Umang Mundhra

The work of this PR has been addressed and continued in PR #930. Closing here

@reski-rukmantiyo after some discussion with the team we have came to a point that at a particular time we don't want to deep dive into all the supported functionalities for...

Sure @reski-rukmantiyo moreover i will suggest let's do the second one first: - support for primary_key, auto_increment, not_null, unique, index, and unique_index,check,comment in gofr tags : change into sql meta...

This issue has been resolved in PR #930.

> @Umang01-hash Can you add more details on sequence of R(Read) and W(Write) cases under which read and write replicas would be selected ? Hey @gizmo-rt we first determine is...

> > > @Umang01-hash Can you add more details on sequence of R(Read) and W(Write) cases under which read and write replicas would be selected ? > > > >...

> Hope we are routing the reads within a transaction to the primary? I couldn't find any the corresponding code for it. Yes @akshat-kumar-singhal Reads within transactions are always routed...

This PR implements SQL-level routing (parsing queries) to split reads vs writes. Another approach is to use the HTTP method (e.g. `GET`=**read**, `POST/PUT`=**write**) to choose the DB. Given cases where...

Hey! Here are my sugegstions for the above problems/comments: 1. **Transaction Scoping:** When developers use `Begin() → Commit() `(database transactions), all queries inside that block would **automatically go to the...

@aryanmehrotra Thanks for your suggestions. Your points seem valid - transaction scoping wouldn't work in three-layer architecture, and global read-after-write tracking would be too expensive. **Request-Scoped Routing:** Instead of query-level...