Signal
19 articles covering "Node.js"
Sources on four 2026 changes show four notice models: a published Node.js calendar, Microsoft's 12-month minimum for ASP.NET Core 2.3, a Python Steering Council window for the JIT, and Kubernetes version targets.
Oracle's JEP 541, Homebrew 7.0.0 and the Node.js 27.0.0-alpha.0.0.0 notes each lower macOS x64 support in different ways, while Apple's macOS 27 does not support Intel Macs.
Reporting on the Brotli decoder flaw CVE-2026-111497 shows fixes, scanner coverage and exploitation status differing across Node.js, .NET, Python packages and others.
Node.js 26.13.0 marks 26.x as LTS with updates through the end of April 2029 and no other changes from 26.12.1; the 27.0.0-alpha.0 pre-release lists V8 15.6.75.13 and breaking changes.
Electron 45.0.0, published October 21, upgrades to Chromium 156.0.8078.48 and Node v24.22.0, lists breaking changes for preload scripts and screen capture, and ends support for Electron 42.
Node.js published its October 21 security releases for the 26.x, 24.x and 22.x lines: ten CVEs, four rated High, plus c-ares, Undici and certificate updates.
Node.js says new 26.x, 24.x and 22.x releases will land on or shortly after October 21 to fix vulnerabilities, highest severity High; 26.12.0 and 24.22.0 shipped October 14.
Node.js 26.11.0 renames DatabaseSync and StatementSync to Database and Statement with aliases kept, adds --process-timeout, and 26.11.1 reverts three commits.
Node.js closed the 19,000-line VFS pull request that split the project over AI-generated code, after its functionality shipped through smaller PRs and the TSC adopted a formal AI use policy.
Node.js 26.9.0 adds a global Worker class implementing the WHATWG Web Worker API and enables the experimental node:ffi module by default.
pnpm 12 replaces the package manager's TypeScript/Node.js implementation with Rust, with independent Vercel Turborepo testing showing median install-time cuts of 64.4% to 90.5%.
Harper's 5.2 release isolates database commits onto dedicated threads and adds a record cache, while a company benchmark argues single-runtime architecture beats a Vercel-based stack on personalized reads.