GitHub has introduced its updated code search engine called Blackbird, which is built using Rust programming language. It aims to provide a faster and more comprehensive exploration of software repositories. Blackbird has been in development for three years and is part of GitHub’s efforts to improve code search capabilities.
Developers can easily search, explore, and comprehend their code with Blackbird. It allows them to find important information, improve productivity, and efficiently work with their code. The new code search and view features are designed to enhance developers’ efficiency.
Previously, GitHub used Apache Solr and later Elasticsearch for code search. However, in 2020, they developed Blackbird from scratch using Rust because they couldn’t find an existing tool with the required capabilities. Blackbird performs significantly better than its predecessors, with the ability to handle up to 640 queries per second and index around 120,000 documents per second.
Blackbird offers various enhancements, including support for substring queries, regular expressions, and symbol search. It contextualizes code and prioritizes the most important results. The new search interface provides suggestions and completions, while the code view combines search, browsing, and code navigation.
It is simple to find certain text across repositories because of the improved code search engine’s ability to perform exact filtering. For example, it can help find values associated with a “memory” key in YAML configuration files for a Kubernetes cluster. It is also useful for identifying the specific part of an application that generated a particular error message.
GitHub’s main objective with Blackbird is to assist developers in finding crucial information scattered throughout their codebase, providing context, and ultimately boosting productivity.
