A knowledge-sharing platform for Sohu's technology products. As a leading Chinese internet brand with media, video, search, and gaming services and over 700 million users, Sohu continuously drives tech innovation and practice. We’ll share practical insights and tech news here.
This article explores the technical architecture of Android dynamic photos, detailing the three‑layer file structure, XMP metadata specifications, and differences among Xiaomi Micro Video, Google Motion Photo, and OPPO O Live Photo, and provides a unified Kotlin solution for detection, parsing, and playback.
PixelPrompt is a Dart-based terminal UI (TUI) framework that brings Flutter‑style declarative UI, component hierarchy, layout engines, and efficient ANSI‑based rendering to the console, using character cells, double buffering, and diff rendering to create interactive, styled terminal applications.
An open-source, Kaggle‑ranked solution for the Mitsui Commodity Prediction Challenge details data preprocessing, feature engineering, and multiple tree‑based modeling strategies—including multi‑target, single‑target, and price‑difference models—with code, evaluation metrics, and suggestions for further improvements.
This article compares MySQL’s SELECT DISTINCT and GROUP BY clauses, explaining their syntax, functional differences, performance implications, and ideal scenarios through detailed examples, index usage analysis, and a feature comparison table, helping developers choose the right approach for deduplication or aggregation tasks.
This tutorial introduces Claude Code’s Model Context Protocol (MCP), compares it with traditional function calling, explains its client‑server architecture, and provides a step‑by‑step hands‑on guide—including Figma token acquisition, server setup, and Flutter code generation—to demonstrate how MCP streamlines AI‑driven development.
This article compares three popular Python OCR frameworks—pytesseract, cnocr, and PaddleOCR—examining their installation ease, Chinese recognition ability, model size, accuracy, and unique features, and provides practical code examples to help developers pick the best tool for their needs.
Riverpod 3.0 introduces automatic retry for failing providers, built‑in pause/resume of listeners, experimental offline caching and mutation support, a unified API that merges AutoDispose and legacy providers, standardized update filtering with ==, and revamped ProviderObserver, all aimed at simplifying state management in Flutter apps.
This article walks through the performance bottlenecks of using Lottie for dual‑state animations in mobile apps and presents a three‑stage progressive optimization—basic synchronous implementation, asynchronous loading with caching, and a state‑machine‑based pending mechanism—complete with Swift code samples, diagrams, and benchmark results.
This guide explains how to communicate effectively with the Cursor AI coding assistant by crafting high‑quality prompts, structuring rules, and using multi‑turn dialogues, enabling developers to boost productivity, maintain code quality, and turn Cursor into a reliable programming partner.
This article explains the GRPO algorithm, an improvement over PPO for large language model training that eliminates the value network, uses group‑relative advantage estimation, and offers flexible supervision, resulting in higher efficiency, stability, and performance on tasks such as mathematical reasoning.
