Coursera Sync

Download Coursera courses → Organize by category → Upload to Baidu Netdisk — fully automated.

---

Quick Start

# 1. Install
cd coursera-sync
pip3 install --break-system-packages -r requirements.txt coursera-helper
bash setup_baidu.sh

# 2. Add credentials to ~/.zshrc (one-time):
#    export CAUTH="..."    # from coursera.org cookies
#    export BDUSS="..."    # from pan.baidu.com cookies
#    export STOKEN="..."   # from pan.baidu.com cookies

# 3. Configure courses in course-data/courses.jsonl (JSON Lines)

# 4. Run
bash run.sh

📖 Full setup guide with credential extraction and troubleshooting: USERGUIDE.md

---

What It Does

run.sh → sync.py → coursera-helper → BaiduPCS-Go → cleanup

Step	Tool	Action
Download	coursera-helper v0.12.3 (Python 3.11)	Fetch videos, slides, quizzes, notebooks, subtitles to ./tmp/
Upload	BaiduPCS-Go v4.0.1	Week-by-week upload to /coursera/{category}/{course}/
Cleanup	sync.py	Remove local tmp after verified upload

Verified working (May 2026, macOS ARM64):

• ✅ Elasticsearch course: download + upload → 4TB Baidu Netdisk
• ✅ Information Theory course: download in progress

Baidu Netdisk result:

/coursera/
  Computer Science /
    Databases and Search /
      Executing Full Text Queries with Elasticsearch /
        01_introduction-to-elasticsearch/
          ... (videos, slides, quizzes)
    Theory /
      Information Theory /
        ... (in progress)

---

Configuration

Course catalog lives in course-data/courses.jsonl (JSON Lines):

{"slug": "machine-learning", "name": "Machine Learning", "category": "Computer Science / AI and Machine Learning", "enabled": true}

config.yaml contains global settings:

Setting	Default	Options
download.resolution	720p	360p / 540p / 720p
download.subtitle_languages	"en"	"en,zh-CN" or "" to skip
baidu.upload_policy	rsync	skip / overwrite / rsync
behavior.max_retries	3	Retry count on failure
behavior.force_redownload	false	Set true for interrupted downloads

⚠️ Category and course names containing &, #, ? break BaiduPCS-Go paths. Use and instead of & (auto-sanitized at upload time, with a warning at load).

---

Requirements

Component	Version	Install
Python	3.9–3.11	(coursera-helper needs distutils, removed in 3.12+)
coursera-helper	0.12.3	pip3 install coursera-helper
BaiduPCS-Go	≥4.0.0	bash setup_baidu.sh
pyyaml	≥6.0	pip3 install pyyaml

---

Project Structure

coursera-sync/
├── sync.py             Orchestrator (77 lines)
├── constants.py        Root path constants
├── config.py           YAML config loading + path resolution
├── courses.py          Course catalog, status/state persistence
├── baidu.py            BaiduPCS-Go login, upload, verify
├── downloader.py       coursera-helper download + tool detection
├── pipeline.py         Upload + process orchestration
├── utils.py            Subprocess runner + Baidu path sanitization
├── run.sh              Launcher (reads ~/.zshrc for credentials)
├── nohup_sync.sh       Background sync with PID lock + log
├── setup_baidu.sh      BaiduPCS-Go installer
├── requirements.txt    pyyaml, requests
├── README.md           This file
├── USERGUIDE.md        Full guide + gotchas
├── config.yaml         Settings (gitignored)
├── course-data/        Course catalog + state + syllabus JSONs
├── bin/                BaiduPCS-Go binary
├── logs/               Log output + upload_state.json
└── tmp/                Temp downloads

License

MIT