Unverified50% confidenceFactExact time
在大型Python项目中，静态调用图的边覆盖率通常仅在60%–80%之间，而Java、Go、Rust等强类型语言可接近95%以上
1
Sources
50%
Confidence
Long-term
Relevance
7/18/2026
First Seen
Sources
Related Claims
UnverifiedSWE-Bench Pro's programming languages are heavily concentrated in Python (36%), Go (38%), and TypeScript (19%).70% similarUnverifiedTongyi Lingma's code optimization is noticeably skewed toward Java and Go as well as the Alibaba Cloud tech stack, with weaker support for Python, Rust, and frontend technologies.67% similarUnverifiedThe Stack等开放代码数据集分析显示Python、JavaScript、Java占据公开代码总量的绝对多数，而Rust、Haskell、OCaml等语言的可用训练数据可能只有前者的1/10甚至1/10065% similarUnverified该数据集覆盖Python和JavaScript两种语言，比例大致为50/5064% similarUnverified完成同样功能，Python代码量通常只有Java、C++等其他语言的一半左右63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/552180API
curl https://kongchang.com/api/v1/knowledge/claims/552180MCP
get_claim(id=552180)