Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Just opened a 1,000,000-line SQL file. Smooth scroll. Full IntelliSense. Zero lag. Pure Rust 🦀 No Electron. No Tauri. Drops soon with Postgres 👇

50,442 Aufrufe • vor 2 Monaten •via X (Twitter)

64 Kommentare

Profilbild von José Gurruchaga
José Gurruchagavor 2 Monaten

Gpui is insane, zed doing the lord’s work

Profilbild von Antfroze
Antfrozevor 2 Monaten

YES 🙌

Profilbild von Antfroze
Antfrozevor 2 Monaten

I don't think there is another editor that can do this

Profilbild von Sahith
Sahithvor 2 Monaten

Why the F do you have a 1M line SQL file?

Profilbild von Antfroze
Antfrozevor 2 Monaten

Valid reaction 😂 They do exist somewhere out there

Profilbild von Json
Jsonvor 2 Monaten

Good to see people use the right technologies for such tools. Great job!

Profilbild von Antfroze
Antfrozevor 2 Monaten

Thank you 🙏 gpui truly is insane

Profilbild von Robert E. Beckner III (Merlin)
Robert E. Beckner III (Merlin)vor 2 Monaten

Legend! Nice moves!!

Profilbild von Antfroze
Antfrozevor 2 Monaten

Thank you 🙏 Much appreciated

Profilbild von Vishnu Dileesh
Vishnu Dileeshvor 2 Monaten

Looks performatively neat

Profilbild von Antfroze
Antfrozevor 2 Monaten

Thank you 🙏

Profilbild von samyak
samyakvor 2 Monaten

@AstraKernel Great work! GPUI definitely is great for such performance usecase. I have a question though, is GPUI better choice than Tauri for software like postman, accounting ?

Profilbild von Antfroze
Antfrozevor 2 Monaten

@AstraKernel Thank you 🙏 If you are up for it, I would recommend using GPUI over Tauri. So many issues came out of Tauri for me, the current version available on the website is built with it

Profilbild von keiko
keikovor 2 Monaten

why would i switch to this from dbeaver?

Profilbild von Antfroze
Antfrozevor 2 Monaten

DBeaver did this to me after trying to replicate the video

Profilbild von keiko
keikovor 2 Monaten

try opening in zed

Profilbild von Antfroze
Antfrozevor 2 Monaten

I did, it struggles to keep up with highlighting (because treesitter) and also there is no good autocomplete for SQL anywhere but in SQL Studio & DataGrip. It's not a fair comparison

Profilbild von 星川 ✨ StarFlow
星川 ✨ StarFlowvor 2 Monaten

这也太丝滑了,这就去试试

Profilbild von Antfroze
Antfrozevor 2 Monaten

Thank you! The version seen in the video isn't out yet but will be soon :)

Profilbild von law
lawvor 2 Monaten

Can feel Zed's UI library from a mile away

Profilbild von Antfroze
Antfrozevor 2 Monaten

GPUI for the win 🏆

Profilbild von Mother Brain ⋈🇵🇹
Mother Brain ⋈🇵🇹vor 2 Monaten

@late4nothing Any GPUI framworks you favor? I've seen a lot come from Zed's, like

Profilbild von Antfroze
Antfrozevor 2 Monaten

@late4nothing I made my own, but if I were to use one I would make sure it is unstyled

Profilbild von Safwan
Safwanvor 2 Monaten

excited for this cuz i look at sql dumps often lol

Profilbild von rusa
rusavor 2 Monaten

a million lines with intellisense and no lag is the demo that matters. that is exactly where electron editors fall over.

Profilbild von Antfroze
Antfrozevor 2 Monaten

Very true, though most people won't have files this big or bigger - keeping it smooth ensures any workload will be

Profilbild von akshay oppiliappan
akshay oppiliappanvor 2 Monaten

gpui?

Profilbild von Antfroze
Antfrozevor 2 Monaten

Yes 🙌

Profilbild von Buswe
Buswevor 2 Monaten

Que sea Rust nativo sin webview es justo lo que deja virtualizar el render y mover un millón de líneas sin lag.

Profilbild von Antfroze
Antfrozevor 2 Monaten

Very true in this case 🙌

Profilbild von Vikrant Guleria
Vikrant Guleriavor 1 Monat

The real test for a SQL editor at this scale isn't rendering, it's whether it can run EXPLAIN ANALYZE on a query against a huge Postgres table and keep the UI responsive while the plan comes back. Curious if streams query results incrementally or waits for the full result set before painting.

Profilbild von Tts
Ttsvor 2 Monaten

who has 1M lines sql file?

Profilbild von Antfroze
Antfrozevor 2 Monaten

Me and many others (most likely) 🙌

Profilbild von Sebastian Buzdugan
Sebastian Buzduganvor 2 Monaten

million-line rendering is easier than keeping incremental parsing fast after edits

Profilbild von Antfroze
Antfrozevor 2 Monaten

Very true, mine does both 🙌

Profilbild von Arik Chakma
Arik Chakmavor 2 Monaten

did you move to gpui by any chance?

Profilbild von Antfroze
Antfrozevor 2 Monaten

Yep 👍

Profilbild von Arik Chakma
Arik Chakmavor 2 Monaten

crazy 🫡

Profilbild von Rahul Rajaram
Rahul Rajaramvor 2 Monaten

What’s are performance numbers like on the file system (and disk if you’re using HDD)? Naiively, I would think this has the ability to trigger a lot of syscall activity upon frequent abrupt cursor pos changes. Should apply to Sublime Text similarly though.

Profilbild von Antfroze
Antfrozevor 2 Monaten

Cursor moves and typing don't hit disk. The file is already in memory. Disk only comes into play on open/save

Profilbild von nik
nikvor 2 Monaten

gpui?

Profilbild von Antfroze
Antfrozevor 2 Monaten

Yes 🙏

Profilbild von Antfroze
Antfrozevor 2 Monaten

It's too good, had to rewrite in it once I built a nice UI library

Profilbild von आशीष खरे
आशीष खरेvor 2 Monaten

Beautiful

Profilbild von Antfroze
Antfrozevor 2 Monaten

Thank you 🙏

Profilbild von आशीष खरे
आशीष खरेvor 2 Monaten

Been a fan since you put up the very first tweet for with only the email bar. It's the most beautiful UI I've ever used.

Profilbild von Antfroze
Antfrozevor 2 Monaten

Thank you 🙏 it means a lot to me. The new version is completely redesigned and faster in every way. Excited for you to try it :)

Profilbild von Antfroze
Antfrozevor 2 Monaten

By completely redesigned I mean improved btw it keeps the same soul

Profilbild von आशीष खरे
आशीष खरेvor 2 Monaten

I'll update to newer version.

Profilbild von Antfroze
Antfrozevor 2 Monaten

It will be out soon :)

Profilbild von Sora
Soravor 2 Monaten

1 Million Lines better not be lagging, or you actively do stuff you shouldn't be doing. This is also the absolute bare minimum.

Profilbild von Antfroze
Antfrozevor 2 Monaten

It should be the bare minimum but editors usually can't keep up with this in many ways. Whether that be intelisense, syntax highlighting, or find & replace or all of them combined hahaha

Profilbild von Sora
Soravor 2 Monaten

Syntax highlighting is not that expensive if you don't need semantic information like in tree-sitter. Intellisense should be done through a fully separate thread anyway and also only has some information about hierarchy...

Profilbild von Droid
Droidvor 2 Monaten

Are there plans for MySQL support?

Profilbild von Antfroze
Antfrozevor 2 Monaten

Yes :)

Profilbild von SyntaxError » CEO of "CEO of..."-names
SyntaxError » CEO of "CEO of..."-namesvor 2 Monaten

wdym 1000000 SQL file

Profilbild von Antfroze
Antfrozevor 2 Monaten

1 million lines, smooth and seamless just like 100 or 10 lines

Profilbild von Rajat Kulkarni
Rajat Kulkarnivor 2 Monaten

GPUI I’m assuming? Seems like the new hype! Unfortunately can’t migrate @tryOctarine due to the editor becoming incompatible

Profilbild von Antfroze
Antfrozevor 2 Monaten

@tryOctarine Yep 👍 it was a harsh migration but very worth it for me. Tauri was holding me back too much, once I prototyped with it I knew I had to continue

Profilbild von Aditya 👀
Aditya 👀vor 2 Monaten

Why would you make a million line sql query anyways 😭

Profilbild von Antfroze
Antfrozevor 2 Monaten

SQL dumps / autogenerated migrations, or some other reason. It's good to have an incremental parser / editor that is capable enough to handle any workload

Profilbild von Aditya 👀
Aditya 👀vor 2 Monaten

alright makes sense, I remember digging through a migration in raw sql file once

Profilbild von Andi Liu
Andi Liuvor 2 Monaten

没有AI,我们真的能处理得了1,000,000 行的 SQL 文件吗?如果有了AI,我们真的有必要处理1,000,000 行的 SQL 文件吗?

Profilbild von rusa
rusavor 1 Monat

the trick with a million lines isn't rust specifically, it's never touching the whole file at once, virtualized rendering for what's visible plus a rope or piece table for edits. native mostly buys you a small constant factor on top. what's backing the intellisense at that size, incremental parse.

Ähnliche Videos

Your agents can't keep up with real-time data. Especially when it's scattered across dozens of sources. Most teams waste weeks building custom connectors for every database, API, and data warehouse. Then they build ETL pipelines to sync everything. By the time your agent retrieves the data, it's already outdated. Picture this: Your Postgres database updated 5 minutes ago. Your MongoDB collection changed 2 minutes ago. Your agent is still pulling from yesterday's snapshot. This is why most production RAG systems fail. There's a better approach: MindsDB is an open-source AI platform with a federated data engine that lets you query multiple data sources in real-time using SQL - without moving any data. Here's what makes it different: ↳ Your data stays in place. No ETL pipelines or data duplication ↳ Query Postgres, MongoDB, REST APIs, and more using consistent SQL ↳ JOIN across different sources in real-time with a unified interface ↳ Works with both structured and un-structured data And here's the best part: You don't even need to write SQL. Just describe what you want in plain English, and MindsDB converts it to SQL automatically. The system does all the heavy lifting. The breakthrough for AI agents is simple: When data updates at the source, your agent gets fresh results immediately. No sync delays. No stale embeddings. No custom code for each integration. You can literally write a SQL query that joins a Postgres table with a MongoDB collection and gets live results. This is what production AI applications need but rarely get. In this video, I give you a complete walkthrough of what we just discussed and how to actually do it. Make sure you watch this till the end. I've shared the link to MindsDB's GitHub repo in the next tweet!

Akshay 🚀

65,672 Aufrufe • vor 11 Monaten