feat: Refactor project structure and enhance functionality

.gitignore

@@ -43,3 +43,5 @@ urls.txt
 # OS specific
 .DS_Store
 Thumbs.db
+temp_*/
+logs/

README.md

@@ -45,13 +45,17 @@
 ![Version](https://img.shields.io/badge/version-1.0.0-blue.svg)
 ![Python](https://img.shields.io/badge/python-3.8+-green.svg)
 
-## 👤 作者信息
+## Huggingface space: [rednote-gen](https://huggingface.co/spaces/windane/rednote-gen)
+PS: 最好还是本地运行，Space 直接下载视频失败率高，可以使用MD文件解析。
+
+## 👤 原作者信息
 
 - **作者**：玄清
 - **博客**：[天天悦读](https://blog.example.com)
 - **Email**：grow8org@gmail.com
 - **GitHub**：[whotto/Video_note_generator](https://github.com/whotto/Video_note_generator)
 
+
 ## 🎯 应用场景
 
 - **内容创作者**：快速将视频/直播内容转换为文章
@@ -96,24 +100,12 @@ graph TD
 
 ## 🚀 使用方式
 
-支持三种使用方式：
-
-1. **处理单个视频**：
-```bash
-python video_note_generator.py https://example.com/video
-```
-
-2. **批量处理 URL 文件**：
-```bash
-# urls.txt 文件，每行一个视频链接
-python video_note_generator.py urls.txt
+启动 WebUI:
+```shell
+$ python web.py
 ```
 
-3. **处理 Markdown 文件**：
-```bash
-# 支持 Markdown 链接和直接 URL
-python video_note_generator.py notes.md
-```
+打开：http://127.0.0.1:7860
 
 ## 🛠️ 使用工具
 
@@ -134,40 +126,13 @@ python video_note_generator.py notes.md
 
 # 安装 Python 依赖
 pip install -r requirements.txt
-
-# 配置环境变量
-cp .env.example .env
-```
-
-### 2. 配置 API 密钥
-
-编辑 `.env` 文件，填入必要的 API 密钥：
-```ini
-# OpenRouter API（必需）
-OPENROUTER_API_KEY=your-api-key-here
-
-# Unsplash API（必需）
-UNSPLASH_ACCESS_KEY=your-unsplash-access-key-here
-UNSPLASH_SECRET_KEY=your-unsplash-secret-key-here
-```
-
-### 3. 开始使用
-
-1. 创建 `urls.txt` 文件，每行一个视频链接
-2. 运行环境检查：
-```bash
-python check_environment.py
-```
-3. 运行生成器：
-```bash
-python video_note_generator.py test.md
 ```
 
 ## 📄 输出文件
 
 每个视频会生成三个文件：
 
-1. **原始笔记** (`YYYYMMDD_HHMMSS.md`)：
+1. **原始笔记** ：
    - 完整的视频转录文本
    - 保留所有细节内容
 
@@ -185,14 +150,18 @@ python video_note_generator.py test.md
 
 ## ⚙️ 配置说明
 
-在 `.env` 文件中可以调整以下参数：
+在设置页面填写必要的 API Key：
+- OpenRouter API（必需）
+- Unsplash API （建议）
+
+还有其他设置：
 
-```ini
-# 内容生成配置
+```
 MAX_TOKENS=2000          # 生成小红书内容的最大长度
 CONTENT_CHUNK_SIZE=2000  # 长文本分块大小（字符数）
 TEMPERATURE=0.7          # AI 创造性程度 (0.0-1.0)
 
+
 # 代理设置（可选）
 # HTTP_PROXY=http://127.0.0.1:7890
 # HTTPS_PROXY=http://127.0.0.1:7890

requirements.txt

@@ -1,17 +1,12 @@
-openai>=1.0.0
-httpx>=0.24.1
-yt-dlp>=2023.11.16
-openai-whisper>=2023.11.17
-python-dotenv>=1.0.0
-requests>=2.31.0
-beautifulsoup4>=4.12.2
-python-unsplash>=1.1.0
-Pillow>=10.1.0
-urllib3>=2.1.0
-certifi>=2023.11.17
-ffmpeg-python>=0.2.0
-
-# Optional dependencies for better performance
-torch>=2.1.0
-pytube>=15.0.0
-you-get>=0.4.1650
+beautifulsoup4==4.12.3
+gradio==5.9.1
+gradio_modal==0.0.4
+httpx==0.28.1
+openai==1.58.1
+openai_whisper==20240930
+python-dotenv==1.0.1
+pytube==15.0.0
+setuptools==75.1.0
+yt_dlp==2024.12.23
+python-unsplash==1.2.5
+moviepy~=2.1.1

src/adapter/ffmpeg.py

@@ -0,0 +1,24 @@
+import subprocess
+
+from src.logger import app_logger
+
+
+class FfmpegAdapter:
+    def __init__(self):
+        ffmpeg_path = None
+        try:
+            subprocess.run(["/opt/homebrew/bin/ffmpeg", "-version"],
+                            stdout=subprocess.PIPE,
+                            stderr=subprocess.PIPE)
+            app_logger.info("✅ ffmpeg is available at /opt/homebrew/bin/ffmpeg")
+            ffmpeg_path = "/opt/homebrew/bin/ffmpeg"
+        except Exception:
+            try:
+                subprocess.run(["ffmpeg", "-version"],
+                                stdout=subprocess.PIPE,
+                                stderr=subprocess.PIPE)
+                app_logger.info("✅ ffmpeg is available (from PATH)")
+                ffmpeg_path = "ffmpeg"
+            except Exception as e:
+                app_logger.warning(f"⚠️ ffmpeg not found: {str(e)}")
+        self.ffmpeg_path = ffmpeg_path

src/adapter/openrouter.py

@@ -0,0 +1,78 @@
+import asyncio
+from typing import Optional
+
+import openai
+
+from src.environment.env import Environment
+from src.logger import app_logger
+from src.video.prompt import share_prompt
+
+
+class OpenRouterAdapter:
+    def __init__(self, env: Environment):
+        self.client = openai.OpenAI(
+            api_key=env.openrouter_api_key,
+            base_url=env.openrouter_api_url or 'https://openrouter.ai/api/v1',
+            default_headers={
+                "HTTP-Referer": env.openrouter_http_referer,
+                "X-Title": env.openrouter_app_name
+            }
+        )
+        self.ai_model = env.openrouter_ai_model
+        self.api_available = False
+        # Initialize the wait queue and flow limiter
+        self.wait_queue = asyncio.Queue()
+        self.flow_limiter = asyncio.Semaphore(5)  # Allow 5 requests per minute
+
+    def connect(self):
+        if self.client.api_key:
+            try:
+                app_logger.info(f"正在测试 OpenRouter API 连接...")
+                response = self.client.models.list()
+                app_logger.info("✅ OpenRouter API 连接测试成功")
+                self.api_available = True
+            except Exception as e:
+                app_logger.error(f"❌ OpenRouter API 连接测试失败: {e}")
+                app_logger.error("将继续尝试使用API，但可能会遇到问题")
+                self.api_available = False
+
+    async def generate(self, system_prompt_type, user_prompt_type, content,
+                 temperature=0.7, max_tokens=4000) -> Optional[str]:
+        if not self.api_available:
+            app_logger.error("OpenRouter API 不可用，无法生成")
+            return None
+
+        # Wait for the flow limiter to allow the request
+        await self.flow_limiter.acquire()
+
+        try:
+            system_prompt = share_prompt(prompt_type=system_prompt_type, content=content)
+            user_prompt = share_prompt(prompt_type=user_prompt_type, content=content) or content
+            app_logger.info('OpenRouter API 开始请求')
+
+            # Add the request to the wait queue
+            await self.wait_queue.put(None)
+
+            # Wait for the previous requests to complete
+            await asyncio.sleep(60 / 5)  # 5 requests per minute
+
+            response = self.client.chat.completions.create(
+                model=self.ai_model,
+                messages=[
+                    {"role": "system", "content": system_prompt},
+                    {"role": "user", "content": user_prompt}
+                ],
+                temperature=temperature,
+                max_tokens=max_tokens
+            )
+            if not response.choices:
+                app_logger.error("OpenRouter API 返回结果为空")
+                return None
+            return response.choices[0].message.content
+        except Exception as e:
+            app_logger.error(f"OpenRouter API 请求失败: {e}")
+            return None
+        finally:
+            # Release the flow limiter
+            self.flow_limiter.release()
+            await self.wait_queue.get()

src/adapter/unsplash.py

@@ -0,0 +1,99 @@
+from typing import List
+
+import httpx
+
+from src.environment.env import Environment
+from unsplash.api import Api as UnsplashApi
+from unsplash.auth import Auth as UnsplashAuth
+
+from src.logger import app_logger
+
+
+class UnsplashAdapter:
+    def __init__(self, env: Environment):
+        self.unsplash_client = None
+        self.unsplash_available = False
+        self.unsplash_access_key = env.unsplash_access_key
+
+        if env.unsplash_access_key:
+            auth = UnsplashAuth(
+                client_id=env.unsplash_access_key,
+                client_secret=None,
+                redirect_uri=None
+            )
+            try:
+                unsplash_client = UnsplashApi(auth)
+                self.unsplash_client = unsplash_client
+                self.unsplash_available = True
+                app_logger.info("✅ Unsplash API 配置成功")
+            except Exception as e:
+                app_logger.error(f"❌ Failed to initialize Unsplash client: {str(e)}")
+                self.unsplash_available = False
+        else:
+            app_logger.warning("⚠️ 未设置 Unsplash API 密钥")
+            self.unsplash_available = False
+
+    def get_images(self, query: str, count: int = 3) -> List[str]:
+        if not self.unsplash_available:
+            app_logger.warning("⚠️ Unsplash API 不可用")
+            return []
+
+        try:
+            headers = {
+                'Authorization': f'Client-ID {self.unsplash_access_key}'
+            }
+            # Query each keyword
+            all_photos = []
+            for keyword in query.split(','):
+                response = httpx.get(
+                    'https://api.unsplash.com/search/photos',
+                    params={
+                        'query': keyword.strip(),
+                        'per_page': count,
+                        'orientation': 'portrait',  # 小红书偏好竖版图片
+                        'content_filter': 'high'  # 只返回高质量图片
+                    },
+                    headers=headers,
+                    verify=False  # 禁用SSL验证
+                )
+
+                if response.status_code == 200:
+                    data = response.json()
+                    if data['results']:
+                        # 获取图片URL，优先使用regular尺寸
+                        photos = [photo['urls'].get('regular', photo['urls']['small'])
+                                  for photo in data['results']]
+                        all_photos.extend(photos)
+
+            # 如果收集到的图片不够，用最后一个关键词继续搜索
+            while len(all_photos) < count and query:
+                response = httpx.get(
+                    'https://api.unsplash.com/search/photos',
+                    params={
+                        'query': query.split(',')[-1].strip(),
+                        'per_page': count - len(all_photos),
+                        'orientation': 'portrait',
+                        'content_filter': 'high',
+                        'page': 2  # 获取下一页的结果
+                    },
+                    headers=headers,
+                    verify=False
+                )
+
+                if response.status_code == 200:
+                    data = response.json()
+                    if data['results']:
+                        photos = [photo['urls'].get('regular', photo['urls']['small'])
+                                  for photo in data['results']]
+                        all_photos.extend(photos)
+                    else:
+                        break
+                else:
+                    break
+
+            # 返回指定数量的图片
+            return all_photos[:count]
+
+        except Exception as e:
+            app_logger.error(f"⚠️ 获取图片失败: {str(e)}")
+            return []