Vector35 / Vector35/binaryninja-api
`pyright` type checking is very slow
还没有人认领这个 Issue。
- 主要语言
- C++
- 星标
- 1.3k
- 派生
- 298
- 平均合并
- 5 天 5 小时
- 30 天内合并 PR
- 19
描述
Version and Platform (required):
- Binary Ninja Version: 6.0.10601 (can't easily test dev for packaging reasons)
- Edition: Personal
- OS: NixOS
- OS Version: 26.05
- CPU Architecture: x64
Bug Description:
Importing and using binaryninja.interaction.show_graph_report causes pyright evaluation time to rise from 6s to ~40s.
Steps To Reproduce:
- Create
test.py - Paste:
from binaryninja.interaction import show_graph_report
show_graph_report(1, 2)
- Run
pyright --outputjson test.py - Observe
timeInSecto be vastly larger than normal (~40s for my system) - Comment out
show_graph_report(1, 2) - Rerun pyright and observe
timeInSecof below 1s
Expected Behavior:
Importing show_graph_report should not cause a drastic increase in pyright evaluation time.
Screenshots/Video Recording:
If applicable, please add screenshots/video recording here to help explain your problem.
Binary:
If applicable, please provide us with the binary to help us work with the issue faster. Here are a few options:
Additional Information:
I'm using this to get pyright to discover the api in my dev env:
BINJA_ROOT=$(cat "$HOME/.binaryninja/lastrun" | tr -d '\n')
export PYTHONPATH="$BINJA_ROOT/python:$PYTHONPATH"
贡献指南
这个仓库没有索引到贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
调研方向
Start by reproducing the issue with the test.py example and pyright --outputjson, comparing results with and without show_graph_report(1, 2). Trace the binaryninja.interaction.show_graph_report API and its exposed type information to identify what causes the evaluation slowdown. Done means importing and using it no longer causes the reported drastic increase in pyright time.
由索引模型根据 Issue 内容生成。
评估
- 技术栈
- python
- 领域
- developer-experience, performance
- Issue 类型
- 缺陷
- 难度
- 4/5
- 预计耗时
- 3-5 天
- 活跃度
- 活跃
- 描述清晰度
- 基本清楚
- 新手友好度
- 52/100