Spaces:

deeme
/

comic

Paused

App Files Files Community

deeme commited on Apr 3, 2025

Commit

c0372f3

verified ·

1 Parent(s): 400e08a

Upload 4 files

Browse files

Files changed (4) hide show

Dockerfile +25 -10
README.md +3 -4
app.py +306 -0
requirements.txt +7 -0

Dockerfile CHANGED Viewed

@@ -1,10 +1,25 @@
-FROM node:20
-# 安装 http-server
-RUN npm install -g http-server
-# 创建一个简单的 HTML 文件来实现跳转
-RUN echo '<!DOCTYPE html><html><head><meta http-equiv="refresh" content="0; url=https://comic.168369.xyz/"></head><body></body></html>' > /index.html
-# 启动 http-server 以提供该文件
-CMD ["http-server", "-p", "3000", "-c-1"]

+FROM python:3.9-slim
+WORKDIR /app
+# 安装FFmpeg
+RUN apt-get update && \
+    apt-get install -y ffmpeg && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+# 安装Python依赖
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+# 复制应用代码
+COPY . .
+# 创建必要的目录
+RUN mkdir -p temp
+# 暴露端口
+EXPOSE 8000
+# 运行应用
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "8000"]

README.md CHANGED Viewed

@@ -1,10 +1,9 @@
 ---
-title: Comic
 emoji: 👩‍🎨
 colorFrom: red
 colorTo: yellow
 sdk: docker
 pinned: false
-app_port: 3000
----

 ---
+title: comic
 emoji: 👩‍🎨
 colorFrom: red
 colorTo: yellow
 sdk: docker
 pinned: false
+app_port: 8000
+---

app.py ADDED Viewed

	@@ -0,0 +1,306 @@

+from fastapi import FastAPI, HTTPException, BackgroundTasks
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+from typing import List
+import os
+import uuid
+import aiohttp
+import asyncio
+import logging
+import tempfile
+import openai
+from pathlib import Path
+import webdav3.client as wc
+import subprocess
+import shutil
+# 配置日志
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+# 环境变量
+OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
+OPENAI_BASE_URL = os.getenv("OPENAI_BASE_URL", "https://api.openai.com/v1")
+WEBDAV_URL = os.getenv("WEBDAV_URL")
+WEBDAV_USERNAME = os.getenv("WEBDAV_USERNAME")
+WEBDAV_PASSWORD = os.getenv("WEBDAV_PASSWORD")
+# 初始化OpenAI
+openai.api_key = OPENAI_API_KEY
+if OPENAI_BASE_URL:
+    openai.api_base = OPENAI_BASE_URL
+app = FastAPI()
+# 配置CORS
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # 生产环境中应该限制来源
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+# 请求模型
+class ComicData(BaseModel):
+    captions: List[str]
+    speeches: List[str]
+    panels: List[str]  # 图片URLs
+# WebDAV客户端配置
+def get_webdav_client():
+    options = {
+        'webdav_hostname': WEBDAV_URL,
+        'webdav_login': WEBDAV_USERNAME,
+        'webdav_password': WEBDAV_PASSWORD
+    }
+    return wc.Client(options)
+# 下载图片
+async def download_image(session, url, output_path):
+    try:
+        async with session.get(url) as response:
+            if response.status == 200:
+                with open(output_path, 'wb') as f:
+                    f.write(await response.read())
+                return output_path
+            else:
+                logger.error(f"Failed to download image: {response.status}")
+                return None
+    except Exception as e:
+        logger.error(f"Error downloading image: {e}")
+        return None
+# 生成语音
+async def generate_speech(text, voice="alloy", output_path=None):
+    try:
+        if not output_path:
+            output_path = f"{uuid.uuid4()}.mp3"
+        response = await openai.audio.speech.create(
+            model="tts-1",
+            voice=voice,
+            input=text
+        )
+        response.stream_to_file(output_path)
+        return output_path
+    except Exception as e:
+        logger.error(f"Error generating speech: {e}")
+        return None
+# 创建视频
+def create_video(project_dir, image_paths, subtitle_file, audio_file, output_video):
+    try:
+        # 创建帧列表文件
+        frames_list = os.path.join(project_dir, "frames.txt")
+        with open(frames_list, "w") as f:
+            for img in image_paths:
+                # 每个图片显示5秒
+                f.write(f"file '{img}'\n")
+                f.write(f"duration 5\n")
+            # 最后一张图片需要单独添加，否则会被忽略
+            f.write(f"file '{image_paths[-1]}'\n")
+        # 使用FFmpeg创建视频
+        cmd = [
+            "ffmpeg", "-y",
+            "-f", "concat", "-safe", "0", "-i", frames_list,
+            "-i", audio_file,
+            "-vf", f"subtitles={subtitle_file}",
+            "-c:v", "libx264", "-pix_fmt", "yuv420p",
+            "-c:a", "aac", "-strict", "experimental",
+            output_video
+        ]
+        subprocess.run(cmd, check=True)
+        return output_video
+    except Exception as e:
+        logger.error(f"Error creating video: {e}")
+        return None
+# 创建字幕文件
+def create_subtitle_file(project_dir, captions, speeches):
+    try:
+        subtitle_file = os.path.join(project_dir, "subtitles.srt")
+        with open(subtitle_file, "w", encoding="utf-8") as f:
+            subtitle_index = 1
+            current_time = 0
+            # 处理每个面板的字幕和对话
+            for i, (caption, speech) in enumerate(zip(captions, speeches)):
+                # 每个面板展示5秒
+                panel_duration = 5
+                # 字幕开始和结束时间
+                start_time = current_time
+                # 字幕显示
+                if caption:
+                    end_time = start_time + 2.5
+                    f.write(f"{subtitle_index}\n")
+                    f.write(f"{format_time(start_time)} --> {format_time(end_time)}\n")
+                    f.write(f"{caption}\n\n")
+                    subtitle_index += 1
+                # 对话显示
+                if speech:
+                    speech_start = start_time + 2.5 if caption else start_time
+                    speech_end = current_time + panel_duration
+                    f.write(f"{subtitle_index}\n")
+                    f.write(f"{format_time(speech_start)} --> {format_time(speech_end)}\n")
+                    f.write(f"{speech}\n\n")
+                    subtitle_index += 1
+                current_time += panel_duration
+        return subtitle_file
+    except Exception as e:
+        logger.error(f"Error creating subtitle file: {e}")
+        return None
+# 格式化时间为SRT格式
+def format_time(seconds):
+    hours = int(seconds / 3600)
+    minutes = int((seconds % 3600) / 60)
+    secs = int(seconds % 60)
+    millisecs = int((seconds - int(seconds)) * 1000)
+    return f"{hours:02}:{minutes:02}:{secs:02},{millisecs:03}"
+# 创建音频文件
+async def create_audio_file(project_dir, captions, speeches):
+    try:
+        audio_parts = []
+        current_time = 0
+        # 为每个面板生成音频
+        for i, (caption, speech) in enumerate(zip(captions, speeches)):
+            # 每个面板的旁白
+            if caption:
+                caption_audio = os.path.join(project_dir, f"caption_{i}.mp3")
+                await generate_speech(caption, "alloy", caption_audio)
+                audio_parts.append(caption_audio)
+            # 每个面板的对话
+            if speech:
+                speech_audio = os.path.join(project_dir, f"speech_{i}.mp3")
+                await generate_speech(speech, "echo", speech_audio)
+                audio_parts.append(speech_audio)
+        # 合并所有音频部分
+        combined_audio = os.path.join(project_dir, "combined_audio.mp3")
+        # 使用FFmpeg合并音频
+        audio_list = os.path.join(project_dir, "audio_list.txt")
+        with open(audio_list, "w") as f:
+            for audio in audio_parts:
+                f.write(f"file '{audio}'\n")
+        subprocess.run([
+            "ffmpeg", "-y", "-f", "concat", "-safe", "0",
+            "-i", audio_list, "-c", "copy", combined_audio
+        ], check=True)
+        return combined_audio
+    except Exception as e:
+        logger.error(f"Error creating audio file: {e}")
+        return None
+# 上传到WebDAV
+def upload_to_webdav(local_path, remote_path):
+    try:
+        client = get_webdav_client()
+        # 确保远程目录存在
+        remote_dir = os.path.dirname(remote_path)
+        if not client.check(remote_dir):
+            client.mkdir(remote_dir)
+        # 上传文件
+        client.upload_sync(local_path=local_path, remote_path=remote_path)
+        # 获取公共URL
+        return f"{WEBDAV_URL}/{remote_path}"
+    except Exception as e:
+        logger.error(f"Error uploading to WebDAV: {e}")
+        return None
+@app.post("/api/generate-video")
+async def generate_video(comic_data: ComicData, background_tasks: BackgroundTasks):
+    # 创建唯一项目ID
+    project_id = str(uuid.uuid4())
+    project_dir = f"temp/{project_id}"
+    os.makedirs(project_dir, exist_ok=True)
+    try:
+        # 下载图片
+        image_paths = []
+        async with aiohttp.ClientSession() as session:
+            download_tasks = []
+            for i, panel_url in enumerate(comic_data.panels):
+                output_path = os.path.join(project_dir, f"panel_{i}.jpg")
+                download_tasks.append(download_image(session, panel_url, output_path))
+            image_paths = await asyncio.gather(*download_tasks)
+            image_paths = [p for p in image_paths if p]  # 过滤失败的下载
+        if not image_paths:
+            raise HTTPException(status_code=500, detail="Failed to download images")
+        # 创建字幕文件
+        subtitle_file = create_subtitle_file(project_dir, comic_data.captions, comic_data.speeches)
+        if not subtitle_file:
+            raise HTTPException(status_code=500, detail="Failed to create subtitle file")
+        # 创建音频文件
+        audio_file = await create_audio_file(project_dir, comic_data.captions, comic_data.speeches)
+        if not audio_file:
+            raise HTTPException(status_code=500, detail="Failed to create audio file")
+        # 创建视频
+        output_video = os.path.join(project_dir, "output.mp4")
+        result = create_video(project_dir, image_paths, subtitle_file, audio_file, output_video)
+        if not result:
+            raise HTTPException(status_code=500, detail="Failed to create video")
+        # 创建WebDAV目录结构
+        webdav_base_path = f"comic_videos/{project_id}"
+        # 上传所有资源到WebDAV
+        video_url = upload_to_webdav(output_video, f"{webdav_base_path}/video.mp4")
+        subtitle_url = upload_to_webdav(subtitle_file, f"{webdav_base_path}/subtitles.srt")
+        audio_url = upload_to_webdav(audio_file, f"{webdav_base_path}/audio.mp3")
+        # 上传图片
+        image_urls = []
+        for i, img_path in enumerate(image_paths):
+            remote_path = f"{webdav_base_path}/images/panel_{i}.jpg"
+            img_url = upload_to_webdav(img_path, remote_path)
+            if img_url:
+                image_urls.append(img_url)
+        # 后台任务清理临时文件
+        background_tasks.add_task(lambda: shutil.rmtree(project_dir, ignore_errors=True))
+        return {
+            "videoUrl": video_url,
+            "subtitleUrl": subtitle_url,
+            "audioUrl": audio_url,
+            "imageUrls": image_urls,
+            "projectId": project_id
+        }
+    except Exception as e:
+        # 清理临时文件
+        shutil.rmtree(project_dir, ignore_errors=True)
+        logger.error(f"Error generating video: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
+# 健康检查端点health
+@app.get("/")
+async def health_check():
+    return {"status": "ok"}
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=8000)

requirements.txt ADDED Viewed

	@@ -0,0 +1,7 @@

+fastapi>=0.95.0
+uvicorn>=0.21.1
+aiohttp>=3.8.4
+openai>=1.2.0
+python-multipart>=0.0.6
+webdavclient3>=3.14.6
+pydantic>=1.10.7