/Fampay-Assesment

Primary LanguageJavaScript

Project Design Overview

##Flow

  • API Layer -> Logic Layer -> Repository Layer
  • Dependencies managed in "di/index.js".

Cron Job

  • Fetches videos every 10 seconds for the predefined text "tech".

Reasons to Choose MongoDB

  • Lower consistency (Eventual Consistency) is acceptable for slight delays in showing the latest videos.
  • Easily scalable using sharding compared to SQL.

Text Searching (Bonus Included)

  • Returns videos if title or description is matched using MongoDB text search operator.
  • Bonus: Partial matches returned by checking if individual words are present in the title/description as substrings.
  • Elastic search can be used to make the searching better in large scale projects

API Fallback Key (Bonus Included)

  • Retries with a new API key if quota is exceeded.
  • Implemented in "repository/YouTubeServices/youtube-YouTubeService.js" with console.log helpers.

Steps to Run Without Docker

  1. Install Node 16.
  2. Run npm i.
  3. Run node index.js (API server on port 3000).
  4. Run node jobs/fetch-videos/index.js (Cron job for YouTube videos).
  5. Use provided CURL commands.

Steps to Run With Docker

  1. docker build -t video-service -f deployment/Dockerfile .
  2. docker build -t job-fetch-videos -f jobs/fetch-videos/deployment/Dockerfile .
  3. docker images (verify images).
  4. docker run -p 3000:3000 -d video-service (API server on port 3000).
  5. docker run -d job-fetch-videos (Cron job for YouTube videos).
  6. docker ps (get container IDs).
  7. docker logs <containerId> -f (check logs).
  8. Use provided CURL commands.

CURLs

API: GET /

Description: Get all videos in descending order of publised time having query as limit and offset. limit can take 15 as the max limit

curl --location 'localhost:3000/video-YouTubeService/videos?limit=15&offset=0'

API: GET /search (Bonus Included)

Description: Search using query q

curl --location 'localhost:3000/video-service/videos/search?q=tech'