Update app.py

app.py

@@ -1,336 +1,59 @@
 import gradio as gr
-import os
-import shutil
-import yaml
-import logging
-import io
-import re
-import img2pdf
-import subprocess  # 用于调用 gallery-dl
-from jmcomic import create_option, JmcomicException
-
-# =================配置区域=================
-BASE_DIR = "download_cache"
-TEMP_YML = "jm_option_temp.yml"
-
-# --- 日志捕获设置 (专用于 JMComic) ---
-log_capture_string = io.StringIO()
-ch = logging.StreamHandler(log_capture_string)
-ch.setLevel(logging.INFO)
-logger = logging.getLogger('jmcomic')
-logger.addHandler(ch)
-logger.setLevel(logging.INFO)
-
-# =================通用工具函数=================
-
-# 1. 自然排序：让 1.jpg, 2.jpg, 10.jpg 正确排序
-def natural_sort_key(s):
-    return [int(text) if text.isdigit() else text.lower()
-            for text in re.split(r'(\d+)', s)]
-
-# 2. 清理文件名：去除非法字符
-def sanitize_filename(name):
-    return re.sub(r'[\\/*?:"<>|]', '_', name)
-
-# 3. 手动合并 PDF (核心通用功能)
-def manual_merge_pdf(album_dir, output_pdf_path):
-    image_paths = []
-    # 递归查找所有图片
-    for root, _, files in os.walk(album_dir):
-        for file in files:
-            if file.lower().endswith(('.jpg', '.jpeg', '.png', '.webp')):
-                image_paths.append(os.path.join(root, file))
-    
-    if not image_paths:
-        raise Exception("目录中未找到任何图片")
-
-    # 按文件名自然排序
-    image_paths.sort(key=natural_sort_key)
-
-    # 转换为 PDF
-    with open(output_pdf_path, "wb") as f:
-        f.write(img2pdf.convert(image_paths))
-
-# =================逻辑模块 1: JMComic 下载=================
-
-def run_jm_download(album_ids_str, auth_type, cookies_input, username_input, password_input):
-    # 1. 解析 ID 列表
-    id_list = [x.strip() for x in re.split(r'[\s,，\n]+', album_ids_str) if x.strip()]
-    
-    if not id_list:
-        yield None, "❌ 错误: 请至少输入一个本子 ID", "未检测到 ID"
-        return
-
-    # 2. 初始化环境
-    log_capture_string.truncate(0)
-    log_capture_string.seek(0)
-    
-    # PDF 存放专用目录
-    pdf_final_dir = os.path.join(BASE_DIR, "final_pdfs")
-    os.makedirs(pdf_final_dir, exist_ok=True)
-    
-    # 原始图片临时目录
-    raw_download_dir = os.path.join(BASE_DIR, "raw_jm")
-
-    # 3. 生成基础配置
-    config = {
-        'client': {'impl': 'api', 'retry_times': 3},
-        'download': {
-            'image': {'suffix': '.jpg', 'quality': 85},
-            'threading': {'batch_count': 5}
-        },
-        'dir_rule': {'base_dir': raw_download_dir},
-    }
-
-    if auth_type == "使用账号密码 (推荐)":
-        if not username_input or not password_input:
-            yield None, "❌ 错误: 请填写账号和密码", "参数错误"
-            return
-        config['client']['username'] = username_input
-        config['client']['password'] = password_input
-
-    with open(TEMP_YML, 'w', encoding='utf-8') as f:
-        yaml.dump(config, f)
-
-    # 4. 创建 Option
-    try:
-        option = create_option(TEMP_YML)
-        if auth_type == "使用 Cookies (手动)" and cookies_input:
-            option.call_once()
-            option.headers['cookie'] = cookies_input.strip()
-    except Exception as e:
-        yield None, f"❌ 配置初始化失败: {e}", "初始化错误"
-        return
-
-    # 5. 循环下载
-    success_files = []
-    total_count = len(id_list)
-    
-    for index, album_id in enumerate(id_list):
-        current_num = index + 1
-        progress_str = f"({current_num}/{total_count})"
-        
-        # 清理旧图片
-        if os.path.exists(raw_download_dir):
-            shutil.rmtree(raw_download_dir)
-        os.makedirs(raw_download_dir, exist_ok=True)
-        
-        yield success_files, f"🔄 {progress_str} 正在处理 ID: {album_id} ...", f"处理中 {current_num}/{total_count}"
-        
-        real_title = f"Unknown_{album_id}"
-        
-        try:
-            # A. 获取信息
-            client = option.build_jm_client()
-            try:
-                album_detail = client.get_album_detail(album_id)
-                real_title = album_detail.title
-            except Exception as e:
-                logger.error(f"ID {album_id} 获取详情失败: {e}")
-                continue
-
-            # B. 下载
-            yield success_files, f"📥 {progress_str} 正在下载图片: {real_title}...", f"下载中 {current_num}/{total_count}"
-            option.download_album(album_id)
-            
-            # C. 合并
-            yield success_files, f"🔨 {progress_str} 正在合并 PDF...", f"合并中 {current_num}/{total_count}"
-            
-            safe_title = sanitize_filename(real_title)
-            final_pdf_name = f"[JM] {safe_title}.pdf" # 加个前缀区分
-            final_pdf_path = os.path.join(pdf_final_dir, final_pdf_name)
-            
-            try:
-                manual_merge_pdf(raw_download_dir, final_pdf_path)
-            except Exception as e:
-                logger.error(f"合并PDF失败: {e}")
-                yield success_files, f"❌ {progress_str} 合并失败: {e}", "合并出错"
-                continue
-
-            success_files.append(final_pdf_path)
-            yield success_files, f"✅ {progress_str} 完成: {real_title}", f"完成 {current_num}/{total_count}"
-
-        except Exception as e:
-            logger.error(f"ID {album_id} 未知错误: {e}")
-            import traceback
-            traceback.print_exc()
-            continue
-
-    full_log = log_capture_string.getvalue()
-    if not success_files:
-        yield None, f"❌ 任务结束，无成功文件。\n日志:\n{full_log[-1000:]}", "全部失败"
-    else:
-        yield success_files, f"🎉 全部处理完毕！成功: {len(success_files)}个", "全部完成"
-
-# =================逻辑模块 2: E-Hentai 下载=================
-
-def run_eh_download(eh_url, cookies_str):
-    # 1. 检查参数
-    if not eh_url:
-        yield None, "❌ 请输入画廊链接", "参数缺失"
-        return
-    
-    # 2. 目录准备
-    eh_base_dir = os.path.join(BASE_DIR, "eh_temp")
-    if os.path.exists(eh_base_dir):
-        try:
-            shutil.rmtree(eh_base_dir)
-        except:
-            pass
-    os.makedirs(eh_base_dir, exist_ok=True)
-    
-    pdf_final_dir = os.path.join(BASE_DIR, "final_pdfs")
-    os.makedirs(pdf_final_dir, exist_ok=True)
-
-    # 3. 处理 Cookies
-    cookie_file_path = os.path.join(BASE_DIR, "eh_cookies.txt")
-    with open(cookie_file_path, "w", encoding="utf-8") as f:
-        f.write(cookies_str)
-
-    yield None, "🚀 启动 gallery-dl 解析...", "开始解析"
-
-    # 4. 构建命令
-    cmd = [
-        "gallery-dl",
-        "--directory", eh_base_dir,
-        "--cookies", cookie_file_path,
-        eh_url
-    ]
-
-    # 5. 执行下载
-    try:
-        process = subprocess.Popen(
-            cmd, 
-            stdout=subprocess.PIPE, 
-            stderr=subprocess.STDOUT,
-            text=True
-        )
-        
-        logs = ""
-        for line in process.stdout:
-            logs += line
-            if "Example" in line or "http" in line or "#" in line:
-                 # 简单的日志过滤，避免刷屏
-                yield None, f"📥 下载中...\n{line.strip()}", "下载中"
-        
-        process.wait()
-
-        if process.returncode != 0:
-            yield None, f"❌ 下载失败，请检查 Cookies 或链接。\n日志片段:\n{logs[-500:]}", "下载失败"
-            return
-
-    except Exception as e:
-        yield None, f"❌ 调用出错: {e}", "系统错误"
-        return
-
-    # 6. 寻找图片目录
-    target_img_dir = None
-    for root, dirs, files in os.walk(eh_base_dir):
-        if any(f.lower().endswith(('.jpg', '.png', '.jpeg')) for f in files):
-            target_img_dir = root
-            break
-    
-    if not target_img_dir:
-        yield None, "❌ 未找到下载的图片 (可能是 Sad Panda 问题)", "未找到图片"
-        return
-
-    # 7. 合并 PDF
-    folder_name = os.path.basename(target_img_dir)
-    safe_title = sanitize_filename(folder_name)
-    final_pdf_name = f"[EH] {safe_title}.pdf"
-    final_pdf_path = os.path.join(pdf_final_dir, final_pdf_name)
-
-    yield None, "🔨 正在合并 PDF...", "正在合并"
-    try:
-        manual_merge_pdf(target_img_dir, final_pdf_path)
-    except Exception as e:
-        yield None, f"❌ 合并失败: {e}", "合并失败"
-        return
-
-    yield [final_pdf_path], f"✅ 处理完成！\n文件: {final_pdf_name}", "完成"
-
-# =================界面构建 (UI)=================
+from jm_logic import run_jm_download
+from eh_logic import run_eh_download
 
+# --- 界面构建 ---
 with gr.Blocks(title="二次元聚合下载器") as demo:
     gr.Markdown("## 📦 二次元聚合下载器 (JMComic + E-Hentai)")
-    gr.Markdown("自动下载图片并合并为 PDF。支持多任务队列。")
+    gr.Markdown("模块化重构版：支持多平台下载与 PDF 自动合并。")
 
     with gr.Tabs():
         
-        # ---------------- Tab 1: JMComic ----------------
+        # === JMComic Tab ===
         with gr.TabItem("🦄 JMComic (ID下载)"):
-            gr.Markdown("一行一个 ID，下载完一个才会开始下一个，稳定防崩。")
-            
             with gr.Row():
-                jm_inp_ids = gr.Textbox(label="输入本子 ID 列表 (回车分隔)", placeholder="123456\n234567", lines=5)
+                jm_inp_ids = gr.Textbox(label="输入本子 ID 列表 (回车分隔)", placeholder="438696\n123456", lines=5)
             
             with gr.Row():
-                jm_auth_select = gr.Radio(
-                    ["使用账号密码 (推荐)", "使用 Cookies (手动)"], 
-                    label="登录方式", 
-                    value="使用账号密码 (推荐)"
-                )
+                jm_auth_select = gr.Radio(["使用账号密码 (推荐)", "使用 Cookies (手动)"], label="登录方式", value="使用账号密码 (推荐)")
             
             with gr.Group(visible=True) as group_user:
                 with gr.Row():
-                    jm_inp_user = gr.Textbox(label="用户名", placeholder="输入你的账号")
-                    jm_inp_pass = gr.Textbox(label="密码", type="password", placeholder="输入你的密码")
-                    
+                    jm_inp_user = gr.Textbox(label="用户名")
+                    jm_inp_pass = gr.Textbox(label="密码", type="password")
             with gr.Group(visible=False) as group_cookie:
-                jm_inp_cookie = gr.Textbox(label="Cookies", placeholder="key=value...")
+                jm_inp_cookie = gr.Textbox(label="Cookies")
 
-            # 切换显示逻辑
             def toggle_auth(choice):
-                if choice == "使用账号密码 (推荐)":
-                    return gr.update(visible=True), gr.update(visible=False)
-                else:
-                    return gr.update(visible=False), gr.update(visible=True)
+                return (gr.update(visible=True), gr.update(visible=False)) if choice == "使用账号密码 (推荐)" else (gr.update(visible=False), gr.update(visible=True))
             
             jm_auth_select.change(fn=toggle_auth, inputs=jm_auth_select, outputs=[group_user, group_cookie])
 
-            jm_btn_run = gr.Button("🚀 JM 批量下载", variant="primary")
-            
+            jm_btn = gr.Button("🚀 开始下载", variant="primary")
             with gr.Row():
-                jm_out_badge = gr.Label(value="等待指令", label="总体进度")
-                jm_out_log = gr.Textbox(label="实时日志", lines=5)
-            
-            jm_out_files = gr.File(label="下载结果", file_count="multiple")
+                jm_badge = gr.Label(value="Ready", label="状态")
+                jm_log = gr.Textbox(label="日志", lines=5)
+            jm_files = gr.File(label="结果")
 
-            jm_btn_run.click(
-                fn=run_jm_download, 
-                inputs=[jm_inp_ids, jm_auth_select, jm_inp_cookie, jm_inp_user, jm_inp_pass], 
-                outputs=[jm_out_files, jm_out_log, jm_out_badge]
-            )
+            jm_btn.click(fn=run_jm_download, 
+                         inputs=[jm_inp_ids, jm_auth_select, jm_inp_cookie, jm_inp_user, jm_inp_pass], 
+                         outputs=[jm_files, jm_log, jm_badge])
 
-        # ---------------- Tab 2: E-Hentai ----------------
+        # === E-Hentai Tab ===
         with gr.TabItem("🐼 E-Hentai (链接下载)"):
-            gr.Markdown("输入 E站/EX站 画廊链接。**必须提供 Cookies** (Netscape格式) 以避免 Sad Panda。")
-            
+            gr.Markdown("**提示**：必须填写 Cookies (Netscape格式) 才能下载 EX 站或原图。")
             with gr.Row():
-                eh_inp_url = gr.Textbox(label="画廊链接 (URL)", placeholder="https://e-hentai.org/g/xxxxx/yyyy/")
-            
+                eh_url = gr.Textbox(label="画廊链接")
             with gr.Row():
-                eh_inp_cookies = gr.Textbox(
-                    label="Cookies (Netscape 格式)", 
-                    placeholder="# Netscape HTTP Cookie File\n.e-hentai.org\tTRUE\t/ ...",
-                    lines=5,
-                    info="请使用 Chrome 插件 'Get cookies.txt LOCALLY' 导出并在粘贴至此。"
-                )
-
-            eh_btn_run = gr.Button("🚀 EH 下载并转 PDF", variant="primary")
+                eh_cookie = gr.Textbox(label="Cookies", lines=5, placeholder="# Netscape HTTP Cookie File...")
             
+            eh_btn = gr.Button("🚀 开始下载", variant="primary")
             with gr.Row():
-                eh_out_badge = gr.Label(value="等待指令", label="状态")
-                eh_out_log = gr.Textbox(label="运行日志", lines=5)
-            
-            eh_out_files = gr.File(label="下载结果")
+                eh_badge = gr.Label(value="Ready", label="状态")
+                eh_log = gr.Textbox(label="日志", lines=5)
+            eh_files = gr.File(label="结果")
 
-            eh_btn_run.click(
-                fn=run_eh_download,
-                inputs=[eh_inp_url, eh_inp_cookies],
-                outputs=[eh_out_files, eh_out_log, eh_out_badge]
-            )
+            eh_btn.click(fn=run_eh_download, inputs=[eh_url, eh_cookie], outputs=[eh_files, eh_log, eh_badge])
 
 if __name__ == "__main__":
-    # 使用 queue() 确保长任务不会超时断开
     demo.queue().launch()