Summary

—Will Add 😊—

Learnings / Journal ➖

Advantages of differentiating backend between different platforms / providers. Example:- AWS free tier EC2 has cold start and limited CPU power, but bigger memory, On the other hand Cloudfare workers has smaller memory but hot start. So for faster operations which require less memory like auth operations and caching I will use Cloudfare Workers, and for more memory extensive tasks like image processing I will be using AWS EC2, as both are free I will be happier
Difference in Databases - I initially thought of using MongoDB as sole database, but I thought of implementing things like rate limiting, email verification rate limiting and as i wanted to try it out for a long time, I decided to use Upstash Redis
Before Starting the build, i am reading the Hono docs for cloudfare workers backend
I came to know about indexes in MongoDB, nice thing
I now know how to use Sentry for logging errors, at least at a very basic level.

I now know that Cloudfare does not support custom TCP connections, and as mongoose works on TCP connection to Atlas, I now need to use a new library to access my mongo database, ie mongodb library, which works on HTTP connections
I have implemented caching of clients as forming new connections for each request separately seems inefficient.
WHY IS MY GITIGNORE NOT WORKING???
It worked, don’t know how, don’t ask how, it did, I’m happy 🙂
I have also completed a basic system design to help me visualize this, it is a bit lacking, but enough to keep me on track

The caching of clients did not work, so I am using the basic approach
I am currently working on my auth endpoints
I learnt about JS Regular expressions or regex, pretty similar to regular expressions from compiler design