Show HN: Datagridxl.js – No-nonsense fast Excel-like data table library

Is there a UI way to quickly sort by a column?

The table has a complex layout:

- One fixed top row (date), two or three fixed columns on the left.

- Data are rendered horizontally; the number of rows does not change, but the number of columns can be in the hundreds.

Does your library support column-oriented data like this (you mention "60+" columns)? Does anyone know of any library which can easily render something like this? It's surprisingly hard to find something which lets you fix left hand columns and also render as many columns as I need.

There are many commercial and open source products on the market, some more established and less expensive than your offering. From the business standpoint, it's easy to sell something like ag-grid because many developers have experience with its open source offering, so we know that we can find experienced people on the market to help if necessary. Your project doesn't benefit from that community and doesn't benefit from having a lower price point (it's comparable to what ag-grid charges)

CSS Scroll Snap?

Nicely done.

This looks like a really great (clean and performant) JS library for interacting with data grid. I'll definitely remember to try it for my next side project.

A quick question: is there a way to make only some of the columns read-only in the grid? I looked through the doc [https://www.datagridxl.com/docs/setting-options] and it seems like user actions for columns are all (applied to every column) or none. If DataGridXL.js allows assigning read-only to a subset of columns, it'd be great to add that in the doc with an example.

Thanks for your hard work and especially for allowing others like me to try it for free!

Since you're building a library, not a tool, whenever I hire an engineer they'll automatically be using this tool and I'd need to remember to send you more money. The only other products that I have to do this with are either:

1. Monthly subscriptions that automatically charge me the correct amount of money.

2. GitLab which tracks how much I owe the developer and charges me the correct amount of money the next time around.

I'm concerned that I'll pay for this software, download a .js file, check it into my repo, hire a new developer, and completely forget they're using this product.

The biggest difference between this tool and my currently paid-for software is that it has no "check-in" period. I could very easily imagine a scenario in which I violate your license. Here are two:

1. I, as an engineer on a single project in a large company, buy your software and use it in a product. I move on to other stuff and get asked to write new features after a long time completely forgetting your software was paid-for. Me and another engineer start working on the project. We now have 2 devs.

2. Another engineer sees me building a cool project. They copy my frontend as a starter for another feature. We now have 2 devs in a multiproject license.

3. We have the correct multi project license for the correct number of devs but we want to hire some contractors for some cheap work. Hiring these contractors just became a lot more expensive.

As for the pricing, at 54,100 EUR ($62,444.87 USD) for a 50 person team, you're competing with a sizable chunk of an engineers salary. Depending on what part of the world you're in you can pay for 1 quarter to 1 years worth of an engineers time with this chunk of change. It seems you're setting yourself aside from the competition because of performance reasons. If that is the large differentiated factor I'd be thinking the following as an engineering manager:

1. Given the timeline (3 months to 1 year) can I build my own implementation of this product?

2. Given the timeline can I take an open source implementation that has more features, use it until there's a performance problem, then have one of my 10x engineers tweak it to make it perform better?

3. Can I find another paid-for version of this library that will be cheaper? In this very HN post there's another implementation of a similar product with a seemingly similar feature set that is substantially cheaper [0] that is also source available.

So, when I'm crunching the numbers, and have to choose to Build, Tweak, or Buy at a cheaper price, it's hard to justify this product. What makes it even harder to justify the product is the support limitations. The main reason to never choose build or tweak or "buy the cheaper one" is because it won't save you any time. Your support page makes it seem like, however, I'll have to put a lot of time and effort into bending over backwards to make a support request that fits your model. This makes the comparison:

1. Build XYZ and support it

2. Tweak XYZ and support it internally

3. Buy your software and get very non-specific support

[0] - https://rowsncolumns.app/pricing

Is that possible in your framework?

I wanted to tackle the performance & reliability issues mainly. My thinking is that most users won't probably need formulas, merged cells, HTML markup. DGXL really shines for larger data sets. Excellent for numeric data input specifically.

Also, I think it implements Excel/G. Sheets controls a little better: when your users are Excel-experts, but keyboard controls work just a little different then they're used to, that's quite annoying. My goal is to stick as best as I can to the original controls.

The actual values are all in HTML. The benefit is that you can use CTRL+F and values are found and selected. You can zoom the page and text will remain crisp.

The values are not in a <table> tag however. <table> values are sorted y,x (rows first, columns second). DGXL sorts values x,y (columns first, rows second). I have to admit that I am not sure what that means for screen readers: I am not an accessibility-expert at this stage. What I do know is that HTML is better than canvas by default in this regard.

Regarding other bullet points:

* Nested Grids: no * Formulas: no / not anytime soon * Column based filtering: not on top of the list, but definitely doable.

There will always be people that find it too expensive or that find it too cheap.

Even though the product looks like a simple <table> tag; the reality is that I spent 2 years working on it. (I am almost embarrassed to admit it.)

Let's say an employee of your company (or perhaps yourself) is a much better programmer than myself. He/she could perhaps build a similar data table in perhaps 6 months, including all these features that don't meet the eye: keyboard controls, context menus, touch events, virtual DOM implementation, clipboard support on all browsers and devices...

That would cost an employer perhaps 6 months worth of programmer salary, which is, depending on where you are, at least $10,000+.

Then 800 euros is not such a bad deal, I believe. At the same time, I hope it's affordable for solo makers as well.

How would you price the product?