1 views collected around this technical thread.
The article demonstrates how detailed Mach‑O and LinkMap analysis combined with custom CocoaPods hooks, Swift‑syntax refactoring, and indexed symbol mapping can systematically shrink an iOS app’s bundle—from 289.3 MB to 259.3 MB—while cutting CI build time and simplifying debugging.
Mach‑O is the macOS and iOS binary format whose header, load commands, and __TEXT/__DATA sections define executable, library, and symbol information, enabling developers to understand loading, linking, and startup performance, and to implement custom initialization hooks or modular launch tasks via custom data sections.
The article explains how Baidu App reduces its iOS package size by analyzing Link Map files and applying six code‑optimization techniques—unused class and module slimming, unused method pruning with LLVM, duplicate‑code detection, utility‑method consolidation, and A/B‑test hardening—yielding up to 8 MB of savings.
The article explains how Baidu’s iOS app reduces package size by examining Mach‑O binaries with tools like otool and MachOView, then applying Python scripts to locate oversized assets, eliminate unused configuration files, and deduplicate resources, achieving a 12 MB reduction and a repeatable optimization pipeline.
This article presents practical methods for reducing iOS app bundle size by detecting and removing unused Objective‑C/Swift classes, redundant resources, and applying binary optimizations such as segment migration and link‑time optimization, illustrated with real‑world experience from 58.com.
This article introduces the Mach-O executable format, explains how its structure can be leveraged to attribute dynamic and static libraries at runtime and during build, and presents two practical projects—library attribution and fast API scanning—complete with implementation details and code snippets.
WBBlades3.0 is a visualized iOS development toolkit that offers unused class detection for Objective‑C and Swift, fast package‑size analysis of static libraries, point‑to‑point crash‑log parsing with symbol restoration, and static‑library dependency inspection, all designed to improve developer efficiency.
This article clarifies three frequent misunderstandings about iOS15+ dynamic linking—namely the encoding of b/bl instructions, the need for rebase/bind in __TEXT segments, and the mismatch between Mach‑O segment VM size and file size—by explaining the underlying ARM64 architecture, PIC concepts, and providing concrete dyld tool examples.
Fixup chain, introduced in iOS 15, replaces the previous compressed‑byte‑stream rebase/bind format with a linked‑list style structure that stores dynamic‑link information more compactly, improves space locality, reduces binary size, and speeds up app launch by allowing a single pass rebase and bind process.
This article explains how to detect and remove unused code in mixed Swift and Objective‑C iOS applications by parsing Mach‑O files, analyzing class metadata, and leveraging AccessFunction calls, while discussing the challenges, implementation details of the WBBlades tool, and practical usage steps.
This article explains the internal structure of Objective‑C categories, how they are represented as structs in the runtime, the process by which the Mach‑O loader discovers and attaches them, and the distinct execution order of their +load and +initialize methods.
This article explains how to use the otool command to parse Mach-O sections, extract __objc_classlist and __objc_classrefs, compute their differences, and locate unused Objective‑C classes and methods for iOS app size optimization, providing sample Objective‑C code and detailed implementation steps.
This article details ByteDance's iOS team approaches to reducing app package size—including resource compression, Assets.car merging, text file zipping, and Mach‑O binary optimizations such as -Oz, LTO, exported symbols, property dynamic handling, __TEXT segment migration, and binary segment compression—resulting in multi‑megabyte savings and improved download performance.
This article explains the fundamentals of hooking on iOS, covering Method Swizzle and Facebook's fishhook, detailing the Mach‑O file structure, PIC technique, and providing complete source code examples that demonstrate how to replace system functions such as NSLog at runtime.
This article examines Swift's binary representation in Mach-O files, compares it with Objective‑C storage, demonstrates runtime dynamic method calls, and analyzes class metadata, VTable construction, and related flags to reveal how Swift functions are located and invoked at runtime.
This article explains the principles of static and dynamic linking on iOS, details how symbol tables, relocation entries, and GOT/PLT work, and demonstrates a practical static‑library hooking technique using relocation tables and the fishhook library.
The 58 client package size detection live series reviewed two open‑source tools—Zucker for Android modular size analysis and a Mach‑O‑based performance suite for iOS—explaining their principles, implementation, and how they enable non‑intrusive monitoring and optimization of app package sizes.
WBBlades is an open‑source iOS toolset that parses Mach‑O files to provide size analysis, high‑precision unused‑class detection, and crash‑log symbolization without requiring symbol tables, offering both a macOS GUI and command‑line interface for developers to optimize and debug their apps.
This article describes how 58.com analyzes the code and resource size of each business component in its iOS app by parsing linkmap files and static libraries, builds version‑wise statistics, and compares component volume changes across releases to support efficient parallel development and app slimming.
This article explains why iOS crash logs are unreadable without symbolication, introduces three symbolication approaches—Xcode's symbolicatecrash, the atos/atosl tools, and dSYM‑based parsing—details the dSYM file structure, Mach‑O headers, load commands, data segments, and how to extract and demangle symbols for effective crash debugging.