caddy 簡單反代谷歌學術

README.md

重點

1. 要反代如 https://scholar.google.com.hk 等的地區域名，而不是 https://scholar.google.com
2. 需將客戶端原始 IP 回傳給 Google，否則會 404
3. 不要用 http.cache，以免被當爬蟲被封 IP

Caddyfile

scholar.example.com {
    timeouts 5m
    proxy / https://scholar.google.com.hk {
        except /robots.txt /usercontent
        header_upstream X-Real-IP {remote}
	header_upstream X-Forwarded-For {remote}
        header_upstream User-Agent {>User-Agent}
        # header_upstream X-Real-IP {>CF-Connecting-IP}
        # header_upstream X-Forwarded-For {>CF-Connecting-IP}
        header_upstream Accept-Language zh-HK
        header_upstream Accept-Encoding identity
        insecure_skip_verify
    }
    proxy /usercontent https://scholar.googleusercontent.com {
	without /usercontent
        except /robots.txt
        header_upstream X-Real-IP {remote}
	header_upstream X-Forwarded-For {remote}
        header_upstream User-Agent {>User-Agent}
        # header_upstream X-Real-IP {>CF-Connecting-IP}
        # header_upstream X-Forwarded-For {>CF-Connecting-IP}
        header_upstream Accept-Language zh-HK
        header_upstream Accept-Encoding identity
        insecure_skip_verify
    }
    filter rule {
        content_type text/.*
        search_pattern scholar\.google\.com
        replacement scholar.example.com
    }
    filter rule {
        content_type text/.*
        search_pattern scholar\.googleusercontent\.com
        replacement scholar.example.com/usercontent
    }
    tls {
        dns cloudflare
    }
}