Official account of the WeChat mobile client development team, sharing development experience, cutting‑edge tech, and little‑known stories across Android, iOS, macOS, Windows Phone, and Windows.
WeChat’s Mars library, a C++‑based cross‑platform network component used across Android, iOS, macOS, Windows and more, offers optimized signaling (STN), logging (XLOG), diagnostics (SDT) and core utilities, enabling reliable mobile connections and simplifying development, and will soon be open‑sourced.
libco is a high‑performance C/C++ coroutine library open‑sourced by Tencent, enabling WeChat’s backend to transform synchronous code into non‑intrusive asynchronous execution, achieve massive concurrency, support shared‑stack coroutines, hook network APIs, and simplify variable handling for million‑scale connections.
This article examines the caching mechanisms of Android's RecyclerView and ListView, comparing their hierarchical caches, view holder strategies, and partial refresh capabilities, and concludes when each component is preferable based on performance, animation support, and data update frequency.
This article examines the design of Mars's cross‑platform xlog module, explaining why traditional file‑based logging harms app smoothness, how memory buffering, mmap, and stream‑compression overcome performance and reliability issues, and what practical challenges remain on Android and iOS.
This article recounts WeChat’s two‑year journey developing the open‑source Tinker hot‑patch framework for Android, detailing the challenges of native vs Java approaches, performance‑critical DexDiff algorithms, platform‑specific issues on Dalvik, ART, and Android N, and the solutions that achieved a high‑availability, low‑overhead patch system.
This article explains how WeChat's iOS client tackled SQLite performance bottlenecks by introducing multi‑handle threading, enabling WAL mode, redesigning the Busy‑Retry mechanism, and applying I/O enhancements such as preserving WAL size and mmap usage, resulting in over 90% reduction in perceived lag.
This article analyzes why Android N’s hybrid AOT/interpret/JIT compilation prevents Tinker hot‑patches from working, explains the underlying ART mechanisms such as profile files and app images, and proposes practical strategies to restore patchability on modern Android devices.
This article explains how to use Android Studio’s Allocation Tracker, automate its start/stop via code using the ddmlib library, and directly access Dalvik’s internal allocation records to retrieve object size, type, and stack information with minimal performance impact, while discussing limitations and compatibility concerns.
This guide explains why high‑DPI screens blur Windows applications, introduces core DPI concepts, and provides step‑by‑step instructions—including manifest declaration, scaling factor calculation, resource handling, and UI testing—to ensure crisp, correctly sized interfaces across all DPI settings.
This article reviews the evolution of Android hot‑patch technologies, compares major solutions such as Dexposed, AndFix, Qzone, and WeChat’s own Tinker, and explains when and how to apply hot patches for lightweight updates, remote debugging, and data‑driven testing.
