【原创】故事分析器v1.0---一键拆文

宋长雨

小说拆文，自己用的，不墨迹，调用的是硅基流动 deepseek 8b的api，有BuG自己找问题，别找我，我用着没毛病。应该很多人在找这种东西吧，趁着api模型免费赶紧用。虽然可能这个目前一直免费，但是，天佑不测风云。。。

自己测试。


测试结果：一本900章左右的小说，大概要拆1个半小时



网盘链接：
百度网盘：链接: https://pan.baidu.com/s/1iP73WOZgsolQnQ1n9zDSrg?pwd=52pj 提取码: 52pj

移动云盘：链接: https://caiyun.139.com/m/i?2jQXiy2UeXVp7  提取码:reio  

宋长雨

Doublevv 发表于 2025-2-27 14:12
挺有意思的应用。
能分享完整源码吗？准备收录到高质量Python源码资源淘专辑中 https://www.52pojie.cn/fo ...

[Python] 纯文本查看 复制代码

?

001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050 051 052 053 054 055 056 057 058 059 060 061 062 063 064 065 066 067 068 069 070 071 072 073 074 075 076 077 078 079 080 081 082 083 084 085 086 087 088 089 090 091 092 093 094 095 096 097 098 099 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334

import re import time import aiohttp import asyncio import tkinter as tk import threading from tkinter import ttk, filedialog, scrolledtext, messagebox from typing import List, Tuple     class ConfigWindow(tk.Tk):     def __init__(self):         super().__init__()         self.title("DeepSeek 故事分析器")         self.geometry("720x700")         self.protocol("WM_DELETE_WINDOW", self.on_close)           # 配置参数初始化         self.api_key = tk.StringVar(value="请输入硅基流动中的API密钥！")         self.input_file = tk.StringVar()         self.output_dir = tk.StringVar()         self.concurrency = tk.IntVar(value=3)         self.prompt_template = tk.StringVar(             value="请以简洁的纯文本格式输出故事八要素，包含：身份、目的、行动、核心问题、阻力、结局、正能量和负能量，不要修饰。"         )         self.content_length = tk.IntVar(value=2000)         self.temperature = tk.DoubleVar(value=0.3)         self.connect_timeout = tk.IntVar(value=35)         self.total_timeout = tk.IntVar(value=160)           self.processing = False         self.processor = None         self.create_widgets()         self.setup_log_area()       def create_widgets(self):         row = 0           # 文件选择部分         ttk.Label(self, text="输入文件:").grid(row=row, column=0, sticky="w", padx=10, pady=5)         ttk.Entry(self, textvariable=self.input_file, width=50).grid(row=row, column=1)         ttk.Button(self, text="浏览...", command=self.select_input_file).grid(row=row, column=2)         row += 1           ttk.Label(self, text="输出目录:").grid(row=row, column=0, sticky="w", padx=10, pady=5)         ttk.Entry(self, textvariable=self.output_dir, width=50).grid(row=row, column=1)         ttk.Button(self, text="浏览...", command=self.select_output_dir).grid(row=row, column=2)         row += 1           # API配置         ttk.Label(self, text="API密钥:").grid(row=row, column=0, sticky="w", padx=10, pady=5)         ttk.Entry(self, textvariable=self.api_key, width=50).grid(row=row, column=1, columnspan=2)         row += 1           # 并发数         ttk.Label(self, text="并发数 (1-5):").grid(row=row, column=0, sticky="w", padx=10, pady=5)         ttk.Spinbox(self, from_=1, to=5, textvariable=self.concurrency, width=5).grid(row=row, column=1, sticky="w")         row += 1           # 内容截取长度         ttk.Label(self, text="每章截取长度:").grid(row=row, column=0, sticky="w", padx=10, pady=5)         ttk.Spinbox(self, from_=100, to=5000, increment=100, textvariable=self.content_length, width=8).grid(row=row,                                                                                                              column=1,                                                                                                              sticky="w")         row += 1           # 温度值         ttk.Label(self, text="温度值 (0.1-1):").grid(row=row, column=0, sticky="w", padx=10, pady=5)         ttk.Spinbox(self, from_=0.1, to=1.0, increment=0.1, format="%.1f",                     textvariable=self.temperature, width=5).grid(row=row, column=1, sticky="w")         row += 1           # 超时设置         ttk.Label(self, text="超时设置(秒):").grid(row=row, column=0, sticky="w", padx=10, pady=5)         ttk.Frame(self).grid(row=row, column=1, sticky="w")         ttk.Label(self, text="连接").grid(row=row, column=1, sticky="w")         ttk.Spinbox(self, from_=10, to=300, textvariable=self.connect_timeout, width=5).grid(row=row, column=2,                                                                                              sticky="w", padx=5)         ttk.Label(self, text="总超时").grid(row=row, column=3, sticky="w")         ttk.Spinbox(self, from_=30, to=600, textvariable=self.total_timeout, width=5).grid(row=row, column=4,                                                                                            sticky="w")         row += 1           # 提示模板         ttk.Label(self, text="提示模板:").grid(row=row, column=0, sticky="nw", padx=10, pady=5)         self.prompt_editor = scrolledtext.ScrolledText(self, width=70, height=8, wrap=tk.WORD)         self.prompt_editor.insert("1.0", self.prompt_template.get())         self.prompt_editor.grid(row=row, column=1, columnspan=4, pady=5, sticky="ew")         row += 1           # 控制按钮         self.btn_frame = ttk.Frame(self)         self.btn_frame.grid(row=row, column=1, pady=10, sticky="e")         self.start_btn = ttk.Button(self.btn_frame, text="开始处理", command=self.toggle_processing)         self.start_btn.pack(side="left", padx=5)         row += 1       def setup_log_area(self):         self.log_area = scrolledtext.ScrolledText(self, width=85, height=12, state="disabled")         self.log_area.grid(row=10, column=0, columnspan=5, padx=10, pady=5, sticky="nsew")         self.grid_rowconfigure(10, weight=1)       def select_input_file(self):         filepath = filedialog.askopenfilename(filetypes=[("文本文件", "*.txt")])         if filepath:             self.input_file.set(filepath)       def select_output_dir(self):         dirpath = filedialog.askdirectory()         if dirpath:             self.output_dir.set(dirpath)       def toggle_processing(self):         if not self.processing:             self.start_processing()         else:             self.cancel_processing()       def start_processing(self):         if not all([self.input_file.get(), self.output_dir.get(), self.api_key.get()]):             messagebox.showerror("错误", "请填写所有必填字段！")             return           try:             if not (0.1 <= self.temperature.get() <= 1.0):                 raise ValueError("温度值必须在0.1到1.0之间")             if self.content_length.get() < 100:                 raise ValueError("截取长度至少100字符")         except Exception as e:             messagebox.showerror("参数错误", str(e))             return           self.processing = True         self.start_btn.config(text="取消处理")         self.log_area.config(state="normal")         self.log_area.delete("1.0", tk.END)         self.log_area.config(state="disabled")           self.processor = TurboProcessor(             api_key=self.api_key.get(),             input_file=self.input_file.get(),             output_dir=self.output_dir.get(),             concurrency=self.concurrency.get(),             prompt_template=self.prompt_editor.get("1.0", tk.END).strip(),             content_length=self.content_length.get(),             temperature=self.temperature.get(),             timeout=(self.connect_timeout.get(), self.total_timeout.get()),             log_callback=self.update_log         )           self.thread = threading.Thread(target=self.run_async_task, daemon=True)         self.thread.start()       def cancel_processing(self):         if self.processor:             self.processor.cancel()         self.processing = False         self.start_btn.config(text="开始处理")         self.update_log("处理已中止")       def run_async_task(self):         try:             loop = asyncio.new_event_loop()             asyncio.set_event_loop(loop)             loop.run_until_complete(self.processor.run())         except Exception as e:             self.update_log(f"错误: {str(e)}")         finally:             self.after(0, lambda: self.cancel_processing())       def update_log(self, message: str):         self.log_area.config(state="normal")         self.log_area.insert(tk.END, message + "\n")         self.log_area.see(tk.END)         self.log_area.config(state="disabled")       def on_close(self):         if self.processing:             if messagebox.askokcancel("退出", "处理正在进行中，确定要退出吗？"):                 self.cancel_processing()                 self.destroy()         else:             self.destroy()     class TurboProcessor:     def __init__(self, api_key, input_file, output_dir, concurrency,                  prompt_template, content_length, temperature, timeout, log_callback):         self.api_key = api_key         self.input_file = input_file         self.output_file = f"{output_dir}/result_{int(time.time())}.txt"         self.concurrency = concurrency         self.prompt_template = f"{prompt_template} 内容：{{content}}"         self.content_length = content_length         self.temperature = temperature         self.timeout = timeout         self.log_callback = log_callback           self.sem = asyncio.Semaphore(concurrency)         self.cancelled = False         self.tasks = []         self.chapters = []         self.results = []         self.total_chapters = 0         self.start_time = 0.0         self.success = 0         self.failed = 0       def cancel(self):         self.cancelled = True         for task in self.tasks:             task.cancel()       def _log(self, message: str):         if self.log_callback:             self.log_callback(message)       async def process_chapter(self, chapter_info: Tuple[int, str]) -> str:         if self.cancelled:             return None           index, content = chapter_info         current = f"{index}/{self.total_chapters}"           payload = {             "model": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",             "messages": [{                 "role": "user",                 "content": self.prompt_template.format(content=content[:self.content_length])             }],             "temperature": self.temperature         }           async with self.sem:             for attempt in range(5 + 1):                 try:                     async with aiohttp.ClientSession(                             headers={"Authorization": f"Bearer {self.api_key}"},                             timeout=aiohttp.ClientTimeout(*self.timeout)                     ) as session:                           if self.cancelled:                             return None                           if attempt == 0:                             self._log(f"[章节 {current}] 开始处理")                         else:                             self._log(f"[章节 {current}] 第{attempt}次重试")                           start = time.monotonic()                         async with session.post(                                 "https://api.siliconflow.cn/v1/chat/completions",                                 json=payload                         ) as resp:                               if resp.status == 200:                                 data = await resp.json()                                 elapsed = time.monotonic() - start                                 self._log(f"[章节 {current}] 成功（耗时 {elapsed:.1f}s）")                                 text = data['choices'][0]['message']['content'].strip()                                 return re.sub(r'\n{3,}', '\n', text)                               elif resp.status == 429:                                 backoff = min(2 ** attempt, 30)                                 self._log(f"[章节 {current}] 频率限制（等待{backoff}s）")                                 await asyncio.sleep(backoff)                               else:                                 self._log(f"[章节 {current}] HTTP错误 {resp.status}")                                 break                   except (asyncio.CancelledError, KeyboardInterrupt):                     raise                 except Exception as e:                     backoff = min(2 ** attempt, 30)                     self._log(f"[章节 {current}] 错误: {type(e).__name__}（{attempt + 1}/5次）")                     await asyncio.sleep(backoff)               self._log(f"[章节 {current}] 处理失败")             return None       async def pipeline(self):         try:             # 读取文件             try:                 with open(self.input_file, 'r', encoding='utf-8') as f:                     text = f.read()             except UnicodeDecodeError:                 with open(self.input_file, 'r', encoding='gbk') as f:                     text = f.read()               # 分割章节             raw_chapters = re.findall(r'(第[^章]+章[\s\S]*?)(?=第[^章]+章|$)', text)             self.total_chapters = len(raw_chapters)             self.chapters = list(enumerate(raw_chapters, start=1))               self._log(f"开始处理 {self.total_chapters} 个章节...")               # 创建任务             self.tasks = [asyncio.create_task(self.process_chapter((idx, ch)))                           for idx, ch in self.chapters]             self.results = await asyncio.gather(*self.tasks, return_exceptions=True)               # 写入结果             with open(self.output_file, 'w', encoding='utf-8') as f:                 for (idx, _), res in zip(self.chapters, self.results):                     if res and not isinstance(res, Exception):                         f.write(f"=== 第{idx}章 ===\n{res}\n\n")                         self.success += 1                     else:                         f.write(f"=== 第{idx}章处理失败 ===\n\n")                         self.failed += 1               # 最终报告             total_time = time.monotonic() - self.start_time             self._log("\n处理完成！\n"                       f"成功: {self.success}\n"                       f"失败: {self.failed}\n"                       f"总耗时: {total_time:.1f}s\n"                       f"输出文件: {self.output_file}")           except asyncio.CancelledError:             self._log("用户中止处理流程")         except Exception as e:             self._log(f"处理出错: {str(e)}")       async def run(self):         self.start_time = time.monotonic()         await self.pipeline()     if __name__ == "__main__":     window = ConfigWindow()     window.mainloop()

宋长雨

核心源码放出来

[Python] 纯文本查看 复制代码

?

001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050 051 052 053 054 055 056 057 058 059 060 061 062 063 064 065 066 067 068 069 070 071 072 073 074 075 076 077 078 079 080 081 082 083 084 085 086 087 088 089 090 091 092 093 094 095 096 097 098 099 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

import re import time import aiohttp  # 改用异步请求库 import asyncio from typing import List, Tuple   # ============== 配置区 ============== API_KEY = "API密钥" INPUT_FILE = "要拆的小说" OUTPUT_FILE = "结果" CONCURRENCY = 3  # 并发请求数（根据API限制调整） MAX_RETRY = 5  # 智能重试次数 TIMEOUT = (35, 160)  # 连接35s/总160s超时     # ====================================   class TurboProcessor:     def __init__(self):         self.log = []         self.success = 0         self.failed = 0         self.sem = asyncio.Semaphore(CONCURRENCY)  # 信号量控制并发         self.chapters: List[Tuple[int, str]] = []         self.results = []       def _log(self, msg: str):         """轻量级日志记录"""         entry = f"[{time.strftime('%H:%M:%S')}] {msg}"         print(entry)         self.log.append(entry)       async def process_chapter(self, index: int, content: str) -> str:         """异步处理核心"""         payload = {             "model": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",             "messages": [{                 "role": "user",                 "content": f"请以简洁的纯文本格式输出故事八要素，包含八要素：身份、目的、行动、核心问题、阻力、结局、正能量和负能量，不要任何修饰。将以下内容精简至100字八要素：{content[:2000]}"             }],             "temperature": 0.3         }           async with self.sem:             for attempt in range(MAX_RETRY):                 try:                     async with aiohttp.ClientSession(                             headers={"Authorization": f"Bearer {API_KEY}"},                             timeout=aiohttp.ClientTimeout(*TIMEOUT)                     ) as session:                           start = time.monotonic()                         async with session.post(                                 "https://api.siliconflow.cn/v1/chat/completions",                                 json=payload                         ) as resp:                               if resp.status == 200:                                 data = await resp.json()                                 elapsed = time.monotonic() - start                                 self._log(f"Ch{index} ✓ ({elapsed:.1f}s)")                                 return re.sub(r'\n{3,}', '\n', data['choices'][0]['message']['content'].strip())                               elif resp.status == 429:                                 backoff = min(2 ** attempt, 10)                                 self._log(f"Ch{index} ⚡ 过载（等待{backoff}s）")                                 await asyncio.sleep(backoff)                                 continue                               else:                                 self._log(f"Ch{index} ✗ HTTP {resp.status}")                                 break                   except (aiohttp.ClientError, asyncio.TimeoutError) as e:                     backoff = min(2 ** attempt, 15)                     self._log(f"Ch{index} ⚡ {type(e).__name__}（重试 {attempt + 1}/{MAX_RETRY}）")                     await asyncio.sleep(backoff)               self._log(f"Ch{index} ✗ 最终失败")             return None       async def pipeline(self):         """异步处理管线"""         # 读取文件         try:             with open(INPUT_FILE, 'r', encoding='utf-8') as f:                 text = f.read()         except UnicodeDecodeError:             with open(INPUT_FILE, 'r', encoding='gbk') as f:                 text = f.read()           # 分割章节         self.chapters = list(enumerate(re.findall(             r'(第[^章]+章[\s\S]*?)(?=第[^章]+章|$)', text         ), start=1))           # 启动异步任务         tasks = [self.process_chapter(idx, cont) for idx, cont in self.chapters]         self.results = await asyncio.gather(*tasks)           # 写入结果         with open(OUTPUT_FILE, 'w', encoding='utf-8') as f:             for (idx, _), res in zip(self.chapters, self.results):                 if res:                     f.write(f"第{idx}章：\n{res}\n\n")                     self.success += 1                 else:                     f.write(f"第{idx}章处理失败\n\n")                     self.failed += 1           # 性能报告         total_time = time.monotonic() - self.start_time         self._log(             f"处理完成 | 成功：{self.success} 失败：{self.failed} | 总耗时：{total_time:.1f}s | 平均速度：{len(self.chapters) / total_time:.2f}章/秒")       async def run(self):         self.start_time = time.monotonic()         await self.pipeline()     if __name__ == "__main__":     processor = TurboProcessor()     asyncio.run(processor.run())

歆风

挺有创意的东西

liangp1979

这个猛啊，多谢分享

dj102007

这个猛啊，多谢分享

hhjjrrtt

这个有点意思，谢谢分享

qinlao123

试一下。谢谢

LUYAJUN

有没有能倒过来的工具，先自己例每个章节的提纲，它按题纲写出几万字的小说出来

halou

大佬NB plus！

自在小鱼

这个有点意思，谢谢分享