python Cmd实例之网络爬虫应用

python Cmd 爬虫

废话少说，直接上代码

# encoding=utf-8
import os
import multiprocessing
from cmd import Cmd
import commands
from mycrawler.dbUtil import DbUtil
import signal
# 下载监控
def run_download_watch():
    os.system("gnome-terminal -x bash -c 'python ./download_process.py' ")
# 下载文件
def run_download():
    os.system("gnome-terminal -x bash -c 'python ./download.py' ")
# 爬虫
def run_spider(arg):
    for i in range(len(arg)):
        os.system("gnome-terminal -x bash -c 'scrapy crawl %s'" % arg[i])
class CLI(Cmd):
    def __init__(self):
        Cmd.__init__(self)
        # 设置命令提示符
        self.prompt = ">>> "
        self.intro = '''
        欢迎进入爬虫控制台
        输入0:查看帮助'''
        self.doc_header = ''
        self.undoc_header = ''
        self.nohelp = "*** 命令%s没有帮助文档"
    def do_download(self, arg):
        '''    下载文件'''
        p2 = multiprocessing.Process(target=run_download)
        p2.start()
    def do_0(self, arg):
        self.do_help(1)
    def do_help(self, arg):
        def ddoc(ss, arg):
            try:
                doc = getattr(ss, 'do_' + arg).__doc__
                if doc:
                    print arg + ":"
                    print doc
                    return
            except AttributeError:
                ss.stdout.write("%s\n" % str(ss.nohelp % (arg,)))
        cmds_doc = []
        for name in self.get_names():
            if name[:3] == 'do_':
                cmds_doc.append(name[3:])
        print self.doc_header
        for c in cmds_doc:
            ddoc(self, c)
    # 添加新的爬虫连接
    def do_add(self, args):
        """    新增链接(厂商网址)到数据库中
    输入格式为:add name abb;start_urls www.baidu.com www.baidu.com www.baidu.com
    add是添加命令，后面的是参数。start_urls后面可以跟随多条数据，空格分开"""
        if not args:
            print "输入内容为空，请查看帮助：help add"
            return
        print args
        data = dict([(bb.split(' ')[0], len(bb.split(' ')[1:]) == 1 and bb.split(
            ' ')[1] or bb.split(' ')[1:]) for bb in args.split(';')])
        print data
        DbUtil().conn().collection('url_items').insert(data)
    # 列出所有的爬虫
    def do_list_spider(self, args):
        '''     列出所有的爬虫'''
        print commands.getoutput("scrapy list")
    # 运行一个爬虫
    def do_run_spider(self, arg):
        '''     运行一个爬虫，例如run_spider abb'''
        p3 = multiprocessing.Process(
            target=run_spider, args=(arg,))
        p3.start()
        # os.system('scrapy crawl ' + arg)
    def do_run(self, args):
        '''    运行所有的程序'''
        # 运行爬虫
        self.do_run_all_spiders(1)
        # 运行下载
        p2 = multiprocessing.Process(target=run_download)
        p2.start()
        # 运行下载监控
        p3 = multiprocessing.Process(target=run_download_watch)
        p3.start()
    # 运行所有的爬虫
    def do_run_all_spiders(self, arg):
        '''    运行所有的爬虫'''
        s = cmd('scrapy list').value().split('\n')
        if not s:
            print "没有爬虫，请检验代码是否正确"
            return
        p = multiprocessing.Process(
            target=run_spider, args=(s,))
        p.start()
        # os.system('./run_spider.sh ' + spider)
    def do_q(self, arg):
        '''    退出系统'''
        return True
    # emptyline
    def emptyline(self):
        os.system('clear')
        print '回车清屏，help帮助,tab补全'
    # 当无法识别输入的command时调用该方法
    def default(self, line):
        print '输入的命令' + repr(line) + '错误，请输入help查看命令帮助'
    # 退出之后调用该方法
    def postloop(self):
        print '谢谢使用'
    def completedefault(self, *ignored):
        return ['add', 'run_spider', 'run_all_spiders', 'list_spider']
if __name__ == "__main__":
    cli = CLI()
    cli.cmdloop()