and dislikes

Efficient Cleanup

Deleted video IDs return to the available pool

Key Takeaway

🎯 Key Insight: Combining min-heap for ID management with hash maps for data storage creates a scalable video platform that efficiently handles millions of videos while maintaining optimal performance for all operations.

Time & Space Complexity

Time Complexity

⏱️

O(log n) for upload, O(1) for other operations

Heap operations for ID management, direct hash access for video operations

⚡ Linearithmic

Space Complexity

O(n)

Space for heap, hash maps, and video content storage

⚡ Linearithmic Space

Constraints

1 ≤ video.length ≤ 10⁵
The sum of video.length over all calls to upload does not exceed 10⁵
video consists of digits
0 ≤ startMinute < endMinute < 10⁵
0 ≤ videoId < 10⁵
At most 10⁴ calls will be made to all functions combined
All video content must consist only of digit characters

Asked in

Y YouTube 85 N Netflix 72 T TikTok 68 a Amazon Prime 45 f Meta 38

The optimal solution uses a min-heap to efficiently manage available video IDs and a hash map for O(1) video operations. This design achieves O(log n) upload time and O(1) for all other operations, making it highly scalable for large video platforms.

Common Approaches

Approach	Time	Space	Notes
✓ Brute Force (Linear Search for IDs)	O(n) for upload, O(1) for other operations	O(n)	Use array/list and linear search for next available ID
Optimized Design (Min-Heap + Hash Map)	O(log n) for upload, O(1) for other operations	O(n)	Use min-heap for ID management and hash map for O(1) video operations

Brute Force (Linear Search for IDs) — Algorithm Steps

Store videos in a map/dictionary with video ID as key
For upload: iterate through IDs 0,1,2,... until finding unused one
For remove: simply delete from the map
For other operations: direct map access

Visualization

Tap to expand

Step-by-Step Walkthrough

Upload Video

Search from ID 0 upward until finding unused slot

Store Video

Place video in the first available ID slot

Track Stats

Initialize engagement metrics for the new video

Code -

solution.c — C

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#define MAX_VIDEOS 10000

typedef struct {
    char* videos[MAX_VIDEOS];
    int views[MAX_VIDEOS];
    int likes[MAX_VIDEOS];
    int dislikes[MAX_VIDEOS];
    int exists[MAX_VIDEOS];
} VideoSharingPlatform;

VideoSharingPlatform* videoSharingPlatformCreate() {
    VideoSharingPlatform* obj = (VideoSharingPlatform*)malloc(sizeof(VideoSharingPlatform));
    for (int i = 0; i < MAX_VIDEOS; i++) {
        obj->videos[i] = NULL;
        obj->views[i] = 0;
        obj->likes[i] = 0;
        obj->dislikes[i] = 0;
        obj->exists[i] = 0;
    }
    return obj;
}

int videoSharingPlatformUpload(VideoSharingPlatform* obj, char* video) {
    int videoId = 0;
    while (videoId < MAX_VIDEOS && obj->exists[videoId]) {
        videoId++;
    }
    
    obj->videos[videoId] = (char*)malloc(strlen(video) + 1);
    strcpy(obj->videos[videoId], video);
    obj->exists[videoId] = 1;
    obj->views[videoId] = 0;
    obj->likes[videoId] = 0;
    obj->dislikes[videoId] = 0;
    
    return videoId;
}

void videoSharingPlatformRemove(VideoSharingPlatform* obj, int videoId) {
    if (videoId >= 0 && videoId < MAX_VIDEOS && obj->exists[videoId]) {
        free(obj->videos[videoId]);
        obj->videos[videoId] = NULL;
        obj->exists[videoId] = 0;
    }
}

char* videoSharingPlatformWatch(VideoSharingPlatform* obj, int videoId, int startMinute, int endMinute) {
    if (videoId < 0 || videoId >= MAX_VIDEOS || !obj->exists[videoId]) {
        char* result = (char*)malloc(3);
        strcpy(result, "-1");
        return result;
    }
    
    obj->views[videoId]++;
    
    char* video = obj->videos[videoId];
    int len = strlen(video);
    
    if (startMinute >= len) {
        char* result = (char*)malloc(1);
        result[0] = '\0';
        return result;
    }
    
    int endIdx = endMinute < len - 1 ? endMinute : len - 1;
    int resultLen = endIdx - startMinute + 1;
    
    char* result = (char*)malloc(resultLen + 1);
    strncpy(result, video + startMinute, resultLen);
    result[resultLen] = '\0';
    
    return result;
}

Time & Space Complexity

Time Complexity

⏱️

O(n) for upload, O(1) for other operations

Finding next available ID requires checking all existing IDs in worst case

✓ Linear Growth

Space Complexity

O(n)

Space for storing video data and metadata

⚡ Linearithmic Space

Constraints

1 ≤ video.length ≤ 10⁵
The sum of video.length over all calls to upload does not exceed 10⁵
video consists of digits
0 ≤ startMinute < endMinute < 10⁵
0 ≤ videoId < 10⁵
At most 10⁴ calls will be made to all functions combined
All video content must consist only of digit characters

67.5K Views

High Frequency

~35 min Avg. Time

2.8K Likes

Ln 1, Col 1

Smart Actions

💡 Explanation

AI Ready

💡 Suggestion Tab to accept Esc to dismiss

// Output will appear here after running code

Code Editor Closed

Click the red button to reopen

Input & Output

Visualization

Time & Space Complexity

Related Problems

Constraints

Common Approaches

Brute Force (Linear Search for IDs) — Algorithm Steps

Visualization

Code -

Time & Space Complexity

Constraints

Select Compiler