开源的多线程爬虫框架

融合了scrapy的架构理念和twisted的任务调度思想，目前只是个精简版的，开源地址

https://github.com/aware-why/multithreaded_crawler/，已附带demo演示的例子。

有兴趣的可以参与进来，有更好想法的人可以直接发起pull request，热烈欢迎大家的贡献。

 

 

multithreaded_crawler

A condensed crawler framework of “multithreaded model”

dependency

At present, the framework depends on nothing except for modules in the python standard libraries.

Usage

cd threaded_spider
python run.py --help

You will see a demo output by python run.py, it crawls the sina.com.cn using five threads and has the crawling depth limited to be 2 by default (It's tested in python2.7).
In threaded_spider directory, there are extra log files whose name like “spider.*.log” respectively generated using python run.py --thread=* command.

Community

QQ Group: 4704309
Your contribute will be welcome.