链接:http://data.eastmoney.com/bbsj/201903/yjbb.html
字体方块
后缀.woff是字体
数字是乱码,需要把乱码改成数字
url="http://dcfm.eastmoney.com/em_mutisvcexpandinterface/api/js"
headers={
"Referer": "http://data.eastmoney.com/bbsj/201903/yjbb.html",
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36"
}
data={
"type": "YJBB21_YJBB",
"token": "70f12f2f4f091e459a279469fe49eca5",
"st": "latestnoticedate",
"sr": "-1",
"p": "1",
"ps": "50",
"js": "var rIKyuApR={pages:(tp),data: (x),font:(font)}",
"filter": "(securitytypecode in (\'058001001\',\'058001002\'))(reportdate=^2018-09-30^)",
"rt": "51838114"
}
r=requests.post(url=url,headers=headers,data=data)
r.encoding=r.apparent_encoding
print(r.text)
r.encoding=r.apparent_encoding
print(r.text)
org_data=r.text[r.text.index("{"):] json_str = or_data.replace(\'data:\', \'"data":\').replace(\'pages:\', \'"pages":\').replace(\'font:\', \'"font":\') json_data = json.loads(json_str)
#字体文件
font_url = json_data[\'font\'][\'WoffUrl\']
数据结果:
字体
cmap表里记录有unicode索引和文字关系https://www.cnblogs.com/shenyiyangle/p/10700156.html
自定义字体的unicode索引和字形映射与标准不一样
解决方法
全局替换
TTFont解析字体
font = requests.get(font_url, headers=header, timeout=30)
font_name = font_url.split("/")[-1]
with codecs.open(font_name, \'wb\') as f:
f.write(font.content)
font_map = TTFont(font_name).getBestCmap()#用getBestCmap()可以得到字体index和glphy的映射关系
“”“{120: \'x\',
57960: \'bgldyy\',
57971: \'qqdwzl\',
58817: \'whyhyx\',
59299: \'wqqdzs\',
60397: \'zbxtdyc\',
60633: \'zwdxtdy\',
60650: \'zrwqqdl\',
61125: \'bdzypyc\',
62069: \'sxyzdxn\',
62669: \'nhpdjl\'}”“”
font_map里是unicode和字形的关系,key是unicode的十进制数字,value表示字形标记,index转换成unicode:
font_index = [hex(key).upper().replace(\'0X\', \'&#x\') +\';\' for key in font_map.keys()]
“”“[\'x\',
\'\',
\'\',
\'\',
\'\',
\'\',
\'\',
\'\',
\'\',
\'\',
\'\']”“”
unicode索引和文字的关系可以需要手动找,但东方财富网页面上有明文。
font_mapping = json_data[\'font\'][\'FontMapping\']
replace_dict= {i[\'code\']: str(i[\'value\']) for i in font_map}
”“”
{\'\': \'7\',
\'\': \'1\',
\'\': \'9\',
\'\': \'3\',
\'\': \'4\',
\'\': \'8\',
\'\': \'2\',
\'\': \'0\',
\'\': \'6\',
\'\': \'5\'}
”“”
#全局替换
for k, v in replace_dict.items():
json_str = json_str.replace(k, v)
finall_data = json.loads(json_str)
结果: