Siddharth Agarwal

@nicholas-a-guerra @puzpuzpuz need a bit of clarity on this.. does **systemic** table store records in descending order of timestamp or ascending in each partition Asking this because : Suppose all...

@nwoolmer Got it! can you point me to the classes which I can explore for this issue, I am new to the codebase and it is complex to understand the...

@nwoolmer @puzpuzpuz I did some code walkthrough basis information conveyed by you Here's what I did Created a table with a designated timestamp column and debugged both queries Observations for...

@nwoolmer current model for (i) SELECT last(timestamp) FROM trades looks like : **select-group-by LAST(ts) LAST from (select [ts] from trades timestamp (ts))** (ii) select ts from trades order by ts...

@puzpuzpuz @nwoolmer I am writing a function like this to optimise aggregation function **LAST** This will basically perform - Addition of limit expressionNode - Addition of order by to nested...

> It'd be great to also apply the same rewrite to `select min(designated_ts) from x;`/`select max(designated_ts) from x;`. @puzpuzpuz queries like **select min(designated_ts) from x; select first(designated_ts) from x;** would...

https://questdb.io/docs/reference/function/window/#avg @nwoolmer I went through this document, and I guess VWMA will be implemented with the `PARTITION BY` clause in the same way, that's how the window will be defined...

@puzpuzpuz @nwoolmer Action items till now : - Create a `VolumeWeightedMovingAverageFunctionFactory` class which implements `FunctionFactory` and returns `isWindow()= true` which accepts 2 sigArgs - `vwma(DD)` - Added it in `module-info.java`...

@nwoolmer I can contribute in the same PR. :)

Hi @nwoolmer I went through the logic of the `(avg)` window function Please validate my understanding here, post that I'll proceed with coding Now basically we have 2 types of...