【爬虫实战】Python爬取知网文献信息

引言

临近毕业季，想必很多今年毕业的朋友们最近都在焦头烂额地忙着撰写论文吧。那么如何高效地了解研究领域的热点问题，学习优秀论文解决问题的思路和方法呢？我们可以使用“知网”这个众所周知的平台来进行文献的检索与阅读。使用python可以更加有效地实现文献信息的爬取。通过快速浏览相关领域文献的基本信息，我们可以从中找出感兴趣的值得我们深入研究的文章再来进行精读，极大地提高了效率。

01 选择合适的待爬取网站

我们日常使用知网的网址为：https://www.cnki.net/。但是该网址难以获取网页源代码。右键“查看网页源代码”后会跳转到如下的页面：

但是，我们发现可以从知网空间获取网页源代码。知网空间是知网的一个搜索入口，常用于文献的快速检索，网址为：https://search.cnki.com.cn/。

02 目标页面分析

以“金融科技”为例，点击搜索，对页面进行分析，按F12选择Fetch/XHR。我们找到了如下发送的请求，且发现是以post方式发送。

且携带的参数为：‘searchType’: ‘MulityTermsSearch’, ‘Article Type’: ‘’, ‘ReSearch’: ‘’, ‘ParamIsNullOrEmpty’: ‘false’, ‘Islegal’: ‘false’, ‘Content’: ‘金融科技’, ‘Theme’: ‘’, ‘Title’: ‘’, ‘KeyWd’: ‘’, ‘Author’: ‘’, ‘SearchFund’: ‘’, ‘Originate’: ‘’, ‘Summary’: ‘’, ‘PublishTimeBegin’: ‘’, ‘PublishTimeEnd’: ‘’, ‘MapNumber’: ‘’, ‘Name’: ‘’, ‘Issn’: ‘’, ‘Cn’: ‘’, ‘Unit’: ‘’, ‘Public’: ‘’, ‘Boss’: ‘’, ‘FirstBoss’: ‘’, ‘Catalog’: ‘’, ‘Reference’: ‘’, ‘Speciality’: ‘’, ‘Type’: ‘’, ‘Subject’: ‘’, ‘SpecialityCode’: ‘’, ‘UnitCode’: ‘’, ‘Year’: ‘’, ‘AcefuthorFilter’: ‘’, ‘BossCode’: ‘’, ‘Fund’: ‘’, ‘Level’: ‘’, ‘Elite’: ‘’, ‘Organization’: ‘’, ‘Order’: ‘1’, ‘Page’: ‘1’, ‘PageIndex’: ‘’, ‘ExcludeField’: ‘’, ‘ZtCode’: ‘’, ‘Smarts’: ‘’,

03 获取相应内容


base_url = 'http://search.cnki.com.cn/Search/ListResult'
headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/110.0.0.0 Safari/537.36',

}


def get_page_text(url, headers, search_word, page_num):
    data = {
        'searchType': 'MulityTermsSearch',
        'ArticleType': '',
        'ReSearch': '',
        'ParamIsNullOrEmpty': 'false',
        'Islegal': 'false',
        'Content': search_word,
        'Theme': '',
        'Title': '',
        'KeyWd': '',
        'Author': '',
        'SearchFund': '',
        'Originate': '',
        'Summary': '',
        'PublishTimeBegin': '',
        'PublishTimeEnd': '',
        'MapNumber': '',
        'Name': '',
        'Issn': '',
        'Cn': '',
        'Unit': '',
        'Public': '',
        'Boss': '',
        'FirstBoss': '',
        'Catalog': '',
        'Reference': '',
        'Speciality': '',
        'Type': '',
        'Subject': '',
        'SpecialityCode': '',
        'UnitCode': '',
        'Year': '',
        'AcefuthorFilter': '',
        'BossCode': '',
        'Fund': '',
        'Level': '',
        'Elite': '',
        'Organization': '',
        'Order': '1',
        'Page': str(page_num),
        'PageIndex': '',
        'ExcludeField': '',
        'ZtCode': '',
        'Smarts': '',
    }

    response = requests.post(url=url, headers=headers, data=data)
    page_text = response.text
    return page_text

04 解析内容

这里我们运用Xpath进行解析：

代码如下：


def list_to_str(my_list):
    my_str = "".join(my_list)
    return my_str

def get_abstract(url):
    response = requests.get(url=url, headers=headers)
    page_text = response.text
    tree = etree.HTML(page_text)
    abstract = tree.xpath('//div[@class="xx_font"]//text()')
    return abstract

def parse_page_text(page_text):
    tree = etree.HTML(page_text)
    item_list = tree.xpath('//div[@class="list-item"]')
    page_info = []
    for item in item_list:
        # 标题
        title = list_to_str(item.xpath(
            './p[@class="tit clearfix"]/a[@class="left"]/@title'))
        # 链接
        link = 'https:' +\
            list_to_str(item.xpath(
                './p[@class="tit clearfix"]/a[@class="left"]/@href'))
        # 作者
        author = list_to_str(item.xpath(
            './p[@class="source"]/span[1]/@title'))
        # 出版日期
        date = list_to_str(item.xpath(
            './p[@class="source"]/span[last()-1]/text() | ./p[@class="source"]/a[2]/span[1]/text() '))
        # 关键词
        keywords = list_to_str(item.xpath(
            './div[@class="info"]/p[@class="info_left left"]/a[1]/@data-key'))
        # 摘要
        abstract = list_to_str(get_abstract(url=link))
        # 文献来源
        paper_source = list_to_str(item.xpath(
            './p[@class="source"]/span[last()-2]/text() | ./p[@class="source"]/a[1]/span[1]/text() '))
        # 文献类型
        paper_type = list_to_str(item.xpath(
            './p[@class="source"]/span[last()]/text()'))
        # 下载量
        download = list_to_str(item.xpath(
            './div[@class="info"]/p[@class="info_right right"]/span[@class="time1"]/text()'))
        # 被引量
        refer = list_to_str(item.xpath(
            './div[@class="info"]/p[@class="info_right right"]/span[@class="time2"]/text()'))

        item_info = [i.strip() for i in [title, author, paper_source, paper_type, date, abstract, keywords, download, refer, link]]
        page_info.append(item_info)
        print(page_info)
    return page_info

运行结果如下：

05 保存数据

这里我们将数据保存至excel表格中，并且实现，每个搜索词都在单独的一个sheet中，具体代码如下：


def write_to_excel(workbook, info,  search_word):

    wb = workbook
    worksheet1 = wb.add_worksheet(search_word)  # 创建子表
    worksheet1.activate()  # 激活表

    title = ['title', 'author', 'paper_source', 'paper_type', 'date', 'abstract', 'keywords', 'download', 'refer', 'link']  # 设置表头
    worksheet1.write_row('A1', title)  # 从A1单元格开始写入表头

    i = 2  # 从第二行开始写入数据
    for j in range(len(info)):
        insert_data = info[j]
        start_pos = 'A' + str(i)
        # print(insert_data)
        worksheet1.write_row(start_pos, insert_data)
        i += 1
    return True