Scrapy之定制命令与自由发挥的信号

定制命令

在项目目录下创建一个脚本

  

1 #!/usr/bin/env python
2 # -*- coding:utf-8 -*-
3 
4 import sys
5 from scrapy.cmdline import execute
6 
7 if __name__ == '__main__':
8     # execute(['scrapy', 'crawl', 'chouti', '--nolog'])
9     execute(['scrapy', 'crawlall', '--nolog'])

在spiders同级目录创建文件夹commands

 1 #!/usr/bin/env python
 2 # -*- coding:utf-8 -*-
 3 
 4 from scrapy.commands import ScrapyCommand
 5 from scrapy.utils.project import get_project_settings
 6 
 7 
 8 class Command(ScrapyCommand):
 9     requires_project = True
10 
11     def syntax(self):
12         return '[options]'
13 
14     def short_desc(self):
15         return 'Runs all of the spiders'
16 
17     def run(self, args, opts):
18         spider_list = self.crawler_process.spiders.list()
19         print(spider_list)
20         for name in spider_list:
21             self.crawler_process.crawl(name, **opts.__dict__)
22         self.crawler_process.start()

配置文件

1 # 自定制命令
2 COMMANDS_MODULE =  "redisdepth.commands"

 信号开发框架的大佬给我们预留的位置自由发挥

 1 #!/usr/bin/env python
 2 # -*- coding:utf-8 -*-
 3 
 4 from scrapy import signals
 5 
 6 
 7 class MyExtension(object):
 8     def __init__(self):
 9         # self.value = value
10         pass
11 
12     @classmethod
13     def from_crawler(cls, crawler):
14         # val = crawler.settings.getint('MMMM')
15         self = cls()
16 
17         crawler.signals.connect(self.spider_opened, signal=signals.spider_opened)
18         crawler.signals.connect(self.spider_closed, signal=signals.spider_closed)
19 
20         return self
21 
22     def spider_opened(self, spider):
23         print('open')
24 
25     def spider_closed(self, spider):
26         print('close')

 配置

1 EXTENSIONS = {
2    # 'scrapy.extensions.telnet.TelnetConsole': None,
3     'redisdepth.ext.MyExtension': 666,
4 }

 内置信号

 1 """
 2 Scrapy signals
 3 
 4 These signals are documented in docs/topics/signals.rst. Please don't add new
 5 signals here without documenting them there.
 6 """
 7 
 8 engine_started = object()
 9 engine_stopped = object()
10 spider_opened = object()
11 spider_idle = object()
12 spider_closed = object()
13 spider_error = object()
14 request_scheduled = object()
15 request_dropped = object()
16 request_reached_downloader = object()
17 response_received = object()
18 response_downloaded = object()
19 item_scraped = object()
20 item_dropped = object()
21 item_error = object()
22 
23 # for backward compatibility
24 stats_spider_opened = spider_opened
25 stats_spider_closing = spider_closed
26 stats_spider_closed = spider_closed
27 
28 item_passed = item_scraped
29 
30 request_received = request_scheduled
原文地址:https://www.cnblogs.com/Alexephor/p/11444002.html