Django接收自定义http header过程详解
addbyzhj:Django将所有httpheader(包括你自定义的httpheader)都放在了HttpRequest.META这个Python标准字典中,当然HttpRequest.META
中还包含其它一些键值对,这些键值对是Django加进去的,如SERVER_PORT等。对于httpheader,Django进行了重命名,规则如下
(1)所有header名大写,将连接符“-”改为下划线“_”
(2)除CONTENT_TYPE和CONTENT_LENGTH,其它的header名称前加“HTTP_”前缀
参见 https://docs.djangoproject.com/en/1.6/ref/request-response/#django.http.HttpRequest.META
我个人比较喜欢跟踪源代码来查看,源代码如下,
classWSGIRequestHandler(BaseHTTPRequestHandler): server_version="WSGIServer/"+__version__ defget_environ(self): env=self.server.base_environ.copy() env['SERVER_PROTOCOL']=self.request_version env['REQUEST_METHOD']=self.command if'?'inself.path: path,query=self.path.split('?',1) else: path,query=self.path,'' env['PATH_INFO']=urllib.unquote(path) env['QUERY_STRING']=query host=self.address_string() ifhost!=self.client_address[0]: env['REMOTE_HOST']=host env['REMOTE_ADDR']=self.client_address[0] ifself.headers.typeheaderisNone: env['CONTENT_TYPE']=self.headers.type else: env['CONTENT_TYPE']=self.headers.typeheader length=self.headers.getheader('content-length') iflength: env['CONTENT_LENGTH']=length forhinself.headers.headers: k,v=h.split(':',1) k=k.replace('-','_').upper();v=v.strip() ifkinenv: continue#skipcontentlength,type,etc. if'HTTP_'+kinenv: env['HTTP_'+k]+=','+v#comma-separatemultipleheaders else: env['HTTP_'+k]=v returnenv defget_stderr(self): returnsys.stderr defhandle(self): """HandleasingleHTTPrequest""" self.raw_requestline=self.rfile.readline() ifnotself.parse_request():#Anerrorcodehasbeensent,justexit return handler=ServerHandler( self.rfile,self.wfile,self.get_stderr(),self.get_environ() ) handler.request_handler=self#backpointerforlogging handler.run(self.server.get_app())
classWSGIRequest(http.HttpRequest): def__init__(self,environ): script_name=base.get_script_name(environ) path_info=base.get_path_info(environ) ifnotpath_info: #SometimesPATH_INFOexists,butisempty(e.g.accessing #theSCRIPT_NAMEURLwithoutatrailingslash).Wereallyneedto #operateasifthey'drequested'/'.Notamazinglynicetoforce #thepathlikethis,butshouldbeharmless. path_info='/' self.environ=environ self.path_info=path_info self.path='%s/%s'%(script_name.rstrip('/'),path_info.lstrip('/')) self.META=environ self.META['PATH_INFO']=path_info self.META['SCRIPT_NAME']=script_name self.method=environ['REQUEST_METHOD'].upper() _,content_params=self._parse_content_type(self.META.get('CONTENT_TYPE','')) if'charset'incontent_params: try: codecs.lookup(content_params['charset']) exceptLookupError: pass else: self.encoding=content_params['charset'] self._post_parse_error=False try: content_length=int(self.environ.get('CONTENT_LENGTH')) except(ValueError,TypeError): content_length=0 self._stream=LimitedStream(self.environ['wsgi.input'],content_length) self._read_started=False self.resolver_match=None
WSGIRequest类实例化方法__init__(self,environ)中第二个参数就是WSGIRequestHandler.get_environ()方法返回的数据
WSGIRequest.META在environ的基础上加了一些键值对
用Django做后台,客户端向Django请求数据,为了区分不同的请求,想把每个请求类别加在HTTP头部(headers)里面。
先做实验,就用Python的httplib库来做模拟客户端,参考网上写出模拟代码如下:
#coding=utf8 importhttplib httpClient=None try: myheaders={"category":"Books", "id":"21", 'My-Agent':"Superbrower" } httpClient=httplib.HTTPConnection('10.14.1XX.XXX',8086,timeout=30) httpClient.request('GET','/headinfo/',headers=myheaders) response=httpClient.getresponse() printresponse.status printresponse.reason printresponse.read() exceptException,e: printe finally: ifhttpClient: httpClient.close()
其中'/headinfo/'为服务器的响应目录。
然后是服务端的响应代码,《TheDjangoBook》第七章有个获取META的例子:
#GOOD(VERSION2) defua_display_good2(request): ua=request.META.get('HTTP_USER_AGENT','unknown') returnHttpResponse("Yourbrowseris%s"%ua)
正好看过这个例子,就模拟上面的这个写了一个能够返回客户端自定义头部的模块:
fromdjango.httpimportHttpResponse defheadinfo(request): category=request.META.get('CATEGORY','unkown') id=request.META.get('ID','unkown') agent=request.META.get('MY-AGENT','unkown') html="Categoryis%s,idis%s,agentis%s"%(category,id,agent) returnHttpResponse(html)
运行结果如下:
$pythonget.py #输出: #200 #OK #Categoryisunkown,idisunkown,agentisunkown
可以看到服务器成功响应了,但是却没有返回自定义的内容。
我以为是客户端模拟headers出问题了,查找和试验了许多次都没有返回正确的结果。后来去查Django的文档,发现了相关的描述:
HttpRequest.META
AstandardPythondictionarycontainingallavailableHTTPheaders.Availableheadersdependontheclientandserver,butherearesomeexamples:
- CONTENT_LENGTH–thelengthoftherequestbody(asastring).
- CONTENT_TYPE–theMIMEtypeoftherequestbody.
- HTTP_ACCEPT_ENCODING–Acceptableencodingsfortheresponse.
- HTTP_ACCEPT_LANGUAGE–Acceptablelanguagesfortheresponse.
- HTTP_HOST–TheHTTPHostheadersentbytheclient.
- HTTP_REFERER–Thereferringpage,ifany.
- HTTP_USER_AGENT–Theclient'suser-agentstring.
- QUERY_STRING–Thequerystring,asasingle(unparsed)string.
- REMOTE_ADDR–TheIPaddressoftheclient.
- REMOTE_HOST–Thehostnameoftheclient.
- REMOTE_USER–TheuserauthenticatedbytheWebserver,ifany.
- REQUEST_METHOD–Astringsuchas"GET"or"POST".
- SERVER_NAME–Thehostnameoftheserver.
- SERVER_PORT–Theportoftheserver(asastring).
WiththeexceptionofCONTENT_LENGTHandCONTENT_TYPE,asgivenabove,anyHTTPheadersintherequestareconvertedtoMETAkeysbyconvertingallcharacterstouppercase,replacinganyhyphenswithunderscoresandaddinganHTTP_prefixtothename.So,forexample,aheadercalledX-BenderwouldbemappedtotheMETAkeyHTTP_X_BENDER.
其中红色的部分说明是说除了两个特例之外,其他的头部在META字典中的key值都会被加上“HTTP_”的前缀,终于找到问题所在了,赶紧修改服务端代码:
category=request.META.get('HTTP_CATEGORY','unkown') id=request.META.get('HTTP_ID','unkown')
果然,执行后返回了想要的结果:
$pythonget.py #正确的输出: #200 #OK #CategoryisBooks,idis21,agentisSuperbrower
得到的经验就是遇到问题要多查文档,搜索引擎并不一定比文档更高效。
以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持毛票票。