Web Extractor return only title for cnb.cool

[ZhulongNT]

🔎 Search before asking | 提交之前请先搜索

• I have searched the MinerU Readme and found no similar bug report.
• I have searched the MinerU Issues and found no similar bug report.
• I have searched the MinerU Discussions and found no similar bug report.

🤖 Consult the online AI assistant for assistance | 在线 AI 助手咨询

• I have consulted the online AI assistant but was unable to obtain a solution to the issue.

Description of the bug | 错误描述

For example,

https://cnb.cool/acidbunny/ChurnPrediction/-/wiki/%E9%A1%B9%E7%9B%AE%E6%A6%82%E8%BF%B0/%E4%B8%9A%E5%8A%A1%E8%83%8C%E6%99%AF

How to reproduce the bug | 如何复现

https://cnb.cool/acidbunny/ChurnPrediction/-/wiki/%E9%A1%B9%E7%9B%AE%E6%A6%82%E8%BF%B0/%E4%B8%9A%E5%8A%A1%E8%83%8C%E6%99%AF

the url is above.

I was using the online version at https://mineru.net/

Operating System Mode | 操作系统类型

Windows

Operating System Version| 操作系统版本

Windows 10 LTSC

I was using the online version at https://mineru.net/

Python version | Python 版本

3.11

Software version | 软件版本 (mineru --version)

>=2.5

Backend name | 解析后端

vlm

Device mode | 设备模式

cpu

[dosubot]

Hi @ZhulongNT! I'm Dosu and I’m helping the MinerU team.

This is a known issue with the VLM backend in MinerU (>=2.5, CPU mode): a bug in the URL handling logic causes requests for web pages with path segments (like cnb.cool wiki pages) to fail, which often results in only the title being extracted and not the full content. The bug has been identified and a fix is planned for a future release. There are also limitations in the VLM backend's error handling—if any part of the extraction fails (such as a 404 or server error), you may get only partial results or just the title returned. This is especially common with complex or protected web pages, or when using the VLM backend in CPU mode, which is less robust with special characters and layouts [source].

Workarounds: try switching to the pipeline backend (which is more stable for web and PDF extraction), or process pages individually in small batches with error handling if you are running MinerU locally. Upgrading to the latest MinerU version (>=2.5.3) may help, but the underlying issue with VLM web extraction is not fully resolved yet [source].

There are no configuration options to force extraction of dynamic content or to bypass this bug in the current online version. If you need reliable extraction from cnb.cool or similar sites, the pipeline backend is recommended until the VLM backend is fixed.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

   

[drunkpig]

我测试的结果是正常的，你是否是copy了源码进行的解析？由于html是动态渲染的，所以源码中没有包含这些正文。请提交渲染之后的html内容