WebCollector-Python基于 Python 的开源网络爬虫框架

联合创作 · 2023-09-29 13:44

WebCollector-Python

WebCollector-Python 是一个无须配置、便于二次开发的 Python 爬虫框架(内核),它提供精简的的 API,只需少量代码即可实现一个功能强大的爬虫。

WebCollector Java版本

WebCollector Java版相比WebCollector-Python具有更高的效率: https://github.com/CrawlScript/WebCollector

安装

pip安装命令

pip install https://github.com/CrawlScript/WebCollector-Python/archive/master.zip

示例

Basic

快速入门

自动探测URL

demo_auto_news_crawler.py:

# coding=utf-8
import webcollector as wc


class NewsCrawler(wc.RamCrawler):
    def __init__(self):
        super().__init__(auto_detect=True)
        self.num_threads = 10
        self.add_seed("https://github.blog/")
        self.add_regex("+https://github.blog/[0-9]+.*")
        self.add_regex("-.*#.*")  # do not detect urls that contain "#"

    def visit(self, page, detected):
        if page.match_url("https://github.blog/[0-9]+.*"):
            title = page.select("h1.lh-condensed")[0].text.strip()
            content = page.select("div.markdown-body")[0].text.replace("\n", " ").strip()
            print("\nURL: ", page.url)
            print("TITLE: ", title)
            print("CONTENT: ", content[:50], "...")


crawler = NewsCrawler()
crawler.start(10)

手动探测URL

demo_manual_news_crawler.py:

# coding=utf-8
import webcollector as wc


class NewsCrawler(wc.RamCrawler):
    def __init__(self):
        super().__init__(auto_detect=False)
        self.num_threads = 10
        self.add_seed("https://github.blog/")

    def visit(self, page, detected):

        detected.extend(page.links("https://github.blog/[0-9]+.*"))

        if page.match_url("https://github.blog/[0-9]+.*"):
            title = page.select("h1.lh-condensed")[0].text.strip()
            content = page.select("div.markdown-body")[0].text.replace("\n", " ").strip()
            print("\nURL: ", page.url)
            print("TITLE: ", title)
            print("CONTENT: ", content[:50], "...")


crawler = NewsCrawler()
crawler.start(10)

用detected_filter插件过滤探测到的URL

demo_detected_filter.py:

# coding=utf-8
import webcollector as wc
from webcollector.filter import Filter
import re


class RegexDetectedFilter(Filter):
    def filter(self, crawl_datum):
        if re.fullmatch("https://github.blog/2019-02.*", crawl_datum.url):
            return crawl_datum
        else:
            print("filtered by detected_filter: {}".format(crawl_datum.brief_info()))
            return None


class NewsCrawler(wc.RamCrawler):
    def __init__(self):
        super().__init__(auto_detect=True, detected_filter=RegexDetectedFilter())
        self.num_threads = 10
        self.add_seed("https://github.blog/")

    def visit(self, page, detected):

        detected.extend(page.links("https://github.blog/[0-9]+.*"))

        if page.match_url("https://github.blog/[0-9]+.*"):
            title = page.select("h1.lh-condensed")[0].text.strip()
            content = page.select("div.markdown-body")[0].text.replace("\n", " ").strip()
            print("\nURL: ", page.url)
            print("TITLE: ", title)
            print("CONTENT: ", content[:50], "...")


crawler = NewsCrawler()
crawler.start(10)

用RedisCrawler进行可断点的采集(可在关闭后恢复)

demo_redis_crawler.py:

# coding=utf-8
from redis import StrictRedis
import webcollector as wc


class NewsCrawler(wc.RedisCrawler):

    def __init__(self):
        super().__init__(redis_client=StrictRedis("127.0.0.1"),
                         db_prefix="news",
                         auto_detect=True)
        self.num_threads = 10
        self.resumable = True # you can resume crawling after shutdown
        self.add_seed("https://github.blog/")
        self.add_regex("+https://github.blog/[0-9]+.*")
        self.add_regex("-.*#.*")  # do not detect urls that contain "#"

    def visit(self, page, detected):
        if page.match_url("https://github.blog/[0-9]+.*"):
            title = page.select("h1.lh-condensed")[0].text.strip()
            content = page.select("div.markdown-body")[0].text.replace("\n", " ").strip()
            print("\nURL: ", page.url)
            print("TITLE: ", title)
            print("CONTENT: ", content[:50], "...")


crawler = NewsCrawler()
crawler.start(10)

用Requests定制Http请求

demo_custom_http_request.py:

# coding=utf-8

import webcollector as wc
from webcollector.model import Page
from webcollector.plugin.net import HttpRequester

import requests


class MyRequester(HttpRequester):
    def get_response(self, crawl_datum):
        # custom http request
        headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/76.0.3809.100 Safari/537.36"
        }

        print("sending request with MyRequester")

        # send request and get response
        response = requests.get(crawl_datum.url, headers=headers)

        # update code
        crawl_datum.code = response.status_code

        # wrap http response as a Page object
        page = Page(crawl_datum,
                    response.content,
                    content_type=response.headers["Content-Type"],
                    http_charset=response.encoding)

        return page


class NewsCrawler(wc.RamCrawler):
    def __init__(self):
        super().__init__(auto_detect=True)
        self.num_threads = 10

        # set requester to enable MyRequester
        self.requester = MyRequester()

        self.add_seed("https://github.blog/")
        self.add_regex("+https://github.blog/[0-9]+.*")
        self.add_regex("-.*#.*")  # do not detect urls that contain "#"

    def visit(self, page, detected):
        if page.match_url("https://github.blog/[0-9]+.*"):
            title = page.select("h1.lh-condensed")[0].text.strip()
            content = page.select("div.markdown-body")[0].text.replace("\n", " ").strip()
            print("\nURL: ", page.url)
            print("TITLE: ", title)
            print("CONTENT: ", content[:50], "...")


crawler = NewsCrawler()
crawler.start(10)
浏览 9
点赞
评论
收藏
分享

手机扫一扫分享

编辑 分享
举报
评论
图片
表情
推荐
点赞
评论
收藏
分享

手机扫一扫分享

编辑 分享
举报