第4.4章 scrapy爬虫lambda传参

拼搏现实的明天。 2022-04-13 02:49 217阅读 0赞

输入地址[聂卫平][Link 1]，我想看看中国棋院棋手到底有哪些特点，看到了下图，决定写个爬虫，一下子下载下来  
![1][]  
这里并不是静态加载的，而是像后台请求的。这里使用了lambda传参

import scrapy
    import json
    from pyquery import PyQuery as pq
    from life_example.items import PersonBaiKeItem
    from life_example.utils.util import get_uuid,remove_js_css
    
    ''' 围棋中国棋院 '''
    class WeiQiZgqySpider(scrapy.Spider):
        name = "zgqy"
        start_urls = [
            "https://baike.baidu.com/guanxi/jsondata?action=getViewLemmaData&args=%5B0%2C8%2C%7B%22fentryTableId%22%3A18311%2C%22lemmaId%22%3A9598014%2C%22subLemmaId%22%3A9598014%7D%2Cfalse%5D",
        ]
    
        def parse(self, response):
            html = json.loads(response.body_as_unicode())
            soup = pq(html['html'])
            # 九段
            d9 = soup('div.relation-unit').eq(0)
            links = pq(d9)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院九段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 八段
            d8 = soup('div.relation-unit').eq(1)
            links = pq(d8)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院八段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 七段
            d7 = soup('div.relation-unit').eq(2)
            links = pq(d7)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院七段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 六段
            d6 = soup('div.relation-unit').eq(3)
            links = pq(d6)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院六段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 五段
            d5 = soup('div.relation-unit').eq(4)
            links = pq(d5)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院五段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 四段
            d4 = soup('div.relation-unit').eq(5)
            links = pq(d4)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院四段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 三段
            d3 = soup('div.relation-unit').eq(6)
            links = pq(d3)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院三段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 二段
            d2 = soup('div.relation-unit').eq(7)
            links = pq(d2)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院二段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
            # 一段
            d1 = soup('div.relation-unit').eq(8)
            links = pq(d1)('a')
            for link in links:
                lo = pq(link)
                if lo.attr('title') =='待创建':
                    continue
                url = lo.attr('href')
                name = lo.text()
                occupation ='中国棋院一段'
                yield scrapy.Request(url, callback=lambda response,name=name,occupation=occupation:self.parse_page(response, name,occupation))
    
        def parse_page(self,response,name,occupation):
            print(name,occupation)
            item = PersonBaiKeItem()
            item['id'] = get_uuid()
            item['name']=name
            item['url']=response.url
            item['category']='围棋'
            item['detail']=remove_js_css(response.body_as_unicode())
            item['occupation']=occupation
            yield item

去除js、css等无用信息，保存内容

def remove_js_css(content):
        """ #删除web中的head,jss,注释、Css和空行等标签 """
        r = re.compile(r'''<script.*?</script>''', re.I | re.M | re.S)
        sc = r.sub('', content)
        r = re.compile(r'''<style.*?</style>''', re.I | re.M | re.S)
        sc = r.sub('', sc)
        r = re.compile(r'''''', re.I | re.M | re.S)
        sc = r.sub('', sc)
        r = re.compile(r'''<meta.*?>''', re.I | re.M | re.S)
        sc = r.sub('', sc)
        # r = re.compile(r'''<a.*?</a>''', re.I | re.M | re.S)
        # sc = r.sub('', sc)
        r = re.compile(r'''<ins.*?</ins>''', re.I | re.M | re.S)
        sc = r.sub('', sc)
        r = re.compile(r'''^\s+$''', re.M | re.S)
        sc = r.sub('', sc)
        r = re.compile(r'''\n+''', re.M | re.S)
        sc = r.sub('\n', sc)
        return sc

url作为参数传递，如果是中文，则会有问题。添加`url=re.sub(r'%(?!%)', '%%', url)`,解决此问题。

def is_exist_baidu_person(self,url):
            url=re.sub(r'%(?!%)', '%%', url)
            sql = "select * from baidu_person where url='{}'".format(url)
            df = pd.read_sql(sql,self.baidu_engine)
            results = json.loads(df.to_json(orient='records'))
            if len(results) == 0:
                return False
            return True
            # return False
    
        def save_baidu_person(self,data):
            if not self.is_exist_baidu_person(data['url']):
            # 将网页内容保存到文件中
                name = data['name']+data['id'][0:2]
                file_name = self.save_file(name,data['detail'])
                data['detail']=file_name
                # 将文件路径写入到数据库中
                df = pd.DataFrame([data])
                df.to_sql('baidu_person',self.baidu_engine,if_exists='append',index=False)
    
        def save_file(self,name,detail):
            ROOT_PATH = '/Users/dzm/Documents/baidu/weiqi_hg'
            file_name = os.path.join(ROOT_PATH,name)
            file = open(file_name,'w')
            file.writelines(detail)
            file.close()
            print('文件保存成功')
            return file_name

[Link 1]: https://baike.baidu.com/item/%E8%81%82%E5%8D%AB%E5%B9%B3/664143
[1]: /images/20220413/2d4f93a7979f41f5831c867c9b6e0449.png