Let's implement a proper open source pivot/datatable library integrated with latest technologies

[davidpelayo]

Introduction

I needed a datatable solution coworking nicely with React. Then one of the views I had to implement was a pivot table layout with a big dataset, aggregating certain columns (sums, averages, totals, etc).
I ended up playing around with many open source solutions, and I forked orb so I could customize it on my needs.

Quickly I realize the open source existing solutions weren't as good as to have in a production ready application, so I ended up paying a license of a commercial lib together with my client.

Nice to have

An open source Pivot table library fully integrated with latest FE solutions on the ecosystem.
Pivot means also spot for datatables if wanted.

Motivation

After the movement I'm observing of a few people here and the potential interest these type of solutions may have, I wonder whether you agree on creating together a proper open source solution for pivot tables, fully working and high-performant with latest technologies available in the FE field (i.e. React, Vuejs, and others).

Additionally, I've seen how poor are the existing open source solutions when it comes to render high volume of datasets or aggregate them on the client side, leading us to go for payed solutions. There are a few open solutions over there, but either they are not properly maintained, or they are just too old.

Proposal

If you find this interesting I can start refactoring this lib from the scratch (already thought of a good name) so we can inherit this codebase to go progressively refactoring until having something useful and easy to plug-in in any project you may need it.

Calling for contributors

I would be nice to have a team of experts willing to collaborate with an open source solution like this.
I spent a couple of days playing around with the existing solutions and I ended up paying a commercial license of a non-open source one, simply because in the end it was cheaper paying than developing the solution by my own.

Feedback is welcome

What do you think?

[jsegura]

Totally agree with this idea.

[turnerniles]

Hi @davidpelayo, @jsegura and all. We faced the same challenge and have been working on a pivot that can handle 100k plus rows. See react-virtualized-pivot and the pivot engine, quick-pivot. We're still trying to reach feature parity with orb but have come quite a long way and are building the pivot with "large" datasets in mind. Let me know what you think. We would appreciate any help if you're interested in the project. I will post on the other thread and some other places over the next couple days to get some more feedback and see if anyone is interested. We still have a ways to go.