我想尝试在 Scrapy 中自动化我的 html 表格抓取。这是我到目前为止所拥有的:import scrapyimport pandas as pdclass XGSpider(scrapy.Spider): name = 'expectedGoals' start_urls = [ 'https://fbref.com/en/comps/9/schedule/Premier-League-Scores-and-Fixtures', ] def parse(self, response): matches = [] for row in response.xpath('//*[@id="sched_ks_3232_1"]//tbody/tr'): match = { 'home': row.xpath('td[4]//text()').extract_first(), 'homeXg': row.xpath('td[5]//text()').extract_first(), 'score': row.xpath('td[6]//text()').extract_first(), 'awayXg': row.xpath('td[7]//text()').extract_first(), 'away': row.xpath('td[8]//text()').extract_first() } matches.append(match) x = pd.DataFrame( matches, columns=['home', 'homeXg', 'score', 'awayXg', 'away']) yield x.to_csv("xG.csv", sep=",", index=False)它工作正常,但是如您所见,我正在对对象的键(home、homeXg等)进行硬编码match。我想自动将键抓取到列表中,然后用所述列表中的键初始化字典。问题是,我不知道如何按索引遍历 xpath。举个例子, headers = [] for row in response.xpath('//*[@id="sched_ks_3260_1"]/thead/tr'): yield{ 'first': row.xpath('th[1]/text()').extract_first(), 'second': row.xpath('th[2]/text()').extract_first() }是否可以将th[1]、th[2]等th[3]插入 for 循环,将数字作为索引,并将值附加到列表中?例如row.xpath('th[i]/text()').extract_first() ?
1 回答

弑天下
TA贡献1818条经验 获得超8个赞
未经测试但应该可以工作:
column_index = 1
columns = {}
for column_node in response.xpath('//*[@id="sched_ks_3260_1"]/thead/tr/th'):
column_name = column_node.xpath('./text()').extract_first()
columns[column_name] = column_index
column_index += 1
matches = []
for row in response.xpath('//*[@id="sched_ks_3232_1"]//tbody/tr'):
match = {}
for column_name in columns.keys():
match[column_name] = row.xpath('./td[{index}]//text()'.format(index=columns[column_name])).extract_first()
matches.append(match)
添加回答
举报
0/150
提交
取消