Python 2.7 在scrapy中提取标记值

我想在scrapy fpr示例中提取xpath标记的值

/html/body/div[3]/ul[1]/li[1]/div/p

q1

ans1

问题2

ans2 链接：

以这样的收益率

 def parse(self, response):
        for quote in response.xpath('//html/body/main'):
            yield {
#question or answer 
#question pattern  li/div/p  or li[1]/div/p
#answer pattern ended with li[2 or higher number]/div/p
#related question and answer both have the same ul for example both are ul[1]
                'type': quote.xpath('i dont know this part').extract_first (),
                'QAnumber': quote.xpath('?').extract(),
                'text': quote.xpath('/html/body/div[3]/*/*/div/p/text()').extract(),


            }

---

我怎样才能提取出这3个部分呢？这3个部分很难理解你的问题。是否要提取问题和答案？会是这样的

from w3lib.html import remove_tags
for qa in response.css('div#product-questions-list ul.c-faq__list'):
    question = qa.css('li.is-question div.section > p::text').get()
    answer = qa.css('li.is-answer div.section > p').get()
    answer = remove_tags(answer) if answer else None
    number = qa.css('li.is-question a::attr(data-question-id)')

---

很难理解你的问题。是否要提取问题和答案？会是这样的

from w3lib.html import remove_tags
for qa in response.css('div#product-questions-list ul.c-faq__list'):
    question = qa.css('li.is-question div.section > p::text').get()
    answer = qa.css('li.is-answer div.section > p').get()
    answer = remove_tags(answer) if answer else None
    number = qa.css('li.is-question a::attr(data-question-id)')

 def parse(self, response):
        for quote in response.xpath('//html/body/main'):
            yield {
#question or answer 
#question pattern  li/div/p  or li[1]/div/p
#answer pattern ended with li[2 or higher number]/div/p
#related question and answer both have the same ul for example both are ul[1]
                'type': quote.xpath('i dont know this part').extract_first (),
                'QAnumber': quote.xpath('?').extract(),
                'text': quote.xpath('/html/body/div[3]/*/*/div/p/text()').extract(),


            }

from w3lib.html import remove_tags
for qa in response.css('div#product-questions-list ul.c-faq__list'):
    question = qa.css('li.is-question div.section > p::text').get()
    answer = qa.css('li.is-answer div.section > p').get()
    answer = remove_tags(answer) if answer else None
    number = qa.css('li.is-question a::attr(data-question-id)')

 def parse(self, response):
     for quote in response.css('#product-questions-list > ul'):
         quest = response.css('.is-question > div.section > div > p::text').extract_first()
         answer = response.css('.is-answer > div.section > div > p::text').extract_first()
         yield {quest: answer}