mirror of
https://github.com/Shineii86/MiruroAPI.git
synced 2026-10-22 10:15:10 +00:00
docs: add bypass solutions for community contributions
Added Contributing Bypass Solutions section to README with: - Free/open source tools (nodriver, Camoufox, curl_cffi, FlareSolverr, etc.) - Paid APIs (Scrape.do 1K free, ZenRows, ScrapingBee) - Implementation guide for contributors - .env.example updated with SCRAPE_DO_TOKEN Encourages community to implement streaming bypass methods.
This commit is contained in:
@@ -27,3 +27,8 @@ SCRAPER_API_KEY=
|
||||
# Deploy on Render.com free tier, Oracle Cloud, or any VPS
|
||||
# Solves Cloudflare Turnstile challenges automatically
|
||||
FLARESOLVERR_URL=
|
||||
|
||||
# Method 4: Scrape.do (1,000 free credits/month)
|
||||
# Sign up at https://scrape.do/ (no credit card required)
|
||||
# Handles Cloudflare, proxy rotation, CAPTCHA solving
|
||||
SCRAPE_DO_TOKEN=
|
||||
|
||||
@@ -1565,6 +1565,49 @@ Expected response when streaming works:
|
||||
}
|
||||
```
|
||||
|
||||
### 🤝 Contributing Bypass Solutions
|
||||
|
||||
Streaming is blocked on all cloud platforms. We need help! If you can implement any of these, PRs are welcome:
|
||||
|
||||
#### Free / Open Source (Best for self-hosted)
|
||||
|
||||
| Tool | Language | How It Works | Notes |
|
||||
|------|----------|-------------|-------|
|
||||
| [nodriver](https://github.com/ultrafunkamsterdam/nodriver) | Python | CDP-based browser automation, bypasses automation detection | Successor to undetected-chromedriver |
|
||||
| [Camoufox](https://github.com/nicandris/camoufox) | Python | Stealth Firefox with fingerprint spoofing | 82% baseline Turnstile bypass |
|
||||
| [curl_cffi](https://github.com/nicandris/curl_cffi) | Python | TLS fingerprint impersonation | Lightweight, no browser needed |
|
||||
| [FlareSolverr](https://github.com/FlareSolverr/FlareSolverr) | Docker | Headless Chrome proxy server | Can't solve interactive CAPTCHAs |
|
||||
| [patchright](https://github.com/nicandris/patchright) | Python | Patched Playwright fork | Drop-in Playwright replacement |
|
||||
| [scrapy-impersonate](https://github.com/driver-utils/scrapy-impersonate) | Python | Scrapy with TLS impersonation | For Scrapy users |
|
||||
|
||||
#### Paid APIs (Good for cloud deployments)
|
||||
|
||||
| Service | Free Tier | Price | API Example |
|
||||
|---------|-----------|-------|-------------|
|
||||
| [Scrape.do](https://scrape.do) | 1,000 credits/mo | $0.08/1K | `https://api.scrape.do/?url=...&token=...` |
|
||||
| [ZenRows](https://zenrows.com) | Trial | $49/mo | `https://api.zenrows.com/?url=...&apikey=...` |
|
||||
| [ScrapingBee](https://scrapingbee.com) | Trial | $49/mo | `https://app.scrapingbee.com/api/v1/?...` |
|
||||
|
||||
#### What to Implement
|
||||
|
||||
1. **Node.js integration** — Add `nodriver` or `Camoufox` as a pipe fallback in `src/helpers/pipe.js`
|
||||
2. **Scrape.do adapter** — Simple HTTP proxy, 1K free credits/mo, no browser needed
|
||||
3. **Playwright stealth mode** — Use `patchright` or `playwright-extra-stealth` for browser-based bypass
|
||||
4. **Residential proxy support** — Add proxy rotation for pipe requests
|
||||
|
||||
#### How to Contribute
|
||||
|
||||
1. Fork the repo
|
||||
2. Add your bypass method as a new fallback in `src/helpers/pipe.js`
|
||||
3. Add your API key to `.env.example`
|
||||
4. Test with: `curl http://localhost:3000/api/episodes/20`
|
||||
5. Submit a PR
|
||||
|
||||
```bash
|
||||
# Example: Add Scrape.do as a fallback
|
||||
SCRAPE_DO_TOKEN=your_token_here
|
||||
```
|
||||
|
||||
### 🐛 Debug Mode
|
||||
|
||||
```bash
|
||||
|
||||
Reference in New Issue
Block a user