def parse(self, response):
item = V_qqItem()
urls = []
url_list = response.xpath(
'//*[@id="mod_main_nav"]/div/div/a/@href').extract()
for url in url_list:
if url[0] == '/' and not url[0:4] == "http":
urls.append(f"http://v.qq.com{url}")
url_list2 = response.xpath(
'//*[@class="mod_row_box"]/div[2]/div/div/a/@href').extract()
urls = urls + url_list2
item['urls'] = urls
for i in range(len(item['urls'])):
try:
res = requests.get(item['urls'][i], headers=HEADERS)
if res.ok:
res.encoding = 'utf-8'
html = res.text
ex = Extractor(threshold=3)
content = ex.filter_tags(html)
data = clean_content(ex.getText(content))
with open(f"E:/c++/毕业设计开发日志/06.文本数据集/娱乐/视频/{i}.txt",
'w',
encoding="utf-8") as txtfile:
txtfile.write(data)
print(f"第{i}个网页爬取完毕")
time.sleep(2)
except Exception as e:
print(f"第{i}个文章错误，链接{item['urls'][i]}，错误原因{e}")
print(f"共{i}个视频页面信息爬取完毕")
def parse_url():
urls = []
data = []
length_all = 0
length_ok = 0
with open(r"C:\Users\叫乌鸦的少年怪\Desktop\历史记录文件.txt", 'r',
encoding='utf-8') as files:
urls += ["http://" + url[:-1] for url in files.readlines()]
length_all = len(urls)
j = 0
for url in urls:
try:
ex = Extractor(threshold=30)
html = ex.getHtml(url)
content = ex.getText(ex.filter_tags(html))
content = clean_content(content)
length_ok += 1
if content != "This page has no content to extract ":
j += 1
data_str = ""
for _ in content.splitlines():
data_str += _
data.append(data_str)
else:
pass
except ConnectionError as errr:
print(errr)
# print("响应失败")
# # TODO 记录下host
except Exception as e:
print(e)
continue
with open(r"C:\Users\叫乌鸦的少年怪\Desktop\content.txt", 'w+',
encoding='utf-8') as rlt_txt:
for single_data in data:
rlt_txt.write(single_data + '\n')
print(f"成功访问的{length_ok}")
print(f"一共{length_all}")
print(f'百分比{length_ok / length_all}')
def parse_urls(url_list: list):
j = 0
for i in range(len(url_list)):
try:
extractor = Extractor(threshold=30)
html = extractor.getHtml(url_list[i])
content = extractor.filter_tags(html)
data = clean_content(extractor.getText(content))
if data != "This page has no content to extract ":
j += 1
with open(f'E:/c++/毕业设计开发日志/06.文本数据集/数据清洗模块测试.txt',
'w+',
encoding='utf-8') as txtfile:
txtfile.write(data)
print(f"第{i+1}篇文章处理完毕")
else:
pass
except Exception as e:
print(e)
print(f"共获取到{i}篇文章")
print(f"成功处理{j}篇文章")
from Settings import HEADERS
import json
from data_cleaning.Extractor import Extractor
from data_cleaning.content_clean import clean_content
if __name__ == '__main__':
urls = []
for i in range(5):
response = get(
"https://www.douyu.com/japi/weblist/apinc/rec/list?uid=8b6321ddbef037034b351cab00081501&num=20",
headers=HEADERS)
data_json = json.loads(response.text)
data_url = (data_json['data'])
for data in data_url:
urls.append(f"https://douyu.com/{data['roomId']}")
print(f"共爬取{len(urls)}条房间")
try:
for i in range(len(urls)):
ex = Extractor(threshold=20)
html = get(urls[i], headers=HEADERS).text
content = ex.filter_tags(html)
data = clean_content(ex.getText(content))
with open(f'E:/c++/毕业设计开发日志/06.文本数据集/娱乐/直播/{i}.txt',
'w',
encoding='utf-8') as txtfile:
txtfile.write(data)
print(f"第{i}个直播间处理完毕")
print(f'共{i}个直播间处理完毕')
except Exception as e:
print(e)
def single_content_analyse(self, corpus):
content = get_txt_to_single(corpus)
cut_content = clean_content(content)
result = self.keams_single_analyse(cut_content)
print(result)
from data_cleaning.Extractor import Extractor
from data_cleaning.content_clean import clean_content
cx = Extractor(threshold=90)
html = cx.getHtml(
"https://blog.csdn.net/Winterto1990/article/details/51220307")
content = cx.filter_tags(html)
# print(content)
s = cx.getText(content)
data = clean_content(s)
print(data)
# TODO: 给爬虫新增代理；
# 增加文本输出；
# 文本语料库！！