from flask import Flask, render_template, Response, request from flask_flatpages import FlatPages from werkzeug.middleware.proxy_fix import ProxyFix import requests app = Flask(__name__) app.wsgi_app = ProxyFix(app.wsgi_app, x_proto=1, x_host=1) app.config.update( FLATPAGES_AUTO_RELOAD = True, FLATPAGES_EXTENSION = '.md', FLATPAGES_MARKDOWN_EXTENSIONS = ['fenced_code', 'tables'] ) pages = FlatPages(app) @app.route('/') def index(): # Sort posts by date metadata posts = sorted(pages, key=lambda p: p.meta.get('date'), reverse=True) return render_template('index.html', posts=posts) @app.route('/about') def about(): return render_template('about.html') @app.route('/proxy-chat', methods=['POST']) def proxy_chat(): target_url = "http://192.168.0.37:5002/v1/chat/completions" try: # We use stream=True so we don't load the whole response into RAM at once response = requests.post( target_url, json=request.json, timeout=300, stream=True ) # Generator to yield chunks of data as they arrive def generate(): for chunk in response.iter_content(chunk_size=1024): yield chunk return Response( generate(), status=response.status_code, content_type=response.headers.get('content-type', 'application/json') ) except requests.exceptions.Timeout: return {"error": "The backend LLM timed out."}, 504 except Exception as e: app.logger.error(f"Proxy error: {str(e)}") return {"error": "Internal server error"}, 500 @app.route('/post//') def post(path): page = pages.get_or_404(path) return render_template('post.html', page=page) @app.route('/tag//') def tag(tag_name): tagged_pages = [p for p in pages if tag_name in p.meta.get('tags', [])] return render_template('tag.html', pages=tagged_pages, tag_name=tag_name) if __name__ == "__main__": app.run(host='0.0.0.0', port=5001, debug=True)