有哪些做婚礼电子请柬的网站,专业建设外贸网站制作江门,养老院服务质量建设专项网站,缪斯设计公司做设计师发展怎么样利用 Python-user-agents 解析 User_Agen
需求分析
近期在尝试做一个登录日志的功能#xff0c;及用户登录成功后我在后台进行一个用户的登录记录#xff0c;两种解决方案#xff1a;
由前端得到用户的手机型号#xff0c;我在后台接收后在数据库进行保存使用User_Agent…利用 Python-user-agents 解析 User_Agen
需求分析
近期在尝试做一个登录日志的功能及用户登录成功后我在后台进行一个用户的登录记录两种解决方案
由前端得到用户的手机型号我在后台接收后在数据库进行保存使用User_Agent 它通过解析(浏览器/HTTP) user agent 字符串提供了一种简单的方法来识别/检测手机、平板等设备及其功能。目标是可靠地检测设备是手机平板还是电脑是否有触摸屏。
用法
各种基本信息可以帮忙识别访问者比如设备操作系统浏览器等属性 from user_agents import parse# iPhones user agent string
ua_string Mozilla/5.0 (iPhone; CPU iPhone OS 5_1 like Mac OS X) AppleWebKit/534.46 (KHTML, like Gecko) Version/5.1 Mobile/9B179 Safari/7534.48.3
user_agent parse(ua_string) # 解析成user_agent# Accessing user agents browser attributes
user_agent.browser # returns Browser(familyuMobile Safari, version(5, 1), version_string5.1)
user_agent.browser.family # returns Mobile Safari
user_agent.browser.version # returns (5, 1)
user_agent.browser.version_string # returns 5.1# Accessing user agents operating system properties
user_agent.os # returns OperatingSystem(familyuiOS, version(5, 1), version_string5.1)
user_agent.os.family # returns iOS
user_agent.os.version # returns (5, 1)
user_agent.os.version_string # returns 5.1# Accessing user agents device properties
user_agent.device # returns Device(familyuiPhone, branduApple, modeluiPhone)
user_agent.device.family # returns iPhone
user_agent.device.brand # returns Apple
user_agent.device.model # returns iPhone# Viewing a pretty string version
str(user_agent) # returns iPhone / iOS 5.1 / Mobile Safari 5.1
# 最后这个最好用目前还支持这些属性
is_mobile判断是不是手机is_tablet判断是不是平板is_pc判断是不是电脑is_touch_capable有没有触屏功能is_bot是不是搜索引擎的爬虫 from user_agents import parse# Lets start from an old, non touch Blackberry device
ua_string BlackBerry9700/5.0.0.862 Profile/MIDP-2.1 Configuration/CLDC-1.1 VendorID/331 UNTRUSTED/1.0 3gpp-gba
user_agent parse(ua_string)
user_agent.is_mobile # returns True
user_agent.is_tablet # returns False
user_agent.is_touch_capable # returns False
user_agent.is_pc # returns False
user_agent.is_bot # returns False
str(user_agent) # returns BlackBerry 9700 / BlackBerry OS 5 / BlackBerry 9700常见机型映射字典
map_phone {Apple: Apple, KIW-AL10: Huawei, PRA-TL10: Huawei, BND-AL00: Huawei, XiaoMi: XiaoMi, MIX 2: XiaoMi, Oppo: Oppo, Oppo: Oppo, Gionee: Gionee, Samsung: Samsung, PRA-AL00X: Huawei, PACM00: Oppo, PBET00: Oppo, R7Plusm: Oppo, PAAT00: Oppo, PBAM00: Oppo, PADM00: Oppo, PAFM00: Oppo, PBEM00: Oppo, PAAM00: Oppo, PBBM00: Oppo, PACT00: Oppo, V1809A: vivo, PBAT00: Oppo, PADT00: Oppo, BND-TL10: Huawei, PBBT00: Oppo, PBCM10: Oppo, Mi Note 3: XiaoMi, V1816A: vivo, V1732T: vivo, V1813A: vivo, V1732A: vivo, V1818A:vivo,CAM-TL00:Huawei,Le X620:leshi,M6 Note:meizu,m3 note:meizu,M5:meizu,M1 E :meizu,BLN-AL10:Huawei,M5 Note:meizu,PRA-AL00:honour,LND-AL30:honour,NEM-AL10:honour,BND-AL10:honour,CAM-AL00:honour,SCL-TL00:honour,LLD-AL30:honour,BLN-AL20:honour,AUM-AL20:honour,JSN-AL00:honour,LLD-AL10:honour,BLN-TL10:honour,LLD-AL20:honour,BLN-AL40:honour,MYA-AL10:honour,LLD-AL00:honour,JSN-AL00a:honour,JMM-AL10:honour,DLI-AL10:honour,JMM-AL00:honour,V1809T:vivo,LND-AL40:honour,PLK-AL10:honour,MX6:meizu,PLK-TL01H:honour,S9:Samsung,KIW-TL00:honour,V1813T:vivo}常见的User_Agent各字段的解释
Mozilla/5.0 网景公司浏览器的标识由于互联网初期浏览器市场主要被网景公司占领很多服务器被设置成仅响应含有标志为Mozilla的浏览器的请求因此新款的浏览器为了打入市场不得不加上这个字段。Windows NT 6.3 : Windows 8.1的标识符WOW64 32位的Windows系统运行在64位的处理器上AppleWebKit/537.36苹果公司开发的呈现引擎KHTML是Linux平台中Konqueror浏览器的呈现引擎KHTMLGeckeo呈现引擎like Gecko表示其行为与Gecko浏览器引擎类似
请求中为什么既含有Chrome/33.0.1750.29又含有Safari/537.36字段 因为AppleWebKit渲染引擎是苹果公司开发的而Google公司要采用它为了获得服务器端的正确响应仅在Safari浏览器UA字段中增加了Chrome字段。 例如
Safari浏览器的UAMozilla/5.0 (平台;加密类型;操作系统或CPU;语言AppleWebKit/AppleWebKit版本号(KHTML, like Gecko) Safari/Safari 版本号Chrome浏览器的UAMozilla/5.0 (平台;加密类型;操作系统或CPU;语言)AppleWebKit/AppleWebKit版本号 (KHTML, like Gecko) Chrome/Chrome 版本号 Safari/Safari 版本号
为什么UA中包含多个浏览器的标识如Mozilla/5.0、Chrome/33.0.1750.29、Safari/537.36以及渲染引擎标识
多增加一些字段都是为了让服务器检测到它支持的浏览器标识以便获得服务器的响应从而提升用户体验。
这里有一个demo代码请参考 Request工具类import json
import loggingfrom django.contrib.auth.models import AbstractBaseUser
from django.contrib.auth.models import AnonymousUser
from django.core.cache import cache
from django.urls.resolvers import ResolverMatch
from user_agents import parsefrom apps.vadmin.utils.authentication import OpAuthJwtAuthenticationlogger logging.getLogger(__name__)def get_request_user(request, authenticateTrue):获取请求user(1)如果request里的user没有认证,那么则手动认证一次:param request::param authenticate::return:user: AbstractBaseUser getattr(request, user, None)if user and user.is_authenticated:return usertry:user, tokrn OpAuthJwtAuthentication().authenticate(request)except Exception as e:passreturn user or AnonymousUser()def get_request_ip(request):获取请求IP:param request::return:ip getattr(request, request_ip, None)if ip:return ipip request.META.get(REMOTE_ADDR, )if not ip:x_forwarded_for request.META.get(HTTP_X_FORWARDED_FOR, )if x_forwarded_for:ip x_forwarded_for.split(,)[-1].strip()else:ip unknownreturn ipdef get_request_data(request):获取请求参数:param request::return:request_data getattr(request, request_data, None)if request_data:return request_datadata: dict {**request.GET.dict(), **request.POST.dict()}if not data:try:body request.bodyif body:data json.loads(body)except Exception as e:passif not isinstance(data, dict):data {data: data}return datadef get_request_path(request, *args, **kwargs):获取请求路径:param request::param args::param kwargs::return:request_path getattr(request, request_path, None)if request_path:return request_pathvalues []for arg in args:if len(arg) 0:continueif isinstance(arg, str):values.append(arg)elif isinstance(arg, (tuple, set, list)):values.extend(arg)elif isinstance(arg, dict):values.extend(arg.values())if len(values) 0:return request.pathpath: str request.pathfor value in values:path path.replace(/ value, / {id})return pathdef get_request_canonical_path(request, *args, **kwargs):获取请求路径:param request::param args::param kwargs::return:request_path getattr(request, request_canonical_path, None)if request_path:return request_pathpath: str request.pathresolver_match: ResolverMatch request.resolver_matchfor value in resolver_match.args:path path.replace(f/{value}, /{id})for key, value in resolver_match.kwargs.items():if key pk:path path.replace(f/{value}, f/{{id}})continuepath path.replace(f/{value}, f/{{{key}}})return pathdef get_browser(request, *args, **kwargs):获取浏览器名:param request::param args::param kwargs::return:ua_string request.META[HTTP_USER_AGENT]user_agent parse(ua_string)return user_agent.get_browser()def get_os(request, *args, **kwargs):获取操作系统:param request::param args::param kwargs::return:ua_string request.META[HTTP_USER_AGENT]user_agent parse(ua_string)return user_agent.get_os()def get_login_location(request, *args, **kwargs):获取ip 登录位置:param request::param args::param kwargs::return:import requestsimport eventlet # 导入eventlet这个模块request_ip get_request_ip(request)# 从缓存中获取location cache.get(request_ip)if location:return location# 通过api 获取再缓存redistry:eventlet.monkey_patch(threadFalse) # 必须加这条代码with eventlet.Timeout(2, False): # 设置超时时间为2秒apiurl http://whois.pconline.com.cn/ip.jsp?ip%s % request_ipr requests.get(apiurl)content r.content.decode(GBK)location str(content).replace(\r, ).replace(\n, )[:64]cache.set(request_ip, location, 86400)return locationexcept Exception as e:passreturn def get_verbose_name(querysetNone, viewNone, modelNone):获取 verbose_name:param request::param view::return:try:if queryset and hasattr(queryset, model):model queryset.modelelif view and hasattr(view.get_queryset(), model):model view.get_queryset().modelelif view and hasattr(view.get_serializer(), Meta) and hasattr(view.get_serializer().Meta, model):model view.get_serializer().Meta.modelif model:return getattr(model, _meta).verbose_nameexcept Exception as e:passreturn
