更新了爬虫第3天的代码

pull/6/head
jackfrued 2018-05-29 21:05:42 +08:00
parent 8adc03157a
commit b8725569cd
4 changed files with 109 additions and 2 deletions

View File

@ -87,7 +87,7 @@
> 说明更多内容可以参考BeautifulSoup的[官方文档]()。
### 例 - 获取知乎发现上的问题链接
### 例 - 获取知乎发现上的问题链接
```Python
from urllib.parse import urljoin

View File

@ -197,5 +197,63 @@ b'admin'
### 实例 - 缓存知乎发现上的链接和页面代码
```Python
from hashlib import sha1
from urllib.parse import urljoin
import pickle
import re
import requests
import zlib
from bs4 import BeautifulSoup
from redis import Redis
def main():
# 指定种子页面
base_url = 'https://www.zhihu.com/'
seed_url = urljoin(base_url, 'explore')
# 创建Redis客户端
client = Redis(host='1.2.3.4', port=6379, password='1qaz2wsx')
# 设置用户代理(否则访问会被拒绝)
headers = {'user-agent': 'Baiduspider'}
# 通过requests模块发送GET请求并指定用户代理
resp = requests.get(seed_url, headers=headers)
# 创建BeautifulSoup对象并指定使用lxml作为解析器
soup = BeautifulSoup(resp.text, 'lxml')
href_regex = re.compile(r'^/question')
# 查找所有href属性以/question打头的a标签
for a_tag in soup.find_all('a', {'href': href_regex}):
# 获取a标签的href属性值并组装完整的URL
href = a_tag.attrs['href']
full_url = urljoin(base_url, href)
# 将URL处理成SHA1摘要(长度固定更简短)
hasher = sha1()
hasher.update(full_url.encode('utf-8'))
field_key = hasher.hexdigest()
# 如果Redis的键'zhihu'对应的hash数据类型中没有URL的摘要就访问页面并缓存
if not client.hexists('zhihu', field_key):
html_page = requests.get(full_url, headers=headers).text
# 对页面进行序列化和压缩操作
zipped_page = zlib.compress(pickle.dumps(html_page))
# 使用hash数据类型保存URL摘要及其对应的页面代码
client.hset('zhihu', field_key, zipped_page)
# 显示总共缓存了多少个页面
print('Total %d question pages found.' % client.hlen('zhihu'))
if __name__ == '__main__':
main()
```

View File

@ -50,7 +50,7 @@ def get_matched_parts(page_html, pattern_str, pattern_ignore_case=re.I):
# 开始执行爬虫程序
def start_crawl(seed_url, match_pattern, *, max_depth=-1):
client = redis.Redis(host='120.77.222.217', port=11223, password='1qaz2wsx')
client = redis.Redis(host='1.2.3.4', port=6379, password='1qaz2wsx')
charsets = ('utf-8', 'gbk', 'gb2312')
logging.info('[Redis ping]', client.ping())
url_list = [seed_url]

View File

@ -0,0 +1,49 @@
from hashlib import sha1
from urllib.parse import urljoin
import pickle
import re
import requests
import zlib
from bs4 import BeautifulSoup
from redis import Redis
def main():
# 指定种子页面
base_url = 'https://www.zhihu.com/'
seed_url = urljoin(base_url, 'explore')
# 创建Redis客户端
client = Redis(host='1.2.3.4', port=6379, password='1qaz2wsx')
# 设置用户代理
headers = {'user-agent': 'Baiduspider'}
# 通过requests模块发送GET请求并指定用户代理
resp = requests.get(seed_url, headers=headers)
# 创建BeautifulSoup对象并指定使用lxml作为解析器
soup = BeautifulSoup(resp.text, 'lxml')
href_regex = re.compile(r'^/question')
# 查找所有href属性以/question打头的a标签
for a_tag in soup.find_all('a', {'href': href_regex}):
# 获取a标签的href属性值并组装完整的URL
href = a_tag.attrs['href']
full_url = urljoin(base_url, href)
# 将URL处理成SHA1摘要(长度固定更简短)
hasher = sha1()
hasher.update(full_url.encode('utf-8'))
field_key = hasher.hexdigest()
# 如果Redis的键'zhihu'对应的hash数据类型中没有URL的摘要就访问页面并缓存
if not client.hexists('zhihu', field_key):
html_page = requests.get(full_url, headers=headers).text
# 对页面进行序列化和压缩操作
zipped_page = zlib.compress(pickle.dumps(html_page))
# 使用hash数据类型保存URL摘要及其对应的页面代码
client.hset('zhihu', field_key, zipped_page)
# 显示总共缓存了多少个页面
print('Total %d question pages found.' % client.hlen('zhihu'))
if __name__ == '__main__':
main()