def get_img(self,
url,
headers=None,
cookies=None,
timeout=60,
verify=False,
proxies=None,
allow_redirects=True,
params=None):
"""
get方式获取 img 二进制信息
:param url: 访问Url
:param headers: 请求头
:param cookies: 请求cookies
:param timeout: 超时时间
:param verify: ssl验证
:param proxies: 代理
:param allow_redirects: 是否允许重定向
:param encoding: 返回的html编码s
:param params: 查询请求参数
:return: 二进制图片数据
"""
if self.session:
r = self.session.get(url,
headers=odict(headers),
cookies=cookies,
timeout=timeout,
verify=verify,
proxies=proxies,
allow_redirects=allow_redirects,
params=params)
else:
r = requests.get(url,
headers=odict(headers),
cookies=cookies,
timeout=timeout,
verify=verify,
proxies=proxies,
allow_redirects=allow_redirects,
params=params)
r.raise_for_status()
return r.content
def classify(in_x, groups, labels, k):
"""分类"""
# 计算欧式距离
gl = array_len(groups)
tmp = tile(in_x, (gl, 1)) - groups
tmp = exponential_operation(tmp, 2)
tmp = array_sum(tmp)
tmp = exponential_operation(tmp, 0.5)
# 得到排序后的数组的索引
arg = argsort(tmp)
# 计算最相似数据的前k个数据的分类次数
cc = odict()
for i in range(k):
# 获得类别次数
la = labels[arg[i]]
cc[la] = cc.get(la, 0) + 1
return max(cc) # 返回最相似数据的前k个数据中出现次数最多的分类作为新数据的分类
def post(self,
url,
headers=None,
cookies=None,
timeout=60,
form_data=None,
verify=False,
proxies=None,
allow_redirects=True,
encoding='utf-8',
params=None,
stream=False):
"""
模拟http post请求
:param url: 访问Url
:param headers: 请求头
:param cookies: 请求cookies
:param timeout: 超时时间
:param verify: ssl验证
:param proxies: 代理
:param allow_redirects: 是否允许重定向
:param encoding: 返回的html编码
:param params: 查询请求参数
:param stream: 是否为流数据
:return: (html, 响应头，响应cookie，访问历史，响应时间（秒）) 或者 流数据
"""
return self.__select('post',
url,
headers=odict(headers),
cookies=cookies,
timeout=timeout,
form_data=form_data,
verify=verify,
proxies=proxies,
allow_redirects=allow_redirects,
encoding=encoding,
params=params,
stream=stream)
def __select(self,
method,
url,
headers=None,
cookies=None,
timeout=60,
verify=False,
proxies=None,
allow_redirects=True,
encoding='utf-8',
params=None,
form_data=None,
stream=False):
"""
对http动作的封装，当传递get，则模拟http get请求，
当传递post, 则模拟http post请求
:param method: 'get' or 'post'
:param url: 访问Url
:param headers: 请求头
:param cookies: 请求cookies
:param timeout: 超时时间
:param verify: ssl验证
:param proxies: 代理
:param allow_redirects: 是否允许重定向
:param encoding: 返回的html编码
:param params: 查询请求参数
:param form_data: 如果是post请求则需要提供有表单提交则需要有这个
:param stream: 是否为流数据
:return: (html, 响应头，响应cookie，访问历史，响应时间（秒）) 或者 流数据
"""
# 增加去重功能，开启后，如若不在history中则，则允许访问，并记住访问过此url
# 下次重复访问时则抛出异常
self.filter_duplicate(url)
r = None
if method == 'get':
if self.session:
r = self.session.get(url,
headers=odict(headers),
cookies=cookies,
timeout=timeout,
verify=verify,
proxies=proxies,
allow_redirects=allow_redirects,
params=params)
else:
r = requests.get(url,
headers=odict(headers),
cookies=cookies,
timeout=timeout,
verify=verify,
proxies=proxies,
allow_redirects=allow_redirects,
params=params)
elif method == 'post':
if self.session:
r = self.session.post(url,
headers=odict(headers),
cookies=cookies,
timeout=timeout,
data=form_data,
verify=verify,
proxies=proxies,
allow_redirects=allow_redirects,
params=params)
else:
r = requests.post(url,
headers=odict(headers),
cookies=cookies,
timeout=timeout,
data=form_data,
verify=verify,
proxies=proxies,
allow_redirects=allow_redirects,
params=params)
else:
raise Exception('unsupported http method')
# 若http状态码如果不正常则抛出异常
r.raise_for_status()
r.encoding = encoding # 设置html编码
if stream:
# 如果为流式数据
return r
return r.text, r.headers, dict_from_cookiejar(
r.cookies), r.history, r.elapsed.microseconds / 1000000