sduoj-autolab

SDUOJ-based pluggable components for in-depth support of course needs, supporting subjective questions, time-limited exams, series of question sets and other functions. Use FastAPI as the backend.

部署方案

1. 建立数据库:

sudo docker-compose -f setup.yaml up
# ctrl + c
sudo docker-compose -f setup.yaml start

redis 的密码设置不上，还需要再修正一下。

sudo docker exec -it id sh
redis-cli
config set requirepass xxx

2. 创建虚拟环境：

本地：

pip freeze > requirements.txt

部署：

conda create --name oj_problem_set python=3.9
conda activate oj_problem_set
python -m pip install -r requirements.txt
#运行 model 中的文件，建立表。
python db.py
mkdir logs
python -m gunicorn -c gunicorn.conf.py main:app

3. 运维

ps -ef |grep main:app |awk '{print $2}'| xargs kill -9
python -m uvicorn main:app --reload --host=0.0.0.0 --port 5005

鸣谢

2024年7月1日至9月1日，SDUOJ团队采用Pengyu Xue, Linhao Wu等人开发的ERICommiter进行提交信息生成，特此致谢。

Acknowledgment

We acknowledge the utilization of ERICommiter, developed by Pengyu Xue, Linhao Wu et al., for commit message generation by the SDUOJ team from July 1 to September 1, 2024. We express our gratitude for this contribution.