pyspider

A Powerful Spider(Web Crawler) System in Python. TRY IT NOW!

• Write script in Python
• Powerful WebUI with script editor, task monitor, project manager and result viewer
• MySQL, MongoDB, Redis, SQLite, Elasticsearch; PostgreSQL with SQLAlchemy as database backend
• RabbitMQ, Beanstalk, Redis and Kombu as message queue
• Task priority, retry, periodical, recrawl by age, etc...
• Distributed architecture, Crawl Javascript pages, Python 2&3, etc...

Tutorial: http://docs.pyspider.org/en/latest/tutorial/
Documentation: http://docs.pyspider.org/
Release notes: https://github.com/binux/pyspider/releases

Sample Code

from pyspider.libs.base_handler import *


class Handler(BaseHandler):
    crawl_config = {
    }

    @every(minutes=24 * 60)
    def on_start(self):
        self.crawl('http://scrapy.org/', callback=self.index_page)

    @config(age=10 * 24 * 60 * 60)
    def index_page(self, response):
        for each in response.doc('a[href^="http"]').items():
            self.crawl(each.attr.href, callback=self.detail_page)

    def detail_page(self, response):
        return {
            "url": response.url,
            "title": response.doc('title').text(),
        }

Installation

• pip install pyspider
• run command pyspider, visit http://localhost:5000/

Quickstart: http://docs.pyspider.org/en/latest/Quickstart/

Contribute

• Use It
• Open Issue, send PR
• User Group
• 中文问答

TODO

v0.4.0

• local mode, load script from file.
• works as a framework (all components running in one process, no threads)
• redis
• shell mode like scrapy shell
• a visual scraping interface like portia

more

• edit script with vim via WebDAV

License

Licensed under the Apache License, Version 2.0