import sys
import os
import re
import requests
from pyquery import PyQuery as pq
headers = {
"Host": 'm.dianping.com',
'Accept-Encoding': 'gzip',
# "Accept": 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
# 'User-Agent': random.choice(utils.ua_mobile),
'User-Agent': 'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.67 Mobile Safari/537.36'
}
LIST_HEADER = {
"Host": 'm.dianping.com',
"Connection": 'keep-alive',
"Cache-Control": 'max-age=0',
"Upgrade-Insecure-Requests": '1',
"Accept": 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
"Accept-Language": 'zh-CN,zh;q=0.8',
# 'User-Agent': random.choice(utils.ua_mobile[0]),
'User-Agent': 'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.67 Mobile Safari/537.36'
}
header_pinlun = {
'Host': 'www.dianping.com',
'Accept-Encoding': 'gzip',
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.102 Safari/537.36'
}
header_css = {
'Host': 's3plus.meituan.net',
'Accept-Encoding': 'gzip',
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.102 Safari/537.36'
}
# 0-详情页
def get_msg():
"""
url: http://www.dianping.com/shop/+ 商铺ID +/review_all
:return:
"""
# url = "http://www.dianping.com/shop/114461956" # 获取人均和评论及评论数
url = "http://www.dianping.com/shop/96658933/review_all"
# url = 'http://m.dianping.com/shop/17149400/map' # 获取经纬度
html = requests.get(url, headers=header_pinlun)
print("1 ===> STATUS", html.status_code)
# print(html.text)
doc = pq(html.text)
# 解析每条评论
pinglunLi = doc("div.reviews-items > ul > li").items()
"""
调用评论里的css样式处理和加密字体svg处理
:return:
dict_svg_text: svg整个加密字库,以字典形式返回
list_svg_y:svg背景中的<path>标签里的[x,y]坐标轴,以[x,y]形式返回
dict_css_x_y:css样式中,每个加密字体的<span> 标签内容,用于匹配dict_svg_text 中的key,以字典形式返回
"""
dict_svg_text, dict_css_x_y = css_get(doc)
for data in pinglunLi:
# 用户名
userName = data("div.main-review > div.dper-info > a").text()
# 用户ID链接
userID = "http://www.dianping.com" + data("div.main-review > div.dper-info > a").attr("href")
# 用户评分星级[10-50]
startShop = str(data("div.review-rank > span").attr("class")).split(" ")[1].replace("sml-str", "")
# 用户描述:机器:非常好 环境:非常好 服务:非常好 人均:0元
describeShop = data("div.review-rank > span.score").text()
# 关键部分,评论HTML,待处理,评论包含隐藏部分和直接展示部分,默认从隐藏部分获取数据,没有则取默认部分。(查看更多)
pinglun = data("div.review-words.Hide").html()
try:
len(pinglun)
except:
pinglun = data("div.review-words").html()
# 该用户喜欢的美食
loveFood = data("div.main-review > div.review-recommend").text()
# 发表评论的时间
pinglunTime = data("div.main-review > div.misc-info.clearfix > span.time").text()
print("userName:", userName)
print("userID:", userID)
print("startShop:", startShop)
print("describeShop:", describeShop)
print("loveFood:", loveFood)
print("pinglunTime:", pinglunTime)
print("pinglun:", css_decode(dict_css_x_y, dict_svg_text, pinglun))
print("*"*100)
# 1-评论隐含部分字体css样式, 获取svg链接,获取加密汉字background
def css_get(doc):
css_link = "http:"+doc("head > link:nth-child(11)").attr("href")
background_link = requests.get(css_link, headers=header_css)
r = r'background-image: url(.*?);'
matchObj = re.compile(r, re.I)
# svg_link = matchObj.findall(background_link.text)[2].replace(")", "").replace("(", "http:") # 会匹配出3个url,需要选取文字最多的,出错可能是这。
svg_links = matchObj.findall(background_link.text)
svg_link = find_font_svg_link(svg_links)
"""
svg_text() 方法:请求svg字库,并抓取加密字
dict_svg_text: svg整个加密字库,以字典形式返回
list_svg_y:svg背景中的<path>标签里的[x,y]坐标轴,以[x,y]形式返回
"""
dict_avg_text = svg_text(svg_link)
"""
css_dict() 方法:生成css样式中background的样式库
dict_css: 返回css字典样式
"""
dict_css = css_dict(background_link.text)
return dict_avg_text, dict_css
def find_font_svg_link(svg_links):
link_url = None
for link in svg_links:
link_url = link.replace(")", "").replace("(", "http:")
res = requests.get(link_url)
text = res.text
height = re.findall('height="(.*?)px"', text)
if height:
height = float(height[0])
if height > 2000:
break
return link_url
# 2-字体库链接
def svg_text(url):
html = requests.get(url)
dict_svg = svg_dict(html.text)
return dict_svg
# 3-生成svg字库字典
def svg_dict(csv_html):
svg_text_r = r'<text.*y="(.*?)">(.*?)</text>'
svg_text_re = re.findall(svg_text_r, csv_html)
dict_avg = {}
if svg_text_re:
# 生成svg加密字体库字典
for data in svg_text_re:
dict_avg[int(data[0])] = list(data[1])
return dict_avg
else:
svg_text_r = r'<textPath xlink:href="#(.*?)" textLength="(.*?)">(.*?)</textPath>'
svg_text_re = re.findall(svg_text_r, csv_html)
dict_1 = {}
for data in svg_text_re:
dict_1[int(data[0])] = list(data[2])
svg_y_r = r'<path id="(.*?)" d="(.*?) (.*?) (.*?)"/>'
svg_y_re = re.findall(svg_y_r, csv_html)
for data in svg_y_re:
if int(data[0]) in dict_1:
dict_avg[int(data[2])] = dict_1[int(data[0])]
return dict_avg
# 4-生成css字库字典
def css_dict(html):
css_text_r = r'.(.*?){background:(.*?)px (.*?)px;}'
css_text_re = re.findall(css_text_r, html)
dict_css = {}
for data in css_text_re:
"""
加密字库.gqi4j {background: -98.0px -745.0px;}与svg文件对应关系,x/14,就是svg文件加密字体下标
y,原样返回,需要在svg函数中做处理
"""
x = int(float(data[1])/-14)
"""
字典参数:{css参数名:(background-x,background-y,background-x/14,background-y)}
"""
dict_css[data[0]] = (x, int(float(data[2]) * -1))
return dict_css
# 5-最终评论汇总
def css_decode(decode_css, decode_svg, pinglun_html):
"""
:param css_html: css 的HTML源码
:param svg_dict: svg加密字库的字典
:param svg_list: svg加密字库对应的坐标数组[x, y]
:param pinglun_html: 评论的HTML源码,对应0-详情页的评论,在此处理
:return: 最终合成的评论
"""
css_dict_text = decode_css
svg_dict_text = decode_svg
pinglun_text = re.sub('<span class="|"/>|">|<br/>|<img class="emoji-img".*?/>|<div class="less-words[\d\D]*?</div>', ',', pinglun_html).strip()
pinglun_list = [x for x in pinglun_text.split(",") if x != '']
pinglun_str = []
for msg in pinglun_list:
if msg in css_dict_text:
x = int(css_dict_text[msg][0])
y = int(css_dict_text[msg][1])
for g in svg_dict_text:
if y < g:
pinglun_str.append(svg_dict_text[g][x])
break
else:
pinglun_str.append(msg.replace("\n", ""))
str_pinglun = ""
for x in pinglun_str:
str_pinglun += x
return str_pinglun
if __name__ == '__main__':
get_msg()
大众点评店铺评论信息获取
最新推荐文章于 2024-07-04 08:27:48 发布