std::mem::drop
@stdmemdrop
Rust 🦀 ∙ Typescript ∙ Go ∙ Python 🐍 Architect + Dev ⋮ UX/CX/DX I ❤️ software
Finally shipped clickhouse-datafusion to crates.io/crates/clickho…! Bridge @ClickHouseDB and @ApacheDataFusio for high-performance analytics. Features: 🔗 Cross-database federation ⚡ Connection pooling 🎯 Arrow-native performance 🛠️ ClickHouse UDF support
Its happening -- DataFusion will (finally) get spilling hash joins. The march to completeness begins
I'd like to start using this platform as a place to post about open source work I do on my off time. To lead it off, I have posted a hash join spilling proposal in Apache Datafusion. Check it out if you're interested 😀: github.com/apache/datafus…
What would it take to see dependent types in Rust? Is it even possible? Is it merely compiler limitations that prevent it thus far?
Lmao
What do you prefer for building backend LLVM 🦅 or Rust 🦀
In the case of a theorem prover, 0 would be my guess. Otherwise I like the idea of being able to produce a +Infinity using an expression. So that’s my choice.
If you were to define what `x/0` did in a programming language, what would you define it as?
Getting ready to release my new Open Source rust crate today: `clickhouse-datafusion`. It allows using @ClickHouseDB as a table provider in @ApacheDataFusio with full set of capabilities. Additionally it introduces a custom Analyzer that allows “pushing down” clickhouse…
Languages w/o match semantics are not serious languages. Such a fundamentally useful paradigm.
I finally have these agentic coders writing better Rust. ...ffs Tell it, "Remember, this is not Java, it's Rust. Be more functional." That's it. You're welcome.
This is getting so much attention bc it’s an utter abomination. Coding is fascinating.
I don't want "smarter" AI. I want robotic, annoyingly pedantic, rule following AI.
I'm building an OS "rust analyzer -> LLM" helper for any @rustlang devs out there. All commits so far have been authored & pushed by @claudeai's ClaudeCode, except first commit. It's an experiment to see if this is possible. So far, so decent. github.com/GeorgeLeePatte…
LLMs are fundamentally flawed. They don't follow directions well. You'd expect a robot-like fidelity to the directions provided, almost to an annoying degree, but that's not what happens. They're more like ADHD teenagers that "kinda" get what you asked for.
Told y’all
United States Trends
- 1. Giants 63.8K posts
- 2. Bills 132K posts
- 3. Bears 56.6K posts
- 4. Caleb 46K posts
- 5. Dart 24.7K posts
- 6. Dolphins 31.2K posts
- 7. Daboll 10.4K posts
- 8. Seahawks 16.1K posts
- 9. Jags 6,291 posts
- 10. Josh Allen 14.8K posts
- 11. Russell Wilson 3,768 posts
- 12. Texans 36.1K posts
- 13. Browns 35.9K posts
- 14. Ravens 35.9K posts
- 15. Patriots 101K posts
- 16. Henderson 16.2K posts
- 17. Trevor Lawrence 2,319 posts
- 18. Bryce 14.9K posts
- 19. Drake Maye 14.8K posts
- 20. Vikings 29.1K posts
Something went wrong.
Something went wrong.