source: draft-ietf-httpbis/latest/p1-messaging.xml @ 1762

Last change on this file since 1762 was 1762, checked in by fielding@…, 10 years ago

(editorial) clean up a few Notes

Created a section on implementation diversity to replace the confusing
and awkward note about user agents not always having interactive users.

  • Property svn:eol-style set to native
  • Property svn:mime-type set to text/xml
File size: 250.2 KB
1<?xml version="1.0" encoding="utf-8"?>
2<?xml-stylesheet type='text/xsl' href='../myxml2rfc.xslt'?>
3<!DOCTYPE rfc [
4  <!ENTITY MAY "<bcp14 xmlns=''>MAY</bcp14>">
5  <!ENTITY MUST "<bcp14 xmlns=''>MUST</bcp14>">
6  <!ENTITY MUST-NOT "<bcp14 xmlns=''>MUST NOT</bcp14>">
7  <!ENTITY OPTIONAL "<bcp14 xmlns=''>OPTIONAL</bcp14>">
8  <!ENTITY RECOMMENDED "<bcp14 xmlns=''>RECOMMENDED</bcp14>">
9  <!ENTITY REQUIRED "<bcp14 xmlns=''>REQUIRED</bcp14>">
10  <!ENTITY SHALL "<bcp14 xmlns=''>SHALL</bcp14>">
11  <!ENTITY SHALL-NOT "<bcp14 xmlns=''>SHALL NOT</bcp14>">
12  <!ENTITY SHOULD "<bcp14 xmlns=''>SHOULD</bcp14>">
13  <!ENTITY SHOULD-NOT "<bcp14 xmlns=''>SHOULD NOT</bcp14>">
14  <!ENTITY ID-VERSION "latest">
15  <!ENTITY ID-MONTH "July">
16  <!ENTITY ID-YEAR "2012">
17  <!ENTITY mdash "&#8212;">
18  <!ENTITY Note "<x:h xmlns:x=''>Note:</x:h>">
19  <!ENTITY caching-overview       "<xref target='Part6' x:rel='#caching.overview' xmlns:x=''/>">
20  <!ENTITY cache-incomplete       "<xref target='Part6' x:rel='#response.cacheability' xmlns:x=''/>">
21  <!ENTITY payload                "<xref target='Part2' xmlns:x=''/>">
22  <!ENTITY media-types            "<xref target='Part2' x:rel='#media.types' xmlns:x=''/>">
23  <!ENTITY content-codings        "<xref target='Part2' x:rel='#content.codings' xmlns:x=''/>">
24  <!ENTITY CONNECT                "<xref target='Part2' x:rel='#CONNECT' xmlns:x=''/>">
25  <!ENTITY content.negotiation    "<xref target='Part2' x:rel='#content.negotiation' xmlns:x=''/>">
26  <!ENTITY diff-mime              "<xref target='Part2' x:rel='#differences.between.http.and.mime' xmlns:x=''/>">
27  <!ENTITY representation         "<xref target='Part2' x:rel='#representation' xmlns:x=''/>">
28  <!ENTITY header-cache-control   "<xref target='Part6' x:rel='#header.cache-control' xmlns:x=''/>">
29  <!ENTITY header-content-encoding    "<xref target='Part2' x:rel='#header.content-encoding' xmlns:x=''/>">
30  <!ENTITY header-content-range   "<xref target='Part5' x:rel='#header.content-range' xmlns:x=''/>">
31  <!ENTITY header-content-type    "<xref target='Part2' x:rel='#header.content-type' xmlns:x=''/>">
32  <!ENTITY header-date            "<xref target='Part2' x:rel='' xmlns:x=''/>">
33  <!ENTITY header-expect          "<xref target='Part2' x:rel='#header.expect' xmlns:x=''/>">
34  <!ENTITY header-expires         "<xref target='Part6' x:rel='#header.expires' xmlns:x=''/>">
35  <!ENTITY header-mime-version    "<xref target='Part2' x:rel='#mime-version' xmlns:x=''/>">
36  <!ENTITY header-pragma          "<xref target='Part6' x:rel='#header.pragma' xmlns:x=''/>">
37  <!ENTITY header-warning         "<xref target='Part6' x:rel='#header.warning' xmlns:x=''/>">
38  <!ENTITY header-proxy-authenticate  "<xref target='Part7' x:rel='#header.proxy-authenticate' xmlns:x=''/>">
39  <!ENTITY header-proxy-authorization "<xref target='Part7' x:rel='#header.proxy-authorization' xmlns:x=''/>">
40  <!ENTITY idempotent-methods     "<xref target='Part2' x:rel='#idempotent.methods' xmlns:x=''/>">
41  <!ENTITY methods                "<xref target='Part2' x:rel='#methods' xmlns:x=''/>">
42  <!ENTITY OPTIONS                "<xref target='Part2' x:rel='#OPTIONS' xmlns:x=''/>">
43  <!ENTITY status-codes           "<xref target='Part2' x:rel='' xmlns:x=''/>">
44  <!ENTITY status-100             "<xref target='Part2' x:rel='#status.100' xmlns:x=''/>">
45  <!ENTITY status-1xx             "<xref target='Part2' x:rel='#status.1xx' xmlns:x=''/>">
46  <!ENTITY status-203             "<xref target='Part2' x:rel='#status.203' xmlns:x=''/>">
47  <!ENTITY status-3xx             "<xref target='Part2' x:rel='#status.3xx' xmlns:x=''/>">
48  <!ENTITY status-304             "<xref target='Part4' x:rel='#status.304' xmlns:x=''/>">
49  <!ENTITY status-4xx             "<xref target='Part2' x:rel='#status.4xx' xmlns:x=''/>">
50  <!ENTITY status-414             "<xref target='Part2' x:rel='#status.414' xmlns:x=''/>">
51  <!ENTITY cons-new-header-fields "<xref target='Part2' x:rel='#considerations.for.creating.header.fields' xmlns:x=''/>">
53<?rfc toc="yes" ?>
54<?rfc symrefs="yes" ?>
55<?rfc sortrefs="yes" ?>
56<?rfc compact="yes"?>
57<?rfc subcompact="no" ?>
58<?rfc linkmailto="no" ?>
59<?rfc editing="no" ?>
60<?rfc comments="yes"?>
61<?rfc inline="yes"?>
62<?rfc rfcedstyle="yes"?>
63<?rfc-ext allow-markup-in-artwork="yes" ?>
64<?rfc-ext include-references-in-index="yes" ?>
65<rfc obsoletes="2145,2616" updates="2817" category="std" x:maturity-level="proposed"
66     ipr="pre5378Trust200902" docName="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"
67     xmlns:x=''>
68<x:link rel="next" basename="p2-semantics"/>
69<x:feedback template="{docname},%20%22{section}%22&amp;body=&lt;{ref}&gt;:"/>
72  <title abbrev="HTTP/1.1, Part 1">HTTP/1.1, part 1: URIs, Connections, and Message Parsing</title>
74  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
75    <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
76    <address>
77      <postal>
78        <street>345 Park Ave</street>
79        <city>San Jose</city>
80        <region>CA</region>
81        <code>95110</code>
82        <country>USA</country>
83      </postal>
84      <email></email>
85      <uri></uri>
86    </address>
87  </author>
89  <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
90    <organization abbrev="W3C">World Wide Web Consortium</organization>
91    <address>
92      <postal>
93        <street>W3C / ERCIM</street>
94        <street>2004, rte des Lucioles</street>
95        <city>Sophia-Antipolis</city>
96        <region>AM</region>
97        <code>06902</code>
98        <country>France</country>
99      </postal>
100      <email></email>
101      <uri></uri>
102    </address>
103  </author>
105  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
106    <organization abbrev="greenbytes">greenbytes GmbH</organization>
107    <address>
108      <postal>
109        <street>Hafenweg 16</street>
110        <city>Muenster</city><region>NW</region><code>48155</code>
111        <country>Germany</country>
112      </postal>
113      <email></email>
114      <uri></uri>
115    </address>
116  </author>
118  <date month="&ID-MONTH;" year="&ID-YEAR;"/>
119  <workgroup>HTTPbis Working Group</workgroup>
123   The Hypertext Transfer Protocol (HTTP) is an application-level protocol for
124   distributed, collaborative, hypertext information systems. HTTP has been in
125   use by the World Wide Web global information initiative since 1990. This
126   document is Part 1 of the seven-part specification that defines the protocol
127   referred to as "HTTP/1.1" and, taken together, obsoletes
128   <xref target="RFC2616" x:fmt="none">RFC 2616</xref> and moves it to historic
129   status, along with its predecessor <xref target="RFC2068" x:fmt="none">RFC
130   2068</xref>.
133   Part 1 provides an overview of HTTP and its associated terminology, defines
134   the "http" and "https" Uniform Resource Identifier (URI) schemes, defines
135   the generic message syntax and parsing requirements for HTTP message frames,
136   and describes general security concerns for implementations.
139   This part also obsoletes RFCs <xref target="RFC2145" x:fmt="none">2145</xref>
140   (on HTTP version numbers) and <xref target="RFC2817" x:fmt="none">2817</xref>
141   (on using CONNECT for TLS upgrades) and moves them to historic status.
145<note title="Editorial Note (To be removed by RFC Editor)">
146  <t>
147    Discussion of this draft takes place on the HTTPBIS working group
148    mailing list (, which is archived at
149    <eref target=""/>.
150  </t>
151  <t>
152    The current issues list is at
153    <eref target=""/> and related
154    documents (including fancy diffs) can be found at
155    <eref target=""/>.
156  </t>
157  <t>
158    The changes in this draft are summarized in <xref target="changes.since.19"/>.
159  </t>
163<section title="Introduction" anchor="introduction">
165   The Hypertext Transfer Protocol (HTTP) is an application-level
166   request/response protocol that uses extensible semantics and MIME-like
167   message payloads for flexible interaction with network-based hypertext
168   information systems. HTTP relies upon the Uniform Resource Identifier (URI)
169   standard <xref target="RFC3986"/> to indicate the target resource
170   (<xref target="target-resource"/>) and relationships between resources.
171   Messages are passed in a format similar to that used by Internet mail
172   <xref target="RFC5322"/> and the Multipurpose Internet Mail Extensions
173   (MIME) <xref target="RFC2045"/> (see &diff-mime; for the differences
174   between HTTP and MIME messages).
177   HTTP is a generic interface protocol for information systems. It is
178   designed to hide the details of how a service is implemented by presenting
179   a uniform interface to clients that is independent of the types of
180   resources provided. Likewise, servers do not need to be aware of each
181   client's purpose: an HTTP request can be considered in isolation rather
182   than being associated with a specific type of client or a predetermined
183   sequence of application steps. The result is a protocol that can be used
184   effectively in many different contexts and for which implementations can
185   evolve independently over time.
188   HTTP is also designed for use as an intermediation protocol for translating
189   communication to and from non-HTTP information systems.
190   HTTP proxies and gateways can provide access to alternative information
191   services by translating their diverse protocols into a hypertext
192   format that can be viewed and manipulated by clients in the same way
193   as HTTP services.
196   One consequence of HTTP flexibility is that the protocol cannot be
197   defined in terms of what occurs behind the interface. Instead, we
198   are limited to defining the syntax of communication, the intent
199   of received communication, and the expected behavior of recipients.
200   If the communication is considered in isolation, then successful
201   actions ought to be reflected in corresponding changes to the
202   observable interface provided by servers. However, since multiple
203   clients might act in parallel and perhaps at cross-purposes, we
204   cannot require that such changes be observable beyond the scope
205   of a single response.
208   This document is Part 1 of the seven-part specification of HTTP,
209   defining the protocol referred to as "HTTP/1.1", obsoleting
210   <xref target="RFC2616"/> and <xref target="RFC2145"/>.
211   Part 1 describes the architectural elements that are used or
212   referred to in HTTP, defines the "http" and "https" URI schemes,
213   describes overall network operation and connection management,
214   and defines HTTP message framing and forwarding requirements.
215   Our goal is to define all of the mechanisms necessary for HTTP message
216   handling that are independent of message semantics, thereby defining the
217   complete set of requirements for message parsers and
218   message-forwarding intermediaries.
221<section title="Requirement Notation" anchor="intro.requirements">
223   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
224   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
225   document are to be interpreted as described in <xref target="RFC2119"/>.
229<section title="Syntax Notation" anchor="notation">
230<iref primary="true" item="Grammar" subitem="ALPHA"/>
231<iref primary="true" item="Grammar" subitem="CR"/>
232<iref primary="true" item="Grammar" subitem="CRLF"/>
233<iref primary="true" item="Grammar" subitem="CTL"/>
234<iref primary="true" item="Grammar" subitem="DIGIT"/>
235<iref primary="true" item="Grammar" subitem="DQUOTE"/>
236<iref primary="true" item="Grammar" subitem="HEXDIG"/>
237<iref primary="true" item="Grammar" subitem="HTAB"/>
238<iref primary="true" item="Grammar" subitem="LF"/>
239<iref primary="true" item="Grammar" subitem="OCTET"/>
240<iref primary="true" item="Grammar" subitem="SP"/>
241<iref primary="true" item="Grammar" subitem="VCHAR"/>
243   This specification uses the Augmented Backus-Naur Form (ABNF) notation
244   of <xref target="RFC5234"/> with the list rule extension defined in
245   <xref target="abnf.extension"/>.  <xref target="collected.abnf"/> shows
246   the collected ABNF with the list rule expanded.
248<t anchor="core.rules">
249  <x:anchor-alias value="ALPHA"/>
250  <x:anchor-alias value="CTL"/>
251  <x:anchor-alias value="CR"/>
252  <x:anchor-alias value="CRLF"/>
253  <x:anchor-alias value="DIGIT"/>
254  <x:anchor-alias value="DQUOTE"/>
255  <x:anchor-alias value="HEXDIG"/>
256  <x:anchor-alias value="HTAB"/>
257  <x:anchor-alias value="LF"/>
258  <x:anchor-alias value="OCTET"/>
259  <x:anchor-alias value="SP"/>
260  <x:anchor-alias value="VCHAR"/>
261   The following core rules are included by
262   reference, as defined in <xref target="RFC5234" x:fmt="," x:sec="B.1"/>:
263   ALPHA (letters), CR (carriage return), CRLF (CR LF), CTL (controls),
264   DIGIT (decimal 0-9), DQUOTE (double quote),
265   HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line feed),
266   OCTET (any 8-bit sequence of data), SP (space), and
267   VCHAR (any visible <xref target="USASCII"/> character).
270   As a convention, ABNF rule names prefixed with "obs-" denote
271   "obsolete" grammar rules that appear for historical reasons.
276<section title="Architecture" anchor="architecture">
278   HTTP was created for the World Wide Web architecture
279   and has evolved over time to support the scalability needs of a worldwide
280   hypertext system. Much of that architecture is reflected in the terminology
281   and syntax productions used to define HTTP.
284<section title="Client/Server Messaging" anchor="operation">
285<iref primary="true" item="client"/>
286<iref primary="true" item="server"/>
287<iref primary="true" item="connection"/>
289   HTTP is a stateless request/response protocol that operates by exchanging
290   <x:dfn>messages</x:dfn> (<xref target="http.message"/>) across a reliable
291   transport or session-layer
292   "<x:dfn>connection</x:dfn>". An HTTP "<x:dfn>client</x:dfn>" is a
293   program that establishes a connection to a server for the purpose of
294   sending one or more HTTP requests.  An HTTP "<x:dfn>server</x:dfn>" is a
295   program that accepts connections in order to service HTTP requests by
296   sending HTTP responses.
298<iref primary="true" item="user agent"/>
299<iref primary="true" item="origin server"/>
300<iref primary="true" item="browser"/>
301<iref primary="true" item="spider"/>
302<iref primary="true" item="sender"/>
303<iref primary="true" item="recipient"/>
305   The terms client and server refer only to the roles that
306   these programs perform for a particular connection.  The same program
307   might act as a client on some connections and a server on others.  We use
308   the term "<x:dfn>user agent</x:dfn>" to refer to the program that initiates a request,
309   such as a WWW browser, editor, or spider (web-traversing robot), and
310   the term "<x:dfn>origin server</x:dfn>" to refer to the program that can originate
311   authoritative responses to a request.  For general requirements, we use
312   the term "<x:dfn>sender</x:dfn>" to refer to whichever component sent a given message
313   and the term "<x:dfn>recipient</x:dfn>" to refer to any component that receives the
314   message.
317   Most HTTP communication consists of a retrieval request (GET) for
318   a representation of some resource identified by a URI.  In the
319   simplest case, this might be accomplished via a single bidirectional
320   connection (===) between the user agent (UA) and the origin server (O).
322<figure><artwork type="drawing">
323         request   &gt;
324    <x:highlight>UA</x:highlight> ======================================= <x:highlight>O</x:highlight>
325                                &lt;   response
327<iref primary="true" item="message"/>
328<iref primary="true" item="request"/>
329<iref primary="true" item="response"/>
331   A client sends an HTTP request to a server in the form of a <x:dfn>request</x:dfn>
332   message, beginning with a request-line that includes a method, URI, and
333   protocol version (<xref target="request.line"/>),
334   followed by MIME-like header fields containing
335   request modifiers, client information, and representation metadata
336   (<xref target="header.fields"/>),
337   an empty line to indicate the end of the header section, and finally
338   a message body containing the payload body (if any,
339   <xref target="message.body"/>).
342   A server responds to a client's request by sending one or more HTTP
343   <x:dfn>response</x:dfn>
344   messages, each beginning with a status line that
345   includes the protocol version, a success or error code, and textual
346   reason phrase (<xref target="status.line"/>),
347   possibly followed by MIME-like header fields containing server
348   information, resource metadata, and representation metadata
349   (<xref target="header.fields"/>),
350   an empty line to indicate the end of the header section, and finally
351   a message body containing the payload body (if any,
352   <xref target="message.body"/>).
355   The following example illustrates a typical message exchange for a
356   GET request on the URI "":
359client request:
360</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
361GET /hello.txt HTTP/1.1
362User-Agent: curl/7.16.3 libcurl/7.16.3 OpenSSL/0.9.7l zlib/1.2.3
364Accept: */*
368server response:
369</preamble><artwork type="message/http; msgtype=&#34;response&#34;" x:indent-with="  ">
370HTTP/1.1 200 OK
371Date: Mon, 27 Jul 2009 12:28:53 GMT
372Server: Apache
373Last-Modified: Wed, 22 Jul 2009 19:15:56 GMT
374ETag: "34aa387-d-1568eb00"
375Accept-Ranges: bytes
376Content-Length: <x:length-of target="exbody"/>
377Vary: Accept-Encoding
378Content-Type: text/plain
380<x:span anchor="exbody">Hello World!
384<section title="Implementation Diversity" anchor="implementation-diversity">
386   When considering the design of HTTP, it is easy to fall into a trap of
387   thinking that all user agents are general-purpose browsers and all origin
388   servers are large public websites. That is not the case in practice.
389   Common HTTP user agents include household appliances, stereos, scales,
390   software/firmware updaters, command-line programs, mobile apps,
391   and communication devices in a multitude of shapes and sizes.  Likewise,
392   common HTTP origin servers include home automation units, configurable
393   networking components, office machines, autonomous robots, news feeds,
394   traffic cameras, ad selectors, and video delivery platforms.
397   The term "user agent" does not imply that there is a human user directly
398   interacting with the software agent at the time of a request. In many
399   cases, a user agent is installed or configured to run in the background
400   and save its results for later inspection (or save only a subset of those
401   results that might be interesting or erroneous). Spiders, for example, are
402   typically given a start URI and configured to follow certain behavior while
403   crawling the Web as a hypertext graph.
406   The implementation diversity of HTTP means that we cannot assume the
407   user agent can make interactive suggestions to a user or provide adequate
408   warning for security or privacy options.  In the few cases where this
409   specification requires reporting of errors to the user, it is acceptable
410   for such reporting to only be visible in an error console or log file.
411   Likewise, requirements that an automated action be confirmed by the user
412   before proceeding can me met via advance configuration choices,
413   run-time options, or simply not proceeding with the unsafe action.
417<section title="Connections and Transport Independence" anchor="transport-independence">
419   HTTP messaging is independent of the underlying transport or
420   session-layer connection protocol(s).  HTTP only presumes a reliable
421   transport with in-order delivery of requests and the corresponding
422   in-order delivery of responses.  The mapping of HTTP request and
423   response structures onto the data units of the underlying transport
424   protocol is outside the scope of this specification.
427   The specific connection protocols to be used for an interaction
428   are determined by client configuration and the target URI
429   (<xref target="target-resource"/>).
430   For example, the "http" URI scheme
431   (<xref target="http.uri"/>) indicates a default connection of TCP
432   over IP, with a default TCP port of 80, but the client might be
433   configured to use a proxy via some other connection port or protocol
434   instead of using the defaults.
437   A connection might be used for multiple HTTP request/response exchanges,
438   as defined in <xref target="persistent.connections"/>.
442<section title="Intermediaries" anchor="intermediaries">
443<iref primary="true" item="intermediary"/>
445   HTTP enables the use of intermediaries to satisfy requests through
446   a chain of connections.  There are three common forms of HTTP
447   <x:dfn>intermediary</x:dfn>: proxy, gateway, and tunnel.  In some cases,
448   a single intermediary might act as an origin server, proxy, gateway,
449   or tunnel, switching behavior based on the nature of each request.
451<figure><artwork type="drawing">
452         &gt;             &gt;             &gt;             &gt;
453    <x:highlight>UA</x:highlight> =========== <x:highlight>A</x:highlight> =========== <x:highlight>B</x:highlight> =========== <x:highlight>C</x:highlight> =========== <x:highlight>O</x:highlight>
454               &lt;             &lt;             &lt;             &lt;
457   The figure above shows three intermediaries (A, B, and C) between the
458   user agent and origin server. A request or response message that
459   travels the whole chain will pass through four separate connections.
460   Some HTTP communication options
461   might apply only to the connection with the nearest, non-tunnel
462   neighbor, only to the end-points of the chain, or to all connections
463   along the chain. Although the diagram is linear, each participant might
464   be engaged in multiple, simultaneous communications. For example, B
465   might be receiving requests from many clients other than A, and/or
466   forwarding requests to servers other than C, at the same time that it
467   is handling A's request.
470<iref primary="true" item="upstream"/><iref primary="true" item="downstream"/>
471<iref primary="true" item="inbound"/><iref primary="true" item="outbound"/>
472   We use the terms "<x:dfn>upstream</x:dfn>" and "<x:dfn>downstream</x:dfn>"
473   to describe various requirements in relation to the directional flow of a
474   message: all messages flow from upstream to downstream.
475   Likewise, we use the terms inbound and outbound to refer to
476   directions in relation to the request path:
477   "<x:dfn>inbound</x:dfn>" means toward the origin server and
478   "<x:dfn>outbound</x:dfn>" means toward the user agent.
480<t><iref primary="true" item="proxy"/>
481   A "<x:dfn>proxy</x:dfn>" is a message forwarding agent that is selected by the
482   client, usually via local configuration rules, to receive requests
483   for some type(s) of absolute URI and attempt to satisfy those
484   requests via translation through the HTTP interface.  Some translations
485   are minimal, such as for proxy requests for "http" URIs, whereas
486   other requests might require translation to and from entirely different
487   application-layer protocols. Proxies are often used to group an
488   organization's HTTP requests through a common intermediary for the
489   sake of security, annotation services, or shared caching.
492<iref primary="true" item="transforming proxy"/>
493<iref primary="true" item="non-transforming proxy"/>
494   An HTTP-to-HTTP proxy is called a "<x:dfn>transforming proxy</x:dfn>" if it is designed
495   or configured to modify request or response messages in a semantically
496   meaningful way (i.e., modifications, beyond those required by normal
497   HTTP processing, that change the message in a way that would be
498   significant to the original sender or potentially significant to
499   downstream recipients).  For example, a transforming proxy might be
500   acting as a shared annotation server (modifying responses to include
501   references to a local annotation database), a malware filter, a
502   format transcoder, or an intranet-to-Internet privacy filter.  Such
503   transformations are presumed to be desired by the client (or client
504   organization) that selected the proxy and are beyond the scope of
505   this specification.  However, when a proxy is not intended to transform
506   a given message, we use the term "<x:dfn>non-transforming proxy</x:dfn>" to target
507   requirements that preserve HTTP message semantics. See &status-203; and
508   &header-warning; for status and warning codes related to transformations.
510<t><iref primary="true" item="gateway"/><iref primary="true" item="reverse proxy"/>
511<iref primary="true" item="accelerator"/>
512   A "<x:dfn>gateway</x:dfn>" (a.k.a., "<x:dfn>reverse proxy</x:dfn>")
513   is a receiving agent that acts
514   as a layer above some other server(s) and translates the received
515   requests to the underlying server's protocol.  Gateways are often
516   used to encapsulate legacy or untrusted information services, to
517   improve server performance through "<x:dfn>accelerator</x:dfn>" caching, and to
518   enable partitioning or load-balancing of HTTP services across
519   multiple machines.
522   A gateway behaves as an origin server on its outbound connection and
523   as a user agent on its inbound connection.
524   All HTTP requirements applicable to an origin server
525   also apply to the outbound communication of a gateway.
526   A gateway communicates with inbound servers using any protocol that
527   it desires, including private extensions to HTTP that are outside
528   the scope of this specification.  However, an HTTP-to-HTTP gateway
529   that wishes to interoperate with third-party HTTP servers &MUST;
530   conform to HTTP user agent requirements on the gateway's inbound
531   connection and &MUST; implement the <x:ref>Connection</x:ref>
532   (<xref target="header.connection"/>) and <x:ref>Via</x:ref>
533   (<xref target="header.via"/>) header fields for both connections.
535<t><iref primary="true" item="tunnel"/>
536   A "<x:dfn>tunnel</x:dfn>" acts as a blind relay between two connections
537   without changing the messages. Once active, a tunnel is not
538   considered a party to the HTTP communication, though the tunnel might
539   have been initiated by an HTTP request. A tunnel ceases to exist when
540   both ends of the relayed connection are closed. Tunnels are used to
541   extend a virtual connection through an intermediary, such as when
542   transport-layer security is used to establish private communication
543   through a shared firewall proxy.
545<t><iref primary="true" item="interception proxy"/>
546<iref primary="true" item="transparent proxy"/>
547<iref primary="true" item="captive portal"/>
548   The above categories for intermediary only consider those acting as
549   participants in the HTTP communication.  There are also intermediaries
550   that can act on lower layers of the network protocol stack, filtering or
551   redirecting HTTP traffic without the knowledge or permission of message
552   senders. Network intermediaries often introduce security flaws or
553   interoperability problems by violating HTTP semantics.  For example, an
554   "<x:dfn>interception proxy</x:dfn>" <xref target="RFC3040"/> (also commonly
555   known as a "<x:dfn>transparent proxy</x:dfn>" <xref target="RFC1919"/> or
556   "<x:dfn>captive portal</x:dfn>")
557   differs from an HTTP proxy because it is not selected by the client.
558   Instead, an interception proxy filters or redirects outgoing TCP port 80
559   packets (and occasionally other common port traffic).
560   Interception proxies are commonly found on public network access points,
561   as a means of enforcing account subscription prior to allowing use of
562   non-local Internet services, and within corporate firewalls to enforce
563   network usage policies.
564   They are indistinguishable from a man-in-the-middle attack.
567   HTTP is defined as a stateless protocol, meaning that each request message
568   can be understood in isolation.  Many implementations depend on HTTP's
569   stateless design in order to reuse proxied connections or dynamically
570   load balance requests across multiple servers.  Hence, servers &MUST-NOT;
571   assume that two requests on the same connection are from the same user
572   agent unless the connection is secured and specific to that agent.
573   Some non-standard HTTP extensions (e.g., <xref target="RFC4559"/>) have
574   been known to violate this requirement, resulting in security and
575   interoperability problems.
579<section title="Caches" anchor="caches">
580<iref primary="true" item="cache"/>
582   A "<x:dfn>cache</x:dfn>" is a local store of previous response messages and the
583   subsystem that controls its message storage, retrieval, and deletion.
584   A cache stores cacheable responses in order to reduce the response
585   time and network bandwidth consumption on future, equivalent
586   requests. Any client or server &MAY; employ a cache, though a cache
587   cannot be used by a server while it is acting as a tunnel.
590   The effect of a cache is that the request/response chain is shortened
591   if one of the participants along the chain has a cached response
592   applicable to that request. The following illustrates the resulting
593   chain if B has a cached copy of an earlier response from O (via C)
594   for a request which has not been cached by UA or A.
596<figure><artwork type="drawing">
597            &gt;             &gt;
598       <x:highlight>UA</x:highlight> =========== <x:highlight>A</x:highlight> =========== <x:highlight>B</x:highlight> - - - - - - <x:highlight>C</x:highlight> - - - - - - <x:highlight>O</x:highlight>
599                  &lt;             &lt;
601<t><iref primary="true" item="cacheable"/>
602   A response is "<x:dfn>cacheable</x:dfn>" if a cache is allowed to store a copy of
603   the response message for use in answering subsequent requests.
604   Even when a response is cacheable, there might be additional
605   constraints placed by the client or by the origin server on when
606   that cached response can be used for a particular request. HTTP
607   requirements for cache behavior and cacheable responses are
608   defined in &caching-overview;. 
611   There are a wide variety of architectures and configurations
612   of caches and proxies deployed across the World Wide Web and
613   inside large organizations. These systems include national hierarchies
614   of proxy caches to save transoceanic bandwidth, systems that
615   broadcast or multicast cache entries, organizations that distribute
616   subsets of cached data via optical media, and so on.
620<section title="Conformance and Error Handling" anchor="intro.conformance.and.error.handling">
622   This specification targets conformance criteria according to the role of
623   a participant in HTTP communication.  Hence, HTTP requirements are placed
624   on senders, recipients, clients, servers, user agents, intermediaries,
625   origin servers, proxies, gateways, or caches, depending on what behavior
626   is being constrained by the requirement. The verb "generate" is used
627   instead of "send" where a requirement differentiates between creating a
628   protocol element and merely forwarding a received element downstream.
631   An implementation is considered conformant if it complies with all of the
632   requirements associated with the roles it partakes in HTTP.
635   A sender &MUST-NOT; generate protocol elements that do not match
636   the grammar defined by the ABNF rules for those protocol elements that
637   are applicable to the sender's role.
638   If a received protocol element is processed, the recipient &MUST; be able
639   to parse any value that would match the ABNF rules for that protocol
640   element, excluding only those rules not applicable to the recipient's role.
643   Unless noted otherwise, a recipient &MAY; attempt to recover a usable
644   protocol element from an invalid construct.  HTTP does not define
645   specific error handling mechanisms except when they have a direct impact
646   on security, since different applications of the protocol require
647   different error handling strategies.  For example, a Web browser might
648   wish to transparently recover from a response where the <x:ref>Location</x:ref>
649   header field doesn't parse according to the ABNF, whereas a systems control
650   client might consider any form of error recovery to be dangerous.
654<section title="Protocol Versioning" anchor="http.version">
655  <x:anchor-alias value="HTTP-version"/>
656  <x:anchor-alias value="HTTP-name"/>
658   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate
659   versions of the protocol. This specification defines version "1.1".
660   The protocol version as a whole indicates the sender's conformance
661   with the set of requirements laid out in that version's corresponding
662   specification of HTTP.
665   The version of an HTTP message is indicated by an HTTP-version field
666   in the first line of the message. HTTP-version is case-sensitive.
668<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-version"/><iref primary="true" item="Grammar" subitem="HTTP-name"/>
669  <x:ref>HTTP-version</x:ref>  = <x:ref>HTTP-name</x:ref> "/" <x:ref>DIGIT</x:ref> "." <x:ref>DIGIT</x:ref>
670  <x:ref>HTTP-name</x:ref>     = <x:abnf-char-sequence>"HTTP"</x:abnf-char-sequence> ; "HTTP", case-sensitive
673   The HTTP version number consists of two decimal digits separated by a "."
674   (period or decimal point).  The first digit ("major version") indicates the
675   HTTP messaging syntax, whereas the second digit ("minor version") indicates
676   the highest minor version to which the sender is
677   conformant and able to understand for future communication.  The minor
678   version advertises the sender's communication capabilities even when the
679   sender is only using a backwards-compatible subset of the protocol,
680   thereby letting the recipient know that more advanced features can
681   be used in response (by servers) or in future requests (by clients).
684   When an HTTP/1.1 message is sent to an HTTP/1.0 recipient
685   <xref target="RFC1945"/> or a recipient whose version is unknown,
686   the HTTP/1.1 message is constructed such that it can be interpreted
687   as a valid HTTP/1.0 message if all of the newer features are ignored.
688   This specification places recipient-version requirements on some
689   new features so that a conformant sender will only use compatible
690   features until it has determined, through configuration or the
691   receipt of a message, that the recipient supports HTTP/1.1.
694   The interpretation of a header field does not change between minor
695   versions of the same major HTTP version, though the default
696   behavior of a recipient in the absence of such a field can change.
697   Unless specified otherwise, header fields defined in HTTP/1.1 are
698   defined for all versions of HTTP/1.x.  In particular, the <x:ref>Host</x:ref>
699   and <x:ref>Connection</x:ref> header fields ought to be implemented by all
700   HTTP/1.x implementations whether or not they advertise conformance with
701   HTTP/1.1.
704   New header fields can be defined such that, when they are
705   understood by a recipient, they might override or enhance the
706   interpretation of previously defined header fields.  When an
707   implementation receives an unrecognized header field, the recipient
708   &MUST; ignore that header field for local processing regardless of
709   the message's HTTP version.  An unrecognized header field received
710   by a proxy &MUST; be forwarded downstream unless the header field's
711   field-name is listed in the message's <x:ref>Connection</x:ref> header field
712   (see <xref target="header.connection"/>).
713   These requirements allow HTTP's functionality to be enhanced without
714   requiring prior update of deployed intermediaries.
717   Intermediaries that process HTTP messages (i.e., all intermediaries
718   other than those acting as tunnels) &MUST; send their own HTTP-version
719   in forwarded messages.  In other words, they &MUST-NOT; blindly
720   forward the first line of an HTTP message without ensuring that the
721   protocol version in that message matches a version to which that
722   intermediary is conformant for both the receiving and
723   sending of messages.  Forwarding an HTTP message without rewriting
724   the HTTP-version might result in communication errors when downstream
725   recipients use the message sender's version to determine what features
726   are safe to use for later communication with that sender.
729   An HTTP client &SHOULD; send a request version equal to the highest
730   version to which the client is conformant and
731   whose major version is no higher than the highest version supported
732   by the server, if this is known.  An HTTP client &MUST-NOT; send a
733   version to which it is not conformant.
736   An HTTP client &MAY; send a lower request version if it is known that
737   the server incorrectly implements the HTTP specification, but only
738   after the client has attempted at least one normal request and determined
739   from the response status or header fields (e.g., <x:ref>Server</x:ref>) that
740   the server improperly handles higher request versions.
743   An HTTP server &SHOULD; send a response version equal to the highest
744   version to which the server is conformant and
745   whose major version is less than or equal to the one received in the
746   request.  An HTTP server &MUST-NOT; send a version to which it is not
747   conformant.  A server &MAY; send a <x:ref>505 (HTTP Version Not
748   Supported)</x:ref> response if it cannot send a response using the
749   major version used in the client's request.
752   An HTTP server &MAY; send an HTTP/1.0 response to an HTTP/1.0 request
753   if it is known or suspected that the client incorrectly implements the
754   HTTP specification and is incapable of correctly processing later
755   version responses, such as when a client fails to parse the version
756   number correctly or when an intermediary is known to blindly forward
757   the HTTP-version even when it doesn't conform to the given minor
758   version of the protocol. Such protocol downgrades &SHOULD-NOT; be
759   performed unless triggered by specific client attributes, such as when
760   one or more of the request header fields (e.g., <x:ref>User-Agent</x:ref>)
761   uniquely match the values sent by a client known to be in error.
764   The intention of HTTP's versioning design is that the major number
765   will only be incremented if an incompatible message syntax is
766   introduced, and that the minor number will only be incremented when
767   changes made to the protocol have the effect of adding to the message
768   semantics or implying additional capabilities of the sender.  However,
769   the minor version was not incremented for the changes introduced between
770   <xref target="RFC2068"/> and <xref target="RFC2616"/>, and this revision
771   is specifically avoiding any such changes to the protocol.
775<section title="Uniform Resource Identifiers" anchor="uri">
776<iref primary="true" item="resource"/>
778   Uniform Resource Identifiers (URIs) <xref target="RFC3986"/> are used
779   throughout HTTP as the means for identifying resources. URI references
780   are used to target requests, indicate redirects, and define relationships.
781   HTTP does not limit what a resource might be; it merely defines an interface
782   that can be used to interact with a resource via HTTP. More information on
783   the scope of URIs and resources can be found in <xref target="RFC3986"/>.
785  <x:anchor-alias value="URI-reference"/>
786  <x:anchor-alias value="absolute-URI"/>
787  <x:anchor-alias value="relative-part"/>
788  <x:anchor-alias value="authority"/>
789  <x:anchor-alias value="path-abempty"/>
790  <x:anchor-alias value="path-absolute"/>
791  <x:anchor-alias value="port"/>
792  <x:anchor-alias value="query"/>
793  <x:anchor-alias value="uri-host"/>
794  <x:anchor-alias value="partial-URI"/>
796   This specification adopts the definitions of "URI-reference",
797   "absolute-URI", "relative-part", "port", "host",
798   "path-abempty", "path-absolute", "query", and "authority" from the
799   URI generic syntax <xref target="RFC3986"/>.
800   In addition, we define a partial-URI rule for protocol elements
801   that allow a relative URI but not a fragment.
803<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="URI-reference"/><iref primary="true" item="Grammar" subitem="absolute-URI"/><iref primary="true" item="Grammar" subitem="authority"/><iref primary="true" item="Grammar" subitem="path-absolute"/><iref primary="true" item="Grammar" subitem="port"/><iref primary="true" item="Grammar" subitem="query"/><iref primary="true" item="Grammar" subitem="uri-host"/>
804  <x:ref>URI-reference</x:ref> = &lt;URI-reference, defined in <xref target="RFC3986" x:fmt="," x:sec="4.1"/>&gt;
805  <x:ref>absolute-URI</x:ref>  = &lt;absolute-URI, defined in <xref target="RFC3986" x:fmt="," x:sec="4.3"/>&gt;
806  <x:ref>relative-part</x:ref> = &lt;relative-part, defined in <xref target="RFC3986" x:fmt="," x:sec="4.2"/>&gt;
807  <x:ref>authority</x:ref>     = &lt;authority, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2"/>&gt;
808  <x:ref>path-abempty</x:ref>  = &lt;path-abempty, defined in <xref target="RFC3986" x:fmt="," x:sec="3.3"/>&gt;
809  <x:ref>path-absolute</x:ref> = &lt;path-absolute, defined in <xref target="RFC3986" x:fmt="," x:sec="3.3"/>&gt;
810  <x:ref>port</x:ref>          = &lt;port, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2.3"/>&gt;
811  <x:ref>query</x:ref>         = &lt;query, defined in <xref target="RFC3986" x:fmt="," x:sec="3.4"/>&gt;
812  <x:ref>uri-host</x:ref>      = &lt;host, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2.2"/>&gt;
814  <x:ref>partial-URI</x:ref>   = relative-part [ "?" query ]
817   Each protocol element in HTTP that allows a URI reference will indicate
818   in its ABNF production whether the element allows any form of reference
819   (URI-reference), only a URI in absolute form (absolute-URI), only the
820   path and optional query components, or some combination of the above.
821   Unless otherwise indicated, URI references are parsed
822   relative to the effective request URI
823   (<xref target="effective.request.uri"/>).
826<section title="http URI scheme" anchor="http.uri">
827  <x:anchor-alias value="http-URI"/>
828  <iref item="http URI scheme" primary="true"/>
829  <iref item="URI scheme" subitem="http" primary="true"/>
831   The "http" URI scheme is hereby defined for the purpose of minting
832   identifiers according to their association with the hierarchical
833   namespace governed by a potential HTTP origin server listening for
834   TCP connections on a given port.
836<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="http-URI"/>
837  <x:ref>http-URI</x:ref> = "http:" "//" <x:ref>authority</x:ref> <x:ref>path-abempty</x:ref> [ "?" <x:ref>query</x:ref> ]
840   The HTTP origin server is identified by the generic syntax's
841   <x:ref>authority</x:ref> component, which includes a host identifier
842   and optional TCP port (<xref target="RFC3986" x:fmt="," x:sec="3.2.2"/>).
843   The remainder of the URI, consisting of both the hierarchical path
844   component and optional query component, serves as an identifier for
845   a potential resource within that origin server's name space.
848   If the host identifier is provided as an IP literal or IPv4 address,
849   then the origin server is any listener on the indicated TCP port at
850   that IP address. If host is a registered name, then that name is
851   considered an indirect identifier and the recipient might use a name
852   resolution service, such as DNS, to find the address of a listener
853   for that host.
854   The host &MUST-NOT; be empty; if an "http" URI is received with an
855   empty host, then it &MUST; be rejected as invalid.
856   If the port subcomponent is empty or not given, then TCP port 80 is
857   assumed (the default reserved port for WWW services).
860   Regardless of the form of host identifier, access to that host is not
861   implied by the mere presence of its name or address. The host might or might
862   not exist and, even when it does exist, might or might not be running an
863   HTTP server or listening to the indicated port. The "http" URI scheme
864   makes use of the delegated nature of Internet names and addresses to
865   establish a naming authority (whatever entity has the ability to place
866   an HTTP server at that Internet name or address) and allows that
867   authority to determine which names are valid and how they might be used.
870   When an "http" URI is used within a context that calls for access to the
871   indicated resource, a client &MAY; attempt access by resolving
872   the host to an IP address, establishing a TCP connection to that address
873   on the indicated port, and sending an HTTP request message
874   (<xref target="http.message"/>) containing the URI's identifying data
875   (<xref target="message.routing"/>) to the server.
876   If the server responds to that request with a non-interim HTTP response
877   message, as described in &status-codes;, then that response
878   is considered an authoritative answer to the client's request.
881   Although HTTP is independent of the transport protocol, the "http"
882   scheme is specific to TCP-based services because the name delegation
883   process depends on TCP for establishing authority.
884   An HTTP service based on some other underlying connection protocol
885   would presumably be identified using a different URI scheme, just as
886   the "https" scheme (below) is used for servers that require an SSL/TLS
887   transport layer on a connection. Other protocols might also be used to
888   provide access to "http" identified resources &mdash; it is only the
889   authoritative interface used for mapping the namespace that is
890   specific to TCP.
893   The URI generic syntax for authority also includes a deprecated
894   userinfo subcomponent (<xref target="RFC3986" x:fmt="," x:sec="3.2.1"/>)
895   for including user authentication information in the URI.  Some
896   implementations make use of the userinfo component for internal
897   configuration of authentication information, such as within command
898   invocation options, configuration files, or bookmark lists, even
899   though such usage might expose a user identifier or password.
900   Senders &MUST-NOT; include a userinfo subcomponent (and its "@"
901   delimiter) when transmitting an "http" URI in a message.  Recipients
902   of HTTP messages that contain a URI reference &SHOULD; parse for the
903   existence of userinfo and treat its presence as an error, likely
904   indicating that the deprecated subcomponent is being used to obscure
905   the authority for the sake of phishing attacks.
909<section title="https URI scheme" anchor="https.uri">
910   <x:anchor-alias value="https-URI"/>
911   <iref item="https URI scheme"/>
912   <iref item="URI scheme" subitem="https"/>
914   The "https" URI scheme is hereby defined for the purpose of minting
915   identifiers according to their association with the hierarchical
916   namespace governed by a potential HTTP origin server listening for
917   SSL/TLS-secured connections on a given TCP port.
920   All of the requirements listed above for the "http" scheme are also
921   requirements for the "https" scheme, except that a default TCP port
922   of 443 is assumed if the port subcomponent is empty or not given,
923   and the TCP connection &MUST; be secured for privacy through the
924   use of strong encryption prior to sending the first HTTP request.
926<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="https-URI"/>
927  <x:ref>https-URI</x:ref> = "https:" "//" <x:ref>authority</x:ref> <x:ref>path-abempty</x:ref> [ "?" <x:ref>query</x:ref> ]
930   Unlike the "http" scheme, responses to "https" identified requests
931   are never "public" and thus &MUST-NOT; be reused for shared caching.
932   They can, however, be reused in a private cache if the message is
933   cacheable by default in HTTP or specifically indicated as such by
934   the Cache-Control header field (&header-cache-control;).
937   Resources made available via the "https" scheme have no shared
938   identity with the "http" scheme even if their resource identifiers
939   indicate the same authority (the same host listening to the same
940   TCP port).  They are distinct name spaces and are considered to be
941   distinct origin servers.  However, an extension to HTTP that is
942   defined to apply to entire host domains, such as the Cookie protocol
943   <xref target="RFC6265"/>, can allow information
944   set by one service to impact communication with other services
945   within a matching group of host domains.
948   The process for authoritative access to an "https" identified
949   resource is defined in <xref target="RFC2818"/>.
953<section title="http and https URI Normalization and Comparison" anchor="uri.comparison">
955   Since the "http" and "https" schemes conform to the URI generic syntax,
956   such URIs are normalized and compared according to the algorithm defined
957   in <xref target="RFC3986" x:fmt="," x:sec="6"/>, using the defaults
958   described above for each scheme.
961   If the port is equal to the default port for a scheme, the normal
962   form is to elide the port subcomponent. Likewise, an empty path
963   component is equivalent to an absolute path of "/", so the normal
964   form is to provide a path of "/" instead. The scheme and host
965   are case-insensitive and normally provided in lowercase; all
966   other components are compared in a case-sensitive manner.
967   Characters other than those in the "reserved" set are equivalent
968   to their percent-encoded octets (see <xref target="RFC3986"
969   x:fmt="," x:sec="2.1"/>): the normal form is to not encode them.
972   For example, the following three URIs are equivalent:
974<figure><artwork type="example">
983<section title="Message Format" anchor="http.message">
984<x:anchor-alias value="generic-message"/>
985<x:anchor-alias value="message.types"/>
986<x:anchor-alias value="HTTP-message"/>
987<x:anchor-alias value="start-line"/>
988<iref item="header section"/>
989<iref item="headers"/>
990<iref item="header field"/>
992   All HTTP/1.1 messages consist of a start-line followed by a sequence of
993   octets in a format similar to the Internet Message Format
994   <xref target="RFC5322"/>: zero or more header fields (collectively
995   referred to as the "headers" or the "header section"), an empty line
996   indicating the end of the header section, and an optional message body.
998<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-message"/>
999  <x:ref>HTTP-message</x:ref>   = <x:ref>start-line</x:ref>
1000                   *( <x:ref>header-field</x:ref> <x:ref>CRLF</x:ref> )
1001                   <x:ref>CRLF</x:ref>
1002                   [ <x:ref>message-body</x:ref> ]
1005   The normal procedure for parsing an HTTP message is to read the
1006   start-line into a structure, read each header field into a hash
1007   table by field name until the empty line, and then use the parsed
1008   data to determine if a message body is expected.  If a message body
1009   has been indicated, then it is read as a stream until an amount
1010   of octets equal to the message body length is read or the connection
1011   is closed.
1014   Recipients &MUST; parse an HTTP message as a sequence of octets in an
1015   encoding that is a superset of US-ASCII <xref target="USASCII"/>.
1016   Parsing an HTTP message as a stream of Unicode characters, without regard
1017   for the specific encoding, creates security vulnerabilities due to the
1018   varying ways that string processing libraries handle invalid multibyte
1019   character sequences that contain the octet LF (%x0A).  String-based
1020   parsers can only be safely used within protocol elements after the element
1021   has been extracted from the message, such as within a header field-value
1022   after message parsing has delineated the individual fields.
1025   An HTTP message can be parsed as a stream for incremental processing or
1026   forwarding downstream.  However, recipients cannot rely on incremental
1027   delivery of partial messages, since some implementations will buffer or
1028   delay message forwarding for the sake of network efficiency, security
1029   checks, or payload transformations.
1032<section title="Start Line" anchor="start.line">
1033  <x:anchor-alias value="Start-Line"/>
1035   An HTTP message can either be a request from client to server or a
1036   response from server to client.  Syntactically, the two types of message
1037   differ only in the start-line, which is either a request-line (for requests)
1038   or a status-line (for responses), and in the algorithm for determining
1039   the length of the message body (<xref target="message.body"/>).
1040   In theory, a client could receive requests and a server could receive
1041   responses, distinguishing them by their different start-line formats,
1042   but in practice servers are implemented to only expect a request
1043   (a response is interpreted as an unknown or invalid request method)
1044   and clients are implemented to only expect a response.
1046<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="start-line"/>
1047  <x:ref>start-line</x:ref>     = <x:ref>request-line</x:ref> / <x:ref>status-line</x:ref>
1050   Implementations &MUST-NOT; send whitespace between the start-line and
1051   the first header field. The presence of such whitespace in a request
1052   might be an attempt to trick a server into ignoring that field or
1053   processing the line after it as a new request, either of which might
1054   result in a security vulnerability if other implementations within
1055   the request chain interpret the same message differently.
1056   Likewise, the presence of such whitespace in a response might be
1057   ignored by some clients or cause others to cease parsing.
1060<section title="Request Line" anchor="request.line">
1061  <x:anchor-alias value="Request"/>
1062  <x:anchor-alias value="request-line"/>
1064   A request-line begins with a method token, followed by a single
1065   space (SP), the request-target, another single space (SP), the
1066   protocol version, and ending with CRLF.
1068<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="request-line"/>
1069  <x:ref>request-line</x:ref>   = <x:ref>method</x:ref> <x:ref>SP</x:ref> <x:ref>request-target</x:ref> <x:ref>SP</x:ref> <x:ref>HTTP-version</x:ref> <x:ref>CRLF</x:ref>
1072   A server &MUST; be able to parse any received message that begins
1073   with a request-line and matches the ABNF rule for HTTP-message.
1075<iref primary="true" item="method"/>
1076<t anchor="method">
1077   The method token indicates the request method to be performed on the
1078   target resource. The request method is case-sensitive.
1080<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="method"/>
1081  <x:ref>method</x:ref>         = <x:ref>token</x:ref>
1084   The methods defined by this specification can be found in
1085   &methods;, along with information regarding the HTTP method registry
1086   and considerations for defining new methods.
1088<iref item="request-target"/>
1090   The request-target identifies the target resource upon which to apply
1091   the request, as defined in <xref target="request-target"/>.
1094   No whitespace is allowed inside the method, request-target, and
1095   protocol version.  Hence, recipients typically parse the request-line
1096   into its component parts by splitting on the SP characters.
1099   Unfortunately, some user agents fail to properly encode hypertext
1100   references that have embedded whitespace, sending the characters
1101   directly instead of properly percent-encoding the disallowed characters.
1102   Recipients of an invalid request-line &SHOULD; respond with either a
1103   <x:ref>400 (Bad Request)</x:ref> error or a <x:ref>301 (Moved Permanently)</x:ref>
1104   redirect with the request-target properly encoded.  Recipients &SHOULD-NOT;
1105   attempt to autocorrect and then process the request without a redirect,
1106   since the invalid request-line might be deliberately crafted to bypass
1107   security filters along the request chain.
1110   HTTP does not place a pre-defined limit on the length of a request-line.
1111   A server that receives a method longer than any that it implements
1112   &SHOULD; respond with either a <x:ref>405 (Method Not Allowed)</x:ref>, if it is an origin
1113   server, or a <x:ref>501 (Not Implemented)</x:ref> status code.
1114   A server &MUST; be prepared to receive URIs of unbounded length and
1115   respond with the <x:ref>414 (URI Too Long)</x:ref> status code if the received
1116   request-target would be longer than the server wishes to handle
1117   (see &status-414;).
1120   Various ad-hoc limitations on request-line length are found in practice.
1121   It is &RECOMMENDED; that all HTTP senders and recipients support, at a
1122   minimum, request-line lengths of up to 8000 octets.
1126<section title="Status Line" anchor="status.line">
1127  <x:anchor-alias value="response"/>
1128  <x:anchor-alias value="status-line"/>
1129  <x:anchor-alias value="status-code"/>
1130  <x:anchor-alias value="reason-phrase"/>
1132   The first line of a response message is the status-line, consisting
1133   of the protocol version, a space (SP), the status code, another space,
1134   a possibly-empty textual phrase describing the status code, and
1135   ending with CRLF.
1137<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="status-line"/>
1138  <x:ref>status-line</x:ref> = <x:ref>HTTP-version</x:ref> <x:ref>SP</x:ref> <x:ref>status-code</x:ref> <x:ref>SP</x:ref> <x:ref>reason-phrase</x:ref> <x:ref>CRLF</x:ref>
1141   A client &MUST; be able to parse any received message that begins
1142   with a status-line and matches the ABNF rule for HTTP-message.
1145   The status-code element is a 3-digit integer code describing the
1146   result of the server's attempt to understand and satisfy the client's
1147   corresponding request. The rest of the response message is to be
1148   interpreted in light of the semantics defined for that status code.
1149   See &status-codes; for information about the semantics of status codes,
1150   including the classes of status code (indicated by the first digit),
1151   the status codes defined by this specification, considerations for the
1152   definition of new status codes, and the IANA registry.
1154<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="status-code"/>
1155  <x:ref>status-code</x:ref>    = 3<x:ref>DIGIT</x:ref>
1158   The reason-phrase element exists for the sole purpose of providing a
1159   textual description associated with the numeric status code, mostly
1160   out of deference to earlier Internet application protocols that were more
1161   frequently used with interactive text clients. A client &SHOULD; ignore
1162   the reason-phrase content.
1164<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="reason-phrase"/>
1165  <x:ref>reason-phrase</x:ref>  = *( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1170<section title="Header Fields" anchor="header.fields">
1171  <x:anchor-alias value="header-field"/>
1172  <x:anchor-alias value="field-content"/>
1173  <x:anchor-alias value="field-name"/>
1174  <x:anchor-alias value="field-value"/>
1175  <x:anchor-alias value="obs-fold"/>
1177   Each HTTP header field consists of a case-insensitive field name
1178   followed by a colon (":"), optional whitespace, and the field value.
1180<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="header-field"/><iref primary="true" item="Grammar" subitem="field-name"/><iref primary="true" item="Grammar" subitem="field-value"/><iref primary="true" item="Grammar" subitem="field-content"/><iref primary="true" item="Grammar" subitem="obs-fold"/>
1181  <x:ref>header-field</x:ref>   = <x:ref>field-name</x:ref> ":" <x:ref>OWS</x:ref> <x:ref>field-value</x:ref> <x:ref>BWS</x:ref>
1182  <x:ref>field-name</x:ref>     = <x:ref>token</x:ref>
1183  <x:ref>field-value</x:ref>    = *( <x:ref>field-content</x:ref> / <x:ref>obs-fold</x:ref> )
1184  <x:ref>field-content</x:ref>  = *( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1185  <x:ref>obs-fold</x:ref>       = <x:ref>CRLF</x:ref> ( <x:ref>SP</x:ref> / <x:ref>HTAB</x:ref> )
1186                 ; obsolete line folding
1187                 ; see <xref target="field.parsing"/>
1190   The field-name token labels the corresponding field-value as having the
1191   semantics defined by that header field.  For example, the <x:ref>Date</x:ref>
1192   header field is defined in &header-date; as containing the origination
1193   timestamp for the message in which it appears.
1196   HTTP header fields are fully extensible: there is no limit on the
1197   introduction of new field names, each presumably defining new semantics,
1198   or on the number of header fields used in a given message.  Existing
1199   fields are defined in each part of this specification and in many other
1200   specifications outside the standards process.
1201   New header fields can be introduced without changing the protocol version
1202   if their defined semantics allow them to be safely ignored by recipients
1203   that do not recognize them.
1206   New HTTP header fields &SHOULD; be registered with IANA according
1207   to the procedures in &cons-new-header-fields;.
1208   Unrecognized header fields &MUST; be forwarded by a proxy unless the
1209   field-name is listed in the <x:ref>Connection</x:ref> header field
1210   (<xref target="header.connection"/>) or the proxy is specifically
1211   configured to block or otherwise transform such fields.
1212   Unrecognized header fields &SHOULD; be ignored by other recipients.
1215   The order in which header fields with differing field names are
1216   received is not significant. However, it is "good practice" to send
1217   header fields that contain control data first, such as <x:ref>Host</x:ref>
1218   on requests and <x:ref>Date</x:ref> on responses, so that implementations
1219   can decide when not to handle a message as early as possible.  A server
1220   &MUST; wait until the entire header section is received before interpreting
1221   a request message, since later header fields might include conditionals,
1222   authentication credentials, or deliberately misleading duplicate
1223   header fields that would impact request processing.
1226   Multiple header fields with the same field name &MUST-NOT; be
1227   sent in a message unless the entire field value for that
1228   header field is defined as a comma-separated list [i.e., #(values)].
1229   Multiple header fields with the same field name can be combined into
1230   one "field-name: field-value" pair, without changing the semantics of the
1231   message, by appending each subsequent field value to the combined
1232   field value in order, separated by a comma. The order in which
1233   header fields with the same field name are received is therefore
1234   significant to the interpretation of the combined field value;
1235   a proxy &MUST-NOT; change the order of these field values when
1236   forwarding a message.
1239  <t>
1240   &Note; The "Set-Cookie" header field as implemented in
1241   practice can occur multiple times, but does not use the list syntax, and
1242   thus cannot be combined into a single line (<xref target="RFC6265"/>). (See Appendix A.2.3 of <xref target="Kri2001"/>
1243   for details.) Also note that the Set-Cookie2 header field specified in
1244   <xref target="RFC2965"/> does not share this problem.
1245  </t>
1248<section title="Whitespace" anchor="whitespace">
1249<t anchor="rule.LWS">
1250   This specification uses three rules to denote the use of linear
1251   whitespace: OWS (optional whitespace), RWS (required whitespace), and
1252   BWS ("bad" whitespace).
1254<t anchor="rule.OWS">
1255   The OWS rule is used where zero or more linear whitespace octets might
1256   appear. OWS &SHOULD; either not be produced or be produced as a single
1257   SP. Multiple OWS octets that occur within field-content &SHOULD; either
1258   be replaced with a single SP or transformed to all SP octets (each
1259   octet other than SP replaced with SP) before interpreting the field value
1260   or forwarding the message downstream.
1262<t anchor="rule.RWS">
1263   RWS is used when at least one linear whitespace octet is required to
1264   separate field tokens. RWS &SHOULD; be produced as a single SP.
1265   Multiple RWS octets that occur within field-content &SHOULD; either
1266   be replaced with a single SP or transformed to all SP octets before
1267   interpreting the field value or forwarding the message downstream.
1269<t anchor="rule.BWS">
1270   BWS is used where the grammar allows optional whitespace for historical
1271   reasons but senders &SHOULD-NOT; produce it in messages. HTTP/1.1
1272   recipients &MUST; accept such bad optional whitespace and remove it before
1273   interpreting the field value or forwarding the message downstream.
1275<t anchor="rule.whitespace">
1276  <x:anchor-alias value="BWS"/>
1277  <x:anchor-alias value="OWS"/>
1278  <x:anchor-alias value="RWS"/>
1280<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="OWS"/><iref primary="true" item="Grammar" subitem="RWS"/><iref primary="true" item="Grammar" subitem="BWS"/>
1281  <x:ref>OWS</x:ref>            = *( <x:ref>SP</x:ref> / <x:ref>HTAB</x:ref> )
1282                 ; "optional" whitespace
1283  <x:ref>RWS</x:ref>            = 1*( <x:ref>SP</x:ref> / <x:ref>HTAB</x:ref> )
1284                 ; "required" whitespace
1285  <x:ref>BWS</x:ref>            = <x:ref>OWS</x:ref>
1286                 ; "bad" whitespace
1290<section title="Field Parsing" anchor="field.parsing">
1292   No whitespace is allowed between the header field-name and colon.
1293   In the past, differences in the handling of such whitespace have led to
1294   security vulnerabilities in request routing and response handling.
1295   Any received request message that contains whitespace between a header
1296   field-name and colon &MUST; be rejected with a response code of 400
1297   (Bad Request).  A proxy &MUST; remove any such whitespace from a response
1298   message before forwarding the message downstream.
1301   A field value &MAY; be preceded by optional whitespace (OWS); a single SP is
1302   preferred. The field value does not include any leading or trailing white
1303   space: OWS occurring before the first non-whitespace octet of the
1304   field value or after the last non-whitespace octet of the field value
1305   is ignored and &SHOULD; be removed before further processing (as this does
1306   not change the meaning of the header field).
1309   Historically, HTTP header field values could be extended over multiple
1310   lines by preceding each extra line with at least one space or horizontal
1311   tab (obs-fold). This specification deprecates such line
1312   folding except within the message/http media type
1313   (<xref target=""/>).
1314   HTTP senders &MUST-NOT; produce messages that include line folding
1315   (i.e., that contain any field-value that matches the obs-fold rule) unless
1316   the message is intended for packaging within the message/http media type.
1317   HTTP recipients &SHOULD; accept line folding and replace any embedded
1318   obs-fold whitespace with either a single SP or a matching number of SP
1319   octets (to avoid buffer copying) prior to interpreting the field value or
1320   forwarding the message downstream.
1323   Historically, HTTP has allowed field content with text in the ISO-8859-1
1324   <xref target="ISO-8859-1"/> character encoding and supported other
1325   character sets only through use of <xref target="RFC2047"/> encoding.
1326   In practice, most HTTP header field values use only a subset of the
1327   US-ASCII character encoding <xref target="USASCII"/>. Newly defined
1328   header fields &SHOULD; limit their field values to US-ASCII octets.
1329   Recipients &SHOULD; treat other (obs-text) octets in field content as
1330   opaque data.
1334<section title="Field Length" anchor="field.length">
1336   HTTP does not place a pre-defined limit on the length of header fields,
1337   either in isolation or as a set. A server &MUST; be prepared to receive
1338   request header fields of unbounded length and respond with a <x:ref>4xx
1339   (Client Error)</x:ref> status code if the received header field(s) would be
1340   longer than the server wishes to handle.
1343   A client that receives response headers that are longer than it wishes to
1344   handle can only treat it as a server error.
1347   Various ad-hoc limitations on header length are found in practice. It is
1348   &RECOMMENDED; that all HTTP senders and recipients support messages whose
1349   combined header fields have 4000 or more octets.
1353<section title="Field value components" anchor="field.components">
1354<t anchor="rule.token.separators">
1355  <x:anchor-alias value="tchar"/>
1356  <x:anchor-alias value="token"/>
1357  <x:anchor-alias value="special"/>
1358  <x:anchor-alias value="word"/>
1359   Many HTTP/1.1 header field values consist of words (token or quoted-string)
1360   separated by whitespace or special characters. These special characters
1361   &MUST; be in a quoted string to be used within a parameter value (as defined
1362   in <xref target="transfer.codings"/>).
1364<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="word"/><iref primary="true" item="Grammar" subitem="token"/><iref primary="true" item="Grammar" subitem="tchar"/><iref primary="true" item="Grammar" subitem="special"/>
1365  <x:ref>word</x:ref>           = <x:ref>token</x:ref> / <x:ref>quoted-string</x:ref>
1367  <x:ref>token</x:ref>          = 1*<x:ref>tchar</x:ref>
1369  IMPORTANT: when editing "tchar" make sure that "special" is updated accordingly!!!
1370 -->
1371  <x:ref>tchar</x:ref>          = "!" / "#" / "$" / "%" / "&amp;" / "'" / "*"
1372                 / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
1373                 / <x:ref>DIGIT</x:ref> / <x:ref>ALPHA</x:ref>
1374                 ; any <x:ref>VCHAR</x:ref>, except <x:ref>special</x:ref>
1376  <x:ref>special</x:ref>        = "(" / ")" / "&lt;" / ">" / "@" / ","
1377                 / ";" / ":" / "\" / DQUOTE / "/" / "["
1378                 / "]" / "?" / "=" / "{" / "}"
1380<t anchor="rule.quoted-string">
1381  <x:anchor-alias value="quoted-string"/>
1382  <x:anchor-alias value="qdtext"/>
1383  <x:anchor-alias value="obs-text"/>
1384   A string of text is parsed as a single word if it is quoted using
1385   double-quote marks.
1387<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-string"/><iref primary="true" item="Grammar" subitem="qdtext"/><iref primary="true" item="Grammar" subitem="obs-text"/>
1388  <x:ref>quoted-string</x:ref>  = <x:ref>DQUOTE</x:ref> *( <x:ref>qdtext</x:ref> / <x:ref>quoted-pair</x:ref> ) <x:ref>DQUOTE</x:ref>
1389  <x:ref>qdtext</x:ref>         = <x:ref>OWS</x:ref> / %x21 / %x23-5B / %x5D-7E / <x:ref>obs-text</x:ref>
1390  <x:ref>obs-text</x:ref>       = %x80-FF
1392<t anchor="rule.quoted-pair">
1393  <x:anchor-alias value="quoted-pair"/>
1394   The backslash octet ("\") can be used as a single-octet
1395   quoting mechanism within quoted-string constructs:
1397<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-pair"/>
1398  <x:ref>quoted-pair</x:ref>    = "\" ( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1401   Recipients that process the value of the quoted-string &MUST; handle a
1402   quoted-pair as if it were replaced by the octet following the backslash.
1405   Senders &SHOULD-NOT; escape octets in quoted-strings that do not require
1406   escaping (i.e., other than DQUOTE and the backslash octet).
1408<t anchor="rule.comment">
1409  <x:anchor-alias value="comment"/>
1410  <x:anchor-alias value="ctext"/>
1411   Comments can be included in some HTTP header fields by surrounding
1412   the comment text with parentheses. Comments are only allowed in
1413   fields containing "comment" as part of their field value definition.
1415<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/>
1416  <x:ref>comment</x:ref>        = "(" *( <x:ref>ctext</x:ref> / <x:ref>quoted-cpair</x:ref> / <x:ref>comment</x:ref> ) ")"
1417  <x:ref>ctext</x:ref>          = <x:ref>OWS</x:ref> / %x21-27 / %x2A-5B / %x5D-7E / <x:ref>obs-text</x:ref>
1419<t anchor="rule.quoted-cpair">
1420  <x:anchor-alias value="quoted-cpair"/>
1421   The backslash octet ("\") can be used as a single-octet
1422   quoting mechanism within comment constructs:
1424<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-cpair"/>
1425  <x:ref>quoted-cpair</x:ref>   = "\" ( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1428   Senders &SHOULD-NOT; escape octets in comments that do not require escaping
1429   (i.e., other than the backslash octet "\" and the parentheses "(" and ")").
1433<section title="ABNF list extension: #rule" anchor="abnf.extension">
1435  A #rule extension to the ABNF rules of <xref target="RFC5234"/> is used to
1436  improve readability in the definitions of some header field values.
1439  A construct "#" is defined, similar to "*", for defining comma-delimited
1440  lists of elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating
1441  at least &lt;n&gt; and at most &lt;m&gt; elements, each separated by a single
1442  comma (",") and optional whitespace (OWS).   
1445  Thus,
1446</preamble><artwork type="example">
1447  1#element =&gt; element *( OWS "," OWS element )
1450  and:
1451</preamble><artwork type="example">
1452  #element =&gt; [ 1#element ]
1455  and for n &gt;= 1 and m &gt; 1:
1456</preamble><artwork type="example">
1457  &lt;n&gt;#&lt;m&gt;element =&gt; element &lt;n-1&gt;*&lt;m-1&gt;( OWS "," OWS element )
1460  For compatibility with legacy list rules, recipients &SHOULD; accept empty
1461  list elements. In other words, consumers would follow the list productions:
1463<figure><artwork type="example">
1464  #element =&gt; [ ( "," / element ) *( OWS "," [ OWS element ] ) ]
1466  1#element =&gt; *( "," OWS ) element *( OWS "," [ OWS element ] )
1469  Note that empty elements do not contribute to the count of elements present,
1470  though.
1473  For example, given these ABNF productions:
1475<figure><artwork type="example">
1476  example-list      = 1#example-list-elmt
1477  example-list-elmt = token ; see <xref target="field.components"/>
1480  Then these are valid values for example-list (not including the double
1481  quotes, which are present for delimitation only):
1483<figure><artwork type="example">
1484  "foo,bar"
1485  "foo ,bar,"
1486  "foo , ,bar,charlie   "
1489  But these values would be invalid, as at least one non-empty element is
1490  required:
1492<figure><artwork type="example">
1493  ""
1494  ","
1495  ",   ,"
1498  <xref target="collected.abnf"/> shows the collected ABNF, with the list rules
1499  expanded as explained above.
1504<section title="Message Body" anchor="message.body">
1505  <x:anchor-alias value="message-body"/>
1507   The message body (if any) of an HTTP message is used to carry the
1508   payload body of that request or response.  The message body is
1509   identical to the payload body unless a transfer coding has been
1510   applied, as described in <xref target="header.transfer-encoding"/>.
1512<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="message-body"/>
1513  <x:ref>message-body</x:ref> = *OCTET
1516   The rules for when a message body is allowed in a message differ for
1517   requests and responses.
1520   The presence of a message body in a request is signaled by a
1521   a <x:ref>Content-Length</x:ref> or <x:ref>Transfer-Encoding</x:ref> header
1522   field. Request message framing is independent of method semantics,
1523   even if the method does not define any use for a message body.
1526   The presence of a message body in a response depends on both
1527   the request method to which it is responding and the response
1528   status code (<xref target="status.line"/>).
1529   Responses to the HEAD request method never include a message body
1530   because the associated response header fields (e.g.,
1531   <x:ref>Transfer-Encoding</x:ref>, <x:ref>Content-Length</x:ref>, etc.) only
1532   indicate what their values would have been if the request method had been
1533   GET. <x:ref>2xx (Successful)</x:ref> responses to CONNECT switch to tunnel
1534   mode instead of having a message body.
1535   All <x:ref>1xx (Informational)</x:ref>, <x:ref>204 (No Content)</x:ref>, and
1536   <x:ref>304 (Not Modified)</x:ref> responses &MUST-NOT; include a message body.
1537   All other responses do include a message body, although the body
1538   &MAY; be of zero length.
1541<section title="Transfer-Encoding" anchor="header.transfer-encoding">
1542  <iref primary="true" item="Transfer-Encoding header field" x:for-anchor=""/>
1543  <iref primary="true" item="Header Fields" subitem="Transfer-Encoding" x:for-anchor=""/>
1544  <x:anchor-alias value="Transfer-Encoding"/>
1546   When one or more transfer codings are applied to a payload body in order
1547   to form the message body, a Transfer-Encoding header field &MUST; be sent
1548   in the message and &MUST; contain the list of corresponding
1549   transfer-coding names in the same order that they were applied.
1550   Transfer codings are defined in <xref target="transfer.codings"/>.
1552<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/>
1553  <x:ref>Transfer-Encoding</x:ref> = 1#<x:ref>transfer-coding</x:ref>
1556   Transfer-Encoding is analogous to the Content-Transfer-Encoding field of
1557   MIME, which was designed to enable safe transport of binary data over a
1558   7-bit transport service (<xref target="RFC2045" x:fmt="," x:sec="6"/>).
1559   However, safe transport has a different focus for an 8bit-clean transfer
1560   protocol. In HTTP's case, Transfer-Encoding is primarily intended to
1561   accurately delimit a dynamically generated payload and to distinguish
1562   payload encodings that are only applied for transport efficiency or
1563   security from those that are characteristics of the target resource.
1566   The "chunked" transfer-coding (<xref target="chunked.encoding"/>)
1567   &MUST; be implemented by all HTTP/1.1 recipients because it plays a
1568   crucial role in delimiting messages when the payload body size is not
1569   known in advance.
1570   When the "chunked" transfer-coding is used, it &MUST; be the last
1571   transfer-coding applied to form the message body and &MUST-NOT;
1572   be applied more than once in a message body.
1573   If any transfer-coding is applied to a request payload body,
1574   the final transfer-coding applied &MUST; be "chunked".
1575   If any transfer-coding is applied to a response payload body, then either
1576   the final transfer-coding applied &MUST; be "chunked" or
1577   the message &MUST; be terminated by closing the connection.
1580   For example,
1581</preamble><artwork type="example">
1582  Transfer-Encoding: gzip, chunked
1584   indicates that the payload body has been compressed using the gzip
1585   coding and then chunked using the chunked coding while forming the
1586   message body.
1589   If more than one Transfer-Encoding header field is present in a message,
1590   the multiple field-values &MUST; be combined into one field-value,
1591   according to the algorithm defined in <xref target="header.fields"/>,
1592   before determining the message body length.
1595   Unlike <x:ref>Content-Encoding</x:ref> (&content-codings;),
1596   Transfer-Encoding is a property of the message, not of the payload, and thus
1597   &MAY; be added or removed by any implementation along the request/response
1598   chain. Additional information about the encoding parameters &MAY; be
1599   provided by other header fields not defined by this specification.
1602   Transfer-Encoding &MAY; be sent in a response to a HEAD request or in a
1603   <x:ref>304 (Not Modified)</x:ref> response (&status-304;) to a GET request,
1604   neither of which includes a message body,
1605   to indicate that the origin server would have applied a transfer coding
1606   to the message body if the request had been an unconditional GET.
1607   This indication is not required, however, because any recipient on
1608   the response chain (including the origin server) can remove transfer
1609   codings when they are not needed.
1612   Transfer-Encoding was added in HTTP/1.1.  It is generally assumed that
1613   implementations advertising only HTTP/1.0 support will not understand
1614   how to process a transfer-encoded payload.
1615   A client &MUST-NOT; send a request containing Transfer-Encoding unless it
1616   knows the server will handle HTTP/1.1 (or later) requests; such knowledge
1617   might be in the form of specific user configuration or by remembering the
1618   version of a prior received response.
1619   A server &MUST-NOT; send a response containing Transfer-Encoding unless
1620   the corresponding request indicates HTTP/1.1 (or later).
1623   A server that receives a request message with a transfer-coding it does
1624   not understand &SHOULD; respond with <x:ref>501 (Not Implemented)</x:ref> and then
1625   close the connection.
1629<section title="Content-Length" anchor="header.content-length">
1630  <iref primary="true" item="Content-Length header field" x:for-anchor=""/>
1631  <iref primary="true" item="Header Fields" subitem="Content-Length" x:for-anchor=""/>
1632  <x:anchor-alias value="Content-Length"/>
1634   When a message does not have a <x:ref>Transfer-Encoding</x:ref> header field
1635   and the payload body length can be determined prior to being transferred, a
1636   Content-Length header field &SHOULD; be sent to indicate the length of the
1637   payload body that is either present as the message body, for requests
1638   and non-HEAD responses other than <x:ref>304 (Not Modified)</x:ref>, or
1639   would have been present had the request been an unconditional GET.  The
1640   length is expressed as a decimal number of octets.
1642<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Length"/>
1643  <x:ref>Content-Length</x:ref> = 1*<x:ref>DIGIT</x:ref>
1646   An example is
1648<figure><artwork type="example">
1649  Content-Length: 3495
1652   In the case of a response to a HEAD request, Content-Length indicates
1653   the size of the payload body (without any potential transfer-coding)
1654   that would have been sent had the request been a GET.
1655   In the case of a <x:ref>304 (Not Modified)</x:ref> response (&status-304;)
1656   to a GET request, Content-Length indicates the size of the payload body (without
1657   any potential transfer-coding) that would have been sent in a <x:ref>200 (OK)</x:ref>
1658   response.
1661   Any Content-Length field value greater than or equal to zero is valid.
1662   Since there is no predefined limit to the length of an HTTP payload,
1663   recipients &SHOULD; anticipate potentially large decimal numerals and
1664   prevent parsing errors due to integer conversion overflows
1665   (<xref target="attack.protocol.element.size.overflows"/>).
1668   If a message is received that has multiple Content-Length header fields
1669   (<xref target="header.content-length"/>) with field-values consisting
1670   of the same decimal value, or a single Content-Length header field with
1671   a field value containing a list of identical decimal values (e.g.,
1672   "Content-Length: 42, 42"), indicating that duplicate Content-Length
1673   header fields have been generated or combined by an upstream message
1674   processor, then the recipient &MUST; either reject the message as invalid
1675   or replace the duplicated field-values with a single valid Content-Length
1676   field containing that decimal value prior to determining the message body
1677   length.
1680  <t>
1681   &Note; HTTP's use of Content-Length for message framing differs
1682   significantly from the same field's use in MIME, where it is an optional
1683   field used only within the "message/external-body" media-type.
1684  </t>
1688<section title="Message Body Length" anchor="message.body.length">
1690   The length of a message body is determined by one of the following
1691   (in order of precedence):
1694  <list style="numbers">
1695    <x:lt><t>
1696     Any response to a HEAD request and any response with a
1697     <x:ref>1xx (Informational)</x:ref>, <x:ref>204 (No Content)</x:ref>, or
1698     <x:ref>304 (Not Modified)</x:ref> status code is always
1699     terminated by the first empty line after the header fields, regardless of
1700     the header fields present in the message, and thus cannot contain a
1701     message body.
1702    </t></x:lt>
1703    <x:lt><t>
1704     Any <x:ref>2xx (Successful)</x:ref> response to a CONNECT request implies that the
1705     connection will become a tunnel immediately after the empty line that
1706     concludes the header fields.  A client &MUST; ignore any
1707     <x:ref>Content-Length</x:ref> or <x:ref>Transfer-Encoding</x:ref> header
1708     fields received in such a message.
1709    </t></x:lt>
1710    <x:lt><t>
1711     If a <x:ref>Transfer-Encoding</x:ref> header field is present
1712     and the "chunked" transfer-coding (<xref target="chunked.encoding"/>)
1713     is the final encoding, the message body length is determined by reading
1714     and decoding the chunked data until the transfer-coding indicates the
1715     data is complete.
1716    </t>
1717    <t>
1718     If a <x:ref>Transfer-Encoding</x:ref> header field is present in a
1719     response and the "chunked" transfer-coding is not the final encoding, the
1720     message body length is determined by reading the connection until it is
1721     closed by the server.
1722     If a Transfer-Encoding header field is present in a request and the
1723     "chunked" transfer-coding is not the final encoding, the message body
1724     length cannot be determined reliably; the server &MUST; respond with
1725     the <x:ref>400 (Bad Request)</x:ref> status code and then close the connection.
1726    </t>
1727    <t>
1728     If a message is received with both a <x:ref>Transfer-Encoding</x:ref>
1729     and a <x:ref>Content-Length</x:ref> header field, the
1730     Transfer-Encoding overrides the Content-Length.
1731     Such a message might indicate an attempt to perform request or response
1732     smuggling (bypass of security-related checks on message routing or content)
1733     and thus ought to be handled as an error.  The provided Content-Length &MUST;
1734     be removed, prior to forwarding the message downstream, or replaced with
1735     the real message body length after the transfer-coding is decoded.
1736    </t></x:lt>
1737    <x:lt><t>
1738     If a message is received without <x:ref>Transfer-Encoding</x:ref> and with
1739     either multiple <x:ref>Content-Length</x:ref> header fields having
1740     differing field-values or a single Content-Length header field having an
1741     invalid value, then the message framing is invalid and &MUST; be treated
1742     as an error to prevent request or response smuggling.
1743     If this is a request message, the server &MUST; respond with
1744     a <x:ref>400 (Bad Request)</x:ref> status code and then close the connection.
1745     If this is a response message received by a proxy, the proxy
1746     &MUST; discard the received response, send a <x:ref>502 (Bad Gateway)</x:ref>
1747     status code as its downstream response, and then close the connection.
1748     If this is a response message received by a user-agent, it &MUST; be
1749     treated as an error by discarding the message and closing the connection.
1750    </t></x:lt>
1751    <x:lt><t>
1752     If a valid <x:ref>Content-Length</x:ref> header field is present without
1753     <x:ref>Transfer-Encoding</x:ref>, its decimal value defines the
1754     message body length in octets.  If the actual number of octets sent in
1755     the message is less than the indicated Content-Length, the recipient
1756     &MUST; consider the message to be incomplete and treat the connection
1757     as no longer usable.
1758     If the actual number of octets sent in the message is more than the indicated
1759     Content-Length, the recipient &MUST; only process the message body up to the
1760     field value's number of octets; the remainder of the message &MUST; either
1761     be discarded or treated as the next message in a pipeline.  For the sake of
1762     robustness, a user-agent &MAY; attempt to detect and correct such an error
1763     in message framing if it is parsing the response to the last request on
1764     a connection and the connection has been closed by the server.
1765    </t></x:lt>
1766    <x:lt><t>
1767     If this is a request message and none of the above are true, then the
1768     message body length is zero (no message body is present).
1769    </t></x:lt>
1770    <x:lt><t>
1771     Otherwise, this is a response message without a declared message body
1772     length, so the message body length is determined by the number of octets
1773     received prior to the server closing the connection.
1774    </t></x:lt>
1775  </list>
1778   Since there is no way to distinguish a successfully completed,
1779   close-delimited message from a partially-received message interrupted
1780   by network failure, implementations &SHOULD; use encoding or
1781   length-delimited messages whenever possible.  The close-delimiting
1782   feature exists primarily for backwards compatibility with HTTP/1.0.
1785   A server &MAY; reject a request that contains a message body but
1786   not a <x:ref>Content-Length</x:ref> by responding with
1787   <x:ref>411 (Length Required)</x:ref>.
1790   Unless a transfer-coding other than "chunked" has been applied,
1791   a client that sends a request containing a message body &SHOULD;
1792   use a valid <x:ref>Content-Length</x:ref> header field if the message body
1793   length is known in advance, rather than the "chunked" encoding, since some
1794   existing services respond to "chunked" with a <x:ref>411 (Length Required)</x:ref>
1795   status code even though they understand the chunked encoding.  This
1796   is typically because such services are implemented via a gateway that
1797   requires a content-length in advance of being called and the server
1798   is unable or unwilling to buffer the entire request before processing.
1801   A client that sends a request containing a message body &MUST; include a
1802   valid <x:ref>Content-Length</x:ref> header field if it does not know the
1803   server will handle HTTP/1.1 (or later) requests; such knowledge can be in
1804   the form of specific user configuration or by remembering the version of a
1805   prior received response.
1810<section anchor="incomplete.messages" title="Handling Incomplete Messages">
1812   Request messages that are prematurely terminated, possibly due to a
1813   cancelled connection or a server-imposed time-out exception, &MUST;
1814   result in closure of the connection; sending an HTTP/1.1 error response
1815   prior to closing the connection is &OPTIONAL;.
1818   Response messages that are prematurely terminated, usually by closure
1819   of the connection prior to receiving the expected number of octets or by
1820   failure to decode a transfer-encoded message body, &MUST; be recorded
1821   as incomplete.  A response that terminates in the middle of the header
1822   block (before the empty line is received) cannot be assumed to convey the
1823   full semantics of the response and &MUST; be treated as an error.
1826   A message body that uses the chunked transfer encoding is
1827   incomplete if the zero-sized chunk that terminates the encoding has not
1828   been received.  A message that uses a valid <x:ref>Content-Length</x:ref> is
1829   incomplete if the size of the message body received (in octets) is less than
1830   the value given by Content-Length.  A response that has neither chunked
1831   transfer encoding nor Content-Length is terminated by closure of the
1832   connection, and thus is considered complete regardless of the number of
1833   message body octets received, provided that the header block was received
1834   intact.
1837   A user agent &MUST-NOT; render an incomplete response message body as if
1838   it were complete (i.e., some indication needs to be given to the user that an
1839   error occurred).  Cache requirements for incomplete responses are defined
1840   in &cache-incomplete;.
1843   A server &MUST; read the entire request message body or close
1844   the connection after sending its response, since otherwise the
1845   remaining data on a persistent connection would be misinterpreted
1846   as the next request.  Likewise,
1847   a client &MUST; read the entire response message body if it intends
1848   to reuse the same connection for a subsequent request.  Pipelining
1849   multiple requests on a connection is described in <xref target="pipelining"/>.
1853<section title="Message Parsing Robustness" anchor="message.robustness">
1855   Older HTTP/1.0 client implementations might send an extra CRLF
1856   after a POST request as a lame workaround for some early server
1857   applications that failed to read message body content that was
1858   not terminated by a line-ending. An HTTP/1.1 client &MUST-NOT;
1859   preface or follow a request with an extra CRLF.  If terminating
1860   the request message body with a line-ending is desired, then the
1861   client &MUST; include the terminating CRLF octets as part of the
1862   message body length.
1865   In the interest of robustness, servers &SHOULD; ignore at least one
1866   empty line received where a request-line is expected. In other words, if
1867   the server is reading the protocol stream at the beginning of a
1868   message and receives a CRLF first, it &SHOULD; ignore the CRLF.
1869   Likewise, although the line terminator for the start-line and header
1870   fields is the sequence CRLF, we recommend that recipients recognize a
1871   single LF as a line terminator and ignore any CR.
1874   When a server listening only for HTTP request messages, or processing
1875   what appears from the start-line to be an HTTP request message,
1876   receives a sequence of octets that does not match the HTTP-message
1877   grammar aside from the robustness exceptions listed above, the
1878   server &MUST; respond with an HTTP/1.1 <x:ref>400 (Bad Request)</x:ref> response. 
1883<section title="Transfer Codings" anchor="transfer.codings">
1884  <x:anchor-alias value="transfer-coding"/>
1885  <x:anchor-alias value="transfer-extension"/>
1887   Transfer-coding values are used to indicate an encoding
1888   transformation that has been, can be, or might need to be applied to a
1889   payload body in order to ensure "safe transport" through the network.
1890   This differs from a content coding in that the transfer-coding is a
1891   property of the message rather than a property of the representation
1892   that is being transferred.
1894<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/>
1895  <x:ref>transfer-coding</x:ref>    = "chunked" ; <xref target="chunked.encoding"/>
1896                     / "compress" ; <xref target="compress.coding"/>
1897                     / "deflate" ; <xref target="deflate.coding"/>
1898                     / "gzip" ; <xref target="gzip.coding"/>
1899                     / <x:ref>transfer-extension</x:ref>
1900  <x:ref>transfer-extension</x:ref> = <x:ref>token</x:ref> *( <x:ref>OWS</x:ref> ";" <x:ref>OWS</x:ref> <x:ref>transfer-parameter</x:ref> )
1902<t anchor="rule.parameter">
1903  <x:anchor-alias value="attribute"/>
1904  <x:anchor-alias value="transfer-parameter"/>
1905  <x:anchor-alias value="value"/>
1906   Parameters are in the form of attribute/value pairs.
1908<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="transfer-parameter"/><iref primary="true" item="Grammar" subitem="attribute"/><iref primary="true" item="Grammar" subitem="value"/><iref primary="true" item="Grammar" subitem="date2"/><iref primary="true" item="Grammar" subitem="date3"/>
1909  <x:ref>transfer-parameter</x:ref> = <x:ref>attribute</x:ref> <x:ref>BWS</x:ref> "=" <x:ref>BWS</x:ref> <x:ref>value</x:ref>
1910  <x:ref>attribute</x:ref>          = <x:ref>token</x:ref>
1911  <x:ref>value</x:ref>              = <x:ref>word</x:ref>
1914   All transfer-coding values are case-insensitive.
1915   The HTTP Transfer Coding registry is defined in
1916   <xref target="transfer.coding.registry"/>.
1917   HTTP/1.1 uses transfer-coding values in the <x:ref>TE</x:ref> header field
1918   (<xref target="header.te"/>) and in the <x:ref>Transfer-Encoding</x:ref>
1919   header field (<xref target="header.transfer-encoding"/>).
1922<section title="Chunked Transfer Coding" anchor="chunked.encoding">
1923  <iref item="chunked (Coding Format)"/>
1924  <iref item="Coding Format" subitem="chunked"/>
1925  <x:anchor-alias value="chunk"/>
1926  <x:anchor-alias value="chunked-body"/>
1927  <x:anchor-alias value="chunk-data"/>
1928  <x:anchor-alias value="chunk-ext"/>
1929  <x:anchor-alias value="chunk-ext-name"/>
1930  <x:anchor-alias value="chunk-ext-val"/>
1931  <x:anchor-alias value="chunk-size"/>
1932  <x:anchor-alias value="last-chunk"/>
1933  <x:anchor-alias value="trailer-part"/>
1934  <x:anchor-alias value="quoted-str-nf"/>
1935  <x:anchor-alias value="qdtext-nf"/>
1937   The chunked encoding modifies the body of a message in order to
1938   transfer it as a series of chunks, each with its own size indicator,
1939   followed by an &OPTIONAL; trailer containing header fields. This
1940   allows dynamically produced content to be transferred along with the
1941   information necessary for the recipient to verify that it has
1942   received the full message.
1944<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="chunked-body"/><iref primary="true" item="Grammar" subitem="chunk"/><iref primary="true" item="Grammar" subitem="chunk-size"/><iref primary="true" item="Grammar" subitem="last-chunk"/><iref primary="true" item="Grammar" subitem="chunk-ext"/><iref primary="true" item="Grammar" subitem="chunk-ext-name"/><iref primary="true" item="Grammar" subitem="chunk-ext-val"/><iref primary="true" item="Grammar" subitem="chunk-data"/><iref primary="true" item="Grammar" subitem="trailer-part"/><iref primary="true" item="Grammar" subitem="quoted-str-nf"/><iref primary="true" item="Grammar" subitem="qdtext-nf"/>
1945  <x:ref>chunked-body</x:ref>   = *<x:ref>chunk</x:ref>
1946                   <x:ref>last-chunk</x:ref>
1947                   <x:ref>trailer-part</x:ref>
1948                   <x:ref>CRLF</x:ref>
1950  <x:ref>chunk</x:ref>          = <x:ref>chunk-size</x:ref> [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>
1951                   <x:ref>chunk-data</x:ref> <x:ref>CRLF</x:ref>
1952  <x:ref>chunk-size</x:ref>     = 1*<x:ref>HEXDIG</x:ref>
1953  <x:ref>last-chunk</x:ref>     = 1*("0") [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>
1955  <x:ref>chunk-ext</x:ref>      = *( ";" <x:ref>chunk-ext-name</x:ref> [ "=" <x:ref>chunk-ext-val</x:ref> ] )
1956  <x:ref>chunk-ext-name</x:ref> = <x:ref>token</x:ref>
1957  <x:ref>chunk-ext-val</x:ref>  = <x:ref>token</x:ref> / <x:ref>quoted-str-nf</x:ref>
1958  <x:ref>chunk-data</x:ref>     = 1*<x:ref>OCTET</x:ref> ; a sequence of chunk-size octets
1959  <x:ref>trailer-part</x:ref>   = *( <x:ref>header-field</x:ref> <x:ref>CRLF</x:ref> )
1961  <x:ref>quoted-str-nf</x:ref>  = <x:ref>DQUOTE</x:ref> *( <x:ref>qdtext-nf</x:ref> / <x:ref>quoted-pair</x:ref> ) <x:ref>DQUOTE</x:ref>
1962                 ; like <x:ref>quoted-string</x:ref>, but disallowing line folding
1963  <x:ref>qdtext-nf</x:ref>      = <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / %x21 / %x23-5B / %x5D-7E / <x:ref>obs-text</x:ref>
1966   The chunk-size field is a string of hex digits indicating the size of
1967   the chunk-data in octets. The chunked encoding is ended by any chunk whose size is
1968   zero, followed by the trailer, which is terminated by an empty line.
1971   The trailer allows the sender to include additional HTTP header
1972   fields at the end of the message. The <x:ref>Trailer</x:ref> header field
1973   can be used to indicate which header fields are included in a trailer (see
1974   <xref target="header.trailer"/>).
1977   A server using chunked transfer-coding in a response &MUST-NOT; use the
1978   trailer for any header fields unless at least one of the following is
1979   true:
1980  <list style="numbers">
1981    <t>the request included a <x:ref>TE</x:ref> header field that indicates
1982    "trailers" is acceptable in the transfer-coding of the response, as
1983    described in <xref target="header.te"/>; or,</t>
1985    <t>the trailer fields consist entirely of optional metadata, and the
1986    recipient could use the message (in a manner acceptable to the server where
1987    the field originated) without receiving it. In other words, the server that
1988    generated the header (often but not always the origin server) is willing to
1989    accept the possibility that the trailer fields might be silently discarded
1990    along the path to the client.</t>
1991  </list>
1994   This requirement prevents an interoperability failure when the
1995   message is being received by an HTTP/1.1 (or later) proxy and
1996   forwarded to an HTTP/1.0 recipient. It avoids a situation where
1997   conformance with the protocol would have necessitated a possibly
1998   infinite buffer on the proxy.
2001   A process for decoding the "chunked" transfer-coding
2002   can be represented in pseudo-code as:
2004<figure><artwork type="code">
2005  length := 0
2006  read chunk-size, chunk-ext (if any) and CRLF
2007  while (chunk-size &gt; 0) {
2008     read chunk-data and CRLF
2009     append chunk-data to decoded-body
2010     length := length + chunk-size
2011     read chunk-size and CRLF
2012  }
2013  read header-field
2014  while (header-field not empty) {
2015     append header-field to existing header fields
2016     read header-field
2017  }
2018  Content-Length := length
2019  Remove "chunked" from Transfer-Encoding
2022   All HTTP/1.1 applications &MUST; be able to receive and decode the
2023   "chunked" transfer-coding and &MUST; ignore chunk-ext extensions
2024   they do not understand.
2027   Use of chunk-ext extensions by senders is deprecated; they &SHOULD-NOT; be
2028   sent and definition of new chunk-extensions is discouraged.
2032<section title="Compression Codings" anchor="compression.codings">
2034   The codings defined below can be used to compress the payload of a
2035   message.
2038   &Note; Use of program names for the identification of encoding formats
2039   is not desirable and is discouraged for future encodings. Their
2040   use here is representative of historical practice, not good
2041   design.
2044   &Note; For compatibility with previous implementations of HTTP,
2045   applications &SHOULD; consider "x-gzip" and "x-compress" to be
2046   equivalent to "gzip" and "compress" respectively.
2049<section title="Compress Coding" anchor="compress.coding">
2050<iref item="compress (Coding Format)"/>
2051<iref item="Coding Format" subitem="compress"/>
2053   The "compress" format is produced by the common UNIX file compression
2054   program "compress". This format is an adaptive Lempel-Ziv-Welch
2055   coding (LZW).
2059<section title="Deflate Coding" anchor="deflate.coding">
2060<iref item="deflate (Coding Format)"/>
2061<iref item="Coding Format" subitem="deflate"/>
2063   The "deflate" format is defined as the "deflate" compression mechanism
2064   (described in <xref target="RFC1951"/>) used inside the "zlib"
2065   data format (<xref target="RFC1950"/>).
2068  <t>
2069    &Note; Some incorrect implementations send the "deflate"
2070    compressed data without the zlib wrapper.
2071   </t>
2075<section title="Gzip Coding" anchor="gzip.coding">
2076<iref item="gzip (Coding Format)"/>
2077<iref item="Coding Format" subitem="gzip"/>
2079   The "gzip" format is produced by the file compression program
2080   "gzip" (GNU zip), as described in <xref target="RFC1952"/>. This format is a
2081   Lempel-Ziv coding (LZ77) with a 32 bit CRC.
2087<section title="TE" anchor="header.te">
2088  <iref primary="true" item="TE header field" x:for-anchor=""/>
2089  <iref primary="true" item="Header Fields" subitem="TE" x:for-anchor=""/>
2090  <x:anchor-alias value="TE"/>
2091  <x:anchor-alias value="t-codings"/>
2092  <x:anchor-alias value="te-params"/>
2093  <x:anchor-alias value="te-ext"/>
2095   The "TE" header field indicates what extension transfer-codings
2096   the client is willing to accept in the response, and whether or not it is
2097   willing to accept trailer fields in a chunked transfer-coding.
2100   Its value consists of the keyword "trailers" and/or a comma-separated
2101   list of extension transfer-coding names with optional accept
2102   parameters (as described in <xref target="transfer.codings"/>).
2104<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="t-codings"/><iref primary="true" item="Grammar" subitem="te-params"/><iref primary="true" item="Grammar" subitem="te-ext"/>
2105  <x:ref>TE</x:ref>        = #<x:ref>t-codings</x:ref>
2106  <x:ref>t-codings</x:ref> = "trailers" / ( <x:ref>transfer-extension</x:ref> [ <x:ref>te-params</x:ref> ] )
2107  <x:ref>te-params</x:ref> = <x:ref>OWS</x:ref> ";" <x:ref>OWS</x:ref> "q=" <x:ref>qvalue</x:ref> *( <x:ref>te-ext</x:ref> )
2108  <x:ref>te-ext</x:ref>    = <x:ref>OWS</x:ref> ";" <x:ref>OWS</x:ref> <x:ref>token</x:ref> [ "=" <x:ref>word</x:ref> ]
2111   The presence of the keyword "trailers" indicates that the client is
2112   willing to accept trailer fields in a chunked transfer-coding, as
2113   defined in <xref target="chunked.encoding"/>. This keyword is reserved for use with
2114   transfer-coding values even though it does not itself represent a
2115   transfer-coding.
2118   Examples of its use are:
2120<figure><artwork type="example">
2121  TE: deflate
2122  TE:
2123  TE: trailers, deflate;q=0.5
2126   The TE header field only applies to the immediate connection.
2127   Therefore, the keyword &MUST; be supplied within a <x:ref>Connection</x:ref>
2128   header field (<xref target="header.connection"/>) whenever TE is present in
2129   an HTTP/1.1 message.
2132   A server tests whether a transfer-coding is acceptable, according to
2133   a TE field, using these rules:
2134  <list style="numbers">
2135    <x:lt>
2136      <t>The "chunked" transfer-coding is always acceptable. If the
2137         keyword "trailers" is listed, the client indicates that it is
2138         willing to accept trailer fields in the chunked response on
2139         behalf of itself and any downstream clients. The implication is
2140         that, if given, the client is stating that either all
2141         downstream clients are willing to accept trailer fields in the
2142         forwarded response, or that it will attempt to buffer the
2143         response on behalf of downstream recipients.
2144      </t><t>
2145         &Note; HTTP/1.1 does not define any means to limit the size of a
2146         chunked response such that a client can be assured of buffering
2147         the entire response.</t>
2148    </x:lt>
2149    <x:lt>
2150      <t>If the transfer-coding being tested is one of the transfer-codings
2151         listed in the TE field, then it is acceptable unless it
2152         is accompanied by a qvalue of 0. (As defined in <xref target="quality.values"/>, a
2153         qvalue of 0 means "not acceptable".)</t>
2154    </x:lt>
2155    <x:lt>
2156      <t>If multiple transfer-codings are acceptable, then the
2157         acceptable transfer-coding with the highest non-zero qvalue is
2158         preferred.  The "chunked" transfer-coding always has a qvalue
2159         of 1.</t>
2160    </x:lt>
2161  </list>
2164   If the TE field-value is empty or if no TE field is present, the only
2165   acceptable transfer-coding is "chunked". A message with no transfer-coding is
2166   always acceptable.
2169<section title="Quality Values" anchor="quality.values">
2170  <x:anchor-alias value="qvalue"/>
2172   Both transfer codings (<x:ref>TE</x:ref> request header field,
2173   <xref target="header.te"/>) and content negotiation (&content.negotiation;)
2174   use short "floating point" numbers to indicate the relative importance
2175   ("weight") of various negotiable parameters.  A weight is normalized to a
2176   real number in the range 0 through 1, where 0 is the minimum and 1 the
2177   maximum value. If a parameter has a quality value of 0, then content with
2178   this parameter is "not acceptable" for the client. HTTP/1.1
2179   applications &MUST-NOT; generate more than three digits after the
2180   decimal point. User configuration of these values &SHOULD; also be
2181   limited in this fashion.
2183<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="qvalue"/>
2184  <x:ref>qvalue</x:ref>         = ( "0" [ "." 0*3<x:ref>DIGIT</x:ref> ] )
2185                 / ( "1" [ "." 0*3("0") ] )
2188  <t>
2189     &Note; "Quality values" is a misnomer, since these values merely represent
2190     relative degradation in desired quality.
2191  </t>
2196<section title="Trailer" anchor="header.trailer">
2197  <iref primary="true" item="Trailer header field" x:for-anchor=""/>
2198  <iref primary="true" item="Header Fields" subitem="Trailer" x:for-anchor=""/>
2199  <x:anchor-alias value="Trailer"/>
2201   The "Trailer" header field indicates that the given set of
2202   header fields is present in the trailer of a message encoded with
2203   chunked transfer-coding.
2205<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Trailer"/>
2206  <x:ref>Trailer</x:ref> = 1#<x:ref>field-name</x:ref>
2209   An HTTP/1.1 message &SHOULD; include a Trailer header field in a
2210   message using chunked transfer-coding with a non-empty trailer. Doing
2211   so allows the recipient to know which header fields to expect in the
2212   trailer.
2215   If no Trailer header field is present, the trailer &SHOULD-NOT;  include
2216   any header fields. See <xref target="chunked.encoding"/> for restrictions on the use of
2217   trailer fields in a "chunked" transfer-coding.
2220   Message header fields listed in the Trailer header field &MUST-NOT;
2221   include the following header fields:
2222  <list style="symbols">
2223    <t><x:ref>Transfer-Encoding</x:ref></t>
2224    <t><x:ref>Content-Length</x:ref></t>
2225    <t><x:ref>Trailer</x:ref></t>
2226  </list>
2231<section title="Message Routing" anchor="message.routing">
2233   HTTP request message routing is determined by each client based on the
2234   target resource, the client's proxy configuration, and
2235   establishment or reuse of an inbound connection.  The corresponding
2236   response routing follows the same connection chain back to the client.
2239<section title="Identifying a Target Resource" anchor="target-resource">
2240  <iref primary="true" item="target resource"/>
2241  <iref primary="true" item="target URI"/>
2243   HTTP is used in a wide variety of applications, ranging from
2244   general-purpose computers to home appliances.  In some cases,
2245   communication options are hard-coded in a client's configuration.
2246   However, most HTTP clients rely on the same resource identification
2247   mechanism and configuration techniques as general-purpose Web browsers.
2250   HTTP communication is initiated by a user agent for some purpose.
2251   The purpose is a combination of request semantics, which are defined in
2252   <xref target="Part2"/>, and a target resource upon which to apply those
2253   semantics.  A URI reference (<xref target="uri"/>) is typically used as
2254   an identifier for the "target resource", which a user agent would resolve
2255   to its absolute form in order to obtain the "target URI".  The target URI
2256   excludes the reference's fragment identifier component, if any,
2257   since fragment identifiers are reserved for client-side processing
2258   (<xref target="RFC3986" x:fmt="," x:sec="3.5"/>).
2261   HTTP intermediaries obtain the request semantics and target URI
2262   from the request-line of an incoming request message.
2266<section title="Connecting Inbound" anchor="connecting.inbound">
2268   Once the target URI is determined, a client needs to decide whether
2269   a network request is necessary to accomplish the desired semantics and,
2270   if so, where that request is to be directed.
2273   If the client has a response cache and the request semantics can be
2274   satisfied by a cache (<xref target="Part6"/>), then the request is
2275   usually directed to the cache first.
2278   If the request is not satisfied by a cache, then a typical client will
2279   check its configuration to determine whether a proxy is to be used to
2280   satisfy the request.  Proxy configuration is implementation-dependent,
2281   but is often based on URI prefix matching, selective authority matching,
2282   or both, and the proxy itself is usually identified by an "http" or
2283   "https" URI.  If a proxy is applicable, the client connects inbound by
2284   establishing (or reusing) a connection to that proxy.
2287   If no proxy is applicable, a typical client will invoke a handler routine,
2288   usually specific to the target URI's scheme, to connect directly
2289   to an authority for the target resource.  How that is accomplished is
2290   dependent on the target URI scheme and defined by its associated
2291   specification, similar to how this specification defines origin server
2292   access for resolution of the "http" (<xref target="http.uri"/>) and
2293   "https" (<xref target="https.uri"/>) schemes.
2297<section title="Request Target" anchor="request-target">
2299   Once an inbound connection is obtained
2300   (<xref target=""/>),
2301   the client sends an HTTP request message (<xref target="http.message"/>)
2302   with a request-target derived from the target URI.
2303   There are four distinct formats for the request-target, depending on both
2304   the method being requested and whether the request is to a proxy.
2306<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="request-target"/><iref primary="true" item="Grammar" subitem="origin-form"/><iref primary="true" item="Grammar" subitem="absolute-form"/><iref primary="true" item="Grammar" subitem="authority-form"/><iref primary="true" item="Grammar" subitem="asterisk-form"/>
2307  <x:ref>request-target</x:ref> = <x:ref>origin-form</x:ref>
2308                 / <x:ref>absolute-form</x:ref>
2309                 / <x:ref>authority-form</x:ref>
2310                 / <x:ref>asterisk-form</x:ref>
2312  <x:ref>origin-form</x:ref>    = <x:ref>path-absolute</x:ref> [ "?" <x:ref>query</x:ref> ]
2313  <x:ref>absolute-form</x:ref>  = <x:ref>absolute-URI</x:ref>
2314  <x:ref>authority-form</x:ref> = <x:ref>authority</x:ref>
2315  <x:ref>asterisk-form</x:ref>  = "*"
2317<t anchor="origin-form"><iref item="origin-form (of request-target)"/>
2318   The most common form of request-target is the origin-form.
2319   When making a request directly to an origin server, other than a CONNECT
2320   or server-wide OPTIONS request (as detailed below),
2321   a client &MUST; send only the absolute path and query components of
2322   the target URI as the request-target.
2323   If the target URI's path component is empty, then the client &MUST; send
2324   "/" as the path within the origin-form of request-target.
2325   A <x:ref>Host</x:ref> header field is also sent, as defined in
2326   <xref target=""/>, containing the target URI's
2327   authority component (excluding any userinfo).
2330   For example, a client wishing to retrieve a representation of the resource
2331   identified as
2333<figure><artwork x:indent-with="  " type="example">
2337   directly from the origin server would open (or reuse) a TCP connection
2338   to port 80 of the host "" and send the lines:
2340<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2341GET /where?q=now HTTP/1.1
2345   followed by the remainder of the request message.
2347<t anchor="absolute-form"><iref item="absolute-form (of request-target)"/>
2348   When making a request to a proxy, other than a CONNECT or server-wide
2349   OPTIONS request (as detailed below), a client &MUST; send the target URI
2350   in absolute-form as the request-target.
2351   The proxy is requested to either service that request from a valid cache,
2352   if possible, or make the same request on the client's behalf to either
2353   the next inbound proxy server or directly to the origin server indicated
2354   by the request-target.  Requirements on such "forwarding" of messages are
2355   defined in <xref target="intermediary.forwarding"/>.
2358   An example absolute-form of request-line would be:
2360<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2361GET HTTP/1.1
2364   To allow for transition to the absolute-form for all requests in some
2365   future version of HTTP, HTTP/1.1 servers &MUST; accept the absolute-form
2366   in requests, even though HTTP/1.1 clients will only send them in requests
2367   to proxies.
2369<t anchor="authority-form"><iref item="authority-form (of request-target)"/>
2370   The authority-form of request-target is only used for CONNECT requests
2371   (&CONNECT;).  When making a CONNECT request to establish a tunnel through
2372   one or more proxies, a client &MUST; send only the target URI's
2373   authority component (excluding any userinfo) as the request-target.
2374   For example,
2376<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2379<t anchor="asterisk-form"><iref item="asterisk-form (of request-target)"/>
2380   The asterisk-form of request-target is only used for a server-wide
2381   OPTIONS request (&OPTIONS;).  When a client wishes to request OPTIONS
2382   for the server as a whole, as opposed to a specific named resource of
2383   that server, the client &MUST; send only "*" (%x2A) as the request-target.
2384   For example,
2386<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2387OPTIONS * HTTP/1.1
2390   If a proxy receives an OPTIONS request with an absolute-form of
2391   request-target in which the URI has an empty path and no query component,
2392   then the last proxy on the request chain &MUST; send a request-target
2393   of "*" when it forwards the request to the indicated origin server.
2396   For example, the request
2397</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2401  would be forwarded by the final proxy as
2402</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2403OPTIONS * HTTP/1.1
2407   after connecting to port 8001 of host "".
2412<section title="Host" anchor="">
2413  <iref primary="true" item="Host header field" x:for-anchor=""/>
2414  <iref primary="true" item="Header Fields" subitem="Host" x:for-anchor=""/>
2415  <x:anchor-alias value="Host"/>
2417   The "Host" header field in a request provides the host and port
2418   information from the target URI, enabling the origin
2419   server to distinguish among resources while servicing requests
2420   for multiple host names on a single IP address.  Since the Host
2421   field-value is critical information for handling a request, it
2422   &SHOULD; be sent as the first header field following the request-line.
2424<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Host"/>
2425  <x:ref>Host</x:ref> = <x:ref>uri-host</x:ref> [ ":" <x:ref>port</x:ref> ] ; <xref target="http.uri"/>
2428   A client &MUST; send a Host header field in all HTTP/1.1 request
2429   messages.  If the target URI includes an authority component, then
2430   the Host field-value &MUST; be identical to that authority component
2431   after excluding any userinfo (<xref target="http.uri"/>).
2432   If the authority component is missing or undefined for the target URI,
2433   then the Host header field &MUST; be sent with an empty field-value.
2436   For example, a GET request to the origin server for
2437   &lt;; would begin with:
2439<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2440GET /pub/WWW/ HTTP/1.1
2444   The Host header field &MUST; be sent in an HTTP/1.1 request even
2445   if the request-target is in the absolute-form, since this
2446   allows the Host information to be forwarded through ancient HTTP/1.0
2447   proxies that might not have implemented Host.
2450   When an HTTP/1.1 proxy receives a request with an absolute-form of
2451   request-target, the proxy &MUST; ignore the received
2452   Host header field (if any) and instead replace it with the host
2453   information of the request-target.  If the proxy forwards the request,
2454   it &MUST; generate a new Host field-value based on the received
2455   request-target rather than forward the received Host field-value.
2458   Since the Host header field acts as an application-level routing
2459   mechanism, it is a frequent target for malware seeking to poison
2460   a shared cache or redirect a request to an unintended server.
2461   An interception proxy is particularly vulnerable if it relies on
2462   the Host field-value for redirecting requests to internal
2463   servers, or for use as a cache key in a shared cache, without
2464   first verifying that the intercepted connection is targeting a
2465   valid IP address for that host.
2468   A server &MUST; respond with a <x:ref>400 (Bad Request)</x:ref> status code
2469   to any HTTP/1.1 request message that lacks a Host header field and
2470   to any request message that contains more than one Host header field
2471   or a Host header field with an invalid field-value.
2475<section title="Effective Request URI" anchor="effective.request.uri">
2476  <iref primary="true" item="effective request URI"/>
2478   A server that receives an HTTP request message &MUST; reconstruct
2479   the user agent's original target URI, based on the pieces of information
2480   learned from the request-target, <x:ref>Host</x:ref> header field, and
2481   connection context, in order to identify the intended target resource and
2482   properly service the request. The URI derived from this reconstruction
2483   process is referred to as the "effective request URI".
2486   For a user agent, the effective request URI is the target URI.
2489   If the request-target is in absolute-form, then the effective request URI
2490   is the same as the request-target.  Otherwise, the effective request URI
2491   is constructed as follows.
2494   If the request is received over an SSL/TLS-secured TCP connection,
2495   then the effective request URI's scheme is "https"; otherwise, the
2496   scheme is "http".
2499   If the request-target is in authority-form, then the effective
2500   request URI's authority component is the same as the request-target.
2501   Otherwise, if a <x:ref>Host</x:ref> header field is supplied with a
2502   non-empty field-value, then the authority component is the same as the
2503   Host field-value. Otherwise, the authority component is the concatenation of
2504   the default host name configured for the server, a colon (":"), and the
2505   connection's incoming TCP port number in decimal form.
2508   If the request-target is in authority-form or asterisk-form, then the
2509   effective request URI's combined path and query component is empty.
2510   Otherwise, the combined path and query component is the same as the
2511   request-target.
2514   The components of the effective request URI, once determined as above,
2515   can be combined into absolute-URI form by concatenating the scheme,
2516   "://", authority, and combined path and query component.
2520   Example 1: the following message received over an insecure TCP connection
2522<artwork type="example" x:indent-with="  ">
2523GET /pub/WWW/TheProject.html HTTP/1.1
2529  has an effective request URI of
2531<artwork type="example" x:indent-with="  ">
2537   Example 2: the following message received over an SSL/TLS-secured TCP
2538   connection
2540<artwork type="example" x:indent-with="  ">
2541OPTIONS * HTTP/1.1
2547  has an effective request URI of
2549<artwork type="example" x:indent-with="  ">
2554   An origin server that does not allow resources to differ by requested
2555   host &MAY; ignore the <x:ref>Host</x:ref> field-value and instead replace it
2556   with a configured server name when constructing the effective request URI.
2559   Recipients of an HTTP/1.0 request that lacks a <x:ref>Host</x:ref> header
2560   field &MAY; attempt to use heuristics (e.g., examination of the URI path for
2561   something unique to a particular host) in order to guess the
2562   effective request URI's authority component.
2566<section title="Intermediary Forwarding" anchor="intermediary.forwarding">
2568   As described in <xref target="intermediaries"/>, intermediaries can serve
2569   a variety of roles in the processing of HTTP requests and responses.
2570   Some intermediaries are used to improve performance or availability.
2571   Others are used for access control or to filter content.
2572   Since an HTTP stream has characteristics similar to a pipe-and-filter
2573   architecture, there are no inherent limits to the extent an intermediary
2574   can enhance (or interfere) with either direction of the stream.
2577   In order to avoid request loops, a proxy that forwards requests to other
2578   proxies &MUST; be able to recognize and exclude all of its own server
2579   names, including any aliases, local variations, or literal IP addresses.
2582   If a proxy receives a request-target with a host name that is not a
2583   fully qualified domain name, it &MAY; add its domain to the host name
2584   it received when forwarding the request.  A proxy &MUST-NOT; change the
2585   host name if it is a fully qualified domain name.
2588   A non-transforming proxy &MUST-NOT; rewrite the "path-absolute" and "query"
2589   parts of the received request-target when forwarding it to the next inbound
2590   server, except as noted above to replace an empty path with "/" or "*".
2593   Intermediaries that forward a message &MUST; implement the
2594   <x:ref>Connection</x:ref> header field as specified in
2595   <xref target="header.connection"/>.
2598<section title="End-to-end and Hop-by-hop Header Fields" anchor="end-to-end.and.hop-by-hop.header-fields">
2600  <cref anchor="TODO-end-to-end" source="jre">
2601    Restored from <eref target=""/>.
2602    See also <eref target=""/>.
2603  </cref>
2606   For the purpose of defining the behavior of caches and non-caching
2607   proxies, we divide HTTP header fields into two categories:
2608  <list style="symbols">
2609      <t>End-to-end header fields, which are  transmitted to the ultimate
2610        recipient of a request or response. End-to-end header fields in
2611        responses &MUST; be stored as part of a cache entry and &MUST; be
2612        transmitted in any response formed from a cache entry.</t>
2614      <t>Hop-by-hop header fields, which are meaningful only for a single
2615        transport-level connection, and are not stored by caches or
2616        forwarded by proxies.</t>
2617  </list>
2620   The following HTTP/1.1 header fields are hop-by-hop header fields:
2621  <list style="symbols">
2622      <t><x:ref>Connection</x:ref></t>
2623      <t>Keep-Alive (<xref target="RFC2068" x:fmt="of" x:sec=""/>)</t>
2624      <t><x:ref>Proxy-Authenticate</x:ref> (&header-proxy-authenticate;)</t>
2625      <t><x:ref>Proxy-Authorization</x:ref> (&header-proxy-authorization;)</t>
2626      <t><x:ref>TE</x:ref></t>
2627      <t><x:ref>Trailer</x:ref></t>
2628      <t><x:ref>Transfer-Encoding</x:ref></t>
2629      <t><x:ref>Upgrade</x:ref></t>
2630  </list>
2633   All other header fields defined by HTTP/1.1 are end-to-end header fields.
2636   Other hop-by-hop header fields &MUST; be listed in a
2637   <x:ref>Connection</x:ref> header field (<xref target="header.connection"/>).
2641<section title="Non-modifiable Header Fields" anchor="non-modifiable.header-fields">
2643  <cref anchor="TODO-non-mod-headers" source="jre">
2644    Restored from <eref target=""/>.
2645    See also <eref target=""/>.
2646  </cref>
2649   Some features of HTTP/1.1, such as Digest Authentication, depend on the
2650   value of certain end-to-end header fields. A non-transforming proxy &SHOULD-NOT;
2651   modify an end-to-end header field unless the definition of that header field requires
2652   or specifically allows that.
2655   A non-transforming proxy &MUST-NOT; modify any of the following fields in a
2656   request or response, and it &MUST-NOT; add any of these fields if not
2657   already present:
2658  <list style="symbols">
2659    <t>Allow</t>
2660    <t>Content-Location</t>
2661    <t>Content-MD5</t>
2662    <t>ETag</t>
2663    <t>Last-Modified</t>
2664    <t>Server</t>
2665  </list>
2668   A non-transforming proxy &MUST-NOT; modify any of the following fields in a
2669   response:
2670  <list style="symbols">
2671    <t><x:ref>Expires</x:ref> (&header-expires;)</t>
2672  </list>
2675   but it &MAY; add any of these fields if not already present. If an
2676   <x:ref>Expires</x:ref> header field is added, it &MUST; be given a
2677   field value identical to that of the <x:ref>Date</x:ref> header field in
2678   that response.
2681   A proxy &MUST-NOT; modify or add any of the following fields in a
2682   message that contains the no-transform cache-control directive, or in
2683   any request:
2684  <list style="symbols">
2685    <t><x:ref>Content-Encoding</x:ref> (&header-content-encoding;)</t>
2686    <t><x:ref>Content-Range</x:ref> (&header-content-range;)</t>
2687    <t><x:ref>Content-Type</x:ref> (&header-content-type;)</t>
2688  </list>
2691   A transforming proxy &MAY; modify or add these fields to a message
2692   that does not include no-transform, but if it does so, it &MUST; add a
2693   Warning 214 (Transformation applied) if one does not already appear
2694   in the message (see &header-warning;).
2697  <t>
2698    <x:h>Warning:</x:h> Unnecessary modification of end-to-end header fields might
2699    cause authentication failures if stronger authentication
2700    mechanisms are introduced in later versions of HTTP. Such
2701    authentication mechanisms &MAY; rely on the values of header fields
2702    not listed here.
2703  </t>
2706   A non-transforming proxy &MUST; preserve the message payload (&payload;),
2707   though it &MAY; change the message body through application or removal
2708   of a transfer-coding (<xref target="transfer.codings"/>).
2714<section title="Associating a Response to a Request" anchor="">
2716   HTTP does not include a request identifier for associating a given
2717   request message with its corresponding one or more response messages.
2718   Hence, it relies on the order of response arrival to correspond exactly
2719   to the order in which requests are made on the same connection.
2720   More than one response message per request only occurs when one or more
2721   informational responses (<x:ref>1xx</x:ref>, see &status-1xx;) precede a final response
2722   to the same request.
2725   A client that uses persistent connections and sends more than one request
2726   per connection &MUST; maintain a list of outstanding requests in the
2727   order sent on that connection and &MUST; associate each received response
2728   message to the highest ordered request that has not yet received a final
2729   (non-<x:ref>1xx</x:ref>) response.
2734<section title="Connection Management" anchor="">
2736<section title="Connection" anchor="header.connection">
2737  <iref primary="true" item="Connection header field" x:for-anchor=""/>
2738  <iref primary="true" item="Header Fields" subitem="Connection" x:for-anchor=""/>
2739  <x:anchor-alias value="Connection"/>
2740  <x:anchor-alias value="connection-option"/>
2742   The "Connection" header field allows the sender to specify
2743   options that are desired only for that particular connection.
2744   Such connection options &MUST; be removed or replaced before the
2745   message can be forwarded downstream by a proxy or gateway.
2746   This mechanism also allows the sender to indicate which HTTP
2747   header fields used in the message are only intended for the
2748   immediate recipient ("hop-by-hop"), as opposed to all recipients
2749   on the chain ("end-to-end"), enabling the message to be
2750   self-descriptive and allowing future connection-specific extensions
2751   to be deployed in HTTP without fear that they will be blindly
2752   forwarded by previously deployed intermediaries.
2755   The Connection header field's value has the following grammar:
2757<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="connection-option"/>
2758  <x:ref>Connection</x:ref>        = 1#<x:ref>connection-option</x:ref>
2759  <x:ref>connection-option</x:ref> = <x:ref>token</x:ref>
2762   Connection options are compared case-insensitively.
2765   A proxy or gateway &MUST; parse a received Connection
2766   header field before a message is forwarded and, for each
2767   connection-option in this field, remove any header field(s) from
2768   the message with the same name as the connection-option, and then
2769   remove the Connection header field itself or replace it with the
2770   sender's own connection options for the forwarded message.
2773   A sender &MUST-NOT; include field-names in the Connection header
2774   field-value for fields that are defined as expressing constraints
2775   for all recipients in the request or response chain, such as the
2776   Cache-Control header field (&header-cache-control;).
2779   The connection options do not have to correspond to a header field
2780   present in the message, since a connection-specific header field
2781   might not be needed if there are no parameters associated with that
2782   connection option.  Recipients that trigger certain connection
2783   behavior based on the presence of connection options &MUST; do so
2784   based on the presence of the connection-option rather than only the
2785   presence of the optional header field.  In other words, if the
2786   connection option is received as a header field but not indicated
2787   within the Connection field-value, then the recipient &MUST; ignore
2788   the connection-specific header field because it has likely been
2789   forwarded by an intermediary that is only partially conformant.
2792   When defining new connection options, specifications ought to
2793   carefully consider existing deployed header fields and ensure
2794   that the new connection option does not share the same name as
2795   an unrelated header field that might already be deployed.
2796   Defining a new connection option essentially reserves that potential
2797   field-name for carrying additional information related to the
2798   connection option, since it would be unwise for senders to use
2799   that field-name for anything else.
2802   HTTP/1.1 defines the "close" connection option for the sender to
2803   signal that the connection will be closed after completion of the
2804   response. For example,
2806<figure><artwork type="example">
2807  Connection: close
2810   in either the request or the response header fields indicates that
2811   the connection &SHOULD-NOT;  be considered "persistent" (<xref target="persistent.connections"/>)
2812   after the current request/response is complete.
2815   An HTTP/1.1 client that does not support persistent connections &MUST;
2816   include the "close" connection option in every request message.
2819   An HTTP/1.1 server that does not support persistent connections &MUST;
2820   include the "close" connection option in every response message that
2821   does not have a <x:ref>1xx (Informational)</x:ref> status code.
2825<section title="Via" anchor="header.via">
2826  <iref primary="true" item="Via header field" x:for-anchor=""/>
2827  <iref primary="true" item="Header Fields" subitem="Via" x:for-anchor=""/>
2828  <x:anchor-alias value="pseudonym"/>
2829  <x:anchor-alias value="received-by"/>
2830  <x:anchor-alias value="received-protocol"/>
2831  <x:anchor-alias value="Via"/>
2833   The "Via" header field &MUST; be sent by a proxy or gateway to
2834   indicate the intermediate protocols and recipients between the user
2835   agent and the server on requests, and between the origin server and
2836   the client on responses. It is analogous to the "Received" field
2837   used by email systems (<xref target="RFC5322" x:fmt="of" x:sec="3.6.7"/>)
2838   and is intended to be used for tracking message forwards,
2839   avoiding request loops, and identifying the protocol capabilities of
2840   all senders along the request/response chain.
2842<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Via"/><iref primary="true" item="Grammar" subitem="received-protocol"/><iref primary="true" item="Grammar" subitem="protocol-name"/><iref primary="true" item="Grammar" subitem="protocol-version"/><iref primary="true" item="Grammar" subitem="received-by"/><iref primary="true" item="Grammar" subitem="pseudonym"/>
2843  <x:ref>Via</x:ref>               = 1#( <x:ref>received-protocol</x:ref> <x:ref>RWS</x:ref> <x:ref>received-by</x:ref>
2844                          [ <x:ref>RWS</x:ref> <x:ref>comment</x:ref> ] )
2845  <x:ref>received-protocol</x:ref> = [ <x:ref>protocol-name</x:ref> "/" ] <x:ref>protocol-version</x:ref>
2846  <x:ref>received-by</x:ref>       = ( <x:ref>uri-host</x:ref> [ ":" <x:ref>port</x:ref> ] ) / <x:ref>pseudonym</x:ref>
2847  <x:ref>pseudonym</x:ref>         = <x:ref>token</x:ref>
2850   The received-protocol indicates the protocol version of the message
2851   received by the server or client along each segment of the
2852   request/response chain. The received-protocol version is appended to
2853   the Via field value when the message is forwarded so that information
2854   about the protocol capabilities of upstream applications remains
2855   visible to all recipients.
2858   The protocol-name is excluded if and only if it would be "HTTP". The
2859   received-by field is normally the host and optional port number of a
2860   recipient server or client that subsequently forwarded the message.
2861   However, if the real host is considered to be sensitive information,
2862   it &MAY; be replaced by a pseudonym. If the port is not given, it &MAY;
2863   be assumed to be the default port of the received-protocol.
2866   Multiple Via field values represent each proxy or gateway that has
2867   forwarded the message. Each recipient &MUST; append its information
2868   such that the end result is ordered according to the sequence of
2869   forwarding applications.
2872   Comments &MAY; be used in the Via header field to identify the software
2873   of each recipient, analogous to the <x:ref>User-Agent</x:ref> and
2874   <x:ref>Server</x:ref> header fields. However, all comments in the Via field
2875   are optional and &MAY; be removed by any recipient prior to forwarding the
2876   message.
2879   For example, a request message could be sent from an HTTP/1.0 user
2880   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
2881   forward the request to a public proxy at, which completes
2882   the request by forwarding it to the origin server at
2883   The request received by would then have the following
2884   Via header field:
2886<figure><artwork type="example">
2887  Via: 1.0 fred, 1.1 (Apache/1.1)
2890   A proxy or gateway used as a portal through a network firewall
2891   &SHOULD-NOT; forward the names and ports of hosts within the firewall
2892   region unless it is explicitly enabled to do so. If not enabled, the
2893   received-by host of any host behind the firewall &SHOULD; be replaced
2894   by an appropriate pseudonym for that host.
2897   For organizations that have strong privacy requirements for hiding
2898   internal structures, a proxy or gateway &MAY; combine an ordered
2899   subsequence of Via header field entries with identical received-protocol
2900   values into a single such entry. For example,
2902<figure><artwork type="example">
2903  Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
2906  could be collapsed to
2908<figure><artwork type="example">
2909  Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
2912   Senders &SHOULD-NOT; combine multiple entries unless they are all
2913   under the same organizational control and the hosts have already been
2914   replaced by pseudonyms. Senders &MUST-NOT; combine entries which
2915   have different received-protocol values.
2919<section title="Persistent Connections" anchor="persistent.connections">
2921<section title="Purpose" anchor="persistent.purpose">
2923   Prior to persistent connections, a separate TCP connection was
2924   established for each request, increasing the load on HTTP servers
2925   and causing congestion on the Internet. The use of inline images and
2926   other associated data often requires a client to make multiple
2927   requests of the same server in a short amount of time. Analysis of
2928   these performance problems and results from a prototype
2929   implementation are available <xref target="Pad1995"/> <xref target="Spe"/>. Implementation experience and
2930   measurements of actual HTTP/1.1 implementations show good
2931   results <xref target="Nie1997"/>. Alternatives have also been explored, for example,
2932   T/TCP <xref target="Tou1998"/>.
2935   Persistent HTTP connections have a number of advantages:
2936  <list style="symbols">
2937      <t>
2938        By opening and closing fewer TCP connections, CPU time is saved
2939        in routers and hosts (clients, servers, proxies, gateways,
2940        tunnels, or caches), and memory used for TCP protocol control
2941        blocks can be saved in hosts.
2942      </t>
2943      <t>
2944        HTTP requests and responses can be pipelined on a connection.
2945        Pipelining allows a client to make multiple requests without
2946        waiting for each response, allowing a single TCP connection to
2947        be used much more efficiently, with much lower elapsed time.
2948      </t>
2949      <t>
2950        Network congestion is reduced by reducing the number of packets
2951        caused by TCP opens, and by allowing TCP sufficient time to
2952        determine the congestion state of the network.
2953      </t>
2954      <t>
2955        Latency on subsequent requests is reduced since there is no time
2956        spent in TCP's connection opening handshake.
2957      </t>
2958      <t>
2959        HTTP can evolve more gracefully, since errors can be reported
2960        without the penalty of closing the TCP connection. Clients using
2961        future versions of HTTP might optimistically try a new feature,
2962        but if communicating with an older server, retry with old
2963        semantics after an error is reported.
2964      </t>
2965    </list>
2968   HTTP implementations &SHOULD; implement persistent connections.
2972<section title="Overall Operation" anchor="persistent.overall">
2974   A significant difference between HTTP/1.1 and earlier versions of
2975   HTTP is that persistent connections are the default behavior of any
2976   HTTP connection. That is, unless otherwise indicated, the client
2977   &SHOULD; assume that the server will maintain a persistent connection,
2978   even after error responses from the server.
2981   Persistent connections provide a mechanism by which a client and a
2982   server can signal the close of a TCP connection. This signaling takes
2983   place using the <x:ref>Connection</x:ref> header field
2984   (<xref target="header.connection"/>). Once a close has been signaled, the
2985   client &MUST-NOT; send any more requests on that
2986   connection.
2989<section title="Negotiation" anchor="persistent.negotiation">
2991   An HTTP/1.1 server &MAY; assume that a HTTP/1.1 client intends to
2992   maintain a persistent connection unless a <x:ref>Connection</x:ref> header
2993   field including the connection option "close" was sent in the request. If
2994   the server chooses to close the connection immediately after sending the
2995   response, it &SHOULD; send a Connection header field including the
2996   connection option "close".
2999   An HTTP/1.1 client &MAY; expect a connection to remain open, but would
3000   decide to keep it open based on whether the response from a server
3001   contains a <x:ref>Connection</x:ref> header field with the connection option
3002   "close". In case the client does not want to maintain a connection for more
3003   than that request, it &SHOULD; send a Connection header field including the
3004   connection option "close".
3007   If either the client or the server sends the "close" option in the
3008   <x:ref>Connection</x:ref> header field, that request becomes the last one
3009   for the connection.
3012   Clients and servers &SHOULD-NOT;  assume that a persistent connection is
3013   maintained for HTTP versions less than 1.1 unless it is explicitly
3014   signaled. See <xref target="compatibility.with.http.1.0.persistent.connections"/> for more information on backward
3015   compatibility with HTTP/1.0 clients.
3018   Each persistent connection applies to only one transport link.
3021   A proxy server &MUST-NOT; establish a HTTP/1.1 persistent connection
3022   with an HTTP/1.0 client (but see <xref x:sec="19.7.1" x:fmt="of" target="RFC2068"/>
3023   for information and discussion of the problems with the Keep-Alive header field
3024   implemented by many HTTP/1.0 clients).
3027   In order to remain persistent, all messages on the connection &MUST;
3028   have a self-defined message length (i.e., one not defined by closure
3029   of the connection), as described in <xref target="message.body"/>.
3033<section title="Pipelining" anchor="pipelining">
3035   A client that supports persistent connections &MAY; "pipeline" its
3036   requests (i.e., send multiple requests without waiting for each
3037   response). A server &MUST; send its responses to those requests in the
3038   same order that the requests were received.
3041   Clients which assume persistent connections and pipeline immediately
3042   after connection establishment &SHOULD; be prepared to retry their
3043   connection if the first pipelined attempt fails. If a client does
3044   such a retry, it &MUST-NOT; pipeline before it knows the connection is
3045   persistent. Clients &MUST; also be prepared to resend their requests if
3046   the server closes the connection before sending all of the
3047   corresponding responses.
3050   Clients &SHOULD-NOT; pipeline requests using non-idempotent request methods or
3051   non-idempotent sequences of request methods (see &idempotent-methods;). Otherwise, a
3052   premature termination of the transport connection could lead to
3053   indeterminate results. A client wishing to send a non-idempotent
3054   request &SHOULD; wait to send that request until it has received the
3055   response status line for the previous request.
3060<section title="Practical Considerations" anchor="persistent.practical">
3062   Servers will usually have some time-out value beyond which they will
3063   no longer maintain an inactive connection. Proxy servers might make
3064   this a higher value since it is likely that the client will be making
3065   more connections through the same server. The use of persistent
3066   connections places no requirements on the length (or existence) of
3067   this time-out for either the client or the server.
3070   When a client or server wishes to time-out it &SHOULD; issue a graceful
3071   close on the transport connection. Clients and servers &SHOULD; both
3072   constantly watch for the other side of the transport close, and
3073   respond to it as appropriate. If a client or server does not detect
3074   the other side's close promptly it could cause unnecessary resource
3075   drain on the network.
3078   A client, server, or proxy &MAY; close the transport connection at any
3079   time. For example, a client might have started to send a new request
3080   at the same time that the server has decided to close the "idle"
3081   connection. From the server's point of view, the connection is being
3082   closed while it was idle, but from the client's point of view, a
3083   request is in progress.
3086   Clients (including proxies) &SHOULD; limit the number of simultaneous
3087   connections that they maintain to a given server (including proxies).
3090   Previous revisions of HTTP gave a specific number of connections as a
3091   ceiling, but this was found to be impractical for many applications. As a
3092   result, this specification does not mandate a particular maximum number of
3093   connections, but instead encourages clients to be conservative when opening
3094   multiple connections.
3097   In particular, while using multiple connections avoids the "head-of-line
3098   blocking" problem (whereby a request that takes significant server-side
3099   processing and/or has a large payload can block subsequent requests on the
3100   same connection), each connection used consumes server resources (sometimes
3101   significantly), and furthermore using multiple connections can cause
3102   undesirable side effects in congested networks.
3105   Note that servers might reject traffic that they deem abusive, including an
3106   excessive number of connections from a client.
3110<section title="Retrying Requests" anchor="persistent.retrying.requests">
3112   Senders can close the transport connection at any time. Therefore,
3113   clients, servers, and proxies &MUST; be able to recover
3114   from asynchronous close events. Client software &MAY; reopen the
3115   transport connection and retransmit the aborted sequence of requests
3116   without user interaction so long as the request sequence is
3117   idempotent (see &idempotent-methods;). Non-idempotent request methods or sequences
3118   &MUST-NOT; be automatically retried, although user agents &MAY; offer a
3119   human operator the choice of retrying the request(s). Confirmation by
3120   user-agent software with semantic understanding of the application
3121   &MAY; substitute for user confirmation. The automatic retry &SHOULD-NOT;
3122   be repeated if the second sequence of requests fails.
3127<section title="Message Transmission Requirements" anchor="message.transmission.requirements">
3129<section title="Persistent Connections and Flow Control" anchor="persistent.flow">
3131   HTTP/1.1 servers &SHOULD; maintain persistent connections and use TCP's
3132   flow control mechanisms to resolve temporary overloads, rather than
3133   terminating connections with the expectation that clients will retry.
3134   The latter technique can exacerbate network congestion.
3138<section title="Monitoring Connections for Error Status Messages" anchor="persistent.monitor">
3140   An HTTP/1.1 (or later) client sending a message body &SHOULD; monitor
3141   the network connection for an error status code while it is transmitting
3142   the request. If the client sees an error status code, it &SHOULD;
3143   immediately cease transmitting the body. If the body is being sent
3144   using a "chunked" encoding (<xref target="transfer.codings"/>), a zero length chunk and
3145   empty trailer &MAY; be used to prematurely mark the end of the message.
3146   If the body was preceded by a Content-Length header field, the client &MUST;
3147   close the connection.
3151<section title="Use of the 100 (Continue) Status" anchor="use.of.the.100.status">
3153   The purpose of the <x:ref>100 (Continue)</x:ref> status code (see &status-100;)
3154   is to allow a client that is sending a request message with a request body
3155   to determine if the origin server is willing to accept the request
3156   (based on the request header fields) before the client sends the request
3157   body. In some cases, it might either be inappropriate or highly
3158   inefficient for the client to send the body if the server will reject
3159   the message without looking at the body.
3162   Requirements for HTTP/1.1 clients:
3163  <list style="symbols">
3164    <t>
3165        If a client will wait for a <x:ref>100 (Continue)</x:ref> response before
3166        sending the request body, it &MUST; send an <x:ref>Expect</x:ref> header
3167        field (&header-expect;) with the "100-continue" expectation.
3168    </t>
3169    <t>
3170        A client &MUST-NOT; send an <x:ref>Expect</x:ref> header field with
3171        the "100-continue" expectation if it does not intend to send a request
3172        body.
3173    </t>
3174  </list>
3177   Because of the presence of older implementations, the protocol allows
3178   ambiguous situations in which a client might send "Expect: 100-continue"
3179   without receiving either a <x:ref>417 (Expectation Failed)</x:ref>
3180   or a <x:ref>100 (Continue)</x:ref> status code. Therefore, when a client sends this
3181   header field to an origin server (possibly via a proxy) from which it
3182   has never seen a <x:ref>100 (Continue)</x:ref> status code, the client &SHOULD-NOT; 
3183   wait for an indefinite period before sending the request body.
3186   Requirements for HTTP/1.1 origin servers:
3187  <list style="symbols">
3188    <t> Upon receiving a request which includes an <x:ref>Expect</x:ref> header
3189        field with the "100-continue" expectation, an origin server &MUST;
3190        either respond with <x:ref>100 (Continue)</x:ref> status code and continue to read
3191        from the input stream, or respond with a final status code. The
3192        origin server &MUST-NOT; wait for the request body before sending
3193        the <x:ref>100 (Continue)</x:ref> response. If it responds with a final status
3194        code, it &MAY; close the transport connection or it &MAY; continue
3195        to read and discard the rest of the request.  It &MUST-NOT;
3196        perform the request method if it returns a final status code.
3197    </t>
3198    <t> An origin server &SHOULD-NOT;  send a <x:ref>100 (Continue)</x:ref> response if
3199        the request message does not include an <x:ref>Expect</x:ref> header
3200        field with the "100-continue" expectation, and &MUST-NOT; send a
3201        <x:ref>100 (Continue)</x:ref> response if such a request comes from an HTTP/1.0
3202        (or earlier) client. There is an exception to this rule: for
3203        compatibility with <xref target="RFC2068"/>, a server &MAY; send a <x:ref>100 (Continue)</x:ref>
3204        status code in response to an HTTP/1.1 PUT or POST request that does
3205        not include an Expect header field with the "100-continue"
3206        expectation. This exception, the purpose of which is
3207        to minimize any client processing delays associated with an
3208        undeclared wait for <x:ref>100 (Continue)</x:ref> status code, applies only to
3209        HTTP/1.1 requests, and not to requests with any other HTTP-version
3210        value.
3211    </t>
3212    <t> An origin server &MAY; omit a <x:ref>100 (Continue)</x:ref> response if it has
3213        already received some or all of the request body for the
3214        corresponding request.
3215    </t>
3216    <t> An origin server that sends a <x:ref>100 (Continue)</x:ref> response &MUST;
3217        ultimately send a final status code, once the request body is
3218        received and processed, unless it terminates the transport
3219        connection prematurely.
3220    </t>
3221    <t> If an origin server receives a request that does not include an
3222        <x:ref>Expect</x:ref> header field with the "100-continue" expectation,
3223        the request includes a request body, and the server responds
3224        with a final status code before reading the entire request body
3225        from the transport connection, then the server &SHOULD-NOT;  close
3226        the transport connection until it has read the entire request,
3227        or until the client closes the connection. Otherwise, the client
3228        might not reliably receive the response message. However, this
3229        requirement ought not be construed as preventing a server from
3230        defending itself against denial-of-service attacks, or from
3231        badly broken client implementations.
3232      </t>
3233    </list>
3236   Requirements for HTTP/1.1 proxies:
3237  <list style="symbols">
3238    <t> If a proxy receives a request that includes an <x:ref>Expect</x:ref>
3239        header field with the "100-continue" expectation, and the proxy
3240        either knows that the next-hop server complies with HTTP/1.1 or
3241        higher, or does not know the HTTP version of the next-hop
3242        server, it &MUST; forward the request, including the Expect header
3243        field.
3244    </t>
3245    <t> If the proxy knows that the version of the next-hop server is
3246        HTTP/1.0 or lower, it &MUST-NOT; forward the request, and it &MUST;
3247        respond with a <x:ref>417 (Expectation Failed)</x:ref> status code.
3248    </t>
3249    <t> Proxies &SHOULD; maintain a record of the HTTP version
3250        numbers received from recently-referenced next-hop servers.
3251    </t>
3252    <t> A proxy &MUST-NOT; forward a <x:ref>100 (Continue)</x:ref> response if the
3253        request message was received from an HTTP/1.0 (or earlier)
3254        client and did not include an <x:ref>Expect</x:ref> header field with
3255        the "100-continue" expectation. This requirement overrides the
3256        general rule for forwarding of <x:ref>1xx</x:ref> responses (see &status-1xx;).
3257    </t>
3258  </list>
3262<section title="Closing Connections on Error" anchor="closing.connections.on.error">
3264   If the client is sending data, a server implementation using TCP
3265   &SHOULD; be careful to ensure that the client acknowledges receipt of
3266   the packet(s) containing the response, before the server closes the
3267   input connection. If the client continues sending data to the server
3268   after the close, the server's TCP stack will send a reset packet to
3269   the client, which might erase the client's unacknowledged input buffers
3270   before they can be read and interpreted by the HTTP application.
3276<section title="Upgrade" anchor="header.upgrade">
3277  <iref primary="true" item="Upgrade header field" x:for-anchor=""/>
3278  <iref primary="true" item="Header Fields" subitem="Upgrade" x:for-anchor=""/>
3279  <x:anchor-alias value="Upgrade"/>
3280  <x:anchor-alias value="protocol"/>
3281  <x:anchor-alias value="protocol-name"/>
3282  <x:anchor-alias value="protocol-version"/>
3284   The "Upgrade" header field allows the client to specify what
3285   additional communication protocols it would like to use, if the server
3286   chooses to switch protocols. Servers can use it to indicate what protocols
3287   they are willing to switch to.
3289<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Upgrade"/>
3290  <x:ref>Upgrade</x:ref>          = 1#<x:ref>protocol</x:ref>
3292  <x:ref>protocol</x:ref>         = <x:ref>protocol-name</x:ref> ["/" <x:ref>protocol-version</x:ref>]
3293  <x:ref>protocol-name</x:ref>    = <x:ref>token</x:ref>
3294  <x:ref>protocol-version</x:ref> = <x:ref>token</x:ref>
3297   For example,
3299<figure><artwork type="example">
3300  Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
3303   The Upgrade header field is intended to provide a simple mechanism
3304   for transitioning from HTTP/1.1 to some other, incompatible protocol. It
3305   does so by allowing the client to advertise its desire to use another
3306   protocol, such as a later version of HTTP with a higher major version
3307   number, even though the current request has been made using HTTP/1.1.
3308   This eases the difficult transition between incompatible protocols by
3309   allowing the client to initiate a request in the more commonly
3310   supported protocol while indicating to the server that it would like
3311   to use a "better" protocol if available (where "better" is determined
3312   by the server, possibly according to the nature of the request method
3313   or target resource).
3316   The Upgrade header field only applies to switching application-layer
3317   protocols upon the existing transport-layer connection. Upgrade
3318   cannot be used to insist on a protocol change; its acceptance and use
3319   by the server is optional. The capabilities and nature of the
3320   application-layer communication after the protocol change is entirely
3321   dependent upon the new protocol chosen, although the first action
3322   after changing the protocol &MUST; be a response to the initial HTTP
3323   request containing the Upgrade header field.
3326   The Upgrade header field only applies to the immediate connection.
3327   Therefore, the upgrade keyword &MUST; be supplied within a
3328   <x:ref>Connection</x:ref> header field (<xref target="header.connection"/>)
3329   whenever Upgrade is present in an HTTP/1.1 message.
3332   The Upgrade header field cannot be used to indicate a switch to a
3333   protocol on a different connection. For that purpose, it is more
3334   appropriate to use a <x:ref>3xx (Redirection)</x:ref> response (&status-3xx;).
3337   Servers &MUST; include the "Upgrade" header field in <x:ref>101 (Switching
3338   Protocols)</x:ref> responses to indicate which protocol(s) are being switched to,
3339   and &MUST; include it in <x:ref>426 (Upgrade Required)</x:ref> responses to indicate
3340   acceptable protocols to upgrade to. Servers &MAY; include it in any other
3341   response to indicate that they are willing to upgrade to one of the
3342   specified protocols.
3345   This specification only defines the protocol name "HTTP" for use by
3346   the family of Hypertext Transfer Protocols, as defined by the HTTP
3347   version rules of <xref target="http.version"/> and future updates to this
3348   specification. Additional tokens can be registered with IANA using the
3349   registration procedure defined in <xref target="upgrade.token.registry"/>.
3355<section title="IANA Considerations" anchor="IANA.considerations">
3357<section title="Header Field Registration" anchor="header.field.registration">
3359   HTTP header fields are registered within the Message Header Field Registry
3360   <xref target="RFC3864"/> maintained by IANA at
3361   <eref target=""/>.
3364   This document defines the following HTTP header fields, so their
3365   associated registry entries shall be updated according to the permanent
3366   registrations below:
3368<?BEGININC p1-messaging.iana-headers ?>
3369<!--AUTOGENERATED FROM extract-header-defs.xslt, do not edit manually-->
3370<texttable align="left" suppress-title="true" anchor="iana.header.registration.table">
3371   <ttcol>Header Field Name</ttcol>
3372   <ttcol>Protocol</ttcol>
3373   <ttcol>Status</ttcol>
3374   <ttcol>Reference</ttcol>
3376   <c>Connection</c>
3377   <c>http</c>
3378   <c>standard</c>
3379   <c>
3380      <xref target="header.connection"/>
3381   </c>
3382   <c>Content-Length</c>
3383   <c>http</c>
3384   <c>standard</c>
3385   <c>
3386      <xref target="header.content-length"/>
3387   </c>
3388   <c>Host</c>
3389   <c>http</c>
3390   <c>standard</c>
3391   <c>
3392      <xref target=""/>
3393   </c>
3394   <c>TE</c>
3395   <c>http</c>
3396   <c>standard</c>
3397   <c>
3398      <xref target="header.te"/>
3399   </c>
3400   <c>Trailer</c>
3401   <c>http</c>
3402   <c>standard</c>
3403   <c>
3404      <xref target="header.trailer"/>
3405   </c>
3406   <c>Transfer-Encoding</c>
3407   <c>http</c>
3408   <c>standard</c>
3409   <c>
3410      <xref target="header.transfer-encoding"/>
3411   </c>
3412   <c>Upgrade</c>
3413   <c>http</c>
3414   <c>standard</c>
3415   <c>
3416      <xref target="header.upgrade"/>
3417   </c>
3418   <c>Via</c>
3419   <c>http</c>
3420   <c>standard</c>
3421   <c>
3422      <xref target="header.via"/>
3423   </c>
3426<?ENDINC p1-messaging.iana-headers ?>
3428   Furthermore, the header field-name "Close" shall be registered as
3429   "reserved", since using that name as an HTTP header field might
3430   conflict with the "close" connection option of the "<x:ref>Connection</x:ref>"
3431   header field (<xref target="header.connection"/>).
3433<texttable align="left" suppress-title="true">
3434   <ttcol>Header Field Name</ttcol>
3435   <ttcol>Protocol</ttcol>
3436   <ttcol>Status</ttcol>
3437   <ttcol>Reference</ttcol>
3439   <c>Close</c>
3440   <c>http</c>
3441   <c>reserved</c>
3442   <c>
3443      <xref target="header.field.registration"/>
3444   </c>
3447   The change controller is: "IETF ( - Internet Engineering Task Force".
3451<section title="URI Scheme Registration" anchor="uri.scheme.registration">
3453   IANA maintains the registry of URI Schemes <xref target="RFC4395"/> at
3454   <eref target=""/>.
3457   This document defines the following URI schemes, so their
3458   associated registry entries shall be updated according to the permanent
3459   registrations below:
3461<texttable align="left" suppress-title="true">
3462   <ttcol>URI Scheme</ttcol>
3463   <ttcol>Description</ttcol>
3464   <ttcol>Reference</ttcol>
3466   <c>http</c>
3467   <c>Hypertext Transfer Protocol</c>
3468   <c><xref target="http.uri"/></c>
3470   <c>https</c>
3471   <c>Hypertext Transfer Protocol Secure</c>
3472   <c><xref target="https.uri"/></c>
3476<section title="Internet Media Type Registrations" anchor="">
3478   This document serves as the specification for the Internet media types
3479   "message/http" and "application/http". The following is to be registered with
3480   IANA (see <xref target="RFC4288"/>).
3482<section title="Internet Media Type message/http" anchor="">
3483<iref item="Media Type" subitem="message/http" primary="true"/>
3484<iref item="message/http Media Type" primary="true"/>
3486   The message/http type can be used to enclose a single HTTP request or
3487   response message, provided that it obeys the MIME restrictions for all
3488   "message" types regarding line length and encodings.
3491  <list style="hanging" x:indent="12em">
3492    <t hangText="Type name:">
3493      message
3494    </t>
3495    <t hangText="Subtype name:">
3496      http
3497    </t>
3498    <t hangText="Required parameters:">
3499      none
3500    </t>
3501    <t hangText="Optional parameters:">
3502      version, msgtype
3503      <list style="hanging">
3504        <t hangText="version:">
3505          The HTTP-version number of the enclosed message
3506          (e.g., "1.1"). If not present, the version can be
3507          determined from the first line of the body.
3508        </t>
3509        <t hangText="msgtype:">
3510          The message type &mdash; "request" or "response". If not
3511          present, the type can be determined from the first
3512          line of the body.
3513        </t>
3514      </list>
3515    </t>
3516    <t hangText="Encoding considerations:">
3517      only "7bit", "8bit", or "binary" are permitted
3518    </t>
3519    <t hangText="Security considerations:">
3520      none
3521    </t>
3522    <t hangText="Interoperability considerations:">
3523      none
3524    </t>
3525    <t hangText="Published specification:">
3526      This specification (see <xref target=""/>).
3527    </t>
3528    <t hangText="Applications that use this media type:">
3529    </t>
3530    <t hangText="Additional information:">
3531      <list style="hanging">
3532        <t hangText="Magic number(s):">none</t>
3533        <t hangText="File extension(s):">none</t>
3534        <t hangText="Macintosh file type code(s):">none</t>
3535      </list>
3536    </t>
3537    <t hangText="Person and email address to contact for further information:">
3538      See Authors Section.
3539    </t>
3540    <t hangText="Intended usage:">
3541      COMMON
3542    </t>
3543    <t hangText="Restrictions on usage:">
3544      none
3545    </t>
3546    <t hangText="Author/Change controller:">
3547      IESG
3548    </t>
3549  </list>
3552<section title="Internet Media Type application/http" anchor="">
3553<iref item="Media Type" subitem="application/http" primary="true"/>
3554<iref item="application/http Media Type" primary="true"/>
3556   The application/http type can be used to enclose a pipeline of one or more
3557   HTTP request or response messages (not intermixed).
3560  <list style="hanging" x:indent="12em">
3561    <t hangText="Type name:">
3562      application
3563    </t>
3564    <t hangText="Subtype name:">
3565      http
3566    </t>
3567    <t hangText="Required parameters:">
3568      none
3569    </t>
3570    <t hangText="Optional parameters:">
3571      version, msgtype
3572      <list style="hanging">
3573        <t hangText="version:">
3574          The HTTP-version number of the enclosed messages
3575          (e.g., "1.1"). If not present, the version can be
3576          determined from the first line of the body.
3577        </t>
3578        <t hangText="msgtype:">
3579          The message type &mdash; "request" or "response". If not
3580          present, the type can be determined from the first
3581          line of the body.
3582        </t>
3583      </list>
3584    </t>
3585    <t hangText="Encoding considerations:">
3586      HTTP messages enclosed by this type
3587      are in "binary" format; use of an appropriate
3588      Content-Transfer-Encoding is required when
3589      transmitted via E-mail.
3590    </t>
3591    <t hangText="Security considerations:">
3592      none
3593    </t>
3594    <t hangText="Interoperability considerations:">
3595      none
3596    </t>
3597    <t hangText="Published specification:">
3598      This specification (see <xref target=""/>).
3599    </t>
3600    <t hangText="Applications that use this media type:">
3601    </t>
3602    <t hangText="Additional information:">
3603      <list style="hanging">
3604        <t hangText="Magic number(s):">none</t>
3605        <t hangText="File extension(s):">none</t>
3606        <t hangText="Macintosh file type code(s):">none</t>
3607      </list>
3608    </t>
3609    <t hangText="Person and email address to contact for further information:">
3610      See Authors Section.
3611    </t>
3612    <t hangText="Intended usage:">
3613      COMMON
3614    </t>
3615    <t hangText="Restrictions on usage:">
3616      none
3617    </t>
3618    <t hangText="Author/Change controller:">
3619      IESG
3620    </t>
3621  </list>
3626<section title="Transfer Coding Registry" anchor="transfer.coding.registry">
3628   The HTTP Transfer Coding Registry defines the name space for transfer
3629   coding names.
3632   Registrations &MUST; include the following fields:
3633   <list style="symbols">
3634     <t>Name</t>
3635     <t>Description</t>
3636     <t>Pointer to specification text</t>
3637   </list>
3640   Names of transfer codings &MUST-NOT; overlap with names of content codings
3641   (&content-codings;) unless the encoding transformation is identical, as it
3642   is the case for the compression codings defined in
3643   <xref target="compression.codings"/>.
3646   Values to be added to this name space require IETF Review (see
3647   <xref target="RFC5226" x:fmt="of" x:sec="4.1"/>), and &MUST;
3648   conform to the purpose of transfer coding defined in this section.
3651   The registry itself is maintained at
3652   <eref target=""/>.
3656<section title="Transfer Coding Registrations" anchor="transfer.coding.registration">
3658   The HTTP Transfer Coding Registry shall be updated with the registrations
3659   below:
3661<texttable align="left" suppress-title="true" anchor="iana.transfer.coding.registration.table">
3662   <ttcol>Name</ttcol>
3663   <ttcol>Description</ttcol>
3664   <ttcol>Reference</ttcol>
3665   <c>chunked</c>
3666   <c>Transfer in a series of chunks</c>
3667   <c>
3668      <xref target="chunked.encoding"/>
3669   </c>
3670   <c>compress</c>
3671   <c>UNIX "compress" program method</c>
3672   <c>
3673      <xref target="compress.coding"/>
3674   </c>
3675   <c>deflate</c>
3676   <c>"deflate" compression mechanism (<xref target="RFC1951"/>) used inside
3677   the "zlib" data format (<xref target="RFC1950"/>)
3678   </c>
3679   <c>
3680      <xref target="deflate.coding"/>
3681   </c>
3682   <c>gzip</c>
3683   <c>Same as GNU zip <xref target="RFC1952"/></c>
3684   <c>
3685      <xref target="gzip.coding"/>
3686   </c>
3690<section title="Upgrade Token Registry" anchor="upgrade.token.registry">
3692   The HTTP Upgrade Token Registry defines the name space for protocol-name
3693   tokens used to identify protocols in the <x:ref>Upgrade</x:ref> header
3694   field. Each registered protocol name is associated with contact information
3695   and an optional set of specifications that details how the connection
3696   will be processed after it has been upgraded.
3699   Registrations happen on a "First Come First Served" basis (see
3700   <xref target="RFC5226" x:sec="4.1" x:fmt="of"/>) and are subject to the
3701   following rules:
3702  <list style="numbers">
3703    <t>A protocol-name token, once registered, stays registered forever.</t>
3704    <t>The registration &MUST; name a responsible party for the
3705       registration.</t>
3706    <t>The registration &MUST; name a point of contact.</t>
3707    <t>The registration &MAY; name a set of specifications associated with
3708       that token. Such specifications need not be publicly available.</t>
3709    <t>The registration &SHOULD; name a set of expected "protocol-version"
3710       tokens associated with that token at the time of registration.</t>
3711    <t>The responsible party &MAY; change the registration at any time.
3712       The IANA will keep a record of all such changes, and make them
3713       available upon request.</t>
3714    <t>The IESG &MAY; reassign responsibility for a protocol token.
3715       This will normally only be used in the case when a
3716       responsible party cannot be contacted.</t>
3717  </list>
3720   This registration procedure for HTTP Upgrade Tokens replaces that
3721   previously defined in <xref target="RFC2817" x:fmt="of" x:sec="7.2"/>.
3725<section title="Upgrade Token Registration" anchor="upgrade.token.registration">
3727   The HTTP Upgrade Token Registry shall be updated with the registration
3728   below:
3730<texttable align="left" suppress-title="true">
3731   <ttcol>Value</ttcol>
3732   <ttcol>Description</ttcol>
3733   <ttcol>Expected Version Tokens</ttcol>
3734   <ttcol>Reference</ttcol>
3736   <c>HTTP</c>
3737   <c>Hypertext Transfer Protocol</c>
3738   <c>any DIGIT.DIGIT (e.g, "2.0")</c>
3739   <c><xref target="http.version"/></c>
3742   The responsible party is: "IETF ( - Internet Engineering Task Force".
3748<section title="Security Considerations" anchor="security.considerations">
3750   This section is meant to inform application developers, information
3751   providers, and users of the security limitations in HTTP/1.1 as
3752   described by this document. The discussion does not include
3753   definitive solutions to the problems revealed, though it does make
3754   some suggestions for reducing security risks.
3757<section title="Personal Information" anchor="personal.information">
3759   HTTP clients are often privy to large amounts of personal information
3760   (e.g., the user's name, location, mail address, passwords, encryption
3761   keys, etc.), and &SHOULD; be very careful to prevent unintentional
3762   leakage of this information.
3763   We very strongly recommend that a convenient interface be provided
3764   for the user to control dissemination of such information, and that
3765   designers and implementors be particularly careful in this area.
3766   History shows that errors in this area often create serious security
3767   and/or privacy problems and generate highly adverse publicity for the
3768   implementor's company.
3772<section title="Abuse of Server Log Information" anchor="abuse.of.server.log.information">
3774   A server is in the position to save personal data about a user's
3775   requests which might identify their reading patterns or subjects of
3776   interest.  In particular, log information gathered at an intermediary
3777   often contains a history of user agent interaction, across a multitude
3778   of sites, that can be traced to individual users.
3781   HTTP log information is confidential in nature; its handling is often
3782   constrained by laws and regulations.  Log information needs to be securely
3783   stored and appropriate guidelines followed for its analysis.
3784   Anonymization of personal information within individual entries helps,
3785   but is generally not sufficient to prevent real log traces from being
3786   re-identified based on correlation with other access characteristics.
3787   As such, access traces that are keyed to a specific client should not
3788   be published even if the key is pseudonymous.
3791   To minimize the risk of theft or accidental publication, log information
3792   should be purged of personally identifiable information, including
3793   user identifiers, IP addresses, and user-provided query parameters,
3794   as soon as that information is no longer necessary to support operational
3795   needs for security, auditing, or fraud control.
3799<section title="Attacks Based On File and Path Names" anchor="attack.pathname">
3801   Implementations of HTTP origin servers &SHOULD; be careful to restrict
3802   the documents returned by HTTP requests to be only those that were
3803   intended by the server administrators. If an HTTP server translates
3804   HTTP URIs directly into file system calls, the server &MUST; take
3805   special care not to serve files that were not intended to be
3806   delivered to HTTP clients. For example, UNIX, Microsoft Windows, and
3807   other operating systems use ".." as a path component to indicate a
3808   directory level above the current one. On such a system, an HTTP
3809   server &MUST; disallow any such construct in the request-target if it
3810   would otherwise allow access to a resource outside those intended to
3811   be accessible via the HTTP server. Similarly, files intended for
3812   reference only internally to the server (such as access control
3813   files, configuration files, and script code) &MUST; be protected from
3814   inappropriate retrieval, since they might contain sensitive
3815   information. Experience has shown that minor bugs in such HTTP server
3816   implementations have turned into security risks.
3820<section title="DNS-related Attacks" anchor="dns.related.attacks">
3822   HTTP clients rely heavily on the Domain Name Service (DNS), and are thus
3823   generally prone to security attacks based on the deliberate misassociation
3824   of IP addresses and DNS names not protected by DNSSec. Clients need to be
3825   cautious in assuming the validity of an IP number/DNS name association unless
3826   the response is protected by DNSSec (<xref target="RFC4033"/>).
3830<section title="Intermediaries and Caching" anchor="attack.intermediaries">
3832   By their very nature, HTTP intermediaries are men-in-the-middle, and
3833   represent an opportunity for man-in-the-middle attacks. Compromise of
3834   the systems on which the intermediaries run can result in serious security
3835   and privacy problems. Intermediaries have access to security-related
3836   information, personal information about individual users and
3837   organizations, and proprietary information belonging to users and
3838   content providers. A compromised intermediary, or an intermediary
3839   implemented or configured without regard to security and privacy
3840   considerations, might be used in the commission of a wide range of
3841   potential attacks.
3844   Intermediaries that contain a shared cache are especially vulnerable
3845   to cache poisoning attacks.
3848   Implementors need to consider the privacy and security
3849   implications of their design and coding decisions, and of the
3850   configuration options they provide to operators (especially the
3851   default configuration).
3854   Users need to be aware that intermediaries are no more trustworthy than
3855   the people who run them; HTTP itself cannot solve this problem.
3858   The judicious use of cryptography, when appropriate, might suffice to
3859   protect against a broad range of security and privacy attacks. Such
3860   cryptography is beyond the scope of the HTTP/1.1 specification.
3864<section title="Protocol Element Size Overflows" anchor="attack.protocol.element.size.overflows">
3866   Because HTTP uses mostly textual, character-delimited fields, attackers can
3867   overflow buffers in implementations, and/or perform a Denial of Service
3868   against implementations that accept fields with unlimited lengths.
3871   To promote interoperability, this specification makes specific
3872   recommendations for minimum size limits on request-line
3873   (<xref target="request.line"/>)
3874   and blocks of header fields (<xref target="header.fields"/>). These are
3875   minimum recommendations, chosen to be supportable even by implementations
3876   with limited resources; it is expected that most implementations will
3877   choose substantially higher limits.
3880   This specification also provides a way for servers to reject messages that
3881   have request-targets that are too long (&status-414;) or request entities
3882   that are too large (&status-4xx;).
3885   Other fields (including but not limited to request methods, response status
3886   phrases, header field-names, and body chunks) &SHOULD; be limited by
3887   implementations carefully, so as to not impede interoperability.
3892<section title="Acknowledgments" anchor="acks">
3894   This edition of HTTP builds on the many contributions that went into
3895   <xref target="RFC1945" format="none">RFC 1945</xref>,
3896   <xref target="RFC2068" format="none">RFC 2068</xref>,
3897   <xref target="RFC2145" format="none">RFC 2145</xref>, and
3898   <xref target="RFC2616" format="none">RFC 2616</xref>, including
3899   substantial contributions made by the previous authors, editors, and
3900   working group chairs: Tim Berners-Lee, Ari Luotonen, Roy T. Fielding,
3901   Henrik Frystyk Nielsen, Jim Gettys, Jeffrey C. Mogul, Larry Masinter,
3902   Paul J. Leach, and Mark Nottingham.
3903   See <xref target="RFC2616" x:fmt="of" x:sec="16"/> for additional
3904   acknowledgements from prior revisions.
3907   Since 1999, the following contributors have helped improve the HTTP
3908   specification by reporting bugs, asking smart questions, drafting or
3909   reviewing text, and evaluating open issues:
3911<?BEGININC acks ?>
3912<t>Adam Barth,
3913Adam Roach,
3914Addison Phillips,
3915Adrian Chadd,
3916Adrien W. de Croy,
3917Alan Ford,
3918Alan Ruttenberg,
3919Albert Lunde,
3920Alek Storm,
3921Alex Rousskov,
3922Alexandre Morgaut,
3923Alexey Melnikov,
3924Alisha Smith,
3925Amichai Rothman,
3926Amit Klein,
3927Amos Jeffries,
3928Andreas Maier,
3929Andreas Petersson,
3930Anne van Kesteren,
3931Anthony Bryan,
3932Asbjorn Ulsberg,
3933Balachander Krishnamurthy,
3934Barry Leiba,
3935Ben Laurie,
3936Benjamin Niven-Jenkins,
3937Bil Corry,
3938Bill Burke,
3939Bjoern Hoehrmann,
3940Bob Scheifler,
3941Boris Zbarsky,
3942Brett Slatkin,
3943Brian Kell,
3944Brian McBarron,
3945Brian Pane,
3946Brian Smith,
3947Bryce Nesbitt,
3948Cameron Heavon-Jones,
3949Carl Kugler,
3950Carsten Bormann,
3951Charles Fry,
3952Chris Newman,
3953Cyrus Daboo,
3954Dale Robert Anderson,
3955Dan Winship,
3956Daniel Stenberg,
3957Dave Cridland,
3958Dave Crocker,
3959Dave Kristol,
3960David Booth,
3961David Singer,
3962David W. Morris,
3963Diwakar Shetty,
3964Dmitry Kurochkin,
3965Drummond Reed,
3966Duane Wessels,
3967Edward Lee,
3968Eliot Lear,
3969Eran Hammer-Lahav,
3970Eric D. Williams,
3971Eric J. Bowman,
3972Eric Lawrence,
3973Eric Rescorla,
3974Erik Aronesty,
3975Florian Weimer,
3976Frank Ellermann,
3977Fred Bohle,
3978Geoffrey Sneddon,
3979Gervase Markham,
3980Greg Wilkins,
3981Harald Tveit Alvestrand,
3982Harry Halpin,
3983Helge Hess,
3984Henrik Nordstrom,
3985Henry S. Thompson,
3986Henry Story,
3987Herbert van de Sompel,
3988Howard Melman,
3989Hugo Haas,
3990Ian Hickson,
3991Ingo Struck,
3992J. Ross Nicoll,
3993James H. Manger,
3994James Lacey,
3995James M. Snell,
3996Jamie Lokier,
3997Jan Algermissen,
3998Jeff Hodges (who came up with the term 'effective Request-URI'),
3999Jeff Walden,
4000Jim Luther,
4001Joe D. Williams,
4002Joe Gregorio,
4003Joe Orton,
4004John C. Klensin,
4005John C. Mallery,
4006John Cowan,
4007John Kemp,
4008John Panzer,
4009John Schneider,
4010John Stracke,
4011John Sullivan,
4012Jonas Sicking,
4013Jonathan Billington,
4014Jonathan Moore,
4015Jonathan Rees,
4016Jordi Ros,
4017Joris Dobbelsteen,
4018Josh Cohen,
4019Julien Pierre,
4020Jungshik Shin,
4021Justin Chapweske,
4022Justin Erenkrantz,
4023Justin James,
4024Kalvinder Singh,
4025Karl Dubost,
4026Keith Hoffman,
4027Keith Moore,
4028Koen Holtman,
4029Konstantin Voronkov,
4030Kris Zyp,
4031Lisa Dusseault,
4032Maciej Stachowiak,
4033Marc Schneider,
4034Marc Slemko,
4035Mark Baker,
4036Mark Pauley,
4037Mark Watson,
4038Markus Isomaki,
4039Markus Lanthaler,
4040Martin J. Duerst,
4041Martin Musatov,
4042Martin Nilsson,
4043Martin Thomson,
4044Matt Lynch,
4045Matthew Cox,
4046Max Clark,
4047Michael Burrows,
4048Michael Hausenblas,
4049Mike Amundsen,
4050Mike Belshe,
4051Mike Kelly,
4052Mike Schinkel,
4053Miles Sabin,
4054Murray S. Kucherawy,
4055Mykyta Yevstifeyev,
4056Nathan Rixham,
4057Nicholas Shanks,
4058Nico Williams,
4059Nicolas Alvarez,
4060Nicolas Mailhot,
4061Noah Slater,
4062Pablo Castro,
4063Pat Hayes,
4064Patrick R. McManus,
4065Paul E. Jones,
4066Paul Hoffman,
4067Paul Marquess,
4068Peter Lepeska,
4069Peter Saint-Andre,
4070Peter Watkins,
4071Phil Archer,
4072Phillip Hallam-Baker,
4073Poul-Henning Kamp,
4074Preethi Natarajan,
4075Ray Polk,
4076Reto Bachmann-Gmuer,
4077Richard Cyganiak,
4078Robert Brewer,
4079Robert Collins,
4080Robert O'Callahan,
4081Robert Olofsson,
4082Robert Sayre,
4083Robert Siemer,
4084Robert de Wilde,
4085Roberto Javier Godoy,
4086Roberto Peon,
4087Ronny Widjaja,
4088S. Mike Dierken,
4089Salvatore Loreto,
4090Sam Johnston,
4091Sam Ruby,
4092Scott Lawrence (who maintained the original issues list),
4093Sean B. Palmer,
4094Shane McCarron,
4095Stefan Eissing,
4096Stefan Tilkov,
4097Stefanos Harhalakis,
4098Stephane Bortzmeyer,
4099Stephen Farrell,
4100Stuart Williams,
4101Subbu Allamaraju,
4102Sylvain Hellegouarch,
4103Tapan Divekar,
4104Ted Hardie,
4105Thomas Broyer,
4106Thomas Nordin,
4107Thomas Roessler,
4108Tim Bray,
4109Tim Morgan,
4110Tim Olsen,
4111Tom Zhou,
4112Travis Snoozy,
4113Tyler Close,
4114Vincent Murphy,
4115Wenbo Zhu,
4116Werner Baumann,
4117Wilbur Streett,
4118Wilfredo Sanchez Vega,
4119William A. Rowe Jr.,
4120William Chan,
4121Willy Tarreau,
4122Xiaoshu Wang,
4123Yaron Goland,
4124Yngve Nysaeter Pettersen,
4125Yoav Nir,
4126Yogesh Bang,
4127Yutaka Oiwa,
4128Zed A. Shaw, and
4129Zhong Yu.
4131<?ENDINC acks ?>
4137<references title="Normative References">
4139<reference anchor="ISO-8859-1">
4140  <front>
4141    <title>
4142     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
4143    </title>
4144    <author>
4145      <organization>International Organization for Standardization</organization>
4146    </author>
4147    <date year="1998"/>
4148  </front>
4149  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
4152<reference anchor="Part2">
4153  <front>
4154    <title>HTTP/1.1, part 2: Message Semantics, Payload and Content Negotiation</title>
4155    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4156      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4157      <address><email></email></address>
4158    </author>
4159    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4160      <organization abbrev="W3C">World Wide Web Consortium</organization>
4161      <address><email></email></address>
4162    </author>
4163    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4164      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4165      <address><email></email></address>
4166    </author>
4167    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
4168  </front>
4169  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-&ID-VERSION;"/>
4170  <x:source href="p2-semantics.xml" basename="p2-semantics">
4171    <x:defines>1xx (Informational)</x:defines>
4172    <x:defines>1xx</x:defines>
4173    <x:defines>100 (Continue)</x:defines>
4174    <x:defines>101 (Switching Protocols)</x:defines>
4175    <x:defines>2xx (Successful)</x:defines>
4176    <x:defines>2xx</x:defines>
4177    <x:defines>200 (OK)</x:defines>
4178    <x:defines>204 (No Content)</x:defines>
4179    <x:defines>3xx (Redirection)</x:defines>
4180    <x:defines>3xx</x:defines>
4181    <x:defines>301 (Moved Permanently)</x:defines>
4182    <x:defines>4xx (Client Error)</x:defines>
4183    <x:defines>4xx</x:defines>
4184    <x:defines>400 (Bad Request)</x:defines>
4185    <x:defines>405 (Method Not Allowed)</x:defines>
4186    <x:defines>411 (Length Required)</x:defines>
4187    <x:defines>414 (URI Too Long)</x:defines>
4188    <x:defines>417 (Expectation Failed)</x:defines>
4189    <x:defines>426 (Upgrade Required)</x:defines>
4190    <x:defines>501 (Not Implemented)</x:defines>
4191    <x:defines>502 (Bad Gateway)</x:defines>
4192    <x:defines>505 (HTTP Version Not Supported)</x:defines>
4193    <x:defines>Content-Encoding</x:defines>
4194    <x:defines>Content-Type</x:defines>
4195    <x:defines>Date</x:defines>
4196    <x:defines>Expect</x:defines>
4197    <x:defines>Location</x:defines>
4198    <x:defines>Server</x:defines>
4199    <x:defines>User-Agent</x:defines>
4200  </x:source>
4203<reference anchor="Part4">
4204  <front>
4205    <title>HTTP/1.1, part 4: Conditional Requests</title>
4206    <author fullname="Roy T. Fielding" initials="R." role="editor" surname="Fielding">
4207      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4208      <address><email></email></address>
4209    </author>
4210    <author fullname="Yves Lafon" initials="Y." role="editor" surname="Lafon">
4211      <organization abbrev="W3C">World Wide Web Consortium</organization>
4212      <address><email></email></address>
4213    </author>
4214    <author fullname="Julian F. Reschke" initials="J. F." role="editor" surname="Reschke">
4215      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4216      <address><email></email></address>
4217    </author>
4218    <date month="&ID-MONTH;" year="&ID-YEAR;" />
4219  </front>
4220  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p4-conditional-&ID-VERSION;" />
4221  <x:source basename="p4-conditional" href="p4-conditional.xml">
4222    <x:defines>304 (Not Modified)</x:defines>
4223  </x:source>
4226<reference anchor="Part5">
4227  <front>
4228    <title>HTTP/1.1, part 5: Range Requests and Partial Responses</title>
4229    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4230      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4231      <address><email></email></address>
4232    </author>
4233    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4234      <organization abbrev="W3C">World Wide Web Consortium</organization>
4235      <address><email></email></address>
4236    </author>
4237    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4238      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4239      <address><email></email></address>
4240    </author>
4241    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
4242  </front>
4243  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-&ID-VERSION;"/>
4244  <x:source href="p5-range.xml" basename="p5-range">
4245    <x:defines>Content-Range</x:defines>
4246  </x:source>
4249<reference anchor="Part6">
4250  <front>
4251    <title>HTTP/1.1, part 6: Caching</title>
4252    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4253      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4254      <address><email></email></address>
4255    </author>
4256    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4257      <organization abbrev="W3C">World Wide Web Consortium</organization>
4258      <address><email></email></address>
4259    </author>
4260    <author initials="M." surname="Nottingham" fullname="Mark Nottingham" role="editor">
4261      <organization>Rackspace</organization>
4262      <address><email></email></address>
4263    </author>
4264    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4265      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4266      <address><email></email></address>
4267    </author>
4268    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
4269  </front>
4270  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-&ID-VERSION;"/>
4271  <x:source href="p6-cache.xml" basename="p6-cache">
4272    <x:defines>Expires</x:defines>
4273  </x:source>
4276<reference anchor="Part7">
4277  <front>
4278    <title>HTTP/1.1, part 7: Authentication</title>
4279    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4280      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4281      <address><email></email></address>
4282    </author>
4283    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4284      <organization abbrev="W3C">World Wide Web Consortium</organization>
4285      <address><email></email></address>
4286    </author>
4287    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4288      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4289      <address><email></email></address>
4290    </author>
4291    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
4292  </front>
4293  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p7-auth-&ID-VERSION;"/>
4294  <x:source href="p7-auth.xml" basename="p7-auth">
4295    <x:defines>Proxy-Authenticate</x:defines>
4296    <x:defines>Proxy-Authorization</x:defines>
4297  </x:source>
4300<reference anchor="RFC5234">
4301  <front>
4302    <title abbrev="ABNF for Syntax Specifications">Augmented BNF for Syntax Specifications: ABNF</title>
4303    <author initials="D." surname="Crocker" fullname="Dave Crocker" role="editor">
4304      <organization>Brandenburg InternetWorking</organization>
4305      <address>
4306        <email></email>
4307      </address> 
4308    </author>
4309    <author initials="P." surname="Overell" fullname="Paul Overell">
4310      <organization>THUS plc.</organization>
4311      <address>
4312        <email></email>
4313      </address>
4314    </author>
4315    <date month="January" year="2008"/>
4316  </front>
4317  <seriesInfo name="STD" value="68"/>
4318  <seriesInfo name="RFC" value="5234"/>
4321<reference anchor="RFC2119">
4322  <front>
4323    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
4324    <author initials="S." surname="Bradner" fullname="Scott Bradner">
4325      <organization>Harvard University</organization>
4326      <address><email></email></address>
4327    </author>
4328    <date month="March" year="1997"/>
4329  </front>
4330  <seriesInfo name="BCP" value="14"/>
4331  <seriesInfo name="RFC" value="2119"/>
4334<reference anchor="RFC3986">
4335 <front>
4336  <title abbrev='URI Generic Syntax'>Uniform Resource Identifier (URI): Generic Syntax</title>
4337  <author initials='T.' surname='Berners-Lee' fullname='Tim Berners-Lee'>
4338    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
4339    <address>
4340       <email></email>
4341       <uri></uri>
4342    </address>
4343  </author>
4344  <author initials='R.' surname='Fielding' fullname='Roy T. Fielding'>
4345    <organization abbrev="Day Software">Day Software</organization>
4346    <address>
4347      <email></email>
4348      <uri></uri>
4349    </address>
4350  </author>
4351  <author initials='L.' surname='Masinter' fullname='Larry Masinter'>
4352    <organization abbrev="Adobe Systems">Adobe Systems Incorporated</organization>
4353    <address>
4354      <email></email>
4355      <uri></uri>
4356    </address>
4357  </author>
4358  <date month='January' year='2005'></date>
4359 </front>
4360 <seriesInfo name="STD" value="66"/>
4361 <seriesInfo name="RFC" value="3986"/>
4364<reference anchor="USASCII">
4365  <front>
4366    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
4367    <author>
4368      <organization>American National Standards Institute</organization>
4369    </author>
4370    <date year="1986"/>
4371  </front>
4372  <seriesInfo name="ANSI" value="X3.4"/>
4375<reference anchor="RFC1950">
4376  <front>
4377    <title>ZLIB Compressed Data Format Specification version 3.3</title>
4378    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4379      <organization>Aladdin Enterprises</organization>
4380      <address><email></email></address>
4381    </author>
4382    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly"/>
4383    <date month="May" year="1996"/>
4384  </front>
4385  <seriesInfo name="RFC" value="1950"/>
4386  <!--<annotation>
4387    RFC 1950 is an Informational RFC, thus it might be less stable than
4388    this specification. On the other hand, this downward reference was
4389    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4390    therefore it is unlikely to cause problems in practice. See also
4391    <xref target="BCP97"/>.
4392  </annotation>-->
4395<reference anchor="RFC1951">
4396  <front>
4397    <title>DEFLATE Compressed Data Format Specification version 1.3</title>
4398    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4399      <organization>Aladdin Enterprises</organization>
4400      <address><email></email></address>
4401    </author>
4402    <date month="May" year="1996"/>
4403  </front>
4404  <seriesInfo name="RFC" value="1951"/>
4405  <!--<annotation>
4406    RFC 1951 is an Informational RFC, thus it might be less stable than
4407    this specification. On the other hand, this downward reference was
4408    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4409    therefore it is unlikely to cause problems in practice. See also
4410    <xref target="BCP97"/>.
4411  </annotation>-->
4414<reference anchor="RFC1952">
4415  <front>
4416    <title>GZIP file format specification version 4.3</title>
4417    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4418      <organization>Aladdin Enterprises</organization>
4419      <address><email></email></address>
4420    </author>
4421    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
4422      <address><email></email></address>
4423    </author>
4424    <author initials="M." surname="Adler" fullname="Mark Adler">
4425      <address><email></email></address>
4426    </author>
4427    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4428      <address><email></email></address>
4429    </author>
4430    <author initials="G." surname="Randers-Pehrson" fullname="Glenn Randers-Pehrson">
4431      <address><email></email></address>
4432    </author>
4433    <date month="May" year="1996"/>
4434  </front>
4435  <seriesInfo name="RFC" value="1952"/>
4436  <!--<annotation>
4437    RFC 1952 is an Informational RFC, thus it might be less stable than
4438    this specification. On the other hand, this downward reference was
4439    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4440    therefore it is unlikely to cause problems in practice. See also
4441    <xref target="BCP97"/>.
4442  </annotation>-->
4447<references title="Informative References">
4449<reference anchor="Nie1997" target="">
4450  <front>
4451    <title>Network Performance Effects of HTTP/1.1, CSS1, and PNG</title>
4452    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen"/>
4453    <author initials="J." surname="Gettys" fullname="J. Gettys"/>
4454    <author initials="E." surname="Prud'hommeaux" fullname="E. Prud'hommeaux"/>
4455    <author initials="H." surname="Lie" fullname="H. Lie"/>
4456    <author initials="C." surname="Lilley" fullname="C. Lilley"/>
4457    <date year="1997" month="September"/>
4458  </front>
4459  <seriesInfo name="ACM" value="Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication SIGCOMM '97"/>
4462<reference anchor="Pad1995" target="">
4463  <front>
4464    <title>Improving HTTP Latency</title>
4465    <author initials="V.N." surname="Padmanabhan" fullname="Venkata N. Padmanabhan"/>
4466    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul"/>
4467    <date year="1995" month="December"/>
4468  </front>
4469  <seriesInfo name="Computer Networks and ISDN Systems" value="v. 28, pp. 25-35"/>
4472<reference anchor='RFC1919'>
4473  <front>
4474    <title>Classical versus Transparent IP Proxies</title>
4475    <author initials='M.' surname='Chatel' fullname='Marc Chatel'>
4476      <address><email></email></address>
4477    </author>
4478    <date year='1996' month='March' />
4479  </front>
4480  <seriesInfo name='RFC' value='1919' />
4483<reference anchor="RFC1945">
4484  <front>
4485    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
4486    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4487      <organization>MIT, Laboratory for Computer Science</organization>
4488      <address><email></email></address>
4489    </author>
4490    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4491      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4492      <address><email></email></address>
4493    </author>
4494    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4495      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
4496      <address><email></email></address>
4497    </author>
4498    <date month="May" year="1996"/>
4499  </front>
4500  <seriesInfo name="RFC" value="1945"/>
4503<reference anchor="RFC2045">
4504  <front>
4505    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
4506    <author initials="N." surname="Freed" fullname="Ned Freed">
4507      <organization>Innosoft International, Inc.</organization>
4508      <address><email></email></address>
4509    </author>
4510    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
4511      <organization>First Virtual Holdings</organization>
4512      <address><email></email></address>
4513    </author>
4514    <date month="November" year="1996"/>
4515  </front>
4516  <seriesInfo name="RFC" value="2045"/>
4519<reference anchor="RFC2047">
4520  <front>
4521    <title abbrev="Message Header Extensions">MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text</title>
4522    <author initials="K." surname="Moore" fullname="Keith Moore">
4523      <organization>University of Tennessee</organization>
4524      <address><email></email></address>
4525    </author>
4526    <date month="November" year="1996"/>
4527  </front>
4528  <seriesInfo name="RFC" value="2047"/>
4531<reference anchor="RFC2068">
4532  <front>
4533    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4534    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
4535      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4536      <address><email></email></address>
4537    </author>
4538    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4539      <organization>MIT Laboratory for Computer Science</organization>
4540      <address><email></email></address>
4541    </author>
4542    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
4543      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
4544      <address><email></email></address>
4545    </author>
4546    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4547      <organization>MIT Laboratory for Computer Science</organization>
4548      <address><email></email></address>
4549    </author>
4550    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4551      <organization>MIT Laboratory for Computer Science</organization>
4552      <address><email></email></address>
4553    </author>
4554    <date month="January" year="1997"/>
4555  </front>
4556  <seriesInfo name="RFC" value="2068"/>
4559<reference anchor="RFC2145">
4560  <front>
4561    <title abbrev="HTTP Version Numbers">Use and Interpretation of HTTP Version Numbers</title>
4562    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
4563      <organization>Western Research Laboratory</organization>
4564      <address><email></email></address>
4565    </author>
4566    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4567      <organization>Department of Information and Computer Science</organization>
4568      <address><email></email></address>
4569    </author>
4570    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4571      <organization>MIT Laboratory for Computer Science</organization>
4572      <address><email></email></address>
4573    </author>
4574    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4575      <organization>W3 Consortium</organization>
4576      <address><email></email></address>
4577    </author>
4578    <date month="May" year="1997"/>
4579  </front>
4580  <seriesInfo name="RFC" value="2145"/>
4583<reference anchor="RFC2616">
4584  <front>
4585    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4586    <author initials="R." surname="Fielding" fullname="R. Fielding">
4587      <organization>University of California, Irvine</organization>
4588      <address><email></email></address>
4589    </author>
4590    <author initials="J." surname="Gettys" fullname="J. Gettys">
4591      <organization>W3C</organization>
4592      <address><email></email></address>
4593    </author>
4594    <author initials="J." surname="Mogul" fullname="J. Mogul">
4595      <organization>Compaq Computer Corporation</organization>
4596      <address><email></email></address>
4597    </author>
4598    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
4599      <organization>MIT Laboratory for Computer Science</organization>
4600      <address><email></email></address>
4601    </author>
4602    <author initials="L." surname="Masinter" fullname="L. Masinter">
4603      <organization>Xerox Corporation</organization>
4604      <address><email></email></address>
4605    </author>
4606    <author initials="P." surname="Leach" fullname="P. Leach">
4607      <organization>Microsoft Corporation</organization>
4608      <address><email></email></address>
4609    </author>
4610    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
4611      <organization>W3C</organization>
4612      <address><email></email></address>
4613    </author>
4614    <date month="June" year="1999"/>
4615  </front>
4616  <seriesInfo name="RFC" value="2616"/>
4619<reference anchor='RFC2817'>
4620  <front>
4621    <title>Upgrading to TLS Within HTTP/1.1</title>
4622    <author initials='R.' surname='Khare' fullname='R. Khare'>
4623      <organization>4K Associates / UC Irvine</organization>
4624      <address><email></email></address>
4625    </author>
4626    <author initials='S.' surname='Lawrence' fullname='S. Lawrence'>
4627      <organization>Agranat Systems, Inc.</organization>
4628      <address><email></email></address>
4629    </author>
4630    <date year='2000' month='May' />
4631  </front>
4632  <seriesInfo name='RFC' value='2817' />
4635<reference anchor='RFC2818'>
4636  <front>
4637    <title>HTTP Over TLS</title>
4638    <author initials='E.' surname='Rescorla' fullname='Eric Rescorla'>
4639      <organization>RTFM, Inc.</organization>
4640      <address><email></email></address>
4641    </author>
4642    <date year='2000' month='May' />
4643  </front>
4644  <seriesInfo name='RFC' value='2818' />
4647<reference anchor='RFC2965'>
4648  <front>
4649    <title>HTTP State Management Mechanism</title>
4650    <author initials='D. M.' surname='Kristol' fullname='David M. Kristol'>
4651      <organization>Bell Laboratories, Lucent Technologies</organization>
4652      <address><email></email></address>
4653    </author>
4654    <author initials='L.' surname='Montulli' fullname='Lou Montulli'>
4655      <organization>, Inc.</organization>
4656      <address><email></email></address>
4657    </author>
4658    <date year='2000' month='October' />
4659  </front>
4660  <seriesInfo name='RFC' value='2965' />
4663<reference anchor='RFC3040'>
4664  <front>
4665    <title>Internet Web Replication and Caching Taxonomy</title>
4666    <author initials='I.' surname='Cooper' fullname='I. Cooper'>
4667      <organization>Equinix, Inc.</organization>
4668    </author>
4669    <author initials='I.' surname='Melve' fullname='I. Melve'>
4670      <organization>UNINETT</organization>
4671    </author>
4672    <author initials='G.' surname='Tomlinson' fullname='G. Tomlinson'>
4673      <organization>CacheFlow Inc.</organization>
4674    </author>
4675    <date year='2001' month='January' />
4676  </front>
4677  <seriesInfo name='RFC' value='3040' />
4680<reference anchor='RFC3864'>
4681  <front>
4682    <title>Registration Procedures for Message Header Fields</title>
4683    <author initials='G.' surname='Klyne' fullname='G. Klyne'>
4684      <organization>Nine by Nine</organization>
4685      <address><email></email></address>
4686    </author>
4687    <author initials='M.' surname='Nottingham' fullname='M. Nottingham'>
4688      <organization>BEA Systems</organization>
4689      <address><email></email></address>
4690    </author>
4691    <author initials='J.' surname='Mogul' fullname='J. Mogul'>
4692      <organization>HP Labs</organization>
4693      <address><email></email></address>
4694    </author>
4695    <date year='2004' month='September' />
4696  </front>
4697  <seriesInfo name='BCP' value='90' />
4698  <seriesInfo name='RFC' value='3864' />
4701<reference anchor='RFC4033'>
4702  <front>
4703    <title>DNS Security Introduction and Requirements</title>
4704    <author initials='R.' surname='Arends' fullname='R. Arends'/>
4705    <author initials='R.' surname='Austein' fullname='R. Austein'/>
4706    <author initials='M.' surname='Larson' fullname='M. Larson'/>
4707    <author initials='D.' surname='Massey' fullname='D. Massey'/>
4708    <author initials='S.' surname='Rose' fullname='S. Rose'/>
4709    <date year='2005' month='March' />
4710  </front>
4711  <seriesInfo name='RFC' value='4033' />
4714<reference anchor="RFC4288">
4715  <front>
4716    <title>Media Type Specifications and Registration Procedures</title>
4717    <author initials="N." surname="Freed" fullname="N. Freed">
4718      <organization>Sun Microsystems</organization>
4719      <address>
4720        <email></email>
4721      </address>
4722    </author>
4723    <author initials="J." surname="Klensin" fullname="J. Klensin">
4724      <address>
4725        <email></email>
4726      </address>
4727    </author>
4728    <date year="2005" month="December"/>
4729  </front>
4730  <seriesInfo name="BCP" value="13"/>
4731  <seriesInfo name="RFC" value="4288"/>
4734<reference anchor='RFC4395'>
4735  <front>
4736    <title>Guidelines and Registration Procedures for New URI Schemes</title>
4737    <author initials='T.' surname='Hansen' fullname='T. Hansen'>
4738      <organization>AT&amp;T Laboratories</organization>
4739      <address>
4740        <email></email>
4741      </address>
4742    </author>
4743    <author initials='T.' surname='Hardie' fullname='T. Hardie'>
4744      <organization>Qualcomm, Inc.</organization>
4745      <address>
4746        <email></email>
4747      </address>
4748    </author>
4749    <author initials='L.' surname='Masinter' fullname='L. Masinter'>
4750      <organization>Adobe Systems</organization>
4751      <address>
4752        <email></email>
4753      </address>
4754    </author>
4755    <date year='2006' month='February' />
4756  </front>
4757  <seriesInfo name='BCP' value='115' />
4758  <seriesInfo name='RFC' value='4395' />
4761<reference anchor='RFC4559'>
4762  <front>
4763    <title>SPNEGO-based Kerberos and NTLM HTTP Authentication in Microsoft Windows</title>
4764    <author initials='K.' surname='Jaganathan' fullname='K. Jaganathan'/>
4765    <author initials='L.' surname='Zhu' fullname='L. Zhu'/>
4766    <author initials='J.' surname='Brezak' fullname='J. Brezak'/>
4767    <date year='2006' month='June' />
4768  </front>
4769  <seriesInfo name='RFC' value='4559' />
4772<reference anchor='RFC5226'>
4773  <front>
4774    <title>Guidelines for Writing an IANA Considerations Section in RFCs</title>
4775    <author initials='T.' surname='Narten' fullname='T. Narten'>
4776      <organization>IBM</organization>
4777      <address><email></email></address>
4778    </author>
4779    <author initials='H.' surname='Alvestrand' fullname='H. Alvestrand'>
4780      <organization>Google</organization>
4781      <address><email></email></address>
4782    </author>
4783    <date year='2008' month='May' />
4784  </front>
4785  <seriesInfo name='BCP' value='26' />
4786  <seriesInfo name='RFC' value='5226' />
4789<reference anchor="RFC5322">
4790  <front>
4791    <title>Internet Message Format</title>
4792    <author initials="P." surname="Resnick" fullname="P. Resnick">
4793      <organization>Qualcomm Incorporated</organization>
4794    </author>
4795    <date year="2008" month="October"/>
4796  </front>
4797  <seriesInfo name="RFC" value="5322"/>
4800<reference anchor="RFC6265">
4801  <front>
4802    <title>HTTP State Management Mechanism</title>
4803    <author initials="A." surname="Barth" fullname="Adam Barth">
4804      <organization abbrev="U.C. Berkeley">
4805        University of California, Berkeley
4806      </organization>
4807      <address><email></email></address>
4808    </author>
4809    <date year="2011" month="April" />
4810  </front>
4811  <seriesInfo name="RFC" value="6265"/>
4814<!--<reference anchor='BCP97'>
4815  <front>
4816    <title>Handling Normative References to Standards-Track Documents</title>
4817    <author initials='J.' surname='Klensin' fullname='J. Klensin'>
4818      <address>
4819        <email></email>
4820      </address>
4821    </author>
4822    <author initials='S.' surname='Hartman' fullname='S. Hartman'>
4823      <organization>MIT</organization>
4824      <address>
4825        <email></email>
4826      </address>
4827    </author>
4828    <date year='2007' month='June' />
4829  </front>
4830  <seriesInfo name='BCP' value='97' />
4831  <seriesInfo name='RFC' value='4897' />
4834<reference anchor="Kri2001" target="">
4835  <front>
4836    <title>HTTP Cookies: Standards, Privacy, and Politics</title>
4837    <author initials="D." surname="Kristol" fullname="David M. Kristol"/>
4838    <date year="2001" month="November"/>
4839  </front>
4840  <seriesInfo name="ACM Transactions on Internet Technology" value="Vol. 1, #2"/>
4843<reference anchor="Spe" target="">
4844  <front>
4845    <title>Analysis of HTTP Performance Problems</title>
4846    <author initials="S." surname="Spero" fullname="Simon E. Spero"/>
4847    <date/>
4848  </front>
4851<reference anchor="Tou1998" target="">
4852  <front>
4853  <title>Analysis of HTTP Performance</title>
4854  <author initials="J." surname="Touch" fullname="Joe Touch">
4855    <organization>USC/Information Sciences Institute</organization>
4856    <address><email></email></address>
4857  </author>
4858  <author initials="J." surname="Heidemann" fullname="John Heidemann">
4859    <organization>USC/Information Sciences Institute</organization>
4860    <address><email></email></address>
4861  </author>
4862  <author initials="K." surname="Obraczka" fullname="Katia Obraczka">
4863    <organization>USC/Information Sciences Institute</organization>
4864    <address><email></email></address>
4865  </author>
4866  <date year="1998" month="Aug"/>
4867  </front>
4868  <seriesInfo name="ISI Research Report" value="ISI/RR-98-463"/>
4869  <annotation>(original report dated Aug. 1996)</annotation>
4875<section title="HTTP Version History" anchor="compatibility">
4877   HTTP has been in use by the World-Wide Web global information initiative
4878   since 1990. The first version of HTTP, later referred to as HTTP/0.9,
4879   was a simple protocol for hypertext data transfer across the Internet
4880   with only a single request method (GET) and no metadata.
4881   HTTP/1.0, as defined by <xref target="RFC1945"/>, added a range of request
4882   methods and MIME-like messaging that could include metadata about the data
4883   transferred and modifiers on the request/response semantics. However,
4884   HTTP/1.0 did not sufficiently take into consideration the effects of
4885   hierarchical proxies, caching, the need for persistent connections, or
4886   name-based virtual hosts. The proliferation of incompletely-implemented
4887   applications calling themselves "HTTP/1.0" further necessitated a
4888   protocol version change in order for two communicating applications
4889   to determine each other's true capabilities.
4892   HTTP/1.1 remains compatible with HTTP/1.0 by including more stringent
4893   requirements that enable reliable implementations, adding only
4894   those new features that will either be safely ignored by an HTTP/1.0
4895   recipient or only sent when communicating with a party advertising
4896   conformance with HTTP/1.1.
4899   It is beyond the scope of a protocol specification to mandate
4900   conformance with previous versions. HTTP/1.1 was deliberately
4901   designed, however, to make supporting previous versions easy.
4902   We would expect a general-purpose HTTP/1.1 server to understand
4903   any valid request in the format of HTTP/1.0 and respond appropriately
4904   with an HTTP/1.1 message that only uses features understood (or
4905   safely ignored) by HTTP/1.0 clients.  Likewise, we would expect
4906   an HTTP/1.1 client to understand any valid HTTP/1.0 response.
4909   Since HTTP/0.9 did not support header fields in a request,
4910   there is no mechanism for it to support name-based virtual
4911   hosts (selection of resource by inspection of the <x:ref>Host</x:ref> header
4912   field).  Any server that implements name-based virtual hosts
4913   ought to disable support for HTTP/0.9.  Most requests that
4914   appear to be HTTP/0.9 are, in fact, badly constructed HTTP/1.x
4915   requests wherein a buggy client failed to properly encode
4916   linear whitespace found in a URI reference and placed in
4917   the request-target.
4920<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
4922   This section summarizes major differences between versions HTTP/1.0
4923   and HTTP/1.1.
4926<section title="Multi-homed Web Servers" anchor="">
4928   The requirements that clients and servers support the <x:ref>Host</x:ref>
4929   header field (<xref target=""/>), report an error if it is
4930   missing from an HTTP/1.1 request, and accept absolute URIs (<xref target="request-target"/>)
4931   are among the most important changes defined by HTTP/1.1.
4934   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
4935   addresses and servers; there was no other established mechanism for
4936   distinguishing the intended server of a request than the IP address
4937   to which that request was directed. The <x:ref>Host</x:ref> header field was
4938   introduced during the development of HTTP/1.1 and, though it was
4939   quickly implemented by most HTTP/1.0 browsers, additional requirements
4940   were placed on all HTTP/1.1 requests in order to ensure complete
4941   adoption.  At the time of this writing, most HTTP-based services
4942   are dependent upon the Host header field for targeting requests.
4946<section title="Keep-Alive Connections" anchor="compatibility.with.http.1.0.persistent.connections">
4948   In HTTP/1.0, each connection is established by the client prior to the
4949   request and closed by the server after sending the response. However, some
4950   implementations implement the explicitly negotiated ("Keep-Alive") version
4951   of persistent connections described in <xref x:sec="19.7.1" x:fmt="of"
4952   target="RFC2068"/>.
4955   Some clients and servers might wish to be compatible with these previous
4956   approaches to persistent connections, by explicitly negotiating for them
4957   with a "Connection: keep-alive" request header field. However, some
4958   experimental implementations of HTTP/1.0 persistent connections are faulty;
4959   for example, if a HTTP/1.0 proxy server doesn't understand
4960   <x:ref>Connection</x:ref>, it will erroneously forward that header to the
4961   next inbound server, which would result in a hung connection.
4964   One attempted solution was the introduction of a Proxy-Connection header,
4965   targeted specifically at proxies. In practice, this was also unworkable,
4966   because proxies are often deployed in multiple layers, bringing about the
4967   same problem discussed above.
4970   As a result, clients are encouraged not to send the Proxy-Connection header
4971   in any requests.
4974   Clients are also encouraged to consider the use of Connection: keep-alive
4975   in requests carefully; while they can enable persistent connections with
4976   HTTP/1.0 servers, clients using them need will need to monitor the
4977   connection for "hung" requests (which indicate that the client ought stop
4978   sending the header), and this mechanism ought not be used by clients at all
4979   when a proxy is being used.
4983<section title="Introduction of Transfer-Encoding" anchor="introduction.of.transfer-encoding">
4985   HTTP/1.1 introduces the <x:ref>Transfer-Encoding</x:ref> header field
4986   (<xref target="header.transfer-encoding"/>). Proxies/gateways &MUST; remove
4987   any transfer-coding prior to forwarding a message via a MIME-compliant
4988   protocol.
4994<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
4996  Clarify that the string "HTTP" in the HTTP-version ABFN production is case
4997  sensitive. Restrict the version numbers to be single digits due to the fact
4998  that implementations are known to handle multi-digit version numbers
4999  incorrectly.
5000  (<xref target="http.version"/>)
5003  Update use of abs_path production from RFC 1808 to the path-absolute + query
5004  components of RFC 3986. State that the asterisk form is allowed for the OPTIONS
5005  request method only.
5006  (<xref target="request-target"/>)
5009  Require that invalid whitespace around field-names be rejected.
5010  (<xref target="header.fields"/>)
5013  Rules about implicit linear whitespace between certain grammar productions
5014  have been removed; now whitespace is only allowed where specifically
5015  defined in the ABNF.
5016  (<xref target="whitespace"/>)
5019  The NUL octet is no longer allowed in comment and quoted-string
5020  text. The quoted-pair rule no longer allows escaping control characters other than HTAB.
5021  Non-ASCII content in header fields and reason phrase has been obsoleted and
5022  made opaque (the TEXT rule was removed).
5023  (<xref target="field.components"/>)
5026  Empty list elements in list productions have been deprecated.
5027  (<xref target="abnf.extension"/>)
5030  Require recipients to handle bogus <x:ref>Content-Length</x:ref> header
5031  fields as errors.
5032  (<xref target="message.body"/>)
5035  Remove reference to non-existent identity transfer-coding value tokens.
5036  (Sections <xref format="counter" target="message.body"/> and
5037  <xref format="counter" target="transfer.codings"/>)
5040  Clarification that the chunk length does not include the count of the octets
5041  in the chunk header and trailer. Furthermore disallowed line folding
5042  in chunk extensions, and deprecate their use.
5043  (<xref target="chunked.encoding"/>)
5046  Registration of Transfer Codings now requires IETF Review
5047  (<xref target="transfer.coding.registry"/>)
5050  Remove hard limit of two connections per server.
5051  Remove requirement to retry a sequence of requests as long it was idempotent.
5052  Remove requirements about when servers are allowed to close connections
5053  prematurely.
5054  (<xref target="persistent.practical"/>)
5057  Remove requirement to retry requests under certain cirumstances when the
5058  server prematurely closes the connection.
5059  (<xref target="message.transmission.requirements"/>)
5062  Change ABNF productions for header fields to only define the field value.
5065  Clarify exactly when "close" connection options have to be sent.
5066  (<xref target="header.connection"/>)
5069  Define the semantics of the <x:ref>Upgrade</x:ref> header field in responses
5070  other than 101 (this was incorporated from <xref target="RFC2817"/>).
5071  (<xref target="header.upgrade"/>)
5074  Take over the Upgrade Token Registry, previously defined in
5075  <xref target="RFC2817" x:fmt="of" x:sec="7.2"/>.
5076  (<xref target="upgrade.token.registry"/>)
5081<?BEGININC p1-messaging.abnf-appendix ?>
5082<section xmlns:x="" title="Collected ABNF" anchor="collected.abnf">
5084<artwork type="abnf" name="p1-messaging.parsed-abnf">
5085<x:ref>BWS</x:ref> = OWS
5087<x:ref>Connection</x:ref> = *( "," OWS ) connection-option *( OWS "," [ OWS
5088 connection-option ] )
5089<x:ref>Content-Length</x:ref> = 1*DIGIT
5091<x:ref>HTTP-message</x:ref> = start-line *( header-field CRLF ) CRLF [ message-body
5092 ]
5093<x:ref>HTTP-name</x:ref> = %x48.54.54.50 ; HTTP
5094<x:ref>HTTP-version</x:ref> = HTTP-name "/" DIGIT "." DIGIT
5095<x:ref>Host</x:ref> = uri-host [ ":" port ]
5097<x:ref>OWS</x:ref> = *( SP / HTAB )
5099<x:ref>RWS</x:ref> = 1*( SP / HTAB )
5101<x:ref>TE</x:ref> = [ ( "," / t-codings ) *( OWS "," [ OWS t-codings ] ) ]
5102<x:ref>Trailer</x:ref> = *( "," OWS ) field-name *( OWS "," [ OWS field-name ] )
5103<x:ref>Transfer-Encoding</x:ref> = *( "," OWS ) transfer-coding *( OWS "," [ OWS
5104 transfer-coding ] )
5106<x:ref>URI-reference</x:ref> = &lt;URI-reference, defined in [RFC3986], Section 4.1&gt;
5107<x:ref>Upgrade</x:ref> = *( "," OWS ) protocol *( OWS "," [ OWS protocol ] )
5109<x:ref>Via</x:ref> = *( "," OWS ) ( received-protocol RWS received-by [ RWS comment
5110 ] ) *( OWS "," [ OWS ( received-protocol RWS received-by [ RWS
5111 comment ] ) ] )
5113<x:ref>absolute-URI</x:ref> = &lt;absolute-URI, defined in [RFC3986], Section 4.3&gt;
5114<x:ref>absolute-form</x:ref> = absolute-URI
5115<x:ref>asterisk-form</x:ref> = "*"
5116<x:ref>attribute</x:ref> = token
5117<x:ref>authority</x:ref> = &lt;authority, defined in [RFC3986], Section 3.2&gt;
5118<x:ref>authority-form</x:ref> = authority
5120<x:ref>chunk</x:ref> = chunk-size [ chunk-ext ] CRLF chunk-data CRLF
5121<x:ref>chunk-data</x:ref> = 1*OCTET
5122<x:ref>chunk-ext</x:ref> = *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
5123<x:ref>chunk-ext-name</x:ref> = token
5124<x:ref>chunk-ext-val</x:ref> = token / quoted-str-nf
5125<x:ref>chunk-size</x:ref> = 1*HEXDIG
5126<x:ref>chunked-body</x:ref> = *chunk last-chunk trailer-part CRLF
5127<x:ref>comment</x:ref> = "(" *( ctext / quoted-cpair / comment ) ")"
5128<x:ref>connection-option</x:ref> = token
5129<x:ref>ctext</x:ref> = OWS / %x21-27 ; '!'-'''
5130 / %x2A-5B ; '*'-'['
5131 / %x5D-7E ; ']'-'~'
5132 / obs-text
5134<x:ref>field-content</x:ref> = *( HTAB / SP / VCHAR / obs-text )
5135<x:ref>field-name</x:ref> = token
5136<x:ref>field-value</x:ref> = *( field-content / obs-fold )
5138<x:ref>header-field</x:ref> = field-name ":" OWS field-value BWS
5139<x:ref>http-URI</x:ref> = "http://" authority path-abempty [ "?" query ]
5140<x:ref>https-URI</x:ref> = "https://" authority path-abempty [ "?" query ]
5142<x:ref>last-chunk</x:ref> = 1*"0" [ chunk-ext ] CRLF
5144<x:ref>message-body</x:ref> = *OCTET
5145<x:ref>method</x:ref> = token
5147<x:ref>obs-fold</x:ref> = CRLF ( SP / HTAB )
5148<x:ref>obs-text</x:ref> = %x80-FF
5149<x:ref>origin-form</x:ref> = path-absolute [ "?" query ]
5151<x:ref>partial-URI</x:ref> = relative-part [ "?" query ]
5152<x:ref>path-abempty</x:ref> = &lt;path-abempty, defined in [RFC3986], Section 3.3&gt;
5153<x:ref>path-absolute</x:ref> = &lt;path-absolute, defined in [RFC3986], Section 3.3&gt;
5154<x:ref>port</x:ref> = &lt;port, defined in [RFC3986], Section 3.2.3&gt;
5155<x:ref>protocol</x:ref> = protocol-name [ "/" protocol-version ]
5156<x:ref>protocol-name</x:ref> = token
5157<x:ref>protocol-version</x:ref> = token
5158<x:ref>pseudonym</x:ref> = token
5160<x:ref>qdtext</x:ref> = OWS / "!" / %x23-5B ; '#'-'['
5161 / %x5D-7E ; ']'-'~'
5162 / obs-text
5163<x:ref>qdtext-nf</x:ref> = HTAB / SP / "!" / %x23-5B ; '#'-'['
5164 / %x5D-7E ; ']'-'~'
5165 / obs-text
5166<x:ref>query</x:ref> = &lt;query, defined in [RFC3986], Section 3.4&gt;
5167<x:ref>quoted-cpair</x:ref> = "\" ( HTAB / SP / VCHAR / obs-text )
5168<x:ref>quoted-pair</x:ref> = "\" ( HTAB / SP / VCHAR / obs-text )
5169<x:ref>quoted-str-nf</x:ref> = DQUOTE *( qdtext-nf / quoted-pair ) DQUOTE
5170<x:ref>quoted-string</x:ref> = DQUOTE *( qdtext / quoted-pair ) DQUOTE
5171<x:ref>qvalue</x:ref> = ( "0" [ "." *3DIGIT ] ) / ( "1" [ "." *3"0" ] )
5173<x:ref>reason-phrase</x:ref> = *( HTAB / SP / VCHAR / obs-text )
5174<x:ref>received-by</x:ref> = ( uri-host [ ":" port ] ) / pseudonym
5175<x:ref>received-protocol</x:ref> = [ protocol-name "/" ] protocol-version
5176<x:ref>relative-part</x:ref> = &lt;relative-part, defined in [RFC3986], Section 4.2&gt;
5177<x:ref>request-line</x:ref> = method SP request-target SP HTTP-version CRLF
5178<x:ref>request-target</x:ref> = origin-form / absolute-form / authority-form /
5179 asterisk-form
5181<x:ref>special</x:ref> = "(" / ")" / "&lt;" / "&gt;" / "@" / "," / ";" / ":" / "\" /
5182 DQUOTE / "/" / "[" / "]" / "?" / "=" / "{" / "}"
5183<x:ref>start-line</x:ref> = request-line / status-line
5184<x:ref>status-code</x:ref> = 3DIGIT
5185<x:ref>status-line</x:ref> = HTTP-version SP status-code SP reason-phrase CRLF
5187<x:ref>t-codings</x:ref> = "trailers" / ( transfer-extension [ te-params ] )
5188<x:ref>tchar</x:ref> = "!" / "#" / "$" / "%" / "&amp;" / "'" / "*" / "+" / "-" / "." /
5189 "^" / "_" / "`" / "|" / "~" / DIGIT / ALPHA
5190<x:ref>te-ext</x:ref> = OWS ";" OWS token [ "=" word ]
5191<x:ref>te-params</x:ref> = OWS ";" OWS "q=" qvalue *te-ext
5192<x:ref>token</x:ref> = 1*tchar
5193<x:ref>trailer-part</x:ref> = *( header-field CRLF )
5194<x:ref>transfer-coding</x:ref> = "chunked" / "compress" / "deflate" / "gzip" /
5195 transfer-extension
5196<x:ref>transfer-extension</x:ref> = token *( OWS ";" OWS transfer-parameter )
5197<x:ref>transfer-parameter</x:ref> = attribute BWS "=" BWS value
5199<x:ref>uri-host</x:ref> = &lt;host, defined in [RFC3986], Section 3.2.2&gt;
5201<x:ref>value</x:ref> = word
5203<x:ref>word</x:ref> = token / quoted-string
5206<figure><preamble>ABNF diagnostics:</preamble><artwork type="inline">
5207; Connection defined but not used
5208; Content-Length defined but not used
5209; HTTP-message defined but not used
5210; Host defined but not used
5211; TE defined but not used
5212; Trailer defined but not used
5213; Transfer-Encoding defined but not used
5214; URI-reference defined but not used
5215; Upgrade defined but not used
5216; Via defined but not used
5217; chunked-body defined but not used
5218; http-URI defined but not used
5219; https-URI defined but not used
5220; partial-URI defined but not used
5221; special defined but not used
5223<?ENDINC p1-messaging.abnf-appendix ?>
5225<section title="Change Log (to be removed by RFC Editor before publication)" anchor="change.log">
5227<section title="Since RFC 2616">
5229  Extracted relevant partitions from <xref target="RFC2616"/>.
5233<section title="Since draft-ietf-httpbis-p1-messaging-00">
5235  Closed issues:
5236  <list style="symbols">
5237    <t>
5238      <eref target=""/>:
5239      "HTTP Version should be case sensitive"
5240      (<eref target=""/>)
5241    </t>
5242    <t>
5243      <eref target=""/>:
5244      "'unsafe' characters"
5245      (<eref target=""/>)
5246    </t>
5247    <t>
5248      <eref target=""/>:
5249      "Chunk Size Definition"
5250      (<eref target=""/>)
5251    </t>
5252    <t>
5253      <eref target=""/>:
5254      "Message Length"
5255      (<eref target=""/>)
5256    </t>
5257    <t>
5258      <eref target=""/>:
5259      "Media Type Registrations"
5260      (<eref target=""/>)
5261    </t>
5262    <t>
5263      <eref target=""/>:
5264      "URI includes query"
5265      (<eref target=""/>)
5266    </t>
5267    <t>
5268      <eref target=""/>:
5269      "No close on 1xx responses"
5270      (<eref target=""/>)
5271    </t>
5272    <t>
5273      <eref target=""/>:
5274      "Remove 'identity' token references"
5275      (<eref target=""/>)
5276    </t>
5277    <t>
5278      <eref target=""/>:
5279      "Import query BNF"
5280    </t>
5281    <t>
5282      <eref target=""/>:
5283      "qdtext BNF"
5284    </t>
5285    <t>
5286      <eref target=""/>:
5287      "Normative and Informative references"
5288    </t>
5289    <t>
5290      <eref target=""/>:
5291      "RFC2606 Compliance"
5292    </t>
5293    <t>
5294      <eref target=""/>:
5295      "RFC977 reference"
5296    </t>
5297    <t>
5298      <eref target=""/>:
5299      "RFC1700 references"
5300    </t>
5301    <t>
5302      <eref target=""/>:
5303      "inconsistency in date format explanation"
5304    </t>
5305    <t>
5306      <eref target=""/>:
5307      "Date reference typo"
5308    </t>
5309    <t>
5310      <eref target=""/>:
5311      "Informative references"
5312    </t>
5313    <t>
5314      <eref target=""/>:
5315      "ISO-8859-1 Reference"
5316    </t>
5317    <t>
5318      <eref target=""/>:
5319      "Normative up-to-date references"
5320    </t>
5321  </list>
5324  Other changes:
5325  <list style="symbols">
5326    <t>
5327      Update media type registrations to use RFC4288 template.
5328    </t>
5329    <t>
5330      Use names of RFC4234 core rules DQUOTE and HTAB,
5331      fix broken ABNF for chunk-data
5332      (work in progress on <eref target=""/>)
5333    </t>
5334  </list>
5338<section title="Since draft-ietf-httpbis-p1-messaging-01">
5340  Closed issues:
5341  <list style="symbols">
5342    <t>
5343      <eref target=""/>:
5344      "Bodies on GET (and other) requests"
5345    </t>
5346    <t>
5347      <eref target=""/>:
5348      "Updating to RFC4288"
5349    </t>
5350    <t>
5351      <eref target=""/>:
5352      "Status Code and Reason Phrase"
5353    </t>
5354    <t>
5355      <eref target=""/>:
5356      "rel_path not used"
5357    </t>
5358  </list>
5361  Ongoing work on ABNF conversion (<eref target=""/>):
5362  <list style="symbols">
5363    <t>
5364      Get rid of duplicate BNF rule names ("host" -> "uri-host", "trailer" ->
5365      "trailer-part").
5366    </t>
5367    <t>
5368      Avoid underscore character in rule names ("http_URL" ->
5369      "http-URL", "abs_path" -> "path-absolute").
5370    </t>
5371    <t>
5372      Add rules for terms imported from URI spec ("absoluteURI", "authority",
5373      "path-absolute", "port", "query", "relativeURI", "host) &mdash; these will
5374      have to be updated when switching over to RFC3986.
5375    </t>
5376    <t>
5377      Synchronize core rules with RFC5234.
5378    </t>
5379    <t>
5380      Get rid of prose rules that span multiple lines.
5381    </t>
5382    <t>
5383      Get rid of unused rules LOALPHA and UPALPHA.
5384    </t>
5385    <t>
5386      Move "Product Tokens" section (back) into Part 1, as "token" is used
5387      in the definition of the Upgrade header field.
5388    </t>
5389    <t>
5390      Add explicit references to BNF syntax and rules imported from other parts of the specification.
5391    </t>
5392    <t>
5393      Rewrite prose rule "token" in terms of "tchar", rewrite prose rule "TEXT".
5394    </t>
5395  </list>
5399<section title="Since draft-ietf-httpbis-p1-messaging-02" anchor="changes.since.02">
5401  Closed issues:
5402  <list style="symbols">
5403    <t>
5404      <eref target=""/>:
5405      "HTTP-date vs. rfc1123-date"
5406    </t>
5407    <t>
5408      <eref target=""/>:
5409      "WS in quoted-pair"
5410    </t>
5411  </list>
5414  Ongoing work on IANA Message Header Field Registration (<eref target=""/>):
5415  <list style="symbols">
5416    <t>
5417      Reference RFC 3984, and update header field registrations for headers defined
5418      in this document.
5419    </t>
5420  </list>
5423  Ongoing work on ABNF conversion (<eref target=""/>):
5424  <list style="symbols">
5425    <t>
5426      Replace string literals when the string really is case-sensitive (HTTP-version).
5427    </t>
5428  </list>
5432<section title="Since draft-ietf-httpbis-p1-messaging-03" anchor="changes.since.03">
5434  Closed issues:
5435  <list style="symbols">
5436    <t>
5437      <eref target=""/>:
5438      "Connection closing"
5439    </t>
5440    <t>
5441      <eref target=""/>:
5442      "Move registrations and registry information to IANA Considerations"
5443    </t>
5444    <t>
5445      <eref target=""/>:
5446      "need new URL for PAD1995 reference"
5447    </t>
5448    <t>
5449      <eref target=""/>:
5450      "IANA Considerations: update HTTP URI scheme registration"
5451    </t>
5452    <t>
5453      <eref target=""/>:
5454      "Cite HTTPS URI scheme definition"
5455    </t>
5456    <t>
5457      <eref target=""/>:
5458      "List-type headers vs Set-Cookie"
5459    </t>
5460  </list>
5463  Ongoing work on ABNF conversion (<eref target=""/>):
5464  <list style="symbols">
5465    <t>
5466      Replace string literals when the string really is case-sensitive (HTTP-Date).
5467    </t>
5468    <t>
5469      Replace HEX by HEXDIG for future consistence with RFC 5234's core rules.
5470    </t>
5471  </list>
5475<section title="Since draft-ietf-httpbis-p1-messaging-04" anchor="changes.since.04">
5477  Closed issues:
5478  <list style="symbols">
5479    <t>
5480      <eref target=""/>:
5481      "Out-of-date reference for URIs"
5482    </t>
5483    <t>
5484      <eref target=""/>:
5485      "RFC 2822 is updated by RFC 5322"
5486    </t>
5487  </list>
5490  Ongoing work on ABNF conversion (<eref target=""/>):
5491  <list style="symbols">
5492    <t>
5493      Use "/" instead of "|" for alternatives.
5494    </t>
5495    <t>
5496      Get rid of RFC822 dependency; use RFC5234 plus extensions instead.
5497    </t>
5498    <t>
5499      Only reference RFC 5234's core rules.
5500    </t>
5501    <t>
5502      Introduce new ABNF rules for "bad" whitespace ("BWS"), optional
5503      whitespace ("OWS") and required whitespace ("RWS").
5504    </t>
5505    <t>
5506      Rewrite ABNFs to spell out whitespace rules, factor out
5507      header field value format definitions.
5508    </t>
5509  </list>
5513<section title="Since draft-ietf-httpbis-p1-messaging-05" anchor="changes.since.05">
5515  Closed issues:
5516  <list style="symbols">
5517    <t>
5518      <eref target=""/>:
5519      "Header LWS"
5520    </t>
5521    <t>
5522      <eref target=""/>:
5523      "Sort 1.3 Terminology"
5524    </t>
5525    <t>
5526      <eref target=""/>:
5527      "RFC2047 encoded words"
5528    </t>
5529    <t>
5530      <eref target=""/>:
5531      "Character Encodings in TEXT"
5532    </t>
5533    <t>
5534      <eref target=""/>:
5535      "Line Folding"
5536    </t>
5537    <t>
5538      <eref target=""/>:
5539      "OPTIONS * and proxies"
5540    </t>
5541    <t>
5542      <eref target=""/>:
5543      "reason-phrase BNF"
5544    </t>
5545    <t>
5546      <eref target=""/>:
5547      "Use of TEXT"
5548    </t>
5549    <t>
5550      <eref target=""/>:
5551      "Join "Differences Between HTTP Entities and RFC 2045 Entities"?"
5552    </t>
5553    <t>
5554      <eref target=""/>:
5555      "RFC822 reference left in discussion of date formats"
5556    </t>
5557  </list>
5560  Final work on ABNF conversion (<eref target=""/>):
5561  <list style="symbols">
5562    <t>
5563      Rewrite definition of list rules, deprecate empty list elements.
5564    </t>
5565    <t>
5566      Add appendix containing collected and expanded ABNF.
5567    </t>
5568  </list>
5571  Other changes:
5572  <list style="symbols">
5573    <t>
5574      Rewrite introduction; add mostly new Architecture Section.
5575    </t>
5576    <t>
5577      Move definition of quality values from Part 3 into Part 1;
5578      make TE request header field grammar independent of accept-params (defined in Part 3).
5579    </t>
5580  </list>
5584<section title="Since draft-ietf-httpbis-p1-messaging-06" anchor="changes.since.06">
5586  Closed issues:
5587  <list style="symbols">
5588    <t>
5589      <eref target=""/>:
5590      "base for numeric protocol elements"
5591    </t>
5592    <t>
5593      <eref target=""/>:
5594      "comment ABNF"
5595    </t>
5596  </list>
5599  Partly resolved issues:
5600  <list style="symbols">
5601    <t>
5602      <eref target=""/>:
5603      "205 Bodies" (took out language that implied that there might be
5604      methods for which a request body MUST NOT be included)
5605    </t>
5606    <t>
5607      <eref target=""/>:
5608      "editorial improvements around HTTP-date"
5609    </t>
5610  </list>
5614<section title="Since draft-ietf-httpbis-p1-messaging-07" anchor="changes.since.07">
5616  Closed issues:
5617  <list style="symbols">
5618    <t>
5619      <eref target=""/>:
5620      "Repeating single-value headers"
5621    </t>
5622    <t>
5623      <eref target=""/>:
5624      "increase connection limit"
5625    </t>
5626    <t>
5627      <eref target=""/>:
5628      "IP addresses in URLs"
5629    </t>
5630    <t>
5631      <eref target=""/>:
5632      "take over HTTP Upgrade Token Registry"
5633    </t>
5634    <t>
5635      <eref target=""/>:
5636      "CR and LF in chunk extension values"
5637    </t>
5638    <t>
5639      <eref target=""/>:
5640      "HTTP/0.9 support"
5641    </t>
5642    <t>
5643      <eref target=""/>:
5644      "pick IANA policy (RFC5226) for Transfer Coding / Content Coding"
5645    </t>
5646    <t>
5647      <eref target=""/>:
5648      "move definitions of gzip/deflate/compress to part 1"
5649    </t>
5650    <t>
5651      <eref target=""/>:
5652      "disallow control characters in quoted-pair"
5653    </t>
5654  </list>
5657  Partly resolved issues:
5658  <list style="symbols">
5659    <t>
5660      <eref target=""/>:
5661      "update IANA requirements wrt Transfer-Coding values" (add the
5662      IANA Considerations subsection)
5663    </t>
5664  </list>
5668<section title="Since draft-ietf-httpbis-p1-messaging-08" anchor="changes.since.08">
5670  Closed issues:
5671  <list style="symbols">
5672    <t>
5673      <eref target=""/>:
5674      "header parsing, treatment of leading and trailing OWS"
5675    </t>
5676  </list>
5679  Partly resolved issues:
5680  <list style="symbols">
5681    <t>
5682      <eref target=""/>:
5683      "Placement of 13.5.1 and 13.5.2"
5684    </t>
5685    <t>
5686      <eref target=""/>:
5687      "use of term "word" when talking about header structure"
5688    </t>
5689  </list>
5693<section title="Since draft-ietf-httpbis-p1-messaging-09" anchor="changes.since.09">
5695  Closed issues:
5696  <list style="symbols">
5697    <t>
5698      <eref target=""/>:
5699      "Clarification of the term 'deflate'"
5700    </t>
5701    <t>
5702      <eref target=""/>:
5703      "OPTIONS * and proxies"
5704    </t>
5705    <t>
5706      <eref target=""/>:
5707      "MIME-Version not listed in P1, general header fields"
5708    </t>
5709    <t>
5710      <eref target=""/>:
5711      "IANA registry for content/transfer encodings"
5712    </t>
5713    <t>
5714      <eref target=""/>:
5715      "Case-sensitivity of HTTP-date"
5716    </t>
5717    <t>
5718      <eref target=""/>:
5719      "use of term "word" when talking about header structure"
5720    </t>
5721  </list>
5724  Partly resolved issues:
5725  <list style="symbols">
5726    <t>
5727      <eref target=""/>:
5728      "Term for the requested resource's URI"
5729    </t>
5730  </list>
5734<section title="Since draft-ietf-httpbis-p1-messaging-10" anchor="changes.since.10">
5736  Closed issues:
5737  <list style="symbols">
5738    <t>
5739      <eref target=""/>:
5740      "Connection Closing"
5741    </t>
5742    <t>
5743      <eref target=""/>:
5744      "Delimiting messages with multipart/byteranges"
5745    </t>
5746    <t>
5747      <eref target=""/>:
5748      "Handling multiple Content-Length headers"
5749    </t>
5750    <t>
5751      <eref target=""/>:
5752      "Clarify entity / representation / variant terminology"
5753    </t>
5754    <t>
5755      <eref target=""/>:
5756      "consider removing the 'changes from 2068' sections"
5757    </t>
5758  </list>
5761  Partly resolved issues:
5762  <list style="symbols">
5763    <t>
5764      <eref target=""/>:
5765      "HTTP(s) URI scheme definitions"
5766    </t>
5767  </list>
5771<section title="Since draft-ietf-httpbis-p1-messaging-11" anchor="changes.since.11">
5773  Closed issues:
5774  <list style="symbols">
5775    <t>
5776      <eref target=""/>:
5777      "Trailer requirements"
5778    </t>
5779    <t>
5780      <eref target=""/>:
5781      "Text about clock requirement for caches belongs in p6"
5782    </t>
5783    <t>
5784      <eref target=""/>:
5785      "effective request URI: handling of missing host in HTTP/1.0"
5786    </t>
5787    <t>
5788      <eref target=""/>:
5789      "confusing Date requirements for clients"
5790    </t>
5791  </list>
5794  Partly resolved issues:
5795  <list style="symbols">
5796    <t>
5797      <eref target=""/>:
5798      "Handling multiple Content-Length headers"
5799    </t>
5800  </list>
5804<section title="Since draft-ietf-httpbis-p1-messaging-12" anchor="changes.since.12">
5806  Closed issues:
5807  <list style="symbols">
5808    <t>
5809      <eref target=""/>:
5810      "RFC2145 Normative"
5811    </t>
5812    <t>
5813      <eref target=""/>:
5814      "HTTP(s) URI scheme definitions" (tune the requirements on userinfo)
5815    </t>
5816    <t>
5817      <eref target=""/>:
5818      "define 'transparent' proxy"
5819    </t>
5820    <t>
5821      <eref target=""/>:
5822      "Header Classification"
5823    </t>
5824    <t>
5825      <eref target=""/>:
5826      "Is * usable as a request-uri for new methods?"
5827    </t>
5828    <t>
5829      <eref target=""/>:
5830      "Migrate Upgrade details from RFC2817"
5831    </t>
5832    <t>
5833      <eref target=""/>:
5834      "untangle ABNFs for header fields"
5835    </t>
5836    <t>
5837      <eref target=""/>:
5838      "update RFC 2109 reference"
5839    </t>
5840  </list>
5844<section title="Since draft-ietf-httpbis-p1-messaging-13" anchor="changes.since.13">
5846  Closed issues:
5847  <list style="symbols">
5848    <t>
5849      <eref target=""/>:
5850      "Allow is not in 13.5.2"
5851    </t>
5852    <t>
5853      <eref target=""/>:
5854      "Handling multiple Content-Length headers"
5855    </t>
5856    <t>
5857      <eref target=""/>:
5858      "untangle ABNFs for header fields"
5859    </t>
5860    <t>
5861      <eref target=""/>:
5862      "Content-Length ABNF broken"
5863    </t>
5864  </list>
5868<section title="Since draft-ietf-httpbis-p1-messaging-14" anchor="changes.since.14">
5870  Closed issues:
5871  <list style="symbols">
5872    <t>
5873      <eref target=""/>:
5874      "HTTP-version should be redefined as fixed length pair of DIGIT . DIGIT"
5875    </t>
5876    <t>
5877      <eref target=""/>:
5878      "Recommend minimum sizes for protocol elements"
5879    </t>
5880    <t>
5881      <eref target=""/>:
5882      "Set expectations around buffering"
5883    </t>
5884    <t>
5885      <eref target=""/>:
5886      "Considering messages in isolation"
5887    </t>
5888  </list>
5892<section title="Since draft-ietf-httpbis-p1-messaging-15" anchor="changes.since.15">
5894  Closed issues:
5895  <list style="symbols">
5896    <t>
5897      <eref target=""/>:
5898      "DNS Spoofing / DNS Binding advice"
5899    </t>
5900    <t>
5901      <eref target=""/>:
5902      "move RFCs 2145, 2616, 2817 to Historic status"
5903    </t>
5904    <t>
5905      <eref target=""/>:
5906      "\-escaping in quoted strings"
5907    </t>
5908    <t>
5909      <eref target=""/>:
5910      "'Close' should be reserved in the HTTP header field registry"
5911    </t>
5912  </list>
5916<section title="Since draft-ietf-httpbis-p1-messaging-16" anchor="changes.since.16">
5918  Closed issues:
5919  <list style="symbols">
5920    <t>
5921      <eref target=""/>:
5922      "Document HTTP's error-handling philosophy"
5923    </t>
5924    <t>
5925      <eref target=""/>:
5926      "Explain header registration"
5927    </t>
5928    <t>
5929      <eref target=""/>:
5930      "Revise Acknowledgements Sections"
5931    </t>
5932    <t>
5933      <eref target=""/>:
5934      "Retrying Requests"
5935    </t>
5936    <t>
5937      <eref target=""/>:
5938      "Closing the connection on server error"
5939    </t>
5940  </list>
5944<section title="Since draft-ietf-httpbis-p1-messaging-17" anchor="changes.since.17">
5946  Closed issues:
5947  <list style="symbols">
5948    <t>
5949      <eref target=""/>:
5950      "Proxy-Connection and Keep-Alive"
5951    </t>
5952    <t>
5953      <eref target=""/>:
5954      "Clarify 'User Agent'"
5955    </t>
5956    <t>
5957      <eref target=""/>:
5958      "Define non-final responses"
5959    </t>
5960    <t>
5961      <eref target=""/>:
5962      "intended maturity level vs normative references"
5963    </t>
5964    <t>
5965      <eref target=""/>:
5966      "Intermediary rewriting of queries"
5967    </t>
5968  </list>
5972<section title="Since draft-ietf-httpbis-p1-messaging-18" anchor="changes.since.18">
5974  Closed issues:
5975  <list style="symbols">
5976    <t>
5977      <eref target=""/>:
5978      "message-body in CONNECT response"
5979    </t>
5980    <t>
5981      <eref target=""/>:
5982      "Misplaced text on connection handling in p2"
5983    </t>
5984    <t>
5985      <eref target=""/>:
5986      "wording of line folding rule"
5987    </t>
5988    <t>
5989      <eref target=""/>:
5990      "chunk-extensions"
5991    </t>
5992    <t>
5993      <eref target=""/>:
5994      "make IANA policy definitions consistent"
5995    </t>
5996  </list>
6000<section title="Since draft-ietf-httpbis-p1-messaging-19" anchor="changes.since.19">
6002  Closed issues:
6003  <list style="symbols">
6004    <t>
6005      <eref target=""/>:
6006      "make IANA policy definitions consistent"
6007    </t>
6008    <t>
6009      <eref target=""/>:
6010      "clarify connection header field values are case-insensitive"
6011    </t>
6012    <t>
6013      <eref target=""/>:
6014      "ABNF requirements for recipients"
6015    </t>
6016    <t>
6017      <eref target=""/>:
6018      "note introduction of new IANA registries as normative changes"
6019    </t>
6020  </list>
Note: See TracBrowser for help on using the repository browser.