一个强大的、高性能的异步Web爬取和抓取框架,基于Python的asyncio生态系统构建。
英文 | 中文
Aioscpy是一个快速的高级web爬行和web抓取框架,用于抓取网站并从其页面提取结构化数据。它受到Scrapy和scrapy_redis的启发,但从头开始设计,充分利用异步编程的全部功能。
- 完全异步:基于Python的asyncio,实现高性能并发操作
- Scrapy风格的API:为来自Scrapy的用户提供熟悉的API
- 分布式爬取:支持使用Redis进行分布式爬取
- 多种HTTP后端:支持aiohttp、httpx和requests
- 动态变量注入:强大的依赖注入系统
- 灵活的中间件系统:可定制的请求/响应处理管道
- 强大的数据处理:用于处理爬取数据的管道
- Python 3.8+
- 支持Linux、Windows、macOS、BSD
pip install aioscpy
pip install aioscpy[all]
pip install aioscpy[aiohttp,httpx]
pip install git+https://github.com/ihandmine/aioscpy
aioscpy startproject myproject
cd myproject
aioscpy genspider myspider
这将在spiders
目录中创建一个基本爬虫。
from aioscpy.spider import Spider
class QuotesSpider(Spider):
name = 'quotes'
custom_settings = {
"SPIDER_IDLE": False
}
start_urls = [
'https://quotes.toscrape.com/tag/humor/',
]
async def parse(self, response):
for quote in response.css('div.quote'):
yield {
'author': quote.xpath('span/small/text()').get(),
'text': quote.css('span.text::text').get(),
}
next_page = response.css('li.next a::attr("href")').get()
if next_page is not None:
yield response.follow(next_page, self.parse)
aioscpy onespider single_quotes
from aioscpy.spider import Spider
from anti_header import Header
from pprint import pprint, pformat
class SingleQuotesSpider(Spider):
name = 'single_quotes'
custom_settings = {
"SPIDER_IDLE": False
}
start_urls = [
'https://quotes.toscrape.com/',
]
async def process_request(self, request):
request.headers = Header(url=request.url, platform='windows', connection=True).random
return request
async def process_response(self, request, response):
if response.status in [404, 503]:
return request
return response
async def process_exception(self, request, exc):
raise exc
async def parse(self, response):
for quote in response.css('div.quote'):
yield {
'author': quote.xpath('span/small/text()').get(),
'text': quote.css('span.text::text').get(),
}
next_page = response.css('li.next a::attr("href")').get()
if next_page is not None:
yield response.follow(next_page, callback=self.parse)
async def process_item(self, item):
self.logger.info("{item}", **{'item': pformat(item)})
if __name__ == '__main__':
quotes = SingleQuotesSpider()
quotes.start()
# 从项目中运行爬虫
aioscpy crawl quotes
# 运行单个爬虫脚本
aioscpy runspider quotes.py
from aioscpy.crawler import call_grace_instance
from aioscpy.utils.tools import get_project_settings
# 方法1:从目录中加载所有爬虫
def load_spiders_from_directory():
process = call_grace_instance("crawler_process", get_project_settings())
process.load_spider(path='./spiders')
process.start()
# 方法2:按名称运行特定爬虫
def run_specific_spider():
process = call_grace_instance("crawler_process", get_project_settings())
process.crawl('myspider')
process.start()
if __name__ == '__main__':
run_specific_spider()
Aioscpy可以通过项目中的settings.py
文件进行配置。以下是最重要的设置:
# 最大并发处理项目数
CONCURRENT_ITEMS = 100
# 最大并发请求数
CONCURRENT_REQUESTS = 16
# 每个域名的最大并发请求数
CONCURRENT_REQUESTS_PER_DOMAIN = 8
# 每个IP的最大并发请求数
CONCURRENT_REQUESTS_PER_IP = 0
# 请求间的延迟(秒)
DOWNLOAD_DELAY = 0
# 请求超时时间(秒)
DOWNLOAD_TIMEOUT = 20
# 是否随机化下载延迟
RANDOMIZE_DOWNLOAD_DELAY = True
# 使用的HTTP后端
DOWNLOAD_HANDLER = "aioscpy.core.downloader.handlers.httpx.HttpxDownloadHandler"
# 其他选项:
# DOWNLOAD_HANDLER = "aioscpy.core.downloader.handlers.aiohttp.AioHttpDownloadHandler"
# DOWNLOAD_HANDLER = "aioscpy.core.downloader.handlers.requests.RequestsDownloadHandler"
# 使用的调度器(基于内存或Redis)
SCHEDULER = "aioscpy.core.scheduler.memory.MemoryScheduler"
# 分布式爬取:
# SCHEDULER = "aioscpy.core.scheduler.redis.RedisScheduler"
# Redis连接设置(用于Redis调度器)
REDIS_URI = "redis://localhost:6379"
QUEUE_KEY = "%(spider)s:queue"
Aioscpy提供了丰富的API来处理响应:
# 使用CSS选择器
title = response.css('title::text').get()
all_links = response.css('a::attr(href)').getall()
# 使用XPath
title = response.xpath('//title/text()').get()
all_links = response.xpath('//a/@href').getall()
# 跟踪链接
yield response.follow('next-page.html', self.parse)
# 使用回调跟踪链接
yield response.follow('details.html', self.parse_details)
# 跟踪所有匹配的CSS选择器的链接
yield from response.follow_all(css='a.product::attr(href)', callback=self.parse_product)
aioscpy -h
要启用基于Redis的分布式爬取:
- 在设置中配置Redis:
SCHEDULER = "aioscpy.core.scheduler.redis.RedisScheduler"
REDIS_URI = "redis://localhost:6379"
QUEUE_KEY = "%(spider)s:queue"
- 在不同的机器上运行多个爬虫实例,全部连接到同一个Redis服务器。
请通过创建issue向项目所有者提交您的建议。