Commit b3a88734 authored by jackfrued's avatar jackfrued

添加了爬虫第2天代码

parent e4204ed9
......@@ -4,8 +4,9 @@
1. 下载数据 - urllib / requests / aiohttp。
2. 解析数据 - re / lxml / beautifulsoup4(bs4)/ pyquery。
3. 持久化 - pymysql / redis / sqlalchemy / peewee / pymongo。
4. 调度器 - 进程 / 线程 / 协程。
3. 缓存和持久化 - pymysql / redis / sqlalchemy / peewee / pymongo。
4. 序列化和压缩 - pickle / json / zlib。
5. 调度器 - 进程 / 线程 / 协程。
### HTML页面分析
......@@ -86,3 +87,40 @@
> 说明:更多内容可以参考BeautifulSoup的[官方文档]()。
### 例子 - 获取知乎发现上的问题链接
```Python
from urllib.parse import urljoin
import re
import requests
from bs4 import BeautifulSoup
def main():
headers = {'user-agent': 'Baiduspider'}
proxies = {
'http': 'http://122.114.31.177:808'
}
base_url = 'https://www.zhihu.com/'
seed_url = urljoin(base_url, 'explore')
resp = requests.get(seed_url,
headers=headers,
proxies=proxies)
soup = BeautifulSoup(resp.text, 'lxml')
href_regex = re.compile(r'^/question')
link_set = set()
for a_tag in soup.find_all('a', {'href': href_regex}):
if 'href' in a_tag.attrs:
href = a_tag.attrs['href']
full_url = urljoin(base_url, href)
link_set.add(full_url)
print('Total %d question pages found.' % len(link_set))
if __name__ == '__main__':
main()
```
## 存储数据
### 缓存和持久化
### 数据缓存
通过前面章节的内容,我们已经知道了如何从指定的页面中抓取数据,以及如何保存抓取的结果,但是我们没有考虑过这么一种情况,就是我们可能需要从已经抓取过的页面中提取出更多的数据,重新去下载这些页面对于规模不大的网站倒是问题也不大,但是如果能够把这些页面缓存起来,对应用的性能会有明显的改善。
### 使用NoSQL
#### Redis
#### Mongo
### 磁盘文件缓存
### 数据库缓存
......@@ -13,7 +13,7 @@ def main():
</head>
<body>
<h1>Hello, world!</h1>
<p>这是一个神奇的网站!</p>
<p>这是一个<em>神奇</em>的网站!</p>
<hr>
<div>
<h2>这是一个例子程序</h2>
......@@ -44,14 +44,28 @@ def main():
print(soup.title)
# JavaScript - document.body.h1
print(soup.body.h1)
print(soup.find_all(re.compile(r'^h')))
print(soup.p)
print(soup.body.p.text)
print(soup.body.p.contents)
for p_child in soup.body.p.children:
print(p_child)
print(len([elem for elem in soup.body.children]))
print(len([elem for elem in soup.body.descendants]))
print(soup.findAll(re.compile(r'^h[1-6]')))
print(soup.body.find_all(r'^h'))
print(soup.body.div.find_all(re.compile(r'^h')))
print(soup.find_all(re.compile(r'r$')))
print(soup.find_all('img', {'src': re.compile(r'\./img/\w+.png')}))
print(soup.find_all(lambda x: len(x.attrs) == 2))
print(soup.find_all(foo))
print(soup.find_all('p', {'class': 'foo'}))
for elem in soup.select('a[href]'):
print(elem.attrs['href'])
def foo(elem):
return len(elem.attrs) == 2
if __name__ == '__main__':
main()
from urllib.parse import urljoin
import re
import requests
from bs4 import BeautifulSoup
def main():
headers = {'user-agent': 'Baiduspider'}
proxies = {
'http': 'http://122.114.31.177:808'
}
base_url = 'https://www.zhihu.com/'
seed_url = urljoin(base_url, 'explore')
resp = requests.get(seed_url,
headers=headers,
proxies=proxies)
soup = BeautifulSoup(resp.text, 'lxml')
href_regex = re.compile(r'^/question')
link_set = set()
for a_tag in soup.find_all('a', {'href': href_regex}):
if 'href' in a_tag.attrs:
href = a_tag.attrs['href']
full_url = urljoin(base_url, href)
link_set.add(full_url)
print('Total %d question pages found.' % len(link_set))
if __name__ == '__main__':
main()
from urllib.error import URLError
from urllib.request import urlopen
import re
import redis
import ssl
import hashlib
import logging
import pickle
import zlib
# Redis有两种持久化方案
# 1. RDB
# 2. AOF
# 通过指定的字符集对页面进行解码(不是每个网站都将字符集设置为utf-8)
def decode_page(page_bytes, charsets=('utf-8',)):
page_html = None
for charset in charsets:
try:
page_html = page_bytes.decode(charset)
break
except UnicodeDecodeError:
pass
# logging.error('[Decode]', err)
return page_html
# 获取页面的HTML代码(通过递归实现指定次数的重试操作)
def get_page_html(seed_url, *, retry_times=3, charsets=('utf-8',)):
page_html = None
try:
if seed_url.startswith('http://') or \
seed_url.startswith('https://'):
page_html = decode_page(urlopen(seed_url).read(), charsets)
except URLError as err:
logging.error('[URL]', err)
if retry_times > 0:
return get_page_html(seed_url, retry_times=retry_times - 1,
charsets=charsets)
return page_html
# 从页面中提取需要的部分(通常是链接也可以通过正则表达式进行指定)
def get_matched_parts(page_html, pattern_str, pattern_ignore_case=re.I):
pattern_regex = re.compile(pattern_str, pattern_ignore_case)
return pattern_regex.findall(page_html) if page_html else []
# 开始执行爬虫程序并对指定的数据进行持久化操作
def start_crawl(seed_url, match_pattern, *, max_depth=-1):
client = redis.Redis(host='120.77.222.217', port=11223, password='1qaz2wsx')
charsets = ('utf-8', 'gbk', 'gb2312')
logging.info('[Redis ping]', client.ping())
url_list = [seed_url]
visited_url_list = {seed_url: 0}
while url_list:
current_url = url_list.pop(0)
depth = visited_url_list[current_url]
if depth != max_depth:
page_html = get_page_html(current_url, charsets=charsets)
links_list = get_matched_parts(page_html, match_pattern)
for link in links_list:
if link not in visited_url_list:
visited_url_list[link] = depth + 1
page_html = get_page_html(link, charsets=charsets)
if page_html:
hasher = hashlib.md5()
hasher.update(link.encode('utf-8'))
zipped_page = zlib.compress(pickle.dumps(page_html))
client.set(hasher.hexdigest(), zipped_page)
def main():
ssl._create_default_https_context = ssl._create_unverified_context
start_crawl('http://sports.sohu.com/nba_a.shtml',
r'<a[^>]+test=a\s[^>]*href=["\'](.*?)["\']',
max_depth=2)
if __name__ == '__main__':
main()
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment