Explore tweets tagged as #SemVer
@ljharb
Jordan Harband
10 days
I made a new thing! like the semver package, but for PURLs: https://t.co/Lc6fVj8YxN `npx purl $specifier` or `npx purl $purl` will validate, normalize, and provide parse info. add `--check` & it'll contact the relevant registry & verify the package and version exist.
0
1
4
@PredragGruevski
Predrag Gruevski
8 days
cargo-semver-checks is growing faster than ever: - 7 new releases, from v0.39 to v0.45 - 122 new lints, more than double last year's count - 4x reduction in lint execution time — some lints became up to 10x faster - across 26 (!!) rustdoc format versions
1
0
4
@Sellmair
Sebastian Sellmair
2 days
The Kotlin ecosystem might not choose a strict 'SemVer' notation (and it's gonna be fine!) https://t.co/fkBtOybZHY
1
1
24
@mandy8055
Manuj Sankrit
8 days
Navigating npm dependencies shouldn't be a rollercoaster. 🎡 New technical guide on: 🔹 PeerDeps: The "Bring Your Own" logic 🔹 ESM vs CJS: Solving the 15A/5A crisis 🔹 SemVer: Demystifying versioning Read on @thepracticaldev: https://t.co/1OVNbQGKNq #Javascript #npm #NodeJS
0
1
3
@Sellmair
Sebastian Sellmair
3 days
I am working on a proposal on making Kotlin's ecosystem versioning more consistent (at least for JetBrains projects at first). It sparked this nice blog post from @martinbonnin . I'll respond on https://t.co/WY5k6jBbWE soon 🔜 https://t.co/d3AXsOiWo0
0
1
9
@atreides_sf
atreides
5 days
every upgrade to my iphone has come at the point of a spear. every iOS patch is a war crime, written in semver. apple in the time cook era has pushed the limits of bullshit jobs for mobile product managers not worth 10 sq feet of cupertino real estate to their asymptotic
0
1
2
@LippetySchtick
Leddy Something
11 days
quick dev note turned my agent workflow on @symphonyio into a mini CI/CD pipeline agents are versioned like code, marketplace publishes semver tags, and my nightly job pulls on‑chain execution receipts to auto‑label training data (latency, route hops, solver id, realized
6
0
10
@gregthwuen
Gregor
5 days
@ArielGMachado @UniverseIce No, this is plain wrong, Linus bumps the first number whenever he feels like the second number is getting too big. This is NOT semver.
0
0
5
@oraclee_xyz
oraclee.btc 🧠
5 days
Copilot failing semver and pretending Rust 1.91.1 isn’t real is wild. Single-file PR reviews are toy vibes, no global context. Real review starts with issue/architecture, then style. Triec CodeRabbit/Codex Cloud - better, but loop hell. holon reviewer mode, ship it
@jolestar
jolestar
13 days
一直默认用 GitHub Copilot 来 review PR,毕竟 GitHub 会自动跑,而且看起来也不收费。 但最近几次 review,让我开始怀疑它到底“看懂”了什么。 比如一个很基础的问题:它依然会把 1.82.0 认为比 1.91.1 高,完全是早期大模型常见的版本号判断错误。 如果说这是模型问题,那它还会认为 rust 1.91.1
6
0
23