source: draft-ietf-httpbis/latest/p1-messaging.xml @ 157

Last change on this file since 157 was 155, checked in by julian.reschke@…, 15 years ago

fix BNF rule for chunk-data; addresses #36.

  • Property svn:eol-style set to native
File size: 175.9 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 "January">
16  <!ENTITY ID-YEAR "2008">
17  <!ENTITY caching                "<xref target='Part6' x:rel='#caching' xmlns:x=''/>">
18  <!ENTITY payload                "<xref target='Part3' xmlns:x=''/>">
19  <!ENTITY media-types            "<xref target='Part3' x:rel='#media.types' xmlns:x=''/>">
20  <!ENTITY content-codings        "<xref target='Part3' x:rel='#content.codings' xmlns:x=''/>">
21  <!ENTITY CONNECT                "<xref target='Part2' x:rel='#CONNECT' xmlns:x=''/>">
22  <!ENTITY content.negotiation    "<xref target='Part3' x:rel='#content.negotiation' xmlns:x=''/>">
23  <!ENTITY diff2045entity         "<xref target='Part3' x:rel='#differences.between.http.entities.and.rfc.2045.entities' xmlns:x=''/>">
24  <!ENTITY entity                 "<xref target='Part3' x:rel='#entity' xmlns:x=''/>">
25  <!ENTITY entity-header-fields   "<xref target='Part3' x:rel='#entity.header.fields' xmlns:x=''/>">
26  <!ENTITY header-cache-control   "<xref target='Part6' x:rel='#header.cache-control' xmlns:x=''/>">
27  <!ENTITY header-expect          "<xref target='Part2' x:rel='#header.expect' xmlns:x=''/>">
28  <!ENTITY header-pragma          "<xref target='Part6' x:rel='#header.pragma' xmlns:x=''/>">
29  <!ENTITY header-warning         "<xref target='Part6' x:rel='#header.warning' xmlns:x=''/>">
30  <!ENTITY idempotent-methods     "<xref target='Part2' x:rel='#idempotent.methods' xmlns:x=''/>">
31  <!ENTITY qvalue                 "<xref target='Part3' x:rel='#quality.values' xmlns:x=''/>">
32  <!ENTITY request-header-fields  "<xref target='Part2' x:rel='#request.header.fields' xmlns:x=''/>">
33  <!ENTITY response-header-fields "<xref target='Part2' x:rel='#response.header.fields' xmlns:x=''/>">
34  <!ENTITY method                 "<xref target='Part2' x:rel='#method' xmlns:x=''/>">
35  <!ENTITY status-codes           "<xref target='Part2' x:rel='' xmlns:x=''/>">
36  <!ENTITY status-100             "<xref target='Part2' x:rel='#status.100' xmlns:x=''/>">
37  <!ENTITY status-1xx             "<xref target='Part2' x:rel='#status.1xx' xmlns:x=''/>">
38  <!ENTITY status-414             "<xref target='Part2' x:rel='#status.414' xmlns:x=''/>">
40<?rfc toc="yes" ?>
41<?rfc symrefs="yes" ?>
42<?rfc sortrefs="yes" ?>
43<?rfc compact="yes"?>
44<?rfc subcompact="no" ?>
45<?rfc linkmailto="no" ?>
46<?rfc editing="no" ?>
47<?rfc-ext allow-markup-in-artwork="yes" ?>
48<?rfc-ext include-references-in-index="yes" ?>
49<rfc obsoletes="2616" category="std"
50     ipr="full3978" docName="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"
51     xmlns:x=''>
54  <title abbrev="HTTP/1.1, Part 1">HTTP/1.1, part 1: URIs, Connections, and Message Parsing</title>
56  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
57    <organization abbrev="Day Software">Day Software</organization>
58    <address>
59      <postal>
60        <street>23 Corporate Plaza DR, Suite 280</street>
61        <city>Newport Beach</city>
62        <region>CA</region>
63        <code>92660</code>
64        <country>USA</country>
65      </postal>
66      <phone>+1-949-706-5300</phone>
67      <facsimile>+1-949-706-5305</facsimile>
68      <email></email>
69      <uri></uri>
70    </address>
71  </author>
73  <author initials="J." surname="Gettys" fullname="Jim Gettys">
74    <organization>One Laptop per Child</organization>
75    <address>
76      <postal>
77        <street>21 Oak Knoll Road</street>
78        <city>Carlisle</city>
79        <region>MA</region>
80        <code>01741</code>
81        <country>USA</country>
82      </postal>
83      <email></email>
84      <uri></uri>
85    </address>
86  </author>
88  <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
89    <organization abbrev="HP">Hewlett-Packard Company</organization>
90    <address>
91      <postal>
92        <street>HP Labs, Large Scale Systems Group</street>
93        <street>1501 Page Mill Road, MS 1177</street>
94        <city>Palo Alto</city>
95        <region>CA</region>
96        <code>94304</code>
97        <country>USA</country>
98      </postal>
99      <email></email>
100    </address>
101  </author>
103  <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
104    <organization abbrev="Microsoft">Microsoft Corporation</organization>
105    <address>
106      <postal>
107        <street>1 Microsoft Way</street>
108        <city>Redmond</city>
109        <region>WA</region>
110        <code>98052</code>
111        <country>USA</country>
112      </postal>
113      <email></email>
114    </address>
115  </author>
117  <author initials="L." surname="Masinter" fullname="Larry Masinter">
118    <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
119    <address>
120      <postal>
121        <street>345 Park Ave</street>
122        <city>San Jose</city>
123        <region>CA</region>
124        <code>95110</code>
125        <country>USA</country>
126      </postal>
127      <email></email>
128      <uri></uri>
129    </address>
130  </author>
132  <author initials="P." surname="Leach" fullname="Paul J. Leach">
133    <organization abbrev="Microsoft">Microsoft Corporation</organization>
134    <address>
135      <postal>
136        <street>1 Microsoft Way</street>
137        <city>Redmond</city>
138        <region>WA</region>
139        <code>98052</code>
140      </postal>
141      <email></email>
142    </address>
143  </author>
145  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
146    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
147    <address>
148      <postal>
149        <street>MIT Computer Science and Artificial Intelligence Laboratory</street>
150        <street>The Stata Center, Building 32</street>
151        <street>32 Vassar Street</street>
152        <city>Cambridge</city>
153        <region>MA</region>
154        <code>02139</code>
155        <country>USA</country>
156      </postal>
157      <email></email>
158      <uri></uri>
159    </address>
160  </author>
162  <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
163    <organization abbrev="W3C">World Wide Web Consortium</organization>
164    <address>
165      <postal>
166        <street>W3C / ERCIM</street>
167        <street>2004, rte des Lucioles</street>
168        <city>Sophia-Antipolis</city>
169        <region>AM</region>
170        <code>06902</code>
171        <country>France</country>
172      </postal>
173      <email></email>
174      <uri></uri>
175    </address>
176  </author>
178  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
179    <organization abbrev="greenbytes">greenbytes GmbH</organization>
180    <address>
181      <postal>
182        <street>Hafenweg 16</street>
183        <city>Muenster</city><region>NW</region><code>48155</code>
184        <country>Germany</country>
185      </postal>
186      <phone>+49 251 2807760</phone>   
187      <facsimile>+49 251 2807761</facsimile>   
188      <email></email>       
189      <uri></uri>     
190    </address>
191  </author>
193  <date month="&ID-MONTH;" year="&ID-YEAR;"/>
197   The Hypertext Transfer Protocol (HTTP) is an application-level
198   protocol for distributed, collaborative, hypermedia information
199   systems. HTTP has been in use by the World Wide Web global information
200   initiative since 1990. This document is Part 1 of the seven-part specification
201   that defines the protocol referred to as "HTTP/1.1" and, taken together,
202   obsoletes RFC 2616.  Part 1 provides an overview of HTTP and
203   its associated terminology, defines the "http" and "https" Uniform
204   Resource Identifier (URI) schemes, defines the generic message syntax
205   and parsing requirements for HTTP message frames, and describes
206   general security concerns for implementations.
210<note title="Editorial Note (To be removed by RFC Editor)">
211  <t>
212    Discussion of this draft should take place on the HTTPBIS working group
213    mailing list ( The current issues list is
214    at <eref target=""/>
215    and related documents (including fancy diffs) can be found at
216    <eref target=""/>.
217  </t>
218  <t>
219    This draft incorporates those issue resolutions that were either
220    collected in the original RFC2616 errata list (<eref target=""/>),
221    or which were agreed upon on the mailing list between October 2006 and
222    November 2007 (as published in "draft-lafon-rfc2616bis-03").
223  </t>
227<section title="Introduction" anchor="introduction">
229   This document will define aspects of HTTP related to overall network
230   operation, message framing, interaction with transport protocols, and
231   URI schemes. Right now it only includes the extracted relevant sections
232   of <xref target="RFC2616"/>.
234<section title="Purpose" anchor="intro.purpose">
236   The Hypertext Transfer Protocol (HTTP) is an application-level
237   protocol for distributed, collaborative, hypermedia information
238   systems. HTTP has been in use by the World-Wide Web global
239   information initiative since 1990. The first version of HTTP,
240   referred to as HTTP/0.9, was a simple protocol for raw data transfer
241   across the Internet. HTTP/1.0, as defined by <xref target="RFC1945"/>, improved
242   the protocol by allowing messages to be in the format of MIME-like
243   messages, containing metainformation about the data transferred and
244   modifiers on the request/response semantics. However, HTTP/1.0 does
245   not sufficiently take into consideration the effects of hierarchical
246   proxies, caching, the need for persistent connections, or virtual
247   hosts. In addition, the proliferation of incompletely-implemented
248   applications calling themselves "HTTP/1.0" has necessitated a
249   protocol version change in order for two communicating applications
250   to determine each other's true capabilities.
253   This specification defines the protocol referred to as "HTTP/1.1".
254   This protocol includes more stringent requirements than HTTP/1.0 in
255   order to ensure reliable implementation of its features.
258   Practical information systems require more functionality than simple
259   retrieval, including search, front-end update, and annotation. HTTP
260   allows an open-ended set of methods and headers that indicate the
261   purpose of a request <xref target="RFC2324"/>. It builds on the discipline of reference
262   provided by the Uniform Resource Identifier (URI) <xref target="RFC1630"/>, as a location
263   (URL) <xref target="RFC1738"/> or name (URN) <xref target="RFC1737"/>, for indicating the resource to which a
264   method is to be applied. Messages are passed in a format similar to
265   that used by Internet mail <xref target="RFC2822"/> as defined by the Multipurpose
266   Internet Mail Extensions (MIME) <xref target="RFC2045"/>.
269   HTTP is also used as a generic protocol for communication between
270   user agents and proxies/gateways to other Internet systems, including
271   those supported by the SMTP <xref target="RFC2821"/>, NNTP <xref target="RFC3977"/>, FTP <xref target="RFC959"/>, Gopher <xref target="RFC1436"/>,
272   and WAIS <xref target="WAIS"/> protocols. In this way, HTTP allows basic hypermedia
273   access to resources available from diverse applications.
277<section title="Requirements" anchor="intro.requirements">
279   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
280   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
281   document are to be interpreted as described in <xref target="RFC2119"/>.
284   An implementation is not compliant if it fails to satisfy one or more
285   of the &MUST; or &REQUIRED; level requirements for the protocols it
286   implements. An implementation that satisfies all the &MUST; or &REQUIRED;
287   level and all the &SHOULD; level requirements for its protocols is said
288   to be "unconditionally compliant"; one that satisfies all the &MUST;
289   level requirements but not all the &SHOULD; level requirements for its
290   protocols is said to be "conditionally compliant."
294<section title="Terminology" anchor="intro.terminology">
296   This specification uses a number of terms to refer to the roles
297   played by participants in, and objects of, the HTTP communication.
300  <iref item="connection"/>
301  <x:dfn>connection</x:dfn>
302  <list>
303    <t>
304      A transport layer virtual circuit established between two programs
305      for the purpose of communication.
306    </t>
307  </list>
310  <iref item="message"/>
311  <x:dfn>message</x:dfn>
312  <list>
313    <t>
314      The basic unit of HTTP communication, consisting of a structured
315      sequence of octets matching the syntax defined in <xref target="http.message"/> and
316      transmitted via the connection.
317    </t>
318  </list>
321  <iref item="request"/>
322  <x:dfn>request</x:dfn>
323  <list>
324    <t>
325      An HTTP request message, as defined in <xref target="request"/>.
326    </t>
327  </list>
330  <iref item="response"/>
331  <x:dfn>response</x:dfn>
332  <list>
333    <t>
334      An HTTP response message, as defined in <xref target="response"/>.
335    </t>
336  </list>
339  <iref item="resource"/>
340  <x:dfn>resource</x:dfn>
341  <list>
342    <t>
343      A network data object or service that can be identified by a URI,
344      as defined in <xref target="uri"/>. Resources may be available in multiple
345      representations (e.g. multiple languages, data formats, size, and
346      resolutions) or vary in other ways.
347    </t>
348  </list>
351  <iref item="entity"/>
352  <x:dfn>entity</x:dfn>
353  <list>
354    <t>
355      The information transferred as the payload of a request or
356      response. An entity consists of metainformation in the form of
357      entity-header fields and content in the form of an entity-body, as
358      described in &entity;.
359    </t>
360  </list>
363  <iref item="representation"/>
364  <x:dfn>representation</x:dfn>
365  <list>
366    <t>
367      An entity included with a response that is subject to content
368      negotiation, as described in &content.negotiation;. There may exist multiple
369      representations associated with a particular response status.
370    </t>
371  </list>
374  <iref item="content negotiation"/>
375  <x:dfn>content negotiation</x:dfn>
376  <list>
377    <t>
378      The mechanism for selecting the appropriate representation when
379      servicing a request, as described in &content.negotiation;. The
380      representation of entities in any response can be negotiated
381      (including error responses).
382    </t>
383  </list>
386  <iref item="variant"/>
387  <x:dfn>variant</x:dfn>
388  <list>
389    <t>
390      A resource may have one, or more than one, representation(s)
391      associated with it at any given instant. Each of these
392      representations is termed a `variant'.  Use of the term `variant'
393      does not necessarily imply that the resource is subject to content
394      negotiation.
395    </t>
396  </list>
399  <iref item="client"/>
400  <x:dfn>client</x:dfn>
401  <list>
402    <t>
403      A program that establishes connections for the purpose of sending
404      requests.
405    </t>
406  </list>
409  <iref item="user agent"/>
410  <x:dfn>user agent</x:dfn>
411  <list>
412    <t>
413      The client which initiates a request. These are often browsers,
414      editors, spiders (web-traversing robots), or other end user tools.
415    </t>
416  </list>
419  <iref item="server"/>
420  <x:dfn>server</x:dfn>
421  <list>
422    <t>
423      An application program that accepts connections in order to
424      service requests by sending back responses. Any given program may
425      be capable of being both a client and a server; our use of these
426      terms refers only to the role being performed by the program for a
427      particular connection, rather than to the program's capabilities
428      in general. Likewise, any server may act as an origin server,
429      proxy, gateway, or tunnel, switching behavior based on the nature
430      of each request.
431    </t>
432  </list>
435  <iref item="origin server"/>
436  <x:dfn>origin server</x:dfn>
437  <list>
438    <t>
439      The server on which a given resource resides or is to be created.
440    </t>
441  </list>
444  <iref item="proxy"/>
445  <x:dfn>proxy</x:dfn>
446  <list>
447    <t>
448      An intermediary program which acts as both a server and a client
449      for the purpose of making requests on behalf of other clients.
450      Requests are serviced internally or by passing them on, with
451      possible translation, to other servers. A proxy &MUST; implement
452      both the client and server requirements of this specification. A
453      "transparent proxy" is a proxy that does not modify the request or
454      response beyond what is required for proxy authentication and
455      identification. A "non-transparent proxy" is a proxy that modifies
456      the request or response in order to provide some added service to
457      the user agent, such as group annotation services, media type
458      transformation, protocol reduction, or anonymity filtering. Except
459      where either transparent or non-transparent behavior is explicitly
460      stated, the HTTP proxy requirements apply to both types of
461      proxies.
462    </t>
463  </list>
466  <iref item="gateway"/>
467  <x:dfn>gateway</x:dfn>
468  <list>
469    <t>
470      A server which acts as an intermediary for some other server.
471      Unlike a proxy, a gateway receives requests as if it were the
472      origin server for the requested resource; the requesting client
473      may not be aware that it is communicating with a gateway.
474    </t>
475  </list>
478  <iref item="tunnel"/>
479  <x:dfn>tunnel</x:dfn>
480  <list>
481    <t>
482      An intermediary program which is acting as a blind relay between
483      two connections. Once active, a tunnel is not considered a party
484      to the HTTP communication, though the tunnel may have been
485      initiated by an HTTP request. The tunnel ceases to exist when both
486      ends of the relayed connections are closed.
487    </t>
488  </list>
491  <iref item="cache"/>
492  <x:dfn>cache</x:dfn>
493  <list>
494    <t>
495      A program's local store of response messages and the subsystem
496      that controls its message storage, retrieval, and deletion. A
497      cache stores cacheable responses in order to reduce the response
498      time and network bandwidth consumption on future, equivalent
499      requests. Any client or server may include a cache, though a cache
500      cannot be used by a server that is acting as a tunnel.
501    </t>
502  </list>
505  <iref item="cacheable"/>
506  <x:dfn>cacheable</x:dfn>
507  <list>
508    <t>
509      A response is cacheable if a cache is allowed to store a copy of
510      the response message for use in answering subsequent requests. The
511      rules for determining the cacheability of HTTP responses are
512      defined in &caching;. Even if a resource is cacheable, there may
513      be additional constraints on whether a cache can use the cached
514      copy for a particular request.
515    </t>
516  </list>
519  <iref item="upstream"/>
520  <iref item="downstream"/>
521  <x:dfn>upstream</x:dfn>/<x:dfn>downstream</x:dfn>
522  <list>
523    <t>
524      Upstream and downstream describe the flow of a message: all
525      messages flow from upstream to downstream.
526    </t>
527  </list>
530  <iref item="inbound"/>
531  <iref item="outbound"/>
532  <x:dfn>inbound</x:dfn>/<x:dfn>outbound</x:dfn>
533  <list>
534    <t>
535      Inbound and outbound refer to the request and response paths for
536      messages: "inbound" means "traveling toward the origin server",
537      and "outbound" means "traveling toward the user agent"
538    </t>
539  </list>
543<section title="Overall Operation" anchor="intro.overall.operation">
545   The HTTP protocol is a request/response protocol. A client sends a
546   request to the server in the form of a request method, URI, and
547   protocol version, followed by a MIME-like message containing request
548   modifiers, client information, and possible body content over a
549   connection with a server. The server responds with a status line,
550   including the message's protocol version and a success or error code,
551   followed by a MIME-like message containing server information, entity
552   metainformation, and possible entity-body content. The relationship
553   between HTTP and MIME is described in &diff2045entity;.
556   Most HTTP communication is initiated by a user agent and consists of
557   a request to be applied to a resource on some origin server. In the
558   simplest case, this may be accomplished via a single connection (v)
559   between the user agent (UA) and the origin server (O).
561<figure><artwork type="drawing">
562       request chain ------------------------&gt;
563    UA -------------------v------------------- O
564       &lt;----------------------- response chain
567   A more complicated situation occurs when one or more intermediaries
568   are present in the request/response chain. There are three common
569   forms of intermediary: proxy, gateway, and tunnel. A proxy is a
570   forwarding agent, receiving requests for a URI in its absolute form,
571   rewriting all or part of the message, and forwarding the reformatted
572   request toward the server identified by the URI. A gateway is a
573   receiving agent, acting as a layer above some other server(s) and, if
574   necessary, translating the requests to the underlying server's
575   protocol. A tunnel acts as a relay point between two connections
576   without changing the messages; tunnels are used when the
577   communication needs to pass through an intermediary (such as a
578   firewall) even when the intermediary cannot understand the contents
579   of the messages.
581<figure><artwork type="drawing">
582       request chain --------------------------------------&gt;
583    UA -----v----- A -----v----- B -----v----- C -----v----- O
584       &lt;------------------------------------- response chain
587   The figure above shows three intermediaries (A, B, and C) between the
588   user agent and origin server. A request or response message that
589   travels the whole chain will pass through four separate connections.
590   This distinction is important because some HTTP communication options
591   may apply only to the connection with the nearest, non-tunnel
592   neighbor, only to the end-points of the chain, or to all connections
593   along the chain. Although the diagram is linear, each participant may
594   be engaged in multiple, simultaneous communications. For example, B
595   may be receiving requests from many clients other than A, and/or
596   forwarding requests to servers other than C, at the same time that it
597   is handling A's request.
600   Any party to the communication which is not acting as a tunnel may
601   employ an internal cache for handling requests. The effect of a cache
602   is that the request/response chain is shortened if one of the
603   participants along the chain has a cached response applicable to that
604   request. The following illustrates the resulting chain if B has a
605   cached copy of an earlier response from O (via C) for a request which
606   has not been cached by UA or A.
608<figure><artwork type="drawing">
609          request chain ----------&gt;
610       UA -----v----- A -----v----- B - - - - - - C - - - - - - O
611          &lt;--------- response chain
614   Not all responses are usefully cacheable, and some requests may
615   contain modifiers which place special requirements on cache behavior.
616   HTTP requirements for cache behavior and cacheable responses are
617   defined in &caching;.
620   In fact, there are a wide variety of architectures and configurations
621   of caches and proxies currently being experimented with or deployed
622   across the World Wide Web. These systems include national hierarchies
623   of proxy caches to save transoceanic bandwidth, systems that
624   broadcast or multicast cache entries, organizations that distribute
625   subsets of cached data via CD-ROM, and so on. HTTP systems are used
626   in corporate intranets over high-bandwidth links, and for access via
627   PDAs with low-power radio links and intermittent connectivity. The
628   goal of HTTP/1.1 is to support the wide diversity of configurations
629   already deployed while introducing protocol constructs that meet the
630   needs of those who build web applications that require high
631   reliability and, failing that, at least reliable indications of
632   failure.
635   HTTP communication usually takes place over TCP/IP connections. The
636   default port is TCP 80 (<eref target=""/>), but other ports can be used. This does
637   not preclude HTTP from being implemented on top of any other protocol
638   on the Internet, or on other networks. HTTP only presumes a reliable
639   transport; any protocol that provides such guarantees can be used;
640   the mapping of the HTTP/1.1 request and response structures onto the
641   transport data units of the protocol in question is outside the scope
642   of this specification.
645   In HTTP/1.0, most implementations used a new connection for each
646   request/response exchange. In HTTP/1.1, a connection may be used for
647   one or more request/response exchanges, although connections may be
648   closed for a variety of reasons (see <xref target="persistent.connections"/>).
653<section title="Notational Conventions and Generic Grammar" anchor="notation">
655<section title="Augmented BNF" anchor="notation.abnf">
657   All of the mechanisms specified in this document are described in
658   both prose and an augmented Backus-Naur Form (BNF) similar to that
659   used by <xref target="RFC822ABNF"/>. Implementors will need to be familiar with the
660   notation in order to understand this specification. The augmented BNF
661   includes the following constructs:
664   name = definition
665  <list>
666    <t>
667      The name of a rule is simply the name itself (without any
668      enclosing "&lt;" and "&gt;") and is separated from its definition by the
669      equal "=" character. White space is only significant in that
670      indentation of continuation lines is used to indicate a rule
671      definition that spans more than one line. Certain basic rules are
672      in uppercase, such as SP, LWS, HTAB, CRLF, DIGIT, ALPHA, etc. Angle
673      brackets are used within definitions whenever their presence will
674      facilitate discerning the use of rule names.
675    </t>
676  </list>
679   "literal"
680  <list>
681    <t>
682      Quotation marks surround literal text. Unless stated otherwise,
683      the text is case-insensitive.
684    </t>
685  </list>
688   rule1 | rule2
689  <list>
690    <t>
691      Elements separated by a bar ("|") are alternatives, e.g., "yes |
692      no" will accept yes or no.
693    </t>
694  </list>
697   (rule1 rule2)
698  <list>
699    <t>
700      Elements enclosed in parentheses are treated as a single element.
701      Thus, "(elem (foo | bar) elem)" allows the token sequences "elem
702      foo elem" and "elem bar elem".
703    </t>
704  </list>
707   *rule
708  <list>
709    <t>
710      The character "*" preceding an element indicates repetition. The
711      full form is "&lt;n&gt;*&lt;m&gt;element" indicating at least &lt;n&gt; and at most
712      &lt;m&gt; occurrences of element. Default values are 0 and infinity so
713      that "*(element)" allows any number, including zero; "1*element"
714      requires at least one; and "1*2element" allows one or two.
715    </t>
716  </list>
719   [rule]
720  <list>
721    <t>
722      Square brackets enclose optional elements; "[foo bar]" is
723      equivalent to "*1(foo bar)".
724    </t>
725  </list>
728   N rule
729  <list>
730    <t>
731      Specific repetition: "&lt;n&gt;(element)" is equivalent to
732      "&lt;n&gt;*&lt;n&gt;(element)"; that is, exactly &lt;n&gt; occurrences of (element).
733      Thus 2DIGIT is a 2-digit number, and 3ALPHA is a string of three
734      alphabetic characters.
735    </t>
736  </list>
739   #rule
740  <list>
741    <t>
742      A construct "#" is defined, similar to "*", for defining lists of
743      elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating at least
744      &lt;n&gt; and at most &lt;m&gt; elements, each separated by one or more commas
745      (",") and &OPTIONAL; linear white space (LWS). This makes the usual
746      form of lists very easy; a rule such as
747    </t>
748    <t>
749         ( *LWS element *( *LWS "," *LWS element ))
750    </t>
751    <t>
752      can be shown as
753    </t>
754    <t>
755         1#element
756    </t>
757    <t>
758      Wherever this construct is used, null elements are allowed, but do
759      not contribute to the count of elements present. That is,
760      "(element), , (element) " is permitted, but counts as only two
761      elements. Therefore, where at least one element is required, at
762      least one non-null element &MUST; be present. Default values are 0
763      and infinity so that "#element" allows any number, including zero;
764      "1#element" requires at least one; and "1#2element" allows one or
765      two.
766    </t>
767  </list>
770   ; comment
771  <list>
772    <t>
773      A semi-colon, set off some distance to the right of rule text,
774      starts a comment that continues to the end of line. This is a
775      simple way of including useful notes in parallel with the
776      specifications.
777    </t>
778  </list>
781   implied *LWS
782  <list>
783    <t>
784      The grammar described by this specification is word-based. Except
785      where noted otherwise, linear white space (LWS) can be included
786      between any two adjacent words (token or quoted-string), and
787      between adjacent words and separators, without changing the
788      interpretation of a field. At least one delimiter (LWS and/or
789      separators) &MUST; exist between any two tokens (for the definition
790      of "token" below), since they would otherwise be interpreted as a
791      single token.
792    </t>
793  </list>
797<section title="Basic Rules" anchor="basic.rules">
798<x:anchor-alias value="OCTET"/>
799<x:anchor-alias value="CHAR"/>
800<x:anchor-alias value="UPALPHA"/>
801<x:anchor-alias value="LOALPHA"/>
802<x:anchor-alias value="ALPHA"/>
803<x:anchor-alias value="DIGIT"/>
804<x:anchor-alias value="CTL"/>
805<x:anchor-alias value="CR"/>
806<x:anchor-alias value="LF"/>
807<x:anchor-alias value="SP"/>
808<x:anchor-alias value="HTAB"/>
809<x:anchor-alias value="CRLF"/>
810<x:anchor-alias value="LWS"/>
811<x:anchor-alias value="TEXT"/>
812<x:anchor-alias value="HEX"/>
813<x:anchor-alias value="token"/>
814<x:anchor-alias value="separators"/>
815<x:anchor-alias value="comment"/>
816<x:anchor-alias value="ctext"/>
817<x:anchor-alias value="quoted-string"/>
818<x:anchor-alias value="qdtext"/>
819<x:anchor-alias value="quoted-pair"/>
821   The following rules are used throughout this specification to
822   describe basic parsing constructs. The US-ASCII coded character set
823   is defined by ANSI X3.4-1986 <xref target="USASCII"/>.
825<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="OCTET"/><iref primary="true" item="Grammar" subitem="CHAR"/><iref primary="true" item="Grammar" subitem="UPALPHA"/><iref primary="true" item="Grammar" subitem="LOALPHA"/><iref primary="true" item="Grammar" subitem="ALPHA"/><iref primary="true" item="Grammar" subitem="DIGIT"/><iref primary="true" item="Grammar" subitem="CTL"/><iref primary="true" item="Grammar" subitem="CR"/><iref primary="true" item="Grammar" subitem="LF"/><iref primary="true" item="Grammar" subitem="SP"/><iref primary="true" item="Grammar" subitem="HTAB"/><iref primary="true" item="Grammar" subitem="DQUOTE"/>
826  OCTET          = &lt;any 8-bit sequence of data&gt;
827  CHAR           = &lt;any US-ASCII character (octets 0 - 127)&gt;
828  UPALPHA        = &lt;any US-ASCII uppercase letter "A".."Z"&gt;
829  LOALPHA        = &lt;any US-ASCII lowercase letter "a".."z"&gt;
830  ALPHA          = UPALPHA | LOALPHA
831  DIGIT          = &lt;any US-ASCII digit "0".."9"&gt;
832  CTL            = &lt;any US-ASCII control character
833                   (octets 0 - 31) and DEL (127)&gt;
834  CR             = &lt;US-ASCII CR, carriage return (13)&gt;
835  LF             = &lt;US-ASCII LF, linefeed (10)&gt;
836  SP             = &lt;US-ASCII SP, space (32)&gt;
837  HTAB           = &lt;US-ASCII HT, horizontal-tab (9)&gt;
838  DQUOTE         = &lt;US-ASCII double-quote mark (34)&gt;
841   HTTP/1.1 defines the sequence CR LF as the end-of-line marker for all
842   protocol elements except the entity-body (see <xref target="tolerant.applications"/> for
843   tolerant applications). The end-of-line marker within an entity-body
844   is defined by its associated media type, as described in &media-types;.
846<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="CRLF"/>
847  CRLF           = CR LF
850   HTTP/1.1 header field values can be folded onto multiple lines if the
851   continuation line begins with a space or horizontal tab. All linear
852   white space, including folding, has the same semantics as SP. A
853   recipient &MAY; replace any linear white space with a single SP before
854   interpreting the field value or forwarding the message downstream.
856<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="LWS"/>
857  LWS            = [CRLF] 1*( SP | HTAB )
860   The TEXT rule is only used for descriptive field contents and values
861   that are not intended to be interpreted by the message parser. Words
862   of *TEXT &MAY; contain characters from character sets other than ISO-8859-1
863   <xref target="ISO-8859-1"/> only when encoded according to the rules of
864   <xref target="RFC2047"/>.
866<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TEXT"/>
867  TEXT           = &lt;any OCTET except CTLs,
868                   but including LWS&gt;
871   A CRLF is allowed in the definition of TEXT only as part of a header
872   field continuation. It is expected that the folding LWS will be
873   replaced with a single SP before interpretation of the TEXT value.
876   Hexadecimal numeric characters are used in several protocol elements.
878<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HEX"/>
879  HEX            = "A" | "B" | "C" | "D" | "E" | "F"
880                 | "a" | "b" | "c" | "d" | "e" | "f" | DIGIT
883   Many HTTP/1.1 header field values consist of words separated by LWS
884   or special characters. These special characters &MUST; be in a quoted
885   string to be used within a parameter value (as defined in
886   <xref target="transfer.codings"/>).
888<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="token"/><iref primary="true" item="Grammar" subitem="separators"/>
889  token          = 1*&lt;any CHAR except CTLs or separators&gt;
890  separators     = "(" | ")" | "&lt;" | "&gt;" | "@"
891                 | "," | ";" | ":" | "\" | DQUOTE
892                 | "/" | "[" | "]" | "?" | "="
893                 | "{" | "}" | SP | HTAB
896   Comments can be included in some HTTP header fields by surrounding
897   the comment text with parentheses. Comments are only allowed in
898   fields containing "comment" as part of their field value definition.
899   In all other fields, parentheses are considered part of the field
900   value.
902<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/>
903  comment        = "(" *( ctext | quoted-pair | comment ) ")"
904  ctext          = &lt;any TEXT excluding "(" and ")"&gt;
907   A string of text is parsed as a single word if it is quoted using
908   double-quote marks.
910<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-string"/><iref primary="true" item="Grammar" subitem="qdtext"/>
911  quoted-string  = ( DQUOTE *(qdtext | quoted-pair ) DQUOTE )
912  qdtext         = &lt;any TEXT excluding DQUOTE and "\">
915   The backslash character ("\") &MAY; be used as a single-character
916   quoting mechanism only within quoted-string and comment constructs.
918<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-pair"/>
919  quoted-pair    = "\" CHAR
924<section title="Protocol Parameters" anchor="protocol.parameters">
926<section title="HTTP Version" anchor="http.version">
928   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate versions
929   of the protocol. The protocol versioning policy is intended to allow
930   the sender to indicate the format of a message and its capacity for
931   understanding further HTTP communication, rather than the features
932   obtained via that communication. No change is made to the version
933   number for the addition of message components which do not affect
934   communication behavior or which only add to extensible field values.
935   The &lt;minor&gt; number is incremented when the changes made to the
936   protocol add features which do not change the general message parsing
937   algorithm, but which may add to the message semantics and imply
938   additional capabilities of the sender. The &lt;major&gt; number is
939   incremented when the format of a message within the protocol is
940   changed. See <xref target="RFC2145"/> for a fuller explanation.
943   The version of an HTTP message is indicated by an HTTP-Version field
944   in the first line of the message. HTTP-Version is case-sensitive.
946<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-Version"/>
947  HTTP-Version   = "HTTP" "/" 1*DIGIT "." 1*DIGIT
950   Note that the major and minor numbers &MUST; be treated as separate
951   integers and that each &MAY; be incremented higher than a single digit.
952   Thus, HTTP/2.4 is a lower version than HTTP/2.13, which in turn is
953   lower than HTTP/12.3. Leading zeros &MUST; be ignored by recipients and
954   &MUST-NOT; be sent.
957   An application that sends a request or response message that includes
958   HTTP-Version of "HTTP/1.1" &MUST; be at least conditionally compliant
959   with this specification. Applications that are at least conditionally
960   compliant with this specification &SHOULD; use an HTTP-Version of
961   "HTTP/1.1" in their messages, and &MUST; do so for any message that is
962   not compatible with HTTP/1.0. For more details on when to send
963   specific HTTP-Version values, see <xref target="RFC2145"/>.
966   The HTTP version of an application is the highest HTTP version for
967   which the application is at least conditionally compliant.
970   Proxy and gateway applications need to be careful when forwarding
971   messages in protocol versions different from that of the application.
972   Since the protocol version indicates the protocol capability of the
973   sender, a proxy/gateway &MUST-NOT; send a message with a version
974   indicator which is greater than its actual version. If a higher
975   version request is received, the proxy/gateway &MUST; either downgrade
976   the request version, or respond with an error, or switch to tunnel
977   behavior.
980   Due to interoperability problems with HTTP/1.0 proxies discovered
981   since the publication of <xref target="RFC2068"/>, caching proxies &MUST;, gateways
982   &MAY;, and tunnels &MUST-NOT; upgrade the request to the highest version
983   they support. The proxy/gateway's response to that request &MUST; be in
984   the same major version as the request.
987  <list>
988    <t>
989      <x:h>Note:</x:h> Converting between versions of HTTP may involve modification
990      of header fields required or forbidden by the versions involved.
991    </t>
992  </list>
996<section title="Uniform Resource Identifiers" anchor="uri">
998   URIs have been known by many names: WWW addresses, Universal Document
999   Identifiers, Universal Resource Identifiers <xref target="RFC1630"/>, and finally the
1000   combination of Uniform Resource Locators (URL) <xref target="RFC1738"/> and Names (URN)
1001   <xref target="RFC1737"/>. As far as HTTP is concerned, Uniform Resource Identifiers are
1002   simply formatted strings which identify--via name, location, or any
1003   other characteristic--a resource.
1006<section title="General Syntax" anchor="general.syntax">
1008   URIs in HTTP can be represented in absolute form or relative to some
1009   known base URI <xref target="RFC1808"/>, depending upon the context of their use. The two
1010   forms are differentiated by the fact that absolute URIs always begin
1011   with a scheme name followed by a colon. For definitive information on
1012   URL syntax and semantics, see "Uniform Resource Identifiers (URI):
1013   Generic Syntax and Semantics," <xref target="RFC2396"/> (which replaces <xref target="RFC1738"/>
1014   and <xref target="RFC1808"/>). This specification adopts the
1015   definitions of "URI-reference", "absoluteURI", "relativeURI", "port",
1016   "host", "abs_path", "rel_path", "query", and "authority" from that
1017   specification.
1020   The HTTP protocol does not place any a priori limit on the length of
1021   a URI. Servers &MUST; be able to handle the URI of any resource they
1022   serve, and &SHOULD; be able to handle URIs of unbounded length if they
1023   provide GET-based forms that could generate such URIs. A server
1024   &SHOULD; return 414 (Request-URI Too Long) status if a URI is longer
1025   than the server can handle (see &status-414;).
1028  <list>
1029    <t>
1030      <x:h>Note:</x:h> Servers ought to be cautious about depending on URI lengths
1031      above 255 bytes, because some older client or proxy
1032      implementations might not properly support these lengths.
1033    </t>
1034  </list>
1038<section title="http URL" anchor="http.url">
1040   The "http" scheme is used to locate network resources via the HTTP
1041   protocol. This section defines the scheme-specific syntax and
1042   semantics for http URLs.
1044<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="http_URL"/>
1045  http_URL = "http:" "//" host [ ":" port ] [ abs_path [ "?" query ]]
1048   If the port is empty or not given, port 80 is assumed. The semantics
1049   are that the identified resource is located at the server listening
1050   for TCP connections on that port of that host, and the Request-URI
1051   for the resource is abs_path (<xref target="request-uri"/>). The use of IP addresses
1052   in URLs &SHOULD; be avoided whenever possible (see <xref target="RFC1900"/>). If
1053   the abs_path is not present in the URL, it &MUST; be given as "/" when
1054   used as a Request-URI for a resource (<xref target="request-uri"/>). If a proxy
1056   receives a host name which is not a fully qualified domain name, it
1057   &MAY; add its domain to the host name it received. If a proxy receives
1058   a fully qualified domain name, the proxy &MUST-NOT; change the host
1059   name.
1063<section title="URI Comparison" anchor="uri.comparison">
1065   When comparing two URIs to decide if they match or not, a client
1066   &SHOULD; use a case-sensitive octet-by-octet comparison of the entire
1067   URIs, with these exceptions:
1068  <list style="symbols">
1069    <t>A port that is empty or not given is equivalent to the default
1070        port for that URI-reference;</t>
1071    <t>Comparisons of host names &MUST; be case-insensitive;</t>
1072    <t>Comparisons of scheme names &MUST; be case-insensitive;</t>
1073    <t>An empty abs_path is equivalent to an abs_path of "/".</t>
1074  </list>
1077   Characters other than those in the "reserved" set (see
1078   <xref target="RFC2396"/>) are equivalent to their ""%" HEX HEX" encoding.
1081   For example, the following three URIs are equivalent:
1083<figure><artwork type="example">
1091<section title="Date/Time Formats" anchor="date.time.formats">
1092<section title="Full Date" anchor="">
1094   HTTP applications have historically allowed three different formats
1095   for the representation of date/time stamps:
1097<figure><artwork type="example">
1098   Sun, 06 Nov 1994 08:49:37 GMT  ; RFC 822, updated by RFC 1123
1099   Sunday, 06-Nov-94 08:49:37 GMT ; obsolete RFC 850 format
1100   Sun Nov  6 08:49:37 1994       ; ANSI C's asctime() format
1103   The first format is preferred as an Internet standard and represents
1104   a fixed-length subset of that defined by <xref target="RFC1123"/> (an update to
1105   <xref target="RFC822"/>). The other formats are described here only for
1106   compatibility with obsolete implementations.
1107   HTTP/1.1 clients and servers that parse the date value &MUST; accept
1108   all three formats (for compatibility with HTTP/1.0), though they &MUST;
1109   only generate the RFC 1123 format for representing HTTP-date values
1110   in header fields. See <xref target="tolerant.applications"/> for further information.
1113      <x:h>Note:</x:h> Recipients of date values are encouraged to be robust in
1114      accepting date values that may have been sent by non-HTTP
1115      applications, as is sometimes the case when retrieving or posting
1116      messages via proxies/gateways to SMTP or NNTP.
1119   All HTTP date/time stamps &MUST; be represented in Greenwich Mean Time
1120   (GMT), without exception. For the purposes of HTTP, GMT is exactly
1121   equal to UTC (Coordinated Universal Time). This is indicated in the
1122   first two formats by the inclusion of "GMT" as the three-letter
1123   abbreviation for time zone, and &MUST; be assumed when reading the
1124   asctime format. HTTP-date is case sensitive and &MUST-NOT; include
1125   additional LWS beyond that specifically included as SP in the
1126   grammar.
1128<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-date"/><iref primary="true" item="Grammar" subitem="rfc1123-date"/><iref primary="true" item="Grammar" subitem="rfc850-date"/><iref primary="true" item="Grammar" subitem="asctime-date"/><iref primary="true" item="Grammar" subitem="date1"/><iref primary="true" item="Grammar" subitem="date2"/><iref primary="true" item="Grammar" subitem="date3"/><iref primary="true" item="Grammar" subitem="time"/><iref primary="true" item="Grammar" subitem="wkday"/><iref primary="true" item="Grammar" subitem="weekday"/><iref primary="true" item="Grammar" subitem="month"/>
1129  HTTP-date    = rfc1123-date | rfc850-date | asctime-date
1130  rfc1123-date = wkday "," SP date1 SP time SP "GMT"
1131  rfc850-date  = weekday "," SP date2 SP time SP "GMT"
1132  asctime-date = wkday SP date3 SP time SP 4DIGIT
1133  date1        = 2DIGIT SP month SP 4DIGIT
1134                 ; day month year (e.g., 02 Jun 1982)
1135  date2        = 2DIGIT "-" month "-" 2DIGIT
1136                 ; day-month-year (e.g., 02-Jun-82)
1137  date3        = month SP ( 2DIGIT | ( SP 1DIGIT ))
1138                 ; month day (e.g., Jun  2)
1139  time         = 2DIGIT ":" 2DIGIT ":" 2DIGIT
1140                 ; 00:00:00 - 23:59:59
1141  wkday        = "Mon" | "Tue" | "Wed"
1142               | "Thu" | "Fri" | "Sat" | "Sun"
1143  weekday      = "Monday" | "Tuesday" | "Wednesday"
1144               | "Thursday" | "Friday" | "Saturday" | "Sunday"
1145  month        = "Jan" | "Feb" | "Mar" | "Apr"
1146               | "May" | "Jun" | "Jul" | "Aug"
1147               | "Sep" | "Oct" | "Nov" | "Dec"
1150      <x:h>Note:</x:h> HTTP requirements for the date/time stamp format apply only
1151      to their usage within the protocol stream. Clients and servers are
1152      not required to use these formats for user presentation, request
1153      logging, etc.
1158<section title="Transfer Codings" anchor="transfer.codings">
1160   Transfer-coding values are used to indicate an encoding
1161   transformation that has been, can be, or may need to be applied to an
1162   entity-body in order to ensure "safe transport" through the network.
1163   This differs from a content coding in that the transfer-coding is a
1164   property of the message, not of the original entity.
1166<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/>
1167  transfer-coding         = "chunked" | transfer-extension
1168  transfer-extension      = token *( ";" parameter )
1171   Parameters are in  the form of attribute/value pairs.
1173<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="parameter"/><iref primary="true" item="Grammar" subitem="attribute"/><iref primary="true" item="Grammar" subitem="value"/>
1174  parameter               = attribute "=" value
1175  attribute               = token
1176  value                   = token | quoted-string
1179   All transfer-coding values are case-insensitive. HTTP/1.1 uses
1180   transfer-coding values in the TE header field (<xref target="header.te"/>) and in
1181   the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1184   Whenever a transfer-coding is applied to a message-body, the set of
1185   transfer-codings &MUST; include "chunked", unless the message is
1186   terminated by closing the connection. When the "chunked" transfer-coding
1187   is used, it &MUST; be the last transfer-coding applied to the
1188   message-body. The "chunked" transfer-coding &MUST-NOT; be applied more
1189   than once to a message-body. These rules allow the recipient to
1190   determine the transfer-length of the message (<xref target="message.length"/>).
1193   Transfer-codings are analogous to the Content-Transfer-Encoding
1194   values of MIME <xref target="RFC2045"/>, which were designed to enable safe transport of
1195   binary data over a 7-bit transport service. However, safe transport
1196   has a different focus for an 8bit-clean transfer protocol. In HTTP,
1197   the only unsafe characteristic of message-bodies is the difficulty in
1198   determining the exact body length (<xref target="message.length"/>), or the desire to
1199   encrypt data over a shared transport.
1202   The Internet Assigned Numbers Authority (IANA) acts as a registry for
1203   transfer-coding value tokens. Initially, the registry contains the
1204   following tokens: "chunked" (<xref target="chunked.transfer.encoding"/>),
1205   "gzip", "compress", and "deflate" (&content-codings;).
1208   New transfer-coding value tokens &SHOULD; be registered in the same way
1209   as new content-coding value tokens (&content-codings;).
1212   A server which receives an entity-body with a transfer-coding it does
1213   not understand &SHOULD; return 501 (Not Implemented), and close the
1214   connection. A server &MUST-NOT; send transfer-codings to an HTTP/1.0
1215   client.
1218<section title="Chunked Transfer Coding" anchor="chunked.transfer.encoding">
1220   The chunked encoding modifies the body of a message in order to
1221   transfer it as a series of chunks, each with its own size indicator,
1222   followed by an &OPTIONAL; trailer containing entity-header fields. This
1223   allows dynamically produced content to be transferred along with the
1224   information necessary for the recipient to verify that it has
1225   received the full message.
1227<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-extension"/><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"/>
1228  Chunked-Body   = *chunk
1229                   last-chunk
1230                   trailer
1231                   CRLF
1233  chunk          = chunk-size [ chunk-extension ] CRLF
1234                   chunk-data CRLF
1235  chunk-size     = 1*HEX
1236  last-chunk     = 1*("0") [ chunk-extension ] CRLF
1238  chunk-extension= *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
1239  chunk-ext-name = token
1240  chunk-ext-val  = token | quoted-string
1241  chunk-data     = 1*OCTET ; a sequence of chunk-size octets
1242  trailer        = *(entity-header CRLF)
1245   The chunk-size field is a string of hex digits indicating the size of
1246   the chunk-data in octets. The chunked encoding is ended by any chunk whose size is
1247   zero, followed by the trailer, which is terminated by an empty line.
1250   The trailer allows the sender to include additional HTTP header
1251   fields at the end of the message. The Trailer header field can be
1252   used to indicate which header fields are included in a trailer (see
1253   <xref target="header.trailer"/>).
1256   A server using chunked transfer-coding in a response &MUST-NOT; use the
1257   trailer for any header fields unless at least one of the following is
1258   true:
1259  <list style="numbers">
1260    <t>the request included a TE header field that indicates "trailers" is
1261     acceptable in the transfer-coding of the  response, as described in
1262     <xref target="header.te"/>; or,</t>
1264    <t>the server is the origin server for the response, the trailer
1265     fields consist entirely of optional metadata, and the recipient
1266     could use the message (in a manner acceptable to the origin server)
1267     without receiving this metadata.  In other words, the origin server
1268     is willing to accept the possibility that the trailer fields might
1269     be silently discarded along the path to the client.</t>
1270  </list>
1273   This requirement prevents an interoperability failure when the
1274   message is being received by an HTTP/1.1 (or later) proxy and
1275   forwarded to an HTTP/1.0 recipient. It avoids a situation where
1276   compliance with the protocol would have necessitated a possibly
1277   infinite buffer on the proxy.
1280   A process for decoding the "chunked" transfer-coding
1281   can be represented in pseudo-code as:
1283<figure><artwork type="code">
1284    length := 0
1285    read chunk-size, chunk-extension (if any) and CRLF
1286    while (chunk-size &gt; 0) {
1287       read chunk-data and CRLF
1288       append chunk-data to entity-body
1289       length := length + chunk-size
1290       read chunk-size and CRLF
1291    }
1292    read entity-header
1293    while (entity-header not empty) {
1294       append entity-header to existing header fields
1295       read entity-header
1296    }
1297    Content-Length := length
1298    Remove "chunked" from Transfer-Encoding
1301   All HTTP/1.1 applications &MUST; be able to receive and decode the
1302   "chunked" transfer-coding, and &MUST; ignore chunk-extension extensions
1303   they do not understand.
1310<section title="HTTP Message" anchor="http.message">
1312<section title="Message Types" anchor="message.types">
1314   HTTP messages consist of requests from client to server and responses
1315   from server to client.
1317<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-message"/>
1318  HTTP-message   = Request | Response     ; HTTP/1.1 messages
1321   Request (<xref target="request"/>) and Response (<xref target="response"/>) messages use the generic
1322   message format of <xref target="RFC2822"/> for transferring entities (the payload
1323   of the message). Both types of message consist of a start-line, zero
1324   or more header fields (also known as "headers"), an empty line (i.e.,
1325   a line with nothing preceding the CRLF) indicating the end of the
1326   header fields, and possibly a message-body.
1328<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="generic-message"/><iref primary="true" item="Grammar" subitem="start-line"/>
1329  generic-message = start-line
1330                    *(message-header CRLF)
1331                    CRLF
1332                    [ message-body ]
1333  start-line      = Request-Line | Status-Line
1336   In the interest of robustness, servers &SHOULD; ignore any empty
1337   line(s) received where a Request-Line is expected. In other words, if
1338   the server is reading the protocol stream at the beginning of a
1339   message and receives a CRLF first, it should ignore the CRLF.
1342   Certain buggy HTTP/1.0 client implementations generate extra CRLF's
1343   after a POST request. To restate what is explicitly forbidden by the
1344   BNF, an HTTP/1.1 client &MUST-NOT; preface or follow a request with an
1345   extra CRLF.
1349<section title="Message Headers" anchor="message.headers">
1351   HTTP header fields, which include general-header (<xref target="general.header.fields"/>),
1352   request-header (&request-header-fields;), response-header (&response-header-fields;), and
1353   entity-header (&entity-header-fields;) fields, follow the same generic format as
1354   that given in <xref target="RFC2822" x:fmt="of" x:sec="2.1"/>. Each header field consists
1355   of a name followed by a colon (":") and the field value. Field names
1356   are case-insensitive. The field value &MAY; be preceded by any amount
1357   of LWS, though a single SP is preferred. Header fields can be
1358   extended over multiple lines by preceding each extra line with at
1359   least one SP or HTAB. Applications ought to follow "common form", where
1360   one is known or indicated, when generating HTTP constructs, since
1361   there might exist some implementations that fail to accept anything
1362   beyond the common forms.
1364<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="message-header"/><iref primary="true" item="Grammar" subitem="field-name"/><iref primary="true" item="Grammar" subitem="field-value"/><iref primary="true" item="Grammar" subitem="field-content"/>
1365  message-header = field-name ":" [ field-value ]
1366  field-name     = token
1367  field-value    = *( field-content | LWS )
1368  field-content  = &lt;the OCTETs making up the field-value
1369                   and consisting of either *TEXT or combinations
1370                   of token, separators, and quoted-string&gt;
1373   The field-content does not include any leading or trailing LWS:
1374   linear white space occurring before the first non-whitespace
1375   character of the field-value or after the last non-whitespace
1376   character of the field-value. Such leading or trailing LWS &MAY; be
1377   removed without changing the semantics of the field value. Any LWS
1378   that occurs between field-content &MAY; be replaced with a single SP
1379   before interpreting the field value or forwarding the message
1380   downstream.
1383   The order in which header fields with differing field names are
1384   received is not significant. However, it is "good practice" to send
1385   general-header fields first, followed by request-header or response-header
1386   fields, and ending with the entity-header fields.
1389   Multiple message-header fields with the same field-name &MAY; be
1390   present in a message if and only if the entire field-value for that
1391   header field is defined as a comma-separated list [i.e., #(values)].
1392   It &MUST; be possible to combine the multiple header fields into one
1393   "field-name: field-value" pair, without changing the semantics of the
1394   message, by appending each subsequent field-value to the first, each
1395   separated by a comma. The order in which header fields with the same
1396   field-name are received is therefore significant to the
1397   interpretation of the combined field value, and thus a proxy &MUST-NOT;
1398   change the order of these field values when a message is forwarded.
1402<section title="Message Body" anchor="message.body">
1404   The message-body (if any) of an HTTP message is used to carry the
1405   entity-body associated with the request or response. The message-body
1406   differs from the entity-body only when a transfer-coding has been
1407   applied, as indicated by the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1409<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="message-body"/>
1410  message-body = entity-body
1411               | &lt;entity-body encoded as per Transfer-Encoding&gt;
1414   Transfer-Encoding &MUST; be used to indicate any transfer-codings
1415   applied by an application to ensure safe and proper transfer of the
1416   message. Transfer-Encoding is a property of the message, not of the
1417   entity, and thus &MAY; be added or removed by any application along the
1418   request/response chain. (However, <xref target="transfer.codings"/> places restrictions on
1419   when certain transfer-codings may be used.)
1422   The rules for when a message-body is allowed in a message differ for
1423   requests and responses.
1426   The presence of a message-body in a request is signaled by the
1427   inclusion of a Content-Length or Transfer-Encoding header field in
1428   the request's message-headers. A message-body &MUST-NOT; be included in
1429   a request if the specification of the request method (&method;)
1430   does not allow sending an entity-body in requests. A server &SHOULD;
1431   read and forward a message-body on any request; if the request method
1432   does not include defined semantics for an entity-body, then the
1433   message-body &SHOULD; be ignored when handling the request.
1436   For response messages, whether or not a message-body is included with
1437   a message is dependent on both the request method and the response
1438   status code (<xref target="status.code.and.reason.phrase"/>). All responses to the HEAD request method
1439   &MUST-NOT; include a message-body, even though the presence of entity-header
1440   fields might lead one to believe they do. All 1xx
1441   (informational), 204 (No Content), and 304 (Not Modified) responses
1442   &MUST-NOT; include a message-body. All other responses do include a
1443   message-body, although it &MAY; be of zero length.
1447<section title="Message Length" anchor="message.length">
1449   The transfer-length of a message is the length of the message-body as
1450   it appears in the message; that is, after any transfer-codings have
1451   been applied. When a message-body is included with a message, the
1452   transfer-length of that body is determined by one of the following
1453   (in order of precedence):
1456  <list style="numbers">
1457    <x:lt><t>
1458     Any response message which "&MUST-NOT;" include a message-body (such
1459     as the 1xx, 204, and 304 responses and any response to a HEAD
1460     request) is always terminated by the first empty line after the
1461     header fields, regardless of the entity-header fields present in
1462     the message.
1463    </t></x:lt>
1464    <x:lt><t>
1465     If a Transfer-Encoding header field (<xref target="header.transfer-encoding"/>)
1466     is present, then the transfer-length is
1467     defined by use of the "chunked" transfer-coding (<xref target="transfer.codings"/>),
1468     unless the message is terminated by closing the connection.
1469    </t></x:lt>
1470    <x:lt><t>
1471     If a Content-Length header field (<xref target="header.content-length"/>) is present, its
1472     decimal value in OCTETs represents both the entity-length and the
1473     transfer-length. The Content-Length header field &MUST-NOT; be sent
1474     if these two lengths are different (i.e., if a Transfer-Encoding
1475     header field is present). If a message is received with both a
1476     Transfer-Encoding header field and a Content-Length header field,
1477     the latter &MUST; be ignored.
1478    </t></x:lt>
1479    <x:lt><t>
1480     If the message uses the media type "multipart/byteranges", and the
1481     transfer-length is not otherwise specified, then this self-delimiting
1482     media type defines the transfer-length. This media type
1483     &MUST-NOT; be used unless the sender knows that the recipient can parse
1484     it; the presence in a request of a Range header with multiple byte-range
1485     specifiers from a 1.1 client implies that the client can parse
1486     multipart/byteranges responses.
1487    <list style="empty"><t>
1488       A range header might be forwarded by a 1.0 proxy that does not
1489       understand multipart/byteranges; in this case the server &MUST;
1490       delimit the message using methods defined in items 1, 3 or 5 of
1491       this section.
1492    </t></list>
1493    </t></x:lt>
1494    <x:lt><t>
1495     By the server closing the connection. (Closing the connection
1496     cannot be used to indicate the end of a request body, since that
1497     would leave no possibility for the server to send back a response.)
1498    </t></x:lt>
1499  </list>
1502   For compatibility with HTTP/1.0 applications, HTTP/1.1 requests
1503   containing a message-body &MUST; include a valid Content-Length header
1504   field unless the server is known to be HTTP/1.1 compliant. If a
1505   request contains a message-body and a Content-Length is not given,
1506   the server &SHOULD; respond with 400 (Bad Request) if it cannot
1507   determine the length of the message, or with 411 (Length Required) if
1508   it wishes to insist on receiving a valid Content-Length.
1511   All HTTP/1.1 applications that receive entities &MUST; accept the
1512   "chunked" transfer-coding (<xref target="transfer.codings"/>), thus allowing this mechanism
1513   to be used for messages when the message length cannot be determined
1514   in advance.
1517   Messages &MUST-NOT; include both a Content-Length header field and a
1518   transfer-coding. If the message does include a
1519   transfer-coding, the Content-Length &MUST; be ignored.
1522   When a Content-Length is given in a message where a message-body is
1523   allowed, its field value &MUST; exactly match the number of OCTETs in
1524   the message-body. HTTP/1.1 user agents &MUST; notify the user when an
1525   invalid length is received and detected.
1529<section title="General Header Fields" anchor="general.header.fields">
1531   There are a few header fields which have general applicability for
1532   both request and response messages, but which do not apply to the
1533   entity being transferred. These header fields apply only to the
1534   message being transmitted.
1536<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="general-header"/>
1537  general-header = Cache-Control            ; &header-cache-control;
1538                 | Connection               ; <xref target="header.connection"/>
1539                 | Date                     ; <xref target=""/>
1540                 | Pragma                   ; &header-pragma;
1541                 | Trailer                  ; <xref target="header.trailer"/>
1542                 | Transfer-Encoding        ; <xref target="header.transfer-encoding"/>
1543                 | Upgrade                  ; <xref target="header.upgrade"/>
1544                 | Via                      ; <xref target="header.via"/>
1545                 | Warning                  ; &header-warning;
1548   General-header field names can be extended reliably only in
1549   combination with a change in the protocol version. However, new or
1550   experimental header fields may be given the semantics of general
1551   header fields if all parties in the communication recognize them to
1552   be general-header fields. Unrecognized header fields are treated as
1553   entity-header fields.
1558<section title="Request" anchor="request">
1560   A request message from a client to a server includes, within the
1561   first line of that message, the method to be applied to the resource,
1562   the identifier of the resource, and the protocol version in use.
1564<!--                 Host                      ; should be moved here eventually -->
1565<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request"/>
1566  Request       = Request-Line              ; <xref target="request-line"/>
1567                  *(( general-header        ; <xref target="general.header.fields"/>
1568                   | request-header         ; &request-header-fields;
1569                   | entity-header ) CRLF)  ; &entity-header-fields;
1570                  CRLF
1571                  [ message-body ]          ; <xref target="message.body"/>
1574<section title="Request-Line" anchor="request-line">
1576   The Request-Line begins with a method token, followed by the
1577   Request-URI and the protocol version, and ending with CRLF. The
1578   elements are separated by SP characters. No CR or LF is allowed
1579   except in the final CRLF sequence.
1581<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request-Line"/>
1582  Request-Line   = Method SP Request-URI SP HTTP-Version CRLF
1585<section title="Method" anchor="method">
1587   The Method  token indicates the method to be performed on the
1588   resource identified by the Request-URI. The method is case-sensitive.
1590<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Method"/><iref primary="true" item="Grammar" subitem="extension-method"/>
1591  Method         = token
1595<section title="Request-URI" anchor="request-uri">
1597   The Request-URI is a Uniform Resource Identifier (<xref target="uri"/>) and
1598   identifies the resource upon which to apply the request.
1600<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request-URI"/>
1601  Request-URI    = "*"
1602                 | absoluteURI
1603                 | ( abs_path [ "?" query ] )
1604                 | authority
1607   The four options for Request-URI are dependent on the nature of the
1608   request. The asterisk "*" means that the request does not apply to a
1609   particular resource, but to the server itself, and is only allowed
1610   when the method used does not necessarily apply to a resource. One
1611   example would be
1613<figure><artwork type="example">
1614    OPTIONS * HTTP/1.1
1617   The absoluteURI form is &REQUIRED; when the request is being made to a
1618   proxy. The proxy is requested to forward the request or service it
1619   from a valid cache, and return the response. Note that the proxy &MAY;
1620   forward the request on to another proxy or directly to the server
1621   specified by the absoluteURI. In order to avoid request loops, a
1622   proxy &MUST; be able to recognize all of its server names, including
1623   any aliases, local variations, and the numeric IP address. An example
1624   Request-Line would be:
1626<figure><artwork type="example">
1627    GET HTTP/1.1
1630   To allow for transition to absoluteURIs in all requests in future
1631   versions of HTTP, all HTTP/1.1 servers &MUST; accept the absoluteURI
1632   form in requests, even though HTTP/1.1 clients will only generate
1633   them in requests to proxies.
1636   The authority form is only used by the CONNECT method (&CONNECT;).
1639   The most common form of Request-URI is that used to identify a
1640   resource on an origin server or gateway. In this case the absolute
1641   path of the URI &MUST; be transmitted (see <xref target="general.syntax"/>, abs_path) as
1642   the Request-URI, and the network location of the URI (authority) &MUST;
1643   be transmitted in a Host header field. For example, a client wishing
1644   to retrieve the resource above directly from the origin server would
1645   create a TCP connection to port 80 of the host "" and send
1646   the lines:
1648<figure><artwork type="example">
1649    GET /pub/WWW/TheProject.html HTTP/1.1
1650    Host:
1653   followed by the remainder of the Request. Note that the absolute path
1654   cannot be empty; if none is present in the original URI, it &MUST; be
1655   given as "/" (the server root).
1658   The Request-URI is transmitted in the format specified in
1659   <xref target="general.syntax"/>. If the Request-URI is encoded using the "% HEX HEX" encoding
1660   <xref target="RFC2396"/>, the origin server &MUST; decode the Request-URI in order to
1661   properly interpret the request. Servers &SHOULD; respond to invalid
1662   Request-URIs with an appropriate status code.
1665   A transparent proxy &MUST-NOT; rewrite the "abs_path" part of the
1666   received Request-URI when forwarding it to the next inbound server,
1667   except as noted above to replace a null abs_path with "/".
1670  <list><t>
1671      <x:h>Note:</x:h> The "no rewrite" rule prevents the proxy from changing the
1672      meaning of the request when the origin server is improperly using
1673      a non-reserved URI character for a reserved purpose.  Implementors
1674      should be aware that some pre-HTTP/1.1 proxies have been known to
1675      rewrite the Request-URI.
1676  </t></list>
1681<section title="The Resource Identified by a Request" anchor="">
1683   The exact resource identified by an Internet request is determined by
1684   examining both the Request-URI and the Host header field.
1687   An origin server that does not allow resources to differ by the
1688   requested host &MAY; ignore the Host header field value when
1689   determining the resource identified by an HTTP/1.1 request. (But see
1690   <xref target=""/>
1691   for other requirements on Host support in HTTP/1.1.)
1694   An origin server that does differentiate resources based on the host
1695   requested (sometimes referred to as virtual hosts or vanity host
1696   names) &MUST; use the following rules for determining the requested
1697   resource on an HTTP/1.1 request:
1698  <list style="numbers">
1699    <t>If Request-URI is an absoluteURI, the host is part of the
1700     Request-URI. Any Host header field value in the request &MUST; be
1701     ignored.</t>
1702    <t>If the Request-URI is not an absoluteURI, and the request includes
1703     a Host header field, the host is determined by the Host header
1704     field value.</t>
1705    <t>If the host as determined by rule 1 or 2 is not a valid host on
1706     the server, the response &MUST; be a 400 (Bad Request) error message.</t>
1707  </list>
1710   Recipients of an HTTP/1.0 request that lacks a Host header field &MAY;
1711   attempt to use heuristics (e.g., examination of the URI path for
1712   something unique to a particular host) in order to determine what
1713   exact resource is being requested.
1720<section title="Response" anchor="response">
1722   After receiving and interpreting a request message, a server responds
1723   with an HTTP response message.
1725<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Response"/>
1726  Response      = Status-Line               ; <xref target="status-line"/>
1727                  *(( general-header        ; <xref target="general.header.fields"/>
1728                   | response-header        ; &response-header-fields;
1729                   | entity-header ) CRLF)  ; &entity-header-fields;
1730                  CRLF
1731                  [ message-body ]          ; <xref target="message.body"/>
1734<section title="Status-Line" anchor="status-line">
1736   The first line of a Response message is the Status-Line, consisting
1737   of the protocol version followed by a numeric status code and its
1738   associated textual phrase, with each element separated by SP
1739   characters. No CR or LF is allowed except in the final CRLF sequence.
1741<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Status-Line"/>
1742  Status-Line = HTTP-Version SP Status-Code SP Reason-Phrase CRLF
1745<section title="Status Code and Reason Phrase" anchor="status.code.and.reason.phrase">
1747   The Status-Code element is a 3-digit integer result code of the
1748   attempt to understand and satisfy the request. These codes are fully
1749   defined in &status-codes;. The Reason-Phrase is intended to give a short
1750   textual description of the Status-Code. The Status-Code is intended
1751   for use by automata and the Reason-Phrase is intended for the human
1752   user. The client is not required to examine or display the Reason-Phrase.
1755   The first digit of the Status-Code defines the class of response. The
1756   last two digits do not have any categorization role. There are 5
1757   values for the first digit:
1758  <list style="symbols">
1759    <t>
1760      1xx: Informational - Request received, continuing process
1761    </t>
1762    <t>
1763      2xx: Success - The action was successfully received,
1764        understood, and accepted
1765    </t>
1766    <t>
1767      3xx: Redirection - Further action must be taken in order to
1768        complete the request
1769    </t>
1770    <t>
1771      4xx: Client Error - The request contains bad syntax or cannot
1772        be fulfilled
1773    </t>
1774    <t>
1775      5xx: Server Error - The server failed to fulfill an apparently
1776        valid request
1777    </t>
1778  </list>
1780<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Status-Code"/><iref primary="true" item="Grammar" subitem="extension-code"/><iref primary="true" item="Grammar" subitem="Reason-Phrase"/>
1781  Status-Code    = 3DIGIT
1782  Reason-Phrase  = *&lt;TEXT, excluding CR, LF&gt;
1790<section title="Connections" anchor="connections">
1792<section title="Persistent Connections" anchor="persistent.connections">
1794<section title="Purpose" anchor="persistent.purpose">
1796   Prior to persistent connections, a separate TCP connection was
1797   established to fetch each URL, increasing the load on HTTP servers
1798   and causing congestion on the Internet. The use of inline images and
1799   other associated data often require a client to make multiple
1800   requests of the same server in a short amount of time. Analysis of
1801   these performance problems and results from a prototype
1802   implementation are available <xref target="Pad1995"/> <xref target="Spe"/>. Implementation experience and
1803   measurements of actual HTTP/1.1 (<xref target="RFC2068" x:fmt="none">RFC 2068</xref>) implementations show good
1804   results <xref target="Nie1997"/>. Alternatives have also been explored, for example,
1805   T/TCP <xref target="Tou1998"/>.
1808   Persistent HTTP connections have a number of advantages:
1809  <list style="symbols">
1810      <t>
1811        By opening and closing fewer TCP connections, CPU time is saved
1812        in routers and hosts (clients, servers, proxies, gateways,
1813        tunnels, or caches), and memory used for TCP protocol control
1814        blocks can be saved in hosts.
1815      </t>
1816      <t>
1817        HTTP requests and responses can be pipelined on a connection.
1818        Pipelining allows a client to make multiple requests without
1819        waiting for each response, allowing a single TCP connection to
1820        be used much more efficiently, with much lower elapsed time.
1821      </t>
1822      <t>
1823        Network congestion is reduced by reducing the number of packets
1824        caused by TCP opens, and by allowing TCP sufficient time to
1825        determine the congestion state of the network.
1826      </t>
1827      <t>
1828        Latency on subsequent requests is reduced since there is no time
1829        spent in TCP's connection opening handshake.
1830      </t>
1831      <t>
1832        HTTP can evolve more gracefully, since errors can be reported
1833        without the penalty of closing the TCP connection. Clients using
1834        future versions of HTTP might optimistically try a new feature,
1835        but if communicating with an older server, retry with old
1836        semantics after an error is reported.
1837      </t>
1838    </list>
1841   HTTP implementations &SHOULD; implement persistent connections.
1845<section title="Overall Operation" anchor="persistent.overall">
1847   A significant difference between HTTP/1.1 and earlier versions of
1848   HTTP is that persistent connections are the default behavior of any
1849   HTTP connection. That is, unless otherwise indicated, the client
1850   &SHOULD; assume that the server will maintain a persistent connection,
1851   even after error responses from the server.
1854   Persistent connections provide a mechanism by which a client and a
1855   server can signal the close of a TCP connection. This signaling takes
1856   place using the Connection header field (<xref target="header.connection"/>). Once a close
1857   has been signaled, the client &MUST-NOT; send any more requests on that
1858   connection.
1861<section title="Negotiation" anchor="persistent.negotiation">
1863   An HTTP/1.1 server &MAY; assume that a HTTP/1.1 client intends to
1864   maintain a persistent connection unless a Connection header including
1865   the connection-token "close" was sent in the request. If the server
1866   chooses to close the connection immediately after sending the
1867   response, it &SHOULD; send a Connection header including the
1868   connection-token close.
1871   An HTTP/1.1 client &MAY; expect a connection to remain open, but would
1872   decide to keep it open based on whether the response from a server
1873   contains a Connection header with the connection-token close. In case
1874   the client does not want to maintain a connection for more than that
1875   request, it &SHOULD; send a Connection header including the
1876   connection-token close.
1879   If either the client or the server sends the close token in the
1880   Connection header, that request becomes the last one for the
1881   connection.
1884   Clients and servers &SHOULD-NOT;  assume that a persistent connection is
1885   maintained for HTTP versions less than 1.1 unless it is explicitly
1886   signaled. See <xref target="compatibility.with.http.1.0.persistent.connections"/> for more information on backward
1887   compatibility with HTTP/1.0 clients.
1890   In order to remain persistent, all messages on the connection &MUST;
1891   have a self-defined message length (i.e., one not defined by closure
1892   of the connection), as described in <xref target="message.length"/>.
1896<section title="Pipelining" anchor="pipelining">
1898   A client that supports persistent connections &MAY; "pipeline" its
1899   requests (i.e., send multiple requests without waiting for each
1900   response). A server &MUST; send its responses to those requests in the
1901   same order that the requests were received.
1904   Clients which assume persistent connections and pipeline immediately
1905   after connection establishment &SHOULD; be prepared to retry their
1906   connection if the first pipelined attempt fails. If a client does
1907   such a retry, it &MUST-NOT; pipeline before it knows the connection is
1908   persistent. Clients &MUST; also be prepared to resend their requests if
1909   the server closes the connection before sending all of the
1910   corresponding responses.
1913   Clients &SHOULD-NOT;  pipeline requests using non-idempotent methods or
1914   non-idempotent sequences of methods (see &idempotent-methods;). Otherwise, a
1915   premature termination of the transport connection could lead to
1916   indeterminate results. A client wishing to send a non-idempotent
1917   request &SHOULD; wait to send that request until it has received the
1918   response status for the previous request.
1923<section title="Proxy Servers" anchor="persistent.proxy">
1925   It is especially important that proxies correctly implement the
1926   properties of the Connection header field as specified in <xref target="header.connection"/>.
1929   The proxy server &MUST; signal persistent connections separately with
1930   its clients and the origin servers (or other proxy servers) that it
1931   connects to. Each persistent connection applies to only one transport
1932   link.
1935   A proxy server &MUST-NOT; establish a HTTP/1.1 persistent connection
1936   with an HTTP/1.0 client (but see <xref target="RFC2068"/> for information and
1937   discussion of the problems with the Keep-Alive header implemented by
1938   many HTTP/1.0 clients).
1942<section title="Practical Considerations" anchor="persistent.practical">
1944   Servers will usually have some time-out value beyond which they will
1945   no longer maintain an inactive connection. Proxy servers might make
1946   this a higher value since it is likely that the client will be making
1947   more connections through the same server. The use of persistent
1948   connections places no requirements on the length (or existence) of
1949   this time-out for either the client or the server.
1952   When a client or server wishes to time-out it &SHOULD; issue a graceful
1953   close on the transport connection. Clients and servers &SHOULD; both
1954   constantly watch for the other side of the transport close, and
1955   respond to it as appropriate. If a client or server does not detect
1956   the other side's close promptly it could cause unnecessary resource
1957   drain on the network.
1960   A client, server, or proxy &MAY; close the transport connection at any
1961   time. For example, a client might have started to send a new request
1962   at the same time that the server has decided to close the "idle"
1963   connection. From the server's point of view, the connection is being
1964   closed while it was idle, but from the client's point of view, a
1965   request is in progress.
1968   This means that clients, servers, and proxies &MUST; be able to recover
1969   from asynchronous close events. Client software &SHOULD; reopen the
1970   transport connection and retransmit the aborted sequence of requests
1971   without user interaction so long as the request sequence is
1972   idempotent (see &idempotent-methods;). Non-idempotent methods or sequences
1973   &MUST-NOT; be automatically retried, although user agents &MAY; offer a
1974   human operator the choice of retrying the request(s). Confirmation by
1975   user-agent software with semantic understanding of the application
1976   &MAY; substitute for user confirmation. The automatic retry &SHOULD-NOT;
1977   be repeated if the second sequence of requests fails.
1980   Servers &SHOULD; always respond to at least one request per connection,
1981   if at all possible. Servers &SHOULD-NOT;  close a connection in the
1982   middle of transmitting a response, unless a network or client failure
1983   is suspected.
1986   Clients that use persistent connections &SHOULD; limit the number of
1987   simultaneous connections that they maintain to a given server. A
1988   single-user client &SHOULD-NOT; maintain more than 2 connections with
1989   any server or proxy. A proxy &SHOULD; use up to 2*N connections to
1990   another server or proxy, where N is the number of simultaneously
1991   active users. These guidelines are intended to improve HTTP response
1992   times and avoid congestion.
1997<section title="Message Transmission Requirements" anchor="message.transmission.requirements">
1999<section title="Persistent Connections and Flow Control" anchor="persistent.flow">
2001   HTTP/1.1 servers &SHOULD; maintain persistent connections and use TCP's
2002   flow control mechanisms to resolve temporary overloads, rather than
2003   terminating connections with the expectation that clients will retry.
2004   The latter technique can exacerbate network congestion.
2008<section title="Monitoring Connections for Error Status Messages" anchor="persistent.monitor">
2010   An HTTP/1.1 (or later) client sending a message-body &SHOULD; monitor
2011   the network connection for an error status while it is transmitting
2012   the request. If the client sees an error status, it &SHOULD;
2013   immediately cease transmitting the body. If the body is being sent
2014   using a "chunked" encoding (<xref target="transfer.codings"/>), a zero length chunk and
2015   empty trailer &MAY; be used to prematurely mark the end of the message.
2016   If the body was preceded by a Content-Length header, the client &MUST;
2017   close the connection.
2021<section title="Use of the 100 (Continue) Status" anchor="use.of.the.100.status">
2023   The purpose of the 100 (Continue) status (see &status-100;) is to
2024   allow a client that is sending a request message with a request body
2025   to determine if the origin server is willing to accept the request
2026   (based on the request headers) before the client sends the request
2027   body. In some cases, it might either be inappropriate or highly
2028   inefficient for the client to send the body if the server will reject
2029   the message without looking at the body.
2032   Requirements for HTTP/1.1 clients:
2033  <list style="symbols">
2034    <t>
2035        If a client will wait for a 100 (Continue) response before
2036        sending the request body, it &MUST; send an Expect request-header
2037        field (&header-expect;) with the "100-continue" expectation.
2038    </t>
2039    <t>
2040        A client &MUST-NOT; send an Expect request-header field (&header-expect;)
2041        with the "100-continue" expectation if it does not intend
2042        to send a request body.
2043    </t>
2044  </list>
2047   Because of the presence of older implementations, the protocol allows
2048   ambiguous situations in which a client may send "Expect: 100-continue"
2049   without receiving either a 417 (Expectation Failed) status
2050   or a 100 (Continue) status. Therefore, when a client sends this
2051   header field to an origin server (possibly via a proxy) from which it
2052   has never seen a 100 (Continue) status, the client &SHOULD-NOT;  wait
2053   for an indefinite period before sending the request body.
2056   Requirements for HTTP/1.1 origin servers:
2057  <list style="symbols">
2058    <t> Upon receiving a request which includes an Expect request-header
2059        field with the "100-continue" expectation, an origin server &MUST;
2060        either respond with 100 (Continue) status and continue to read
2061        from the input stream, or respond with a final status code. The
2062        origin server &MUST-NOT; wait for the request body before sending
2063        the 100 (Continue) response. If it responds with a final status
2064        code, it &MAY; close the transport connection or it &MAY; continue
2065        to read and discard the rest of the request.  It &MUST-NOT;
2066        perform the requested method if it returns a final status code.
2067    </t>
2068    <t> An origin server &SHOULD-NOT;  send a 100 (Continue) response if
2069        the request message does not include an Expect request-header
2070        field with the "100-continue" expectation, and &MUST-NOT; send a
2071        100 (Continue) response if such a request comes from an HTTP/1.0
2072        (or earlier) client. There is an exception to this rule: for
2073        compatibility with <xref target="RFC2068"/>, a server &MAY; send a 100 (Continue)
2074        status in response to an HTTP/1.1 PUT or POST request that does
2075        not include an Expect request-header field with the "100-continue"
2076        expectation. This exception, the purpose of which is
2077        to minimize any client processing delays associated with an
2078        undeclared wait for 100 (Continue) status, applies only to
2079        HTTP/1.1 requests, and not to requests with any other HTTP-version
2080        value.
2081    </t>
2082    <t> An origin server &MAY; omit a 100 (Continue) response if it has
2083        already received some or all of the request body for the
2084        corresponding request.
2085    </t>
2086    <t> An origin server that sends a 100 (Continue) response &MUST;
2087    ultimately send a final status code, once the request body is
2088        received and processed, unless it terminates the transport
2089        connection prematurely.
2090    </t>
2091    <t> If an origin server receives a request that does not include an
2092        Expect request-header field with the "100-continue" expectation,
2093        the request includes a request body, and the server responds
2094        with a final status code before reading the entire request body
2095        from the transport connection, then the server &SHOULD-NOT;  close
2096        the transport connection until it has read the entire request,
2097        or until the client closes the connection. Otherwise, the client
2098        might not reliably receive the response message. However, this
2099        requirement is not be construed as preventing a server from
2100        defending itself against denial-of-service attacks, or from
2101        badly broken client implementations.
2102      </t>
2103    </list>
2106   Requirements for HTTP/1.1 proxies:
2107  <list style="symbols">
2108    <t> If a proxy receives a request that includes an Expect request-header
2109        field with the "100-continue" expectation, and the proxy
2110        either knows that the next-hop server complies with HTTP/1.1 or
2111        higher, or does not know the HTTP version of the next-hop
2112        server, it &MUST; forward the request, including the Expect header
2113        field.
2114    </t>
2115    <t> If the proxy knows that the version of the next-hop server is
2116        HTTP/1.0 or lower, it &MUST-NOT; forward the request, and it &MUST;
2117        respond with a 417 (Expectation Failed) status.
2118    </t>
2119    <t> Proxies &SHOULD; maintain a cache recording the HTTP version
2120        numbers received from recently-referenced next-hop servers.
2121    </t>
2122    <t> A proxy &MUST-NOT; forward a 100 (Continue) response if the
2123        request message was received from an HTTP/1.0 (or earlier)
2124        client and did not include an Expect request-header field with
2125        the "100-continue" expectation. This requirement overrides the
2126        general rule for forwarding of 1xx responses (see &status-1xx;).
2127    </t>
2128  </list>
2132<section title="Client Behavior if Server Prematurely Closes Connection" anchor="connection.premature">
2134   If an HTTP/1.1 client sends a request which includes a request body,
2135   but which does not include an Expect request-header field with the
2136   "100-continue" expectation, and if the client is not directly
2137   connected to an HTTP/1.1 origin server, and if the client sees the
2138   connection close before receiving any status from the server, the
2139   client &SHOULD; retry the request.  If the client does retry this
2140   request, it &MAY; use the following "binary exponential backoff"
2141   algorithm to be assured of obtaining a reliable response:
2142  <list style="numbers">
2143    <t>
2144      Initiate a new connection to the server
2145    </t>
2146    <t>
2147      Transmit the request-headers
2148    </t>
2149    <t>
2150      Initialize a variable R to the estimated round-trip time to the
2151         server (e.g., based on the time it took to establish the
2152         connection), or to a constant value of 5 seconds if the round-trip
2153         time is not available.
2154    </t>
2155    <t>
2156       Compute T = R * (2**N), where N is the number of previous
2157         retries of this request.
2158    </t>
2159    <t>
2160       Wait either for an error response from the server, or for T
2161         seconds (whichever comes first)
2162    </t>
2163    <t>
2164       If no error response is received, after T seconds transmit the
2165         body of the request.
2166    </t>
2167    <t>
2168       If client sees that the connection is closed prematurely,
2169         repeat from step 1 until the request is accepted, an error
2170         response is received, or the user becomes impatient and
2171         terminates the retry process.
2172    </t>
2173  </list>
2176   If at any point an error status is received, the client
2177  <list style="symbols">
2178      <t>&SHOULD-NOT;  continue and</t>
2180      <t>&SHOULD; close the connection if it has not completed sending the
2181        request message.</t>
2182    </list>
2189<section title="Header Field Definitions" anchor="header.fields">
2191   This section defines the syntax and semantics of HTTP/1.1 header fields
2192   related to message framing and transport protocols.
2195   For entity-header fields, both sender and recipient refer to either the
2196   client or the server, depending on who sends and who receives the entity.
2199<section title="Connection" anchor="header.connection">
2200  <iref primary="true" item="Connection header" x:for-anchor=""/>
2201  <iref primary="true" item="Headers" subitem="Connection" x:for-anchor=""/>
2203   The Connection general-header field allows the sender to specify
2204   options that are desired for that particular connection and &MUST-NOT;
2205   be communicated by proxies over further connections.
2208   The Connection header has the following grammar:
2210<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="connection-token"/>
2211  Connection = "Connection" ":" 1#(connection-token)
2212  connection-token  = token
2215   HTTP/1.1 proxies &MUST; parse the Connection header field before a
2216   message is forwarded and, for each connection-token in this field,
2217   remove any header field(s) from the message with the same name as the
2218   connection-token. Connection options are signaled by the presence of
2219   a connection-token in the Connection header field, not by any
2220   corresponding additional header field(s), since the additional header
2221   field may not be sent if there are no parameters associated with that
2222   connection option.
2225   Message headers listed in the Connection header &MUST-NOT; include
2226   end-to-end headers, such as Cache-Control.
2229   HTTP/1.1 defines the "close" connection option for the sender to
2230   signal that the connection will be closed after completion of the
2231   response. For example,
2233<figure><artwork type="example">
2234    Connection: close
2237   in either the request or the response header fields indicates that
2238   the connection &SHOULD-NOT;  be considered `persistent' (<xref target="persistent.connections"/>)
2239   after the current request/response is complete.
2242   An HTTP/1.1 client that does not support persistent connections &MUST;
2243   include the "close" connection option in every request message.
2246   An HTTP/1.1 server that does not support persistent connections &MUST;
2247   include the "close" connection option in every response message that
2248   does not have a 1xx (informational) status code.
2251   A system receiving an HTTP/1.0 (or lower-version) message that
2252   includes a Connection header &MUST;, for each connection-token in this
2253   field, remove and ignore any header field(s) from the message with
2254   the same name as the connection-token. This protects against mistaken
2255   forwarding of such header fields by pre-HTTP/1.1 proxies. See <xref target="compatibility.with.http.1.0.persistent.connections"/>.
2259<section title="Content-Length" anchor="header.content-length">
2260  <iref primary="true" item="Content-Length header" x:for-anchor=""/>
2261  <iref primary="true" item="Headers" subitem="Content-Length" x:for-anchor=""/>
2263   The Content-Length entity-header field indicates the size of the
2264   entity-body, in decimal number of OCTETs, sent to the recipient or,
2265   in the case of the HEAD method, the size of the entity-body that
2266   would have been sent had the request been a GET.
2268<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Length"/>
2269  Content-Length    = "Content-Length" ":" 1*DIGIT
2272   An example is
2274<figure><artwork type="example">
2275    Content-Length: 3495
2278   Applications &SHOULD; use this field to indicate the transfer-length of
2279   the message-body, unless this is prohibited by the rules in <xref target="message.length"/>.
2282   Any Content-Length greater than or equal to zero is a valid value.
2283   <xref target="message.length"/> describes how to determine the length of a message-body
2284   if a Content-Length is not given.
2287   Note that the meaning of this field is significantly different from
2288   the corresponding definition in MIME, where it is an optional field
2289   used within the "message/external-body" content-type. In HTTP, it
2290   &SHOULD; be sent whenever the message's length can be determined prior
2291   to being transferred, unless this is prohibited by the rules in
2292   <xref target="message.length"/>.
2296<section title="Date" anchor="">
2297  <iref primary="true" item="Date header" x:for-anchor=""/>
2298  <iref primary="true" item="Headers" subitem="Date" x:for-anchor=""/>
2300   The Date general-header field represents the date and time at which
2301   the message was originated, having the same semantics as orig-date in
2302   <xref target="RFC2822" x:fmt="of" x:sec="3.6.1"/>. The field value is an HTTP-date, as described in <xref target=""/>;
2303   it &MUST; be sent in rfc1123-date format.
2305<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Date"/>
2306  Date  = "Date" ":" HTTP-date
2309   An example is
2311<figure><artwork type="example">
2312    Date: Tue, 15 Nov 1994 08:12:31 GMT
2315   Origin servers &MUST; include a Date header field in all responses,
2316   except in these cases:
2317  <list style="numbers">
2318      <t>If the response status code is 100 (Continue) or 101 (Switching
2319         Protocols), the response &MAY; include a Date header field, at
2320         the server's option.</t>
2322      <t>If the response status code conveys a server error, e.g. 500
2323         (Internal Server Error) or 503 (Service Unavailable), and it is
2324         inconvenient or impossible to generate a valid Date.</t>
2326      <t>If the server does not have a clock that can provide a
2327         reasonable approximation of the current time, its responses
2328         &MUST-NOT; include a Date header field. In this case, the rules
2329         in <xref target="clockless.origin.server.operation"/> &MUST; be followed.</t>
2330  </list>
2333   A received message that does not have a Date header field &MUST; be
2334   assigned one by the recipient if the message will be cached by that
2335   recipient or gatewayed via a protocol which requires a Date. An HTTP
2336   implementation without a clock &MUST-NOT; cache responses without
2337   revalidating them on every use. An HTTP cache, especially a shared
2338   cache, &SHOULD; use a mechanism, such as NTP <xref target="RFC1305"/>, to synchronize its
2339   clock with a reliable external standard.
2342   Clients &SHOULD; only send a Date header field in messages that include
2343   an entity-body, as in the case of the PUT and POST requests, and even
2344   then it is optional. A client without a clock &MUST-NOT; send a Date
2345   header field in a request.
2348   The HTTP-date sent in a Date header &SHOULD-NOT;  represent a date and
2349   time subsequent to the generation of the message. It &SHOULD; represent
2350   the best available approximation of the date and time of message
2351   generation, unless the implementation has no means of generating a
2352   reasonably accurate date and time. In theory, the date ought to
2353   represent the moment just before the entity is generated. In
2354   practice, the date can be generated at any time during the message
2355   origination without affecting its semantic value.
2358<section title="Clockless Origin Server Operation" anchor="clockless.origin.server.operation">
2360   Some origin server implementations might not have a clock available.
2361   An origin server without a clock &MUST-NOT; assign Expires or Last-Modified
2362   values to a response, unless these values were associated
2363   with the resource by a system or user with a reliable clock. It &MAY;
2364   assign an Expires value that is known, at or before server
2365   configuration time, to be in the past (this allows "pre-expiration"
2366   of responses without storing separate Expires values for each
2367   resource).
2372<section title="Host" anchor="">
2373  <iref primary="true" item="Host header" x:for-anchor=""/>
2374  <iref primary="true" item="Headers" subitem="Host" x:for-anchor=""/>
2376   The Host request-header field specifies the Internet host and port
2377   number of the resource being requested, as obtained from the original
2378   URI given by the user or referring resource (generally an HTTP URL,
2379   as described in <xref target="http.url"/>). The Host field value &MUST; represent
2380   the naming authority of the origin server or gateway given by the
2381   original URL. This allows the origin server or gateway to
2382   differentiate between internally-ambiguous URLs, such as the root "/"
2383   URL of a server for multiple host names on a single IP address.
2385<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Host"/>
2386  Host = "Host" ":" host [ ":" port ] ; <xref target="http.url"/>
2389   A "host" without any trailing port information implies the default
2390   port for the service requested (e.g., "80" for an HTTP URL). For
2391   example, a request on the origin server for
2392   &lt;; would properly include:
2394<figure><artwork type="example">
2395    GET /pub/WWW/ HTTP/1.1
2396    Host:
2399   A client &MUST; include a Host header field in all HTTP/1.1 request
2400   messages. If the requested URI does not include an Internet host
2401   name for the service being requested, then the Host header field &MUST;
2402   be given with an empty value. An HTTP/1.1 proxy &MUST; ensure that any
2403   request message it forwards does contain an appropriate Host header
2404   field that identifies the service being requested by the proxy. All
2405   Internet-based HTTP/1.1 servers &MUST; respond with a 400 (Bad Request)
2406   status code to any HTTP/1.1 request message which lacks a Host header
2407   field.
2410   See Sections <xref target="" format="counter"/>
2411   and <xref target="" format="counter"/>
2412   for other requirements relating to Host.
2416<section title="TE" anchor="header.te">
2417  <iref primary="true" item="TE header" x:for-anchor=""/>
2418  <iref primary="true" item="Headers" subitem="TE" x:for-anchor=""/>
2420   The TE request-header field indicates what extension transfer-codings
2421   it is willing to accept in the response and whether or not it is
2422   willing to accept trailer fields in a chunked transfer-coding. Its
2423   value may consist of the keyword "trailers" and/or a comma-separated
2424   list of extension transfer-coding names with optional accept
2425   parameters (as described in <xref target="transfer.codings"/>).
2427<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="t-codings"/>
2428  TE        = "TE" ":" #( t-codings )
2429  t-codings = "trailers" | ( transfer-extension [ accept-params ] )
2432   The presence of the keyword "trailers" indicates that the client is
2433   willing to accept trailer fields in a chunked transfer-coding, as
2434   defined in <xref target="chunked.transfer.encoding"/>. This keyword is reserved for use with
2435   transfer-coding values even though it does not itself represent a
2436   transfer-coding.
2439   Examples of its use are:
2441<figure><artwork type="example">
2442    TE: deflate
2443    TE:
2444    TE: trailers, deflate;q=0.5
2447   The TE header field only applies to the immediate connection.
2448   Therefore, the keyword &MUST; be supplied within a Connection header
2449   field (<xref target="header.connection"/>) whenever TE is present in an HTTP/1.1 message.
2452   A server tests whether a transfer-coding is acceptable, according to
2453   a TE field, using these rules:
2454  <list style="numbers">
2455    <x:lt>
2456      <t>The "chunked" transfer-coding is always acceptable. If the
2457         keyword "trailers" is listed, the client indicates that it is
2458         willing to accept trailer fields in the chunked response on
2459         behalf of itself and any downstream clients. The implication is
2460         that, if given, the client is stating that either all
2461         downstream clients are willing to accept trailer fields in the
2462         forwarded response, or that it will attempt to buffer the
2463         response on behalf of downstream recipients.
2464      </t><t>
2465         <x:h>Note:</x:h> HTTP/1.1 does not define any means to limit the size of a
2466         chunked response such that a client can be assured of buffering
2467         the entire response.</t>
2468    </x:lt>
2469    <x:lt>
2470      <t>If the transfer-coding being tested is one of the transfer-codings
2471         listed in the TE field, then it is acceptable unless it
2472         is accompanied by a qvalue of 0. (As defined in &qvalue;, a
2473         qvalue of 0 means "not acceptable.")</t>
2474    </x:lt>
2475    <x:lt>
2476      <t>If multiple transfer-codings are acceptable, then the
2477         acceptable transfer-coding with the highest non-zero qvalue is
2478         preferred.  The "chunked" transfer-coding always has a qvalue
2479         of 1.</t>
2480    </x:lt>
2481  </list>
2484   If the TE field-value is empty or if no TE field is present, the only
2485   transfer-coding  is "chunked". A message with no transfer-coding is
2486   always acceptable.
2490<section title="Trailer" anchor="header.trailer">
2491  <iref primary="true" item="Trailer header" x:for-anchor=""/>
2492  <iref primary="true" item="Headers" subitem="Trailer" x:for-anchor=""/>
2494   The Trailer general field value indicates that the given set of
2495   header fields is present in the trailer of a message encoded with
2496   chunked transfer-coding.
2498<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Trailer"/>
2499  Trailer  = "Trailer" ":" 1#field-name
2502   An HTTP/1.1 message &SHOULD; include a Trailer header field in a
2503   message using chunked transfer-coding with a non-empty trailer. Doing
2504   so allows the recipient to know which header fields to expect in the
2505   trailer.
2508   If no Trailer header field is present, the trailer &SHOULD-NOT;  include
2509   any header fields. See <xref target="chunked.transfer.encoding"/> for restrictions on the use of
2510   trailer fields in a "chunked" transfer-coding.
2513   Message header fields listed in the Trailer header field &MUST-NOT;
2514   include the following header fields:
2515  <list style="symbols">
2516    <t>Transfer-Encoding</t>
2517    <t>Content-Length</t>
2518    <t>Trailer</t>
2519  </list>
2523<section title="Transfer-Encoding" anchor="header.transfer-encoding">
2524  <iref primary="true" item="Transfer-Encoding header" x:for-anchor=""/>
2525  <iref primary="true" item="Headers" subitem="Transfer-Encoding" x:for-anchor=""/>
2527   The Transfer-Encoding general-header field indicates what (if any)
2528   type of transformation has been applied to the message body in order
2529   to safely transfer it between the sender and the recipient. This
2530   differs from the content-coding in that the transfer-coding is a
2531   property of the message, not of the entity.
2533<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/>
2534  Transfer-Encoding       = "Transfer-Encoding" ":" 1#transfer-coding
2537   Transfer-codings are defined in <xref target="transfer.codings"/>. An example is:
2539<figure><artwork type="example">
2540  Transfer-Encoding: chunked
2543   If multiple encodings have been applied to an entity, the transfer-codings
2544   &MUST; be listed in the order in which they were applied.
2545   Additional information about the encoding parameters &MAY; be provided
2546   by other entity-header fields not defined by this specification.
2549   Many older HTTP/1.0 applications do not understand the Transfer-Encoding
2550   header.
2554<section title="Upgrade" anchor="header.upgrade">
2555  <iref primary="true" item="Upgrade header" x:for-anchor=""/>
2556  <iref primary="true" item="Headers" subitem="Upgrade" x:for-anchor=""/>
2558   The Upgrade general-header allows the client to specify what
2559   additional communication protocols it supports and would like to use
2560   if the server finds it appropriate to switch protocols. The server
2561   &MUST; use the Upgrade header field within a 101 (Switching Protocols)
2562   response to indicate which protocol(s) are being switched.
2564<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Upgrade"/>
2565  Upgrade        = "Upgrade" ":" 1#product
2568   For example,
2570<figure><artwork type="example">
2571    Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
2574   The Upgrade header field is intended to provide a simple mechanism
2575   for transition from HTTP/1.1 to some other, incompatible protocol. It
2576   does so by allowing the client to advertise its desire to use another
2577   protocol, such as a later version of HTTP with a higher major version
2578   number, even though the current request has been made using HTTP/1.1.
2579   This eases the difficult transition between incompatible protocols by
2580   allowing the client to initiate a request in the more commonly
2581   supported protocol while indicating to the server that it would like
2582   to use a "better" protocol if available (where "better" is determined
2583   by the server, possibly according to the nature of the method and/or
2584   resource being requested).
2587   The Upgrade header field only applies to switching application-layer
2588   protocols upon the existing transport-layer connection. Upgrade
2589   cannot be used to insist on a protocol change; its acceptance and use
2590   by the server is optional. The capabilities and nature of the
2591   application-layer communication after the protocol change is entirely
2592   dependent upon the new protocol chosen, although the first action
2593   after changing the protocol &MUST; be a response to the initial HTTP
2594   request containing the Upgrade header field.
2597   The Upgrade header field only applies to the immediate connection.
2598   Therefore, the upgrade keyword &MUST; be supplied within a Connection
2599   header field (<xref target="header.connection"/>) whenever Upgrade is present in an
2600   HTTP/1.1 message.
2603   The Upgrade header field cannot be used to indicate a switch to a
2604   protocol on a different connection. For that purpose, it is more
2605   appropriate to use a 301, 302, 303, or 305 redirection response.
2608   This specification only defines the protocol name "HTTP" for use by
2609   the family of Hypertext Transfer Protocols, as defined by the HTTP
2610   version rules of <xref target="http.version"/> and future updates to this
2611   specification. Any token can be used as a protocol name; however, it
2612   will only be useful if both the client and server associate the name
2613   with the same protocol.
2617<section title="Via" anchor="header.via">
2618  <iref primary="true" item="Via header" x:for-anchor=""/>
2619  <iref primary="true" item="Headers" subitem="Via" x:for-anchor=""/>
2621   The Via general-header field &MUST; be used by gateways and proxies to
2622   indicate the intermediate protocols and recipients between the user
2623   agent and the server on requests, and between the origin server and
2624   the client on responses. It is analogous to the "Received" field of
2625   <xref target="RFC2822"/> and is intended to be used for tracking message forwards,
2626   avoiding request loops, and identifying the protocol capabilities of
2627   all senders along the request/response chain.
2629<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"/>
2630  Via =  "Via" ":" 1#( received-protocol received-by [ comment ] )
2631  received-protocol = [ protocol-name "/" ] protocol-version
2632  protocol-name     = token
2633  protocol-version  = token
2634  received-by       = ( host [ ":" port ] ) | pseudonym
2635  pseudonym         = token
2638   The received-protocol indicates the protocol version of the message
2639   received by the server or client along each segment of the
2640   request/response chain. The received-protocol version is appended to
2641   the Via field value when the message is forwarded so that information
2642   about the protocol capabilities of upstream applications remains
2643   visible to all recipients.
2646   The protocol-name is optional if and only if it would be "HTTP". The
2647   received-by field is normally the host and optional port number of a
2648   recipient server or client that subsequently forwarded the message.
2649   However, if the real host is considered to be sensitive information,
2650   it &MAY; be replaced by a pseudonym. If the port is not given, it &MAY;
2651   be assumed to be the default port of the received-protocol.
2654   Multiple Via field values represents each proxy or gateway that has
2655   forwarded the message. Each recipient &MUST; append its information
2656   such that the end result is ordered according to the sequence of
2657   forwarding applications.
2660   Comments &MAY; be used in the Via header field to identify the software
2661   of the recipient proxy or gateway, analogous to the User-Agent and
2662   Server header fields. However, all comments in the Via field are
2663   optional and &MAY; be removed by any recipient prior to forwarding the
2664   message.
2667   For example, a request message could be sent from an HTTP/1.0 user
2668   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
2669   forward the request to a public proxy at, which completes
2670   the request by forwarding it to the origin server at
2671   The request received by would then have the following
2672   Via header field:
2674<figure><artwork type="example">
2675    Via: 1.0 fred, 1.1 (Apache/1.1)
2678   Proxies and gateways used as a portal through a network firewall
2679   &SHOULD-NOT;, by default, forward the names and ports of hosts within
2680   the firewall region. This information &SHOULD; only be propagated if
2681   explicitly enabled. If not enabled, the received-by host of any host
2682   behind the firewall &SHOULD; be replaced by an appropriate pseudonym
2683   for that host.
2686   For organizations that have strong privacy requirements for hiding
2687   internal structures, a proxy &MAY; combine an ordered subsequence of
2688   Via header field entries with identical received-protocol values into
2689   a single such entry. For example,
2691<figure><artwork type="example">
2692    Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
2695        could be collapsed to
2697<figure><artwork type="example">
2698    Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
2701   Applications &SHOULD-NOT;  combine multiple entries unless they are all
2702   under the same organizational control and the hosts have already been
2703   replaced by pseudonyms. Applications &MUST-NOT; combine entries which
2704   have different received-protocol values.
2710<section title="IANA Considerations" anchor="IANA.considerations">
2712   TBD.
2716<section title="Security Considerations" anchor="security.considerations">
2718   This section is meant to inform application developers, information
2719   providers, and users of the security limitations in HTTP/1.1 as
2720   described by this document. The discussion does not include
2721   definitive solutions to the problems revealed, though it does make
2722   some suggestions for reducing security risks.
2725<section title="Personal Information" anchor="personal.information">
2727   HTTP clients are often privy to large amounts of personal information
2728   (e.g. the user's name, location, mail address, passwords, encryption
2729   keys, etc.), and &SHOULD; be very careful to prevent unintentional
2730   leakage of this information via the HTTP protocol to other sources.
2731   We very strongly recommend that a convenient interface be provided
2732   for the user to control dissemination of such information, and that
2733   designers and implementors be particularly careful in this area.
2734   History shows that errors in this area often create serious security
2735   and/or privacy problems and generate highly adverse publicity for the
2736   implementor's company.
2740<section title="Abuse of Server Log Information" anchor="abuse.of.server.log.information">
2742   A server is in the position to save personal data about a user's
2743   requests which might identify their reading patterns or subjects of
2744   interest. This information is clearly confidential in nature and its
2745   handling can be constrained by law in certain countries. People using
2746   the HTTP protocol to provide data are responsible for ensuring that
2747   such material is not distributed without the permission of any
2748   individuals that are identifiable by the published results.
2752<section title="Attacks Based On File and Path Names" anchor="attack.pathname">
2754   Implementations of HTTP origin servers &SHOULD; be careful to restrict
2755   the documents returned by HTTP requests to be only those that were
2756   intended by the server administrators. If an HTTP server translates
2757   HTTP URIs directly into file system calls, the server &MUST; take
2758   special care not to serve files that were not intended to be
2759   delivered to HTTP clients. For example, UNIX, Microsoft Windows, and
2760   other operating systems use ".." as a path component to indicate a
2761   directory level above the current one. On such a system, an HTTP
2762   server &MUST; disallow any such construct in the Request-URI if it
2763   would otherwise allow access to a resource outside those intended to
2764   be accessible via the HTTP server. Similarly, files intended for
2765   reference only internally to the server (such as access control
2766   files, configuration files, and script code) &MUST; be protected from
2767   inappropriate retrieval, since they might contain sensitive
2768   information. Experience has shown that minor bugs in such HTTP server
2769   implementations have turned into security risks.
2773<section title="DNS Spoofing" anchor="dns.spoofing">
2775   Clients using HTTP rely heavily on the Domain Name Service, and are
2776   thus generally prone to security attacks based on the deliberate
2777   mis-association of IP addresses and DNS names. Clients need to be
2778   cautious in assuming the continuing validity of an IP number/DNS name
2779   association.
2782   In particular, HTTP clients &SHOULD; rely on their name resolver for
2783   confirmation of an IP number/DNS name association, rather than
2784   caching the result of previous host name lookups. Many platforms
2785   already can cache host name lookups locally when appropriate, and
2786   they &SHOULD; be configured to do so. It is proper for these lookups to
2787   be cached, however, only when the TTL (Time To Live) information
2788   reported by the name server makes it likely that the cached
2789   information will remain useful.
2792   If HTTP clients cache the results of host name lookups in order to
2793   achieve a performance improvement, they &MUST; observe the TTL
2794   information reported by DNS.
2797   If HTTP clients do not observe this rule, they could be spoofed when
2798   a previously-accessed server's IP address changes. As network
2799   renumbering is expected to become increasingly common <xref target="RFC1900"/>, the
2800   possibility of this form of attack will grow. Observing this
2801   requirement thus reduces this potential security vulnerability.
2804   This requirement also improves the load-balancing behavior of clients
2805   for replicated servers using the same DNS name and reduces the
2806   likelihood of a user's experiencing failure in accessing sites which
2807   use that strategy.
2811<section title="Proxies and Caching" anchor="attack.proxies">
2813   By their very nature, HTTP proxies are men-in-the-middle, and
2814   represent an opportunity for man-in-the-middle attacks. Compromise of
2815   the systems on which the proxies run can result in serious security
2816   and privacy problems. Proxies have access to security-related
2817   information, personal information about individual users and
2818   organizations, and proprietary information belonging to users and
2819   content providers. A compromised proxy, or a proxy implemented or
2820   configured without regard to security and privacy considerations,
2821   might be used in the commission of a wide range of potential attacks.
2824   Proxy operators should protect the systems on which proxies run as
2825   they would protect any system that contains or transports sensitive
2826   information. In particular, log information gathered at proxies often
2827   contains highly sensitive personal information, and/or information
2828   about organizations. Log information should be carefully guarded, and
2829   appropriate guidelines for use developed and followed. (<xref target="abuse.of.server.log.information"/>).
2832   Proxy implementors should consider the privacy and security
2833   implications of their design and coding decisions, and of the
2834   configuration options they provide to proxy operators (especially the
2835   default configuration).
2838   Users of a proxy need to be aware that they are no trustworthier than
2839   the people who run the proxy; HTTP itself cannot solve this problem.
2842   The judicious use of cryptography, when appropriate, may suffice to
2843   protect against a broad range of security and privacy attacks. Such
2844   cryptography is beyond the scope of the HTTP/1.1 specification.
2848<section title="Denial of Service Attacks on Proxies" anchor="attack.DoS">
2850   They exist. They are hard to defend against. Research continues.
2851   Beware.
2856<section title="Acknowledgments" anchor="ack">
2858   This specification makes heavy use of the augmented BNF and generic
2859   constructs defined by David H. Crocker for <xref target="RFC822ABNF"/>. Similarly, it
2860   reuses many of the definitions provided by Nathaniel Borenstein and
2861   Ned Freed for MIME <xref target="RFC2045"/>. We hope that their inclusion in this
2862   specification will help reduce past confusion over the relationship
2863   between HTTP and Internet mail message formats.
2866   The HTTP protocol has evolved considerably over the years. It has
2867   benefited from a large and active developer community--the many
2868   people who have participated on the www-talk mailing list--and it is
2869   that community which has been most responsible for the success of
2870   HTTP and of the World-Wide Web in general. Marc Andreessen, Robert
2871   Cailliau, Daniel W. Connolly, Bob Denny, John Franks, Jean-Francois
2872   Groff, Phillip M. Hallam-Baker, Hakon W. Lie, Ari Luotonen, Rob
2873   McCool, Lou Montulli, Dave Raggett, Tony Sanders, and Marc
2874   VanHeyningen deserve special recognition for their efforts in
2875   defining early aspects of the protocol.
2878   This document has benefited greatly from the comments of all those
2879   participating in the HTTP-WG. In addition to those already mentioned,
2880   the following individuals have contributed to this specification:
2883   Gary Adams, Harald Tveit Alvestrand, Keith Ball, Brian Behlendorf,
2884   Paul Burchard, Maurizio Codogno, Mike Cowlishaw, Roman Czyborra,
2885   Michael A. Dolan, Daniel DuBois, David J. Fiander, Alan Freier, Marc Hedlund, Greg Herlihy,
2886   Koen Holtman, Alex Hopmann, Bob Jernigan, Shel Kaphan, Rohit Khare,
2887   John Klensin, Martijn Koster, Alexei Kosut, David M. Kristol,
2888   Daniel LaLiberte, Ben Laurie, Paul J. Leach, Albert Lunde,
2889   John C. Mallery, Jean-Philippe Martin-Flatin, Mitra, David Morris,
2890   Gavin Nicol, Ross Patterson, Bill Perry, Jeffrey Perry, Scott Powers, Owen Rees,
2891   Luigi Rizzo, David Robinson, Marc Salomon, Rich Salz,
2892   Allan M. Schiffman, Jim Seidman, Chuck Shotton, Eric W. Sink,
2893   Simon E. Spero, Richard N. Taylor, Robert S. Thau,
2894   Bill (BearHeart) Weinman, Francois Yergeau, Mary Ellen Zurko,
2895   Josh Cohen.
2898   Thanks to the "cave men" of Palo Alto. You know who you are.
2901   Jim Gettys (the editor of <xref target="RFC2616"/>) wishes particularly
2902   to thank Roy Fielding, the editor of <xref target="RFC2068"/>, along
2903   with John Klensin, Jeff Mogul, Paul Leach, Dave Kristol, Koen
2904   Holtman, John Franks, Josh Cohen, Alex Hopmann, Scott Lawrence, and
2905   Larry Masinter for their help. And thanks go particularly to Jeff
2906   Mogul and Scott Lawrence for performing the "MUST/MAY/SHOULD" audit.
2909   The Apache Group, Anselm Baird-Smith, author of Jigsaw, and Henrik
2910   Frystyk implemented RFC 2068 early, and we wish to thank them for the
2911   discovery of many of the problems that this document attempts to
2912   rectify.
2919<references title="Normative References">
2921<reference anchor="ISO-8859-1">
2922  <front>
2923    <title>
2924     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
2925    </title>
2926    <author>
2927      <organization>International Organization for Standardization</organization>
2928    </author>
2929    <date year="1998"/>
2930  </front>
2931  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
2934<reference anchor="Part2">
2935  <front>
2936    <title abbrev="HTTP/1.1">HTTP/1.1, part 2: Message Semantics</title>
2937    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
2938      <organization abbrev="Day Software">Day Software</organization>
2939      <address><email></email></address>
2940    </author>
2941    <author initials="J." surname="Gettys" fullname="Jim Gettys">
2942      <organization>One Laptop per Child</organization>
2943      <address><email></email></address>
2944    </author>
2945    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
2946      <organization abbrev="HP">Hewlett-Packard Company</organization>
2947      <address><email></email></address>
2948    </author>
2949    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
2950      <organization abbrev="Microsoft">Microsoft Corporation</organization>
2951      <address><email></email></address>
2952    </author>
2953    <author initials="L." surname="Masinter" fullname="Larry Masinter">
2954      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
2955      <address><email></email></address>
2956    </author>
2957    <author initials="P." surname="Leach" fullname="Paul J. Leach">
2958      <organization abbrev="Microsoft">Microsoft Corporation</organization>
2959      <address><email></email></address>
2960    </author>
2961    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
2962      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
2963      <address><email></email></address>
2964    </author>
2965    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
2966      <organization abbrev="W3C">World Wide Web Consortium</organization>
2967      <address><email></email></address>
2968    </author>
2969    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
2970      <organization abbrev="greenbytes">greenbytes GmbH</organization>
2971      <address><email></email></address>
2972    </author>
2973    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
2974  </front>
2975  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-&ID-VERSION;"/>
2976  <x:source href="p2-semantics.xml" basename="p2-semantics"/>
2979<reference anchor="Part3">
2980  <front>
2981    <title abbrev="HTTP/1.1">HTTP/1.1, part 3: Message Payload and Content Negotiation</title>
2982    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
2983      <organization abbrev="Day Software">Day Software</organization>
2984      <address><email></email></address>
2985    </author>
2986    <author initials="J." surname="Gettys" fullname="Jim Gettys">
2987      <organization>One Laptop per Child</organization>
2988      <address><email></email></address>
2989    </author>
2990    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
2991      <organization abbrev="HP">Hewlett-Packard Company</organization>
2992      <address><email></email></address>
2993    </author>
2994    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
2995      <organization abbrev="Microsoft">Microsoft Corporation</organization>
2996      <address><email></email></address>
2997    </author>
2998    <author initials="L." surname="Masinter" fullname="Larry Masinter">
2999      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3000      <address><email></email></address>
3001    </author>
3002    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3003      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3004      <address><email></email></address>
3005    </author>
3006    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3007      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3008      <address><email></email></address>
3009    </author>
3010    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
3011      <organization abbrev="W3C">World Wide Web Consortium</organization>
3012      <address><email></email></address>
3013    </author>
3014    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3015      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3016      <address><email></email></address>
3017    </author>
3018    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3019  </front>
3020  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p3-payload-&ID-VERSION;"/>
3021  <x:source href="p3-payload.xml" basename="p3-payload"/>
3024<reference anchor="Part5">
3025  <front>
3026    <title abbrev="HTTP/1.1">HTTP/1.1, part 5: Range Requests and Partial Responses</title>
3027    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3028      <organization abbrev="Day Software">Day Software</organization>
3029      <address><email></email></address>
3030    </author>
3031    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3032      <organization>One Laptop per Child</organization>
3033      <address><email></email></address>
3034    </author>
3035    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3036      <organization abbrev="HP">Hewlett-Packard Company</organization>
3037      <address><email></email></address>
3038    </author>
3039    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
3040      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3041      <address><email></email></address>
3042    </author>
3043    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3044      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3045      <address><email></email></address>
3046    </author>
3047    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3048      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3049      <address><email></email></address>
3050    </author>
3051    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3052      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3053      <address><email></email></address>
3054    </author>
3055    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
3056      <organization abbrev="W3C">World Wide Web Consortium</organization>
3057      <address><email></email></address>
3058    </author>
3059    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3060      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3061      <address><email></email></address>
3062    </author>
3063    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3064  </front>
3065  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-&ID-VERSION;"/>
3066  <x:source href="p5-range.xml" basename="p5-range"/>
3069<reference anchor="Part6">
3070  <front>
3071    <title abbrev="HTTP/1.1">HTTP/1.1, part 6: Caching</title>
3072    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3073      <organization abbrev="Day Software">Day Software</organization>
3074      <address><email></email></address>
3075    </author>
3076    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3077      <organization>One Laptop per Child</organization>
3078      <address><email></email></address>
3079    </author>
3080    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3081      <organization abbrev="HP">Hewlett-Packard Company</organization>
3082      <address><email></email></address>
3083    </author>
3084    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
3085      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3086      <address><email></email></address>
3087    </author>
3088    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3089      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3090      <address><email></email></address>
3091    </author>
3092    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3093      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3094      <address><email></email></address>
3095    </author>
3096    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3097      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3098      <address><email></email></address>
3099    </author>
3100    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
3101      <organization abbrev="W3C">World Wide Web Consortium</organization>
3102      <address><email></email></address>
3103    </author>
3104    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3105      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3106      <address><email></email></address>
3107    </author>
3108    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3109  </front>
3110  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-&ID-VERSION;"/>
3111  <x:source href="p6-cache.xml" basename="p6-cache"/>
3114<reference anchor="RFC822ABNF">
3115  <front>
3116    <title abbrev="Standard for ARPA Internet Text Messages">Standard for the format of ARPA Internet text messages</title>
3117    <author initials="D.H." surname="Crocker" fullname="David H. Crocker">
3118      <organization>University of Delaware, Dept. of Electrical Engineering</organization>
3119      <address><email>DCrocker@UDel-Relay</email></address>
3120    </author>
3121    <date month="August" day="13" year="1982"/>
3122  </front>
3123  <seriesInfo name="STD" value="11"/>
3124  <seriesInfo name="RFC" value="822"/>
3127<reference anchor="RFC2045">
3128  <front>
3129    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
3130    <author initials="N." surname="Freed" fullname="Ned Freed">
3131      <organization>Innosoft International, Inc.</organization>
3132      <address><email></email></address>
3133    </author>
3134    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
3135      <organization>First Virtual Holdings</organization>
3136      <address><email></email></address>
3137    </author>
3138    <date month="November" year="1996"/>
3139  </front>
3140  <seriesInfo name="RFC" value="2045"/>
3143<reference anchor="RFC2047">
3144  <front>
3145    <title abbrev="Message Header Extensions">MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text</title>
3146    <author initials="K." surname="Moore" fullname="Keith Moore">
3147      <organization>University of Tennessee</organization>
3148      <address><email></email></address>
3149    </author>
3150    <date month="November" year="1996"/>
3151  </front>
3152  <seriesInfo name="RFC" value="2047"/>
3155<reference anchor="RFC2119">
3156  <front>
3157    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
3158    <author initials="S." surname="Bradner" fullname="Scott Bradner">
3159      <organization>Harvard University</organization>
3160      <address><email></email></address>
3161    </author>
3162    <date month="March" year="1997"/>
3163  </front>
3164  <seriesInfo name="BCP" value="14"/>
3165  <seriesInfo name="RFC" value="2119"/>
3168<reference anchor="RFC2396">
3169  <front>
3170    <title abbrev="URI Generic Syntax">Uniform Resource Identifiers (URI): Generic Syntax</title>
3171    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3172      <organization abbrev="MIT/LCS">World Wide Web Consortium</organization>
3173      <address><email></email></address>
3174    </author>
3175    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3176      <organization abbrev="U.C. Irvine">Department of Information and Computer Science</organization>
3177      <address><email></email></address>
3178    </author>
3179    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3180      <organization abbrev="Xerox Corporation">Xerox PARC</organization>
3181      <address><email></email></address>
3182    </author>
3183    <date month="August" year="1998"/>
3184  </front>
3185  <seriesInfo name="RFC" value="2396"/>
3188<reference anchor="RFC4288">
3189  <front>
3190    <title>Media Type Specifications and Registration Procedures</title>
3191    <author initials="N." surname="Freed" fullname="N. Freed">
3192      <organization>Sun Microsystems</organization>
3193      <address>
3194        <email></email>
3195      </address>
3196    </author>
3197    <author initials="J." surname="Klensin" fullname="J. Klensin">
3198      <organization/>
3199      <address>
3200        <email></email>
3201      </address>
3202    </author>
3203    <date year="2005" month="December"/>
3204  </front>
3205  <seriesInfo name="BCP" value="13"/>
3206  <seriesInfo name="RFC" value="4288"/>
3209<reference anchor="USASCII">
3210  <front>
3211    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
3212    <author>
3213      <organization>American National Standards Institute</organization>
3214    </author>
3215    <date year="1986"/>
3216  </front>
3217  <seriesInfo name="ANSI" value="X3.4"/>
3222<references title="Informative References">
3224<reference anchor="Nie1997" target="">
3225  <front>
3226    <title>Network Performance Effects of HTTP/1.1, CSS1, and PNG</title>
3227    <author initials="H.F.." surname="Nielsen" fullname="H.F. Nielsen">
3228      <organization/>
3229    </author>
3230    <author initials="J." surname="Gettys" fullname="J. Gettys">
3231      <organization/>
3232    </author>
3233    <author initials="E." surname="Prud'hommeaux" fullname="E. Prud'hommeaux">
3234      <organization/>
3235    </author>
3236    <author initials="H." surname="Lie" fullname="H. Lie">
3237      <organization/>
3238    </author>
3239    <author initials="C." surname="Lilley" fullname="C. Lilley">
3240      <organization/>
3241    </author>
3242    <date year="1997" month="September"/>
3243  </front>
3244  <seriesInfo name="ACM" value="Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication SIGCOMM '97"/>
3247<reference anchor="Pad1995">
3248  <front>
3249    <title>Improving HTTP Latency</title>
3250    <author initials="V.N." surname="Padmanabhan" fullname="Venkata N. Padmanabhan">
3251      <organization/>
3252    </author>
3253    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
3254      <organization/>
3255    </author>
3256    <date year="1995" month="December"/>
3257  </front>
3258  <seriesInfo name="Computer Networks and ISDN Systems" value="v. 28, pp. 25-35"/>
3259  <annotation>
3260    Slightly revised version of paper in Proc. 2nd International WWW Conference '94: Mosaic and the Web, Oct. 1994,
3261    which is available at <eref target=""/>.
3262  </annotation>
3265<reference anchor="RFC822">
3266  <front>
3267    <title abbrev="Standard for ARPA Internet Text Messages">Standard for the format of ARPA Internet text messages</title>
3268    <author initials="D.H." surname="Crocker" fullname="David H. Crocker">
3269      <organization>University of Delaware, Dept. of Electrical Engineering</organization>
3270      <address><email>DCrocker@UDel-Relay</email></address>
3271    </author>
3272    <date month="August" day="13" year="1982"/>
3273  </front>
3274  <seriesInfo name="STD" value="11"/>
3275  <seriesInfo name="RFC" value="822"/>
3278<reference anchor="RFC959">
3279  <front>
3280    <title abbrev="File Transfer Protocol">File Transfer Protocol</title>
3281    <author initials="J." surname="Postel" fullname="J. Postel">
3282      <organization>Information Sciences Institute (ISI)</organization>
3283    </author>
3284    <author initials="J." surname="Reynolds" fullname="J. Reynolds">
3285      <organization/>
3286    </author>
3287    <date month="October" year="1985"/>
3288  </front>
3289  <seriesInfo name="STD" value="9"/>
3290  <seriesInfo name="RFC" value="959"/>
3293<reference anchor="RFC1123">
3294  <front>
3295    <title>Requirements for Internet Hosts - Application and Support</title>
3296    <author initials="R." surname="Braden" fullname="Robert Braden">
3297      <organization>University of Southern California (USC), Information Sciences Institute</organization>
3298      <address><email>Braden@ISI.EDU</email></address>
3299    </author>
3300    <date month="October" year="1989"/>
3301  </front>
3302  <seriesInfo name="STD" value="3"/>
3303  <seriesInfo name="RFC" value="1123"/>
3306<reference anchor="RFC1305">
3307  <front>
3308    <title>Network Time Protocol (Version 3) Specification, Implementation</title>
3309    <author initials="D." surname="Mills" fullname="David L. Mills">
3310      <organization>University of Delaware, Electrical Engineering Department</organization>
3311      <address><email></email></address>
3312    </author>
3313    <date month="March" year="1992"/>
3314  </front>
3315  <seriesInfo name="RFC" value="1305"/>
3318<reference anchor="RFC1436">
3319  <front>
3320    <title abbrev="Gopher">The Internet Gopher Protocol (a distributed document search and retrieval protocol)</title>
3321    <author initials="F." surname="Anklesaria" fullname="Farhad Anklesaria">
3322      <organization>University of Minnesota, Computer and Information Services</organization>
3323      <address><email></email></address>
3324    </author>
3325    <author initials="M." surname="McCahill" fullname="Mark McCahill">
3326      <organization>University of Minnesota, Computer and Information Services</organization>
3327      <address><email></email></address>
3328    </author>
3329    <author initials="P." surname="Lindner" fullname="Paul Lindner">
3330      <organization>University of Minnesota, Computer and Information Services</organization>
3331      <address><email></email></address>
3332    </author>
3333    <author initials="D." surname="Johnson" fullname="David Johnson">
3334      <organization>University of Minnesota, Computer and Information Services</organization>
3335      <address><email></email></address>
3336    </author>
3337    <author initials="D." surname="Torrey" fullname="Daniel Torrey">
3338      <organization>University of Minnesota, Computer and Information Services</organization>
3339      <address><email></email></address>
3340    </author>
3341    <author initials="B." surname="Alberti" fullname="Bob Alberti">
3342      <organization>University of Minnesota, Computer and Information Services</organization>
3343      <address><email></email></address>
3344    </author>
3345    <date month="March" year="1993"/>
3346  </front>
3347  <seriesInfo name="RFC" value="1436"/>
3350<reference anchor="RFC1630">
3351  <front>
3352    <title abbrev="URIs in WWW">Universal Resource Identifiers in WWW: A Unifying Syntax for the Expression of Names and Addresses of Objects on the Network as used in the World-Wide Web</title>
3353    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3354      <organization>CERN, World-Wide Web project</organization>
3355      <address><email></email></address>
3356    </author>
3357    <date month="June" year="1994"/>
3358  </front>
3359  <seriesInfo name="RFC" value="1630"/>
3362<reference anchor="RFC1737">
3363  <front>
3364    <title abbrev="Requirements for Uniform Resource Names">Functional Requirements for Uniform Resource Names</title>
3365    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3366      <organization>Xerox Palo Alto Research Center</organization>
3367      <address><email></email></address>
3368    </author>
3369    <author initials="K." surname="Sollins" fullname="Karen Sollins">
3370      <organization>MIT Laboratory for Computer Science</organization>
3371      <address><email></email></address>
3372    </author>
3373    <date month="December" year="1994"/>
3374  </front>
3375  <seriesInfo name="RFC" value="1737"/>
3378<reference anchor="RFC1738">
3379  <front>
3380    <title>Uniform Resource Locators (URL)</title>
3381    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3382      <organization>CERN, World-Wide Web project</organization>
3383      <address><email></email></address>
3384    </author>
3385    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3386      <organization>Xerox PARC</organization>
3387      <address><email></email></address>
3388    </author>
3389    <author initials="M." surname="McCahill" fullname="Mark McCahill">
3390      <organization>University of Minnesota, Computer and Information Services</organization>
3391      <address><email></email></address>
3392    </author>
3393    <date month="December" year="1994"/>
3394  </front>
3395  <seriesInfo name="RFC" value="1738"/>
3398<reference anchor="RFC1808">
3399  <front>
3400    <title>Relative Uniform Resource Locators</title>
3401    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
3402      <organization>University of California Irvine, Department of Information and Computer Science</organization>
3403      <address><email></email></address>
3404    </author>
3405    <date month="June" year="1995"/>
3406  </front>
3407  <seriesInfo name="RFC" value="1808"/>
3410<reference anchor="RFC1900">
3411  <front>
3412    <title>Renumbering Needs Work</title>
3413    <author initials="B." surname="Carpenter" fullname="Brian E. Carpenter">
3414      <organization>CERN, Computing and Networks Division</organization>
3415      <address><email></email></address>
3416    </author>
3417    <author initials="Y." surname="Rekhter" fullname="Yakov Rekhter">
3418      <organization>cisco Systems</organization>
3419      <address><email></email></address>
3420    </author>
3421    <date month="February" year="1996"/>
3422  </front>
3423  <seriesInfo name="RFC" value="1900"/>
3426<reference anchor="RFC1945">
3427  <front>
3428    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
3429    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3430      <organization>MIT, Laboratory for Computer Science</organization>
3431      <address><email></email></address>
3432    </author>
3433    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3434      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
3435      <address><email></email></address>
3436    </author>
3437    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3438      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
3439      <address><email></email></address>
3440    </author>
3441    <date month="May" year="1996"/>
3442  </front>
3443  <seriesInfo name="RFC" value="1945"/>
3446<reference anchor="RFC2068">
3447  <front>
3448    <title abbrev="HTTP/1.1">Hypertext Transfer Protocol -- HTTP/1.1</title>
3449    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
3450      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
3451      <address><email></email></address>
3452    </author>
3453    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3454      <organization>MIT Laboratory for Computer Science</organization>
3455      <address><email></email></address>
3456    </author>
3457    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3458      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
3459      <address><email></email></address>
3460    </author>
3461    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3462      <organization>MIT Laboratory for Computer Science</organization>
3463      <address><email></email></address>
3464    </author>
3465    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3466      <organization>MIT Laboratory for Computer Science</organization>
3467      <address><email></email></address>
3468    </author>
3469    <date month="January" year="1997"/>
3470  </front>
3471  <seriesInfo name="RFC" value="2068"/>
3474<reference anchor="RFC2145">
3475  <front>
3476    <title abbrev="HTTP Version Numbers">Use and Interpretation of HTTP Version Numbers</title>
3477    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
3478      <organization>Western Research Laboratory</organization>
3479      <address><email></email></address>
3480    </author>
3481    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3482      <organization>Department of Information and Computer Science</organization>
3483      <address><email></email></address>
3484    </author>
3485    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3486      <organization>MIT Laboratory for Computer Science</organization>
3487      <address><email></email></address>
3488    </author>
3489    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3490      <organization>W3 Consortium</organization>
3491      <address><email></email></address>
3492    </author>
3493    <date month="May" year="1997"/>
3494  </front>
3495  <seriesInfo name="RFC" value="2145"/>
3498<reference anchor="RFC2324">
3499  <front>
3500    <title abbrev="HTCPCP/1.0">Hyper Text Coffee Pot Control Protocol (HTCPCP/1.0)</title>
3501    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3502      <organization>Xerox Palo Alto Research Center</organization>
3503      <address><email></email></address>
3504    </author>
3505    <date month="April" day="1" year="1998"/>
3506  </front>
3507  <seriesInfo name="RFC" value="2324"/>
3510<reference anchor="RFC2616">
3511  <front>
3512    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
3513    <author initials="R." surname="Fielding" fullname="R. Fielding">
3514      <organization>University of California, Irvine</organization>
3515      <address><email></email></address>
3516    </author>
3517    <author initials="J." surname="Gettys" fullname="J. Gettys">
3518      <organization>W3C</organization>
3519      <address><email></email></address>
3520    </author>
3521    <author initials="J." surname="Mogul" fullname="J. Mogul">
3522      <organization>Compaq Computer Corporation</organization>
3523      <address><email></email></address>
3524    </author>
3525    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
3526      <organization>MIT Laboratory for Computer Science</organization>
3527      <address><email></email></address>
3528    </author>
3529    <author initials="L." surname="Masinter" fullname="L. Masinter">
3530      <organization>Xerox Corporation</organization>
3531      <address><email></email></address>
3532    </author>
3533    <author initials="P." surname="Leach" fullname="P. Leach">
3534      <organization>Microsoft Corporation</organization>
3535      <address><email></email></address>
3536    </author>
3537    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
3538      <organization>W3C</organization>
3539      <address><email></email></address>
3540    </author>
3541    <date month="June" year="1999"/>
3542  </front>
3543  <seriesInfo name="RFC" value="2616"/>
3546<reference anchor="RFC2821">
3547  <front>
3548    <title>Simple Mail Transfer Protocol</title>
3549    <author initials="J." surname="Klensin" fullname="J. Klensin">
3550      <organization>AT&amp;T Laboratories</organization>
3551      <address><email></email></address>
3552    </author>
3553    <date year="2001" month="April"/>
3554  </front>
3555  <seriesInfo name="RFC" value="2821"/>
3558<reference anchor="RFC2822">
3559  <front>
3560    <title>Internet Message Format</title>
3561    <author initials="P." surname="Resnick" fullname="P. Resnick">
3562      <organization>QUALCOMM Incorporated</organization>
3563    </author>
3564    <date year="2001" month="April"/>
3565  </front>
3566  <seriesInfo name="RFC" value="2822"/>
3569<reference anchor='RFC3977'>
3570  <front>
3571    <title>Network News Transfer Protocol (NNTP)</title>
3572    <author initials='C.' surname='Feather' fullname='C. Feather'>
3573      <organization>THUS plc</organization>
3574      <address><email></email></address>
3575    </author>
3576    <date year='2006' month='October' />
3577  </front>
3578  <seriesInfo name="RFC" value="3977"/>
3581<reference anchor="Spe" target="">
3582  <front>
3583  <title>Analysis of HTTP Performance Problems</title>
3584  <author initials="S." surname="Spero" fullname="Simon E. Spero">
3585    <organization/>
3586  </author>
3587  <date/>
3588  </front>
3591<reference anchor="Tou1998" target="">
3592  <front>
3593  <title>Analysis of HTTP Performance</title>
3594  <author initials="J." surname="Touch" fullname="Joe Touch">
3595    <organization>USC/Information Sciences Institute</organization>
3596    <address><email></email></address>
3597  </author>
3598  <author initials="J." surname="Heidemann" fullname="John Heidemann">
3599    <organization>USC/Information Sciences Institute</organization>
3600    <address><email></email></address>
3601  </author>
3602  <author initials="K." surname="Obraczka" fullname="Katia Obraczka">
3603    <organization>USC/Information Sciences Institute</organization>
3604    <address><email></email></address>
3605  </author>
3606  <date year="1998" month="Aug"/>
3607  </front>
3608  <seriesInfo name="ISI Research Report" value="ISI/RR-98-463"/>
3609  <annotation>(original report dated Aug. 1996)</annotation>
3612<reference anchor="WAIS">
3613  <front>
3614    <title>WAIS Interface Protocol Prototype Functional Specification (v1.5)</title>
3615    <author initials="F." surname="Davis" fullname="F. Davis">
3616      <organization>Thinking Machines Corporation</organization>
3617    </author>
3618    <author initials="B." surname="Kahle" fullname="B. Kahle">
3619      <organization>Thinking Machines Corporation</organization>
3620    </author>
3621    <author initials="H." surname="Morris" fullname="H. Morris">
3622      <organization>Thinking Machines Corporation</organization>
3623    </author>
3624    <author initials="J." surname="Salem" fullname="J. Salem">
3625      <organization>Thinking Machines Corporation</organization>
3626    </author>
3627    <author initials="T." surname="Shen" fullname="T. Shen">
3628      <organization>Thinking Machines Corporation</organization>
3629    </author>
3630    <author initials="R." surname="Wang" fullname="R. Wang">
3631      <organization>Thinking Machines Corporation</organization>
3632    </author>
3633    <author initials="J." surname="Sui" fullname="J. Sui">
3634      <organization>Thinking Machines Corporation</organization>
3635    </author>
3636    <author initials="M." surname="Grinbaum" fullname="M. Grinbaum">
3637      <organization>Thinking Machines Corporation</organization>
3638    </author>
3639    <date month="April" year="1990"/>
3640  </front>
3641  <seriesInfo name="Thinking Machines Corporation" value=""/>
3647<section title="Internet Media Types" anchor="">
3649   In addition to defining the HTTP/1.1 protocol, this document serves
3650   as the specification for the Internet media type "message/http" and
3651   "application/http". The following is to be registered with IANA <xref target="RFC4288"/>.
3653<section title="Internet Media Type message/http" anchor="">
3654<iref item="Media Type" subitem="message/http" primary="true"/>
3655<iref item="message/http Media Type" primary="true"/>
3657   The message/http type can be used to enclose a single HTTP request or
3658   response message, provided that it obeys the MIME restrictions for all
3659   "message" types regarding line length and encodings.
3662  <list style="hanging" x:indent="12em">
3663    <t hangText="Type name:">
3664      message
3665    </t>
3666    <t hangText="Subtype name:">
3667      http
3668    </t>
3669    <t hangText="Required parameters:">
3670      none
3671    </t>
3672    <t hangText="Optional parameters:">
3673      version, msgtype
3674      <list style="hanging">
3675        <t hangText="version:">
3676          The HTTP-Version number of the enclosed message
3677          (e.g., "1.1"). If not present, the version can be
3678          determined from the first line of the body.
3679        </t>
3680        <t hangText="msgtype:">
3681          The message type -- "request" or "response". If not
3682          present, the type can be determined from the first
3683          line of the body.
3684        </t>
3685      </list>
3686    </t>
3687    <t hangText="Encoding considerations:">
3688      only "7bit", "8bit", or "binary" are permitted
3689    </t>
3690    <t hangText="Security considerations:">
3691      none
3692    </t>
3693    <t hangText="Interoperability considerations:">
3694      none
3695    </t>
3696    <t hangText="Published specification:">
3697      This specification (see <xref target=""/>).
3698    </t>
3699    <t hangText="Applications that use this media type:">
3700    </t>
3701    <t hangText="Additional information:">
3702      <list style="hanging">
3703        <t hangText="Magic number(s):">none</t>
3704        <t hangText="File extension(s):">none</t>
3705        <t hangText="Macintosh file type code(s):">none</t>
3706      </list>
3707    </t>
3708    <t hangText="Person and email address to contact for further information:">
3709      See Authors Section.
3710    </t>
3711                <t hangText="Intended usage:">
3712                  COMMON
3713    </t>
3714                <t hangText="Restrictions on usage:">
3715                  none
3716    </t>
3717    <t hangText="Author/Change controller:">
3718      IESG
3719    </t>
3720  </list>
3723<section title="Internet Media Type application/http" anchor="">
3724<iref item="Media Type" subitem="application/http" primary="true"/>
3725<iref item="application/http Media Type" primary="true"/>
3727   The application/http type can be used to enclose a pipeline of one or more
3728   HTTP request or response messages (not intermixed).
3731  <list style="hanging" x:indent="12em">
3732    <t hangText="Type name:">
3733      application
3734    </t>
3735    <t hangText="Subtype name:">
3736      http
3737    </t>
3738    <t hangText="Required parameters:">
3739      none
3740    </t>
3741    <t hangText="Optional parameters:">
3742      version, msgtype
3743      <list style="hanging">
3744        <t hangText="version:">
3745          The HTTP-Version number of the enclosed messages
3746          (e.g., "1.1"). If not present, the version can be
3747          determined from the first line of the body.
3748        </t>
3749        <t hangText="msgtype:">
3750          The message type -- "request" or "response". If not
3751          present, the type can be determined from the first
3752          line of the body.
3753        </t>
3754      </list>
3755    </t>
3756    <t hangText="Encoding considerations:">
3757      HTTP messages enclosed by this type
3758      are in "binary" format; use of an appropriate
3759      Content-Transfer-Encoding is required when
3760      transmitted via E-mail.
3761    </t>
3762    <t hangText="Security considerations:">
3763      none
3764    </t>
3765    <t hangText="Interoperability considerations:">
3766      none
3767    </t>
3768    <t hangText="Published specification:">
3769      This specification (see <xref target=""/>).
3770    </t>
3771    <t hangText="Applications that use this media type:">
3772    </t>
3773    <t hangText="Additional information:">
3774      <list style="hanging">
3775        <t hangText="Magic number(s):">none</t>
3776        <t hangText="File extension(s):">none</t>
3777        <t hangText="Macintosh file type code(s):">none</t>
3778      </list>
3779    </t>
3780    <t hangText="Person and email address to contact for further information:">
3781      See Authors Section.
3782    </t>
3783                <t hangText="Intended usage:">
3784                  COMMON
3785    </t>
3786                <t hangText="Restrictions on usage:">
3787                  none
3788    </t>
3789    <t hangText="Author/Change controller:">
3790      IESG
3791    </t>
3792  </list>
3797<section title="Tolerant Applications" anchor="tolerant.applications">
3799   Although this document specifies the requirements for the generation
3800   of HTTP/1.1 messages, not all applications will be correct in their
3801   implementation. We therefore recommend that operational applications
3802   be tolerant of deviations whenever those deviations can be
3803   interpreted unambiguously.
3806   Clients &SHOULD; be tolerant in parsing the Status-Line and servers
3807   tolerant when parsing the Request-Line. In particular, they &SHOULD;
3808   accept any amount of SP or HTAB characters between fields, even though
3809   only a single SP is required.
3812   The line terminator for message-header fields is the sequence CRLF.
3813   However, we recommend that applications, when parsing such headers,
3814   recognize a single LF as a line terminator and ignore the leading CR.
3817   The character set of an entity-body &SHOULD; be labeled as the lowest
3818   common denominator of the character codes used within that body, with
3819   the exception that not labeling the entity is preferred over labeling
3820   the entity with the labels US-ASCII or ISO-8859-1. See &payload;.
3823   Additional rules for requirements on parsing and encoding of dates
3824   and other potential problems with date encodings include:
3827  <list style="symbols">
3828     <t>HTTP/1.1 clients and caches &SHOULD; assume that an RFC-850 date
3829        which appears to be more than 50 years in the future is in fact
3830        in the past (this helps solve the "year 2000" problem).</t>
3832     <t>An HTTP/1.1 implementation &MAY; internally represent a parsed
3833        Expires date as earlier than the proper value, but &MUST-NOT;
3834        internally represent a parsed Expires date as later than the
3835        proper value.</t>
3837     <t>All expiration-related calculations &MUST; be done in GMT. The
3838        local time zone &MUST-NOT; influence the calculation or comparison
3839        of an age or expiration time.</t>
3841     <t>If an HTTP header incorrectly carries a date value with a time
3842        zone other than GMT, it &MUST; be converted into GMT using the
3843        most conservative possible conversion.</t>
3844  </list>
3848<section title="Conversion of Date Formats" anchor="">
3850   HTTP/1.1 uses a restricted set of date formats (<xref target=""/>) to
3851   simplify the process of date comparison. Proxies and gateways from
3852   other protocols &SHOULD; ensure that any Date header field present in a
3853   message conforms to one of the HTTP/1.1 formats and rewrite the date
3854   if necessary.
3858<section title="Compatibility with Previous Versions" anchor="compatibility">
3860   It is beyond the scope of a protocol specification to mandate
3861   compliance with previous versions. HTTP/1.1 was deliberately
3862   designed, however, to make supporting previous versions easy. It is
3863   worth noting that, at the time of composing this specification
3864   (1996), we would expect commercial HTTP/1.1 servers to:
3865  <list style="symbols">
3866     <t>recognize the format of the Request-Line for HTTP/0.9, 1.0, and
3867        1.1 requests;</t>
3869     <t>understand any valid request in the format of HTTP/0.9, 1.0, or
3870        1.1;</t>
3872     <t>respond appropriately with a message in the same major version
3873        used by the client.</t>
3874  </list>
3877   And we would expect HTTP/1.1 clients to:
3878  <list style="symbols">
3879     <t>recognize the format of the Status-Line for HTTP/1.0 and 1.1
3880        responses;</t>
3882     <t>understand any valid response in the format of HTTP/0.9, 1.0, or
3883        1.1.</t>
3884  </list>
3887   For most implementations of HTTP/1.0, each connection is established
3888   by the client prior to the request and closed by the server after
3889   sending the response. Some implementations implement the Keep-Alive
3890   version of persistent connections described in <xref x:sec="19.7.1" x:fmt="of" target="RFC2068"/>.
3893<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
3895   This section summarizes major differences between versions HTTP/1.0
3896   and HTTP/1.1.
3899<section title="Changes to Simplify Multi-homed Web Servers and Conserve IP Addresses" anchor="">
3901   The requirements that clients and servers support the Host request-header,
3902   report an error if the Host request-header (<xref target=""/>) is
3903   missing from an HTTP/1.1 request, and accept absolute URIs (<xref target="request-uri"/>)
3904   are among the most important changes defined by this
3905   specification.
3908   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
3909   addresses and servers; there was no other established mechanism for
3910   distinguishing the intended server of a request than the IP address
3911   to which that request was directed. The changes outlined above will
3912   allow the Internet, once older HTTP clients are no longer common, to
3913   support multiple Web sites from a single IP address, greatly
3914   simplifying large operational Web servers, where allocation of many
3915   IP addresses to a single host has created serious problems. The
3916   Internet will also be able to recover the IP addresses that have been
3917   allocated for the sole purpose of allowing special-purpose domain
3918   names to be used in root-level HTTP URLs. Given the rate of growth of
3919   the Web, and the number of servers already deployed, it is extremely
3920   important that all implementations of HTTP (including updates to
3921   existing HTTP/1.0 applications) correctly implement these
3922   requirements:
3923  <list style="symbols">
3924     <t>Both clients and servers &MUST; support the Host request-header.</t>
3926     <t>A client that sends an HTTP/1.1 request &MUST; send a Host header.</t>
3928     <t>Servers &MUST; report a 400 (Bad Request) error if an HTTP/1.1
3929        request does not include a Host request-header.</t>
3931     <t>Servers &MUST; accept absolute URIs.</t>
3932  </list>
3937<section title="Compatibility with HTTP/1.0 Persistent Connections" anchor="compatibility.with.http.1.0.persistent.connections">
3939   Some clients and servers might wish to be compatible with some
3940   previous implementations of persistent connections in HTTP/1.0
3941   clients and servers. Persistent connections in HTTP/1.0 are
3942   explicitly negotiated as they are not the default behavior. HTTP/1.0
3943   experimental implementations of persistent connections are faulty,
3944   and the new facilities in HTTP/1.1 are designed to rectify these
3945   problems. The problem was that some existing 1.0 clients may be
3946   sending Keep-Alive to a proxy server that doesn't understand
3947   Connection, which would then erroneously forward it to the next
3948   inbound server, which would establish the Keep-Alive connection and
3949   result in a hung HTTP/1.0 proxy waiting for the close on the
3950   response. The result is that HTTP/1.0 clients must be prevented from
3951   using Keep-Alive when talking to proxies.
3954   However, talking to proxies is the most important use of persistent
3955   connections, so that prohibition is clearly unacceptable. Therefore,
3956   we need some other mechanism for indicating a persistent connection
3957   is desired, which is safe to use even when talking to an old proxy
3958   that ignores Connection. Persistent connections are the default for
3959   HTTP/1.1 messages; we introduce a new keyword (Connection: close) for
3960   declaring non-persistence. See <xref target="header.connection"/>.
3963   The original HTTP/1.0 form of persistent connections (the Connection:
3964   Keep-Alive and Keep-Alive header) is documented in <xref target="RFC2068"/>.
3968<section title="Changes from RFC 2068" anchor="changes.from.rfc.2068">
3970   This specification has been carefully audited to correct and
3971   disambiguate key word usage; RFC 2068 had many problems in respect to
3972   the conventions laid out in <xref target="RFC2119"/>.
3975   Transfer-coding and message lengths all interact in ways that
3976   required fixing exactly when chunked encoding is used (to allow for
3977   transfer encoding that may not be self delimiting); it was important
3978   to straighten out exactly how message lengths are computed. (Sections
3979   <xref target="transfer.codings" format="counter"/>, <xref target="message.length" format="counter"/>,
3980   <xref target="header.content-length" format="counter"/>,
3981   see also <xref target="Part3"/>, <xref target="Part5"/> and <xref target="Part6"/>)
3984   The use and interpretation of HTTP version numbers has been clarified
3985   by <xref target="RFC2145"/>. Require proxies to upgrade requests to highest protocol
3986   version they support to deal with problems discovered in HTTP/1.0
3987   implementations (<xref target="http.version"/>)
3990   Transfer-coding had significant problems, particularly with
3991   interactions with chunked encoding. The solution is that transfer-codings
3992   become as full fledged as content-codings. This involves
3993   adding an IANA registry for transfer-codings (separate from content
3994   codings), a new header field (TE) and enabling trailer headers in the
3995   future. Transfer encoding is a major performance benefit, so it was
3996   worth fixing <xref target="Nie1997"/>. TE also solves another, obscure, downward
3997   interoperability problem that could have occurred due to interactions
3998   between authentication trailers, chunked encoding and HTTP/1.0
3999   clients.(Section <xref target="transfer.codings" format="counter"/>, <xref target="chunked.transfer.encoding" format="counter"/>,
4000   and <xref target="header.te" format="counter"/>)
4004<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
4006  Clarify that HTTP-Version is case sensitive.
4007  (<xref target="http.version"/>)
4010  Remove reference to non-existant identity transfer-coding value tokens.
4011  (Sections <xref format="counter" target="transfer.codings"/> and
4012  <xref format="counter" target="message.length"/>)
4015  Clarification that the chunk length does not include
4016  the count of the octets in the chunk header and trailer.
4017  (<xref target="chunked.transfer.encoding"/>)
4020  Fix BNF to add query, as the abs_path production in
4021  <xref x:sec="3" x:fmt="of" target="RFC2396"/> doesn't define it.
4022  (<xref target="request-uri"/>)
4025  Clarify exactly when close connection options must be sent.
4026  (<xref target="header.connection"/>)
4031<section title="Change Log (to be removed by RFC Editor before publication)">
4033<section title="Since RFC2616">
4035  Extracted relevant partitions from <xref target="RFC2616"/>.
4039<section title="Since draft-ietf-httpbis-p1-messaging-00">
4041  Closed issues:
4042  <list style="symbols">
4043    <t>
4044      <eref target=""/>:
4045      "HTTP Version should be case sensitive"
4046      (<eref target=""/>)
4047    </t>
4048    <t>
4049      <eref target=""/>:
4050      "'unsafe' characters"
4051      (<eref target=""/>)
4052    </t>
4053    <t>
4054      <eref target=""/>:
4055      "Chunk Size Definition"
4056      (<eref target=""/>)
4057    </t>
4058    <t>
4059      <eref target=""/>:
4060      "Message Length"
4061      (<eref target=""/>)
4062    </t>
4063    <t>
4064      <eref target=""/>:
4065      "Media Type Registrations"
4066      (<eref target=""/>)
4067    </t>
4068    <t>
4069      <eref target=""/>:
4070      "URI includes query"
4071      (<eref target=""/>)
4072    </t>
4073    <t>
4074      <eref target=""/>:
4075      "No close on 1xx responses"
4076      (<eref target=""/>)
4077    </t>
4078    <t>
4079      <eref target=""/>:
4080      "Remove 'identity' token references"
4081      (<eref target=""/>)
4082    </t>
4083    <t>
4084      <eref target=""/>:
4085      "Import query BNF"
4086    </t>
4087    <t>
4088      <eref target=""/>:
4089      "qdtext BNF"
4090    </t>
4091    <t>
4092      <eref target=""/>:
4093      "Normative and Informative references"
4094    </t>
4095    <t>
4096      <eref target=""/>:
4097      "RFC2606 Compliance"
4098    </t>
4099    <t>
4100      <eref target=""/>:
4101      "RFC977 reference"
4102    </t>
4103    <t>
4104      <eref target=""/>:
4105      "RFC1700 references"
4106    </t>
4107    <t>
4108      <eref target=""/>:
4109      "inconsistency in date format explanation"
4110    </t>
4111    <t>
4112      <eref target=""/>:
4113      "Date reference typo"
4114    </t>
4115    <t>
4116      <eref target=""/>:
4117      "Informative references"
4118    </t>
4119    <t>
4120      <eref target=""/>:
4121      "ISO-8859-1 Reference"
4122    </t>
4123    <t>
4124      <eref target=""/>:
4125      "Normative up-to-date references"
4126    </t>
4127  </list>
4130  Other changes:
4131  <list style="symbols">
4132    <t>
4133      Update media type registrations to use RFC4288 template.
4134    </t>
4135    <t>
4136      Use names of RFC4234 core rules DQUOTE and HTAB,
4137      fix broken ABNF for chunk-data
4138      (work in progress on <eref target=""/>)
4139    </t>
4140  </list>
Note: See TracBrowser for help on using the repository browser.