import os from lxml import etree import json import re target_file_path = r"E:\yuxin\nuofang-data\structure\0420\webpage" xpath_parse_dict = { 'title': "//div[@class='good-info-box']/div[@class='top cls']/span/text()", 'base_info': "//div[@class='tab '][1]/div[@id='tab1_info']/div[@class='cItem active']/div[@class='layui-row txt']/div[1]/p", 'leader_info': "//div[@class='good-info-box']/div[@class='tab '][1]/div[@id='tab1_info']/div[@class='cItem active']/div[@id='leader_list']//tbody/tr", 'website_info': "//div[@class='good-info-box']/div[@class='tab '][2]/div[@class='tabInfo sourcetabInfo']/div", } def trim_space(input_str): result = re.sub(' {2,}', ' ', input_str) return result def parse_base_info(current_selector): base_info_line = current_selector.xpath(xpath_parse_dict['base_info']) base_info = {} for line in base_info_line: line_text = ''.join(line.xpath(".//text()")) label = line_text.split('：')[0].strip() value = ''.join(line_text.split('：')[1:]).strip() base_info[label] = trim_space(value) # if current_title != '' and len(base_info) > 0: # print("[No. {}] {}: {}".format(count, current_title, json.dumps(base_info, ensure_ascii=False))) # pass # else: # print("[No. {}] {} 解析不成功".format(count, _path)) return base_info def parse_leader_info(current_selector): leader_info_line = current_selector.xpath(xpath_parse_dict['leader_info']) leader_info_list = [] ths = [] for i in range(len(leader_info_line)): leader_info = {} line = leader_info_line[i] if i == 0: ths = line.xpath("./th/text()") else: for th in ths: leader_info[th] = '' tds = line.xpath("./td") for tdi in range(len(tds)): td_txt = ''.join(tds[tdi].xpath(".//text()")) leader_info[ths[tdi]] = trim_space(td_txt.strip()) leader_info_list.append(leader_info) # if current_title != '' and len(leader_info_list) > 0: # print("[No. {}] {}: {}".format(count, current_title, json.dumps(leader_info_list, ensure_ascii=False))) # pass return leader_info_list def parse_website_info(current_selector): tab_name_xpath = "//div[@class='good-info-box']/div[@class='tab ']/div[@id='tabNav']/p/text()" tab_names = current_selector.xpath(tab_name_xpath) if len(tab_names) <= 0: return [] website_info_tab = current_selector.xpath("//div[@class='good-info-box']/div[@class='tab ']/div[@id='tabInfo']/div") website_info = {} for i in range(len(website_info_tab)): tab_dict = {} tab_name = tab_names[i] tab_content = website_info_tab[i] overview_box = tab_content.xpath("./dl[@class='overview box']") if overview_box is not None and len(overview_box) > 0: overview_info = overview_box[0].xpath(".//p/text()") tab_dict["概述"] = trim_space(''.join(overview_info).strip()) other_box = tab_content.xpath("./div") if other_box is not None and len(other_box) > 0: for box in other_box: subtitle_info = box.xpath(".//h2/text()") if len(subtitle_info) > 0: subtitle = ''.join(subtitle_info).replace("：", "").strip() if subtitle not in ['入驻单位', '主要设施']: content_info = box.xpath(".//text()") content = ''.join([_.strip() for _ in content_info]) tab_dict[subtitle] = trim_space(content) website_info[tab_name] = tab_dict # if current_title != '' and len(website_info) > 0: # print("[No. {}] {}: {}".format(count, current_title, json.dumps(website_info, ensure_ascii=False))) # pass return website_info def parse_contact_info(current_selector): tab_name_xpath = "//div[@class='good-info-box']/div[@id='tabContact']/div[@class='tabNav nav2 cls']/p/text()" tab_names = current_selector.xpath(tab_name_xpath) if len(tab_names) <= 0: return [] contact_info_line = current_selector.xpath( "//div[@class='good-info-box']/div[@id='tabContact']/div[@id='tabContactText']/div//p") contact_info = {} for i in range(len(contact_info_line)): contact_item = contact_info_line[i] contact_item_text = "".join(contact_item.xpath(".//text()")) contact_item_name = contact_item_text.split("：")[0] contact_item_value = "".join(contact_item_text.split("：")[1:]) contact_info[contact_item_name] = contact_item_value # if current_title != '' and len(website_info) > 0: # print("[No. {}] {}: {}".format(count, current_title, json.dumps(website_info, ensure_ascii=False))) # pass return contact_info if __name__ == '__main__': count = 0 for _path in os.listdir(target_file_path): if _path.endswith('.html'): # if _path.endswith('美国联邦调查局（FBI）_组织结构_全球军事态势情报数据库.html'): count += 1 webpage_info = {} target_file = open(target_file_path + '\\' + _path, mode='r', encoding='utf-8') html_content = target_file.read().replace('\n', '') target_file.close() selector = etree.HTML(html_content) title_info = selector.xpath(xpath_parse_dict['title']) title = ''.join(title_info).strip() url_info = selector.xpath('//ul[@class="osc-dropMenu"]/li[last()]/a/@href')[0] webpage_info["url_info"] = url_info.replace("#", "") # demo_info = parse_contact_info(selector) # if demo_info is not None and len(demo_info) > 0: # print("[No. {}] {} 读取完成，{}".format(count, title, json.dumps(demo_info, ensure_ascii=False))) # pass # else: # print("[No. {}] {} 内容为空".format(count, title)) base_info = parse_base_info(selector) if len(base_info) > 0: webpage_info["base_info"] = base_info leader_info = parse_leader_info(selector) if len(leader_info) > 0: webpage_info["leader_info"] = leader_info website_info = parse_website_info(selector) if len(website_info) > 0: webpage_info["website_info"] = website_info contact_info = parse_contact_info(selector) if len(contact_info) > 0: webpage_info["contact_info"] = contact_info result_file_name = '%s-' % count + title.replace("/", "-") + '.json' result_file_path = 'E:/yuxin/nuofang-data/structure/result0512/' + result_file_name result_file = open(result_file_path, 'w', encoding='utf-8') result_file.write(json.dumps(webpage_info, ensure_ascii=False)) result_file.close() print("[No. {}] {} 写入完成，原始url是{}".format(count, title, webpage_info["url_info"]))