Important points

Part-1 (Initial Setup)

• Used the github.com/joho/godotenv package in order to pull the PORT information from the .env to the current shell session
• Used go mod vendor to act as a local storage for modules, allowing Go to use these dependencies directly without needing to fetch them from remote repositories (kinnda like npm install)
• Used go mod tidy to clean up the go.mod and go.sum files by removing any unnecessary dependencies and adding any missing ones. This command is especially useful for maintaining a clean and optimized module configuration for your project

Part-2 (Setting up the chi router)

• Installed the chi-router using github.com/go-chi/chi for routing and github.com/go-chi/cors for cors
• So, we have to first use some bits, then do go mod vendor which will bring the necessary code, and then we have to do go mod tidy to do some cleanups (yes it is confusing)
• Wrote some simple code for creating a router, createing a server and then make it listen to the port coming from the .env file
• Added some cors instructions which will help us to hit the server

Part-3 (Setting up a helper function)

• Making a helper function that will help in returning the JSON respose as per the request
• First, we'll marshal the data, if it fails we'll write status code 500 to the client otherwise, we'll first add a header to make the client side browser know that we are sending some JSON data, then we'll write the successful code as well as the data
• Added a handler_readiness function, which is just to check whether the server is health is good or not
• Then in the main file, we made a new router using chi (we made this because we want to have sub routes, well technically we don't but just for versioning and stuff), then we just mounted this new router (v1Router) to the original base route that we created
• Then we made an error response helper function, which will help us to know the error, (we used struct tag to mention the error for JSON), then we used the pervious function as well, then we finally made a new handler function and used it on the v1 router that we made previously

Part-4 (Setting up database)

• Setted up a postgres database using docker, and added it's connection string to the .env file
• We'll use some libraries(ORM like but not exactly an ORM) which are github.com/kyleconroy/sqlc/cmd/sqlc@latest and github.com/pressly/goose/v3/cmd/goose@latest.
• There is no harm in commiting the vendor folder, as it isn't as large as the node module folder
• SQLC handles our queries and Goose handles our migrations
• We, then proceeded to writing the migratations for the database in the sql/schema folder, here we'll start with 001_users.sql where we created the migrations (both up and down) then we ran those migrations via the following command goose postgres postgresql://user:password@localhost:5432/postgres up (we can write down migrations as well using the same command but with down instead of up) NOTE: we have to be in the same directory
• We made a new file in the root direction named sqlc.yaml and we'll also create the queries folder and mention the query in the sql folder, and then we'll finally run the sqlc by the following command sqlc generate

Part-5 (Using the database)

• First, we'll load the database url into the environment, then we'll use the sql package from the Standard library to open the database with the mentioned syntax conn, err := sql.Open("postgres", dbURL)
• We would need some driver as well (which is weird) and it comes via github.com/lib/pq and make sure to add the import to the top of the main file as well like this _ github.com/lib.pq, the underscore means that include the code in my program even though I am not calling it directly
• Then we have to create an apiCfg which include conversion of types as well, now we can pass this apiCfg (actually we'll make the handlers part of the apiCfg stuct)to different handlers so that they can have access to the database
• Now, in the handler function we'll first define in what format we expect the data, then we'll do some encoding and decoding stuff and then use the function that is being generated by SQLC
• After that, whatever we get according to that, we'll decide whether to respond with some error or with a success object
• Made a new file models, which will contain the structure of the response of the api, this would make sure that the control for the response object is in our hand, not in the code generated by sqlc