backend_task.md

Criteria

Your work will be evaluated primarily on:

• Consistency of coding style (ideally in harmony with our JavaScript style guide
• Idiomatic use of express,node, mongodb, and lodash (Don't use for loops, lodash is mandatory)
• Absence of callback hell
• Documentation of API end-points
• Efficient MongoDB queries
• General quality of code and technical communication.

Task

1. Create MongoDB using the csv files. One collection per file. Define database schema. Data is given in Here

2. Create end-point to add new entry

   Input format

   {
   	"movie_id" : String,
   	"title" : String,
   	.
   	.
   	.
   	"budget" : Number,
   	"genres" : Array,
   	"homepage" : Array,
   	.
   	.
   	.
   	"vote_average" : Number,
   	"vote_count" : Number
   }
   

3. Create end-pont which populates profit earned by production house in a given year. Here profit is calculated as [Revenue - Budget].

   Output format

   {
   	"<name of production_company_1>" : <profit_1>,
   	"<name of production_company_2>" : <profit_2>,
   	.
   	.
   	.
   	"<name of production_company_n>" : <profit_n>
   }
   Note: Assume, Revenue and Budget are ditributed equaly over all production companies for a given movie.
   

4. For a given movie find N similar movies. Similarity can be checked using what number of keywords are matching between input movie and avaialable movies. Sort the result in descending order. if count are same then combine those movie names and make an array, if not then make an array of single movie name.

   Output format

   [
   	{
   		"movie_names" : [<movie_1>,<movie_2>],
   		"Number of keywords matched" : <count_1>
   	},
   	{
   		"movie_names" : [<movie_3>],
   		"Number of keywords matched" : <count_2>
   	},
   	.
   	.
   	.
   	{
   		"movie_name" : [<movie_N>],
   		"Number of keywords matched" : <count_N>
   	}
   ]
   

5. Given a movie name, for each cast member of the movie, find the number of movies per genre in which the actor casted. If a movie has multiple genre then count it as one for each genre.

   Output format

   [
   	{
   		"actor_name": <actor_1>,
   		"movies_per_genre": [
   			{
   				"genre": <actor_1_genre_1>, // genre_1 for actor_1
   				"number of movies": <count_1>
   			},
   			{
   				"genre": <actor_1_genre_2>, // genre_2 for actor_1
   				"number of movies": <count_2>
   			},
   			.
   			.
   			.
   			{
   				"genre": <actor_1_genre_n>, // genre_n for actor_1
   				"number of movies": <count_n>
   			}
   		]
   	},
   	{
   		"actor_name": <actor_2>,
   		"movies_per_genre": [
   			{
   				"genre": <actor_2_genre_1>, // genre_1 for actor_2
   				"number of movies": <count_1>
   			},
   			{
   				"genre": <actor_2_genre_2>, // genre_2 for actor_2
   				"number of movies": <count_2>
   			},
   			.
   			.
   			.
   			{
   				"genre": <actor_2_genre_n>, // genre_n for actor_2
   				"number of movies": <count_n>
   			}
   		]
   	}
   
   ]
   

Extra credit

• Integrate Apollo and Express to build a Node.js GraphQL API istead of REST API
• Dockerize your app
  • Note: You can build your image in alpine or debian

How to submit your work

Fork this project on github. When you're finished, push final code in to the public repository. Consider using .gitignore to avoid putting node_modules