Languages
View on GitHub →C++33.9%
Kotlin19.6%
JavaScript18.9%
Objective-C++11.2%
Objective-C6.9%
Java3.7%
Community health
6 of 6 standards met
87
✓README✓License✓Contributing✓Code of Conduct✓Issue Template✓PR Template
Recent PRs & issues
Active · Last activity 18m ago
See all on GitHub →Summary: The existing range guard in the UITextView delegate (added in #24084) only checks range.location + range.length > text.length. It doesn't handle range.location > text.length, which causes an unsigned integer underflow in the clamped length, still crashing with NSRangeException. See #45050 for another report of this. The UITextField delegate has no range validation at all. This PR adds a range.location > textLength early return to both delegate adapters, ahead of the existing length clamping. Changelog: [IOS] [FIXED] - Fix NSRangeException crash in RCTBackedTextInputDelegateAdapter when text range is out of bounds Test Plan: This is a race condition between iOS computing the text range and React Native updating the backing text (e.g. controlled TextInput state updates, maxLength truncation, autocorrect). It's difficult to reproduce deterministically but shows up in production crash logs. The fix is straightforward defensive bounds checking before calling replaceCharactersInRange:.
Recent fixes
Summary: Adds to the Fantom custom implementation, matching Jest's inline snapshot behavior. On first run, snapshot values are written directly into the test source file as template literal arguments. On subsequent runs, values are compared and the test fails on mismatch unless the flag is passed to force an update. The implementation spans the two-process Fantom architecture: Runtime (Hermes VM): : New method that compares received values against existing inline snapshots, captures stack traces for new/mismatched snapshots, and tracks results per test. : New method on the class that serializes values with and delegates to snapshot context. : Plumbs through . Runner (Node.js): : New resolves VM stack traces back to original source locations via source map symbolication. New directly rewrites the test source file by finding each call and replacing its argument with the formatted template literal. This avoids jest-snapshot's which requires Babel and Prettier 2.x. : Generates source maps when inline snapshots need updating, collects pending snapshots across test configs, and writes them after all configs complete. Changelog: [Internal] Differential Revision: D95077617
andrewdacenko3h ago
Structured data for AI agents
Repository: facebook/react-native. Description: A framework for building native applications using React Stars: 125504, Forks: 25106. Primary language: C++. Languages: C++ (33.9%), Kotlin (19.6%), JavaScript (18.9%), Objective-C++ (11.2%), Objective-C (6.9%). License: MIT. Homepage: https://reactnative.dev Topics: android, app-framework, cross-platform, ios, mobile, mobile-development, react, react-native. Latest release: v0.84.1 (6d ago). Open PRs: 100, open issues: 1129. Last activity: 18m ago. Community health: 87%. Top contributors: mdvacca, shergin, cortinico, sammy-SC, javache, RSNara, cipolleschi, rubennorte, NickGerleman, JoshuaGross and others.