source: draft-ietf-httpbis/latest/p1-messaging.xml @ 31

Last change on this file since 31 was 31, checked in by julian.reschke@…, 12 years ago

Insert cross-document references.

  • Property svn:eol-style set to native
File size: 175.2 KB
Line 
1<?xml version="1.0" encoding="utf-8"?>
2<!DOCTYPE rfc [
3  <!ENTITY MAY "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MAY</bcp14>">
4  <!ENTITY MUST "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST</bcp14>">
5  <!ENTITY MUST-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST NOT</bcp14>">
6  <!ENTITY OPTIONAL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>OPTIONAL</bcp14>">
7  <!ENTITY RECOMMENDED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>RECOMMENDED</bcp14>">
8  <!ENTITY REQUIRED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>REQUIRED</bcp14>">
9  <!ENTITY SHALL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL</bcp14>">
10  <!ENTITY SHALL-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL NOT</bcp14>">
11  <!ENTITY SHOULD "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD</bcp14>">
12  <!ENTITY SHOULD-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD NOT</bcp14>">
13  <!ENTITY ID-VERSION "latest">
14  <!ENTITY ID-MONTH "December">
15  <!ENTITY ID-YEAR "2007">
16  <!ENTITY caching                "<xref target='Part6' x:rel='#caching' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
17  <!ENTITY payload                "<xref target='Part3' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
18  <!ENTITY CONNECT                "<xref target='Part2' x:rel='#CONNECT' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
19  <!ENTITY content.negotiation    "<xref target='Part3' x:rel='#content.negotiation' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
20  <!ENTITY diff2045entity         "<xref target='Part3' x:rel='#differences.between.http.entities.and.rfc.2045.entities' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
21  <!ENTITY entity                 "<xref target='Part3' x:rel='#entity' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
22  <!ENTITY entity-header-fields   "<xref target='Part3' x:rel='#entity.header.fields' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
23  <!ENTITY header-cache-control   "<xref target='Part6' x:rel='#header.cache-control' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
24  <!ENTITY header-expect          "<xref target='Part2' x:rel='#header.expect' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
25  <!ENTITY header-pragma          "<xref target='Part6' x:rel='#header.pragma' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
26  <!ENTITY header-warning         "<xref target='Part6' x:rel='#header.warning' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
27  <!ENTITY idempotent-methods     "<xref target='Part2' x:rel='#idempotent.methods' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
28  <!ENTITY qvalue                 "<xref target='Part3' x:rel='#quality.values' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
29  <!ENTITY request-header-fields  "<xref target='Part2' x:rel='#request.header.fields' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
30  <!ENTITY response-header-fields "<xref target='Part2' x:rel='#response.header.fields' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
31  <!ENTITY method                 "<xref target='Part2' x:rel='#method' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
32  <!ENTITY status-codes           "<xref target='Part2' x:rel='#status.codes' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
33  <!ENTITY status-100             "<xref target='Part2' x:rel='#status.100' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
34  <!ENTITY status-1xx             "<xref target='Part2' x:rel='#status.1xx' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
35  <!ENTITY status-414             "<xref target='Part2' x:rel='#status.414' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
36]>
37<?rfc toc="yes" ?>
38<?rfc symrefs="yes" ?>
39<?rfc sortrefs="yes" ?>
40<?rfc compact="yes"?>
41<?rfc subcompact="no" ?>
42<?rfc linkmailto="no" ?>
43<?rfc editing="no" ?>
44<?rfc-ext allow-markup-in-artwork="yes" ?>
45<?rfc-ext include-references-in-index="yes" ?>
46<rfc obsoletes="2068, 2616, 2617" category="std"
47     ipr="full3978" docName="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"
48     xmlns:x='http://purl.org/net/xml2rfc/ext' xmlns:ed="http://greenbytes.de/2002/rfcedit">
49<front>
50
51  <title abbrev="HTTP/1.1">HTTP/1.1, part 1: URIs, Connections, and Message Parsing</title>
52
53  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
54    <organization abbrev="Day Software">Day Software</organization>
55    <address>
56      <postal>
57        <street>23 Corporate Plaza DR, Suite 280</street>
58        <city>Newport Beach</city>
59        <region>CA</region>
60        <code>92660</code>
61        <country>USA</country>
62      </postal>
63      <phone>+1-949-706-5300</phone>
64      <facsimile>+1-949-706-5305</facsimile>
65      <email>fielding@gbiv.com</email>
66      <uri>http://roy.gbiv.com/</uri>
67    </address>
68  </author>
69
70  <author initials="J." surname="Gettys" fullname="Jim Gettys">
71    <organization>One Laptop per Child</organization>
72    <address>
73      <postal>
74        <street>21 Oak Knoll Road</street>
75        <city>Carlisle</city>
76        <region>MA</region>
77        <code>01741</code>
78        <country>USA</country>
79      </postal>
80      <email>jg@laptop.org</email>
81      <uri>http://www.laptop.org/</uri>
82    </address>
83  </author>
84 
85  <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
86    <organization abbrev="HP">Hewlett-Packard Company</organization>
87    <address>
88      <postal>
89        <street>HP Labs, Large Scale Systems Group</street>
90        <street>1501 Page Mill Road, MS 1177</street>
91        <city>Palo Alto</city>
92        <region>CA</region>
93        <code>94304</code>
94        <country>USA</country>
95      </postal>
96      <email>JeffMogul@acm.org</email>
97    </address>
98  </author>
99
100  <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
101    <organization abbrev="Microsoft">Microsoft Corporation</organization>
102    <address>
103      <postal>
104        <street>1 Microsoft Way</street>
105        <city>Redmond</city>
106        <region>WA</region>
107        <code>98052</code>
108        <country>USA</country>
109      </postal>
110      <email>henrikn@microsoft.com</email>
111    </address>
112  </author>
113
114  <author initials="L." surname="Masinter" fullname="Larry Masinter">
115    <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
116    <address>
117      <postal>
118        <street>345 Park Ave</street>
119        <city>San Jose</city>
120        <region>CA</region>
121        <code>95110</code>
122        <country>USA</country>
123      </postal>
124      <email>LMM@acm.org</email>
125      <uri>http://larry.masinter.net/</uri>
126    </address>
127  </author>
128 
129  <author initials="P." surname="Leach" fullname="Paul J. Leach">
130    <organization abbrev="Microsoft">Microsoft Corporation</organization>
131    <address>
132      <postal>
133        <street>1 Microsoft Way</street>
134        <city>Redmond</city>
135        <region>WA</region>
136        <code>98052</code>
137      </postal>
138      <email>paulle@microsoft.com</email>
139    </address>
140  </author>
141   
142  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
143    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
144    <address>
145      <postal>
146        <street>MIT Laboratory for Computer Science</street>
147        <street>545 Technology Square</street>
148        <city>Cambridge</city>
149        <region>MA</region>
150        <code>02139</code>
151        <country>USA</country>
152      </postal>
153      <facsimile>+1 (617) 258 8682</facsimile>
154      <email>timbl@w3.org</email>
155    </address>
156  </author>
157
158  <date month="&ID-MONTH;" year="&ID-YEAR;"/>
159
160<abstract>
161<t>
162   The Hypertext Transfer Protocol (HTTP) is an application-level
163   protocol for distributed, collaborative, hypermedia information
164   systems. HTTP has been in use by the World Wide Web global information
165   initiative since 1990. This document is Part 1 of the eight-part specification
166   that defines the protocol referred to as "HTTP/1.1" and, taken together,
167   updates RFC 2616 and RFC 2617.  Part 1 provides an overview of HTTP and
168   its associated terminology, defines the "http" and "https" Uniform
169   Resource Identifier (URI) schemes, defines the generic message syntax
170   and parsing requirements for HTTP message frames, and describes
171   general security concerns for implementations.
172</t>
173</abstract>
174</front>
175<middle>
176<section title="Introduction" anchor="introduction">
177<t>
178   This document will define aspects of HTTP related to overall network
179   operation, message framing, interaction with transport protocols, and
180   URI schemes. Right now it only includes the extracted relevant sections
181   of <xref target="RFC2616"/> and <xref target="RFC2617"/>.
182</t>
183<section title="Purpose" anchor="intro.purpose">
184<t>
185   The Hypertext Transfer Protocol (HTTP) is an application-level
186   protocol for distributed, collaborative, hypermedia information
187   systems. HTTP has been in use by the World-Wide Web global
188   information initiative since 1990. The first version of HTTP,
189   referred to as HTTP/0.9, was a simple protocol for raw data transfer
190   across the Internet. HTTP/1.0, as defined by RFC 1945 <xref target="RFC1945"/>, improved
191   the protocol by allowing messages to be in the format of MIME-like
192   messages, containing metainformation about the data transferred and
193   modifiers on the request/response semantics. However, HTTP/1.0 does
194   not sufficiently take into consideration the effects of hierarchical
195   proxies, caching, the need for persistent connections, or virtual
196   hosts. In addition, the proliferation of incompletely-implemented
197   applications calling themselves "HTTP/1.0" has necessitated a
198   protocol version change in order for two communicating applications
199   to determine each other's true capabilities.
200</t>
201<t>
202   This specification defines the protocol referred to as "HTTP/1.1".
203   This protocol includes more stringent requirements than HTTP/1.0 in
204   order to ensure reliable implementation of its features.
205</t>
206<t>
207   Practical information systems require more functionality than simple
208   retrieval, including search, front-end update, and annotation. HTTP
209   allows an open-ended set of methods and headers that indicate the
210   purpose of a request <xref target="RFC2324"/>. It builds on the discipline of reference
211   provided by the Uniform Resource Identifier (URI) <xref target="RFC1630"/>, as a location
212   (URL) <xref target="RFC1738"/> or name (URN) <xref target="RFC1737"/>, for indicating the resource to which a
213   method is to be applied. Messages are passed in a format similar to
214   that used by Internet mail <xref target="RFC822"/> as defined by the Multipurpose
215   Internet Mail Extensions (MIME) <xref target="RFC2045"/>.
216</t>
217<t>
218   HTTP is also used as a generic protocol for communication between
219   user agents and proxies/gateways to other Internet systems, including
220   those supported by the SMTP <xref target="RFC821"/>, NNTP <xref target="RFC977"/>, FTP <xref target="RFC959"/>, Gopher <xref target="RFC1436"/>,
221   and WAIS <xref target="WAIS"/> protocols. In this way, HTTP allows basic hypermedia
222   access to resources available from diverse applications.
223</t>
224</section>
225
226<section title="Requirements" anchor="intro.requirements">
227<t>
228   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
229   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
230   document are to be interpreted as described in RFC 2119 <xref target="RFC2119"/>.
231</t>
232<t>
233   An implementation is not compliant if it fails to satisfy one or more
234   of the &MUST; or &REQUIRED; level requirements for the protocols it
235   implements. An implementation that satisfies all the &MUST; or &REQUIRED;
236   level and all the &SHOULD; level requirements for its protocols is said
237   to be "unconditionally compliant"; one that satisfies all the &MUST;
238   level requirements but not all the &SHOULD; level requirements for its
239   protocols is said to be "conditionally compliant."
240</t>
241</section>
242
243<section title="Terminology" anchor="intro.terminology">
244<t>
245   This specification uses a number of terms to refer to the roles
246   played by participants in, and objects of, the HTTP communication.
247</t>
248<t>
249  <iref item="connection"/>
250  <x:dfn>connection</x:dfn>
251  <list>
252    <t>
253      A transport layer virtual circuit established between two programs
254      for the purpose of communication.
255    </t>
256  </list>
257</t>
258<t>
259  <iref item="message"/>
260  <x:dfn>message</x:dfn>
261  <list>
262    <t>
263      The basic unit of HTTP communication, consisting of a structured
264      sequence of octets matching the syntax defined in <xref target="http.message"/> and
265      transmitted via the connection.
266    </t>
267  </list>
268</t>
269<t>
270  <iref item="request"/>
271  <x:dfn>request</x:dfn>
272  <list>
273    <t>
274      An HTTP request message, as defined in <xref target="request"/>.
275    </t>
276  </list>
277</t>
278<t>
279  <iref item="response"/>
280  <x:dfn>response</x:dfn>
281  <list>
282    <t>
283      An HTTP response message, as defined in <xref target="response"/>.
284    </t>
285  </list>
286</t>
287<t>
288  <iref item="resource"/>
289  <x:dfn>resource</x:dfn>
290  <list>
291    <t>
292      A network data object or service that can be identified by a URI,
293      as defined in <xref target="uri"/>. Resources may be available in multiple
294      representations (e.g. multiple languages, data formats, size, and
295      resolutions) or vary in other ways.
296    </t>
297  </list>
298</t>
299<t>
300  <iref item="entity"/>
301  <x:dfn>entity</x:dfn>
302  <list>
303    <t>
304      The information transferred as the payload of a request or
305      response. An entity consists of metainformation in the form of
306      entity-header fields and content in the form of an entity-body, as
307      described in &entity;.
308    </t>
309  </list>
310</t>
311<t>
312  <iref item="representation"/>
313  <x:dfn>representation</x:dfn>
314  <list>
315    <t>
316      An entity included with a response that is subject to content
317      negotiation, as described in &content.negotiation;. There may exist multiple
318      representations associated with a particular response status.
319    </t>
320  </list>
321</t>
322<t>
323  <iref item="content negotiation"/>
324  <x:dfn>content negotiation</x:dfn>
325  <list>
326    <t>
327      The mechanism for selecting the appropriate representation when
328      servicing a request, as described in &content.negotiation;. The
329      representation of entities in any response can be negotiated
330      (including error responses).
331    </t>
332  </list>
333</t>
334<t>
335  <iref item="variant"/>
336  <x:dfn>variant</x:dfn>
337  <list>
338    <t>
339      A resource may have one, or more than one, representation(s)
340      associated with it at any given instant. Each of these
341      representations is termed a `varriant'.  Use of the term `variant'
342      does not necessarily imply that the resource is subject to content
343      negotiation.
344    </t>
345  </list>
346</t>
347<t>
348  <iref item="client"/>
349  <x:dfn>client</x:dfn>
350  <list>
351    <t>
352      A program that establishes connections for the purpose of sending
353      requests.
354    </t>
355  </list>
356</t>
357<t>
358  <iref item="user agent"/>
359  <x:dfn>user agent</x:dfn>
360  <list>
361    <t>
362      The client which initiates a request. These are often browsers,
363      editors, spiders (web-traversing robots), or other end user tools.
364    </t>
365  </list>
366</t>
367<t>
368  <iref item="server"/>
369  <x:dfn>server</x:dfn>
370  <list>
371    <t>
372      An application program that accepts connections in order to
373      service requests by sending back responses. Any given program may
374      be capable of being both a client and a server; our use of these
375      terms refers only to the role being performed by the program for a
376      particular connection, rather than to the program's capabilities
377      in general. Likewise, any server may act as an origin server,
378      proxy, gateway, or tunnel, switching behavior based on the nature
379      of each request.
380    </t>
381  </list>
382</t>
383<t>
384  <iref item="origin server"/>
385  <x:dfn>origin server</x:dfn>
386  <list>
387    <t>
388      The server on which a given resource resides or is to be created.
389    </t>
390  </list>
391</t>
392<t>
393  <iref item="proxy"/>
394  <x:dfn>proxy</x:dfn>
395  <list>
396    <t>
397      An intermediary program which acts as both a server and a client
398      for the purpose of making requests on behalf of other clients.
399      Requests are serviced internally or by passing them on, with
400      possible translation, to other servers. A proxy &MUST; implement
401      both the client and server requirements of this specification. A
402      "transparent proxy" is a proxy that does not modify the request or
403      response beyond what is required for proxy authentication and
404      identification. A "non-transparent proxy" is a proxy that modifies
405      the request or response in order to provide some added service to
406      the user agent, such as group annotation services, media type
407      transformation, protocol reduction, or anonymity filtering. Except
408      where either transparent or non-transparent behavior is explicitly
409      stated, the HTTP proxy requirements apply to both types of
410      proxies.
411    </t>
412  </list>
413</t>
414<t>
415  <iref item="gateway"/>
416  <x:dfn>gateway</x:dfn>
417  <list>
418    <t>
419      A server which acts as an intermediary for some other server.
420      Unlike a proxy, a gateway receives requests as if it were the
421      origin server for the requested resource; the requesting client
422      may not be aware that it is communicating with a gateway.
423    </t>
424  </list>
425</t>
426<t>
427  <iref item="tunnel"/>
428  <x:dfn>tunnel</x:dfn>
429  <list>
430    <t>
431      An intermediary program which is acting as a blind relay between
432      two connections. Once active, a tunnel is not considered a party
433      to the HTTP communication, though the tunnel may have been
434      initiated by an HTTP request. The tunnel ceases to exist when both
435      ends of the relayed connections are closed.
436    </t>
437  </list>
438</t>
439<t>
440  <iref item="cache"/>
441  <x:dfn>cache</x:dfn>
442  <list>
443    <t>
444      A program's local store of response messages and the subsystem
445      that controls its message storage, retrieval, and deletion. A
446      cache stores cacheable responses in order to reduce the response
447      time and network bandwidth consumption on future, equivalent
448      requests. Any client or server may include a cache, though a cache
449      cannot be used by a server that is acting as a tunnel.
450    </t>
451  </list>
452</t>
453<t>
454  <iref item="cacheable"/>
455  <x:dfn>cacheable</x:dfn>
456  <list>
457    <t>
458      A response is cacheable if a cache is allowed to store a copy of
459      the response message for use in answering subsequent requests. The
460      rules for determining the cacheability of HTTP responses are
461      defined in &caching;. Even if a resource is cacheable, there may
462      be additional constraints on whether a cache can use the cached
463      copy for a particular request.
464    </t>
465  </list>
466</t>
467<t>
468  <iref item="upstream"/>
469  <iref item="downstream"/>
470  <x:dfn>upstream</x:dfn>/<x:dfn>downstream</x:dfn>
471  <list>
472    <t>
473      Upstream and downstream describe the flow of a message: all
474      messages flow from upstream to downstream.
475    </t>
476  </list>
477</t>
478<t>
479  <iref item="inbound"/>
480  <iref item="outbound"/>
481  <x:dfn>inbound</x:dfn>/<x:dfn>outbound</x:dfn>
482  <list>
483    <t>
484      Inbound and outbound refer to the request and response paths for
485      messages: "inbound" means "traveling toward the origin server",
486      and "outbound" means "traveling toward the user agent"
487    </t>
488  </list>
489</t>
490</section>
491
492<section title="Overall Operation" anchor="intro.overall.operation">
493<t>
494   The HTTP protocol is a request/response protocol. A client sends a
495   request to the server in the form of a request method, URI, and
496   protocol version, followed by a MIME-like message containing request
497   modifiers, client information, and possible body content over a
498   connection with a server. The server responds with a status line,
499   including the message's protocol version and a success or error code,
500   followed by a MIME-like message containing server information, entity
501   metainformation, and possible entity-body content. The relationship
502   between HTTP and MIME is described in &diff2045entity;.
503</t>
504<t>
505   Most HTTP communication is initiated by a user agent and consists of
506   a request to be applied to a resource on some origin server. In the
507   simplest case, this may be accomplished via a single connection (v)
508   between the user agent (UA) and the origin server (O).
509</t>
510<figure><artwork type="drawing">
511       request chain ------------------------&gt;
512    UA -------------------v------------------- O
513       &lt;----------------------- response chain
514</artwork></figure>
515<t>
516   A more complicated situation occurs when one or more intermediaries
517   are present in the request/response chain. There are three common
518   forms of intermediary: proxy, gateway, and tunnel. A proxy is a
519   forwarding agent, receiving requests for a URI in its absolute form,
520   rewriting all or part of the message, and forwarding the reformatted
521   request toward the server identified by the URI. A gateway is a
522   receiving agent, acting as a layer above some other server(s) and, if
523   necessary, translating the requests to the underlying server's
524   protocol. A tunnel acts as a relay point between two connections
525   without changing the messages; tunnels are used when the
526   communication needs to pass through an intermediary (such as a
527   firewall) even when the intermediary cannot understand the contents
528   of the messages.
529</t>
530<figure><artwork type="drawing">
531       request chain --------------------------------------&gt;
532    UA -----v----- A -----v----- B -----v----- C -----v----- O
533       &lt;------------------------------------- response chain
534</artwork></figure>
535<t>
536   The figure above shows three intermediaries (A, B, and C) between the
537   user agent and origin server. A request or response message that
538   travels the whole chain will pass through four separate connections.
539   This distinction is important because some HTTP communication options
540   may apply only to the connection with the nearest, non-tunnel
541   neighbor, only to the end-points of the chain, or to all connections
542   along the chain. Although the diagram is linear, each participant may
543   be engaged in multiple, simultaneous communications. For example, B
544   may be receiving requests from many clients other than A, and/or
545   forwarding requests to servers other than C, at the same time that it
546   is handling A's request.
547</t>
548<t>
549   Any party to the communication which is not acting as a tunnel may
550   employ an internal cache for handling requests. The effect of a cache
551   is that the request/response chain is shortened if one of the
552   participants along the chain has a cached response applicable to that
553   request. The following illustrates the resulting chain if B has a
554   cached copy of an earlier response from O (via C) for a request which
555   has not been cached by UA or A.
556</t>
557<figure><artwork type="drawing">
558          request chain ----------&gt;
559       UA -----v----- A -----v----- B - - - - - - C - - - - - - O
560          &lt;--------- response chain
561</artwork></figure>
562<t>
563   Not all responses are usefully cacheable, and some requests may
564   contain modifiers which place special requirements on cache behavior.
565   HTTP requirements for cache behavior and cacheable responses are
566   defined in &caching;.
567</t>
568<t>
569   In fact, there are a wide variety of architectures and configurations
570   of caches and proxies currently being experimented with or deployed
571   across the World Wide Web. These systems include national hierarchies
572   of proxy caches to save transoceanic bandwidth, systems that
573   broadcast or multicast cache entries, organizations that distribute
574   subsets of cached data via CD-ROM, and so on. HTTP systems are used
575   in corporate intranets over high-bandwidth links, and for access via
576   PDAs with low-power radio links and intermittent connectivity. The
577   goal of HTTP/1.1 is to support the wide diversity of configurations
578   already deployed while introducing protocol constructs that meet the
579   needs of those who build web applications that require high
580   reliability and, failing that, at least reliable indications of
581   failure.
582</t>
583<t>
584   HTTP communication usually takes place over TCP/IP connections. The
585   default port is TCP 80 <xref target="RFC1700"/>, but other ports can be used. This does
586   not preclude HTTP from being implemented on top of any other protocol
587   on the Internet, or on other networks. HTTP only presumes a reliable
588   transport; any protocol that provides such guarantees can be used;
589   the mapping of the HTTP/1.1 request and response structures onto the
590   transport data units of the protocol in question is outside the scope
591   of this specification.
592</t>
593<t>
594   In HTTP/1.0, most implementations used a new connection for each
595   request/response exchange. In HTTP/1.1, a connection may be used for
596   one or more request/response exchanges, although connections may be
597   closed for a variety of reasons (see <xref target="persistent.connections"/>).
598</t>
599</section>
600</section>
601
602<section title="Notational Conventions and Generic Grammar" anchor="notation">
603
604<section title="Augmented BNF" anchor="notation.abnf">
605<t>
606   All of the mechanisms specified in this document are described in
607   both prose and an augmented Backus-Naur Form (BNF) similar to that
608   used by RFC 822 <xref target="RFC822"/>. Implementors will need to be familiar with the
609   notation in order to understand this specification. The augmented BNF
610   includes the following constructs:
611</t>
612<t>
613   name = definition
614  <list>
615    <t>
616      The name of a rule is simply the name itself (without any
617      enclosing "&lt;" and "&gt;") and is separated from its definition by the
618      equal "=" character. White space is only significant in that
619      indentation of continuation lines is used to indicate a rule
620      definition that spans more than one line. Certain basic rules are
621      in uppercase, such as SP, LWS, HT, CRLF, DIGIT, ALPHA, etc. Angle
622      brackets are used within definitions whenever their presence will
623      facilitate discerning the use of rule names.
624    </t>
625  </list>
626</t>
627<t>
628   "literal"
629  <list>
630    <t>
631      Quotation marks surround literal text. Unless stated otherwise,
632      the text is case-insensitive.
633    </t>
634  </list>
635</t>
636<t>
637   rule1 | rule2
638  <list>
639    <t>
640      Elements separated by a bar ("|") are alternatives, e.g., "yes |
641      no" will accept yes or no.
642    </t>
643  </list>
644</t>
645<t>
646   (rule1 rule2)
647  <list>
648    <t>
649      Elements enclosed in parentheses are treated as a single element.
650      Thus, "(elem (foo | bar) elem)" allows the token sequences "elem
651      foo elem" and "elem bar elem".
652    </t>
653  </list>
654</t>
655<t>
656   *rule
657  <list>
658    <t>
659      The character "*" preceding an element indicates repetition. The
660      full form is "&lt;n&gt;*&lt;m&gt;element" indicating at least &lt;n&gt; and at most
661      &lt;m&gt; occurrences of element. Default values are 0 and infinity so
662      that "*(element)" allows any number, including zero; "1*element"
663      requires at least one; and "1*2element" allows one or two.
664    </t>
665  </list>
666</t>
667<t>
668   [rule]
669  <list>
670    <t>
671      Square brackets enclose optional elements; "[foo bar]" is
672      equivalent to "*1(foo bar)".
673    </t>
674  </list>
675</t>
676<t>
677   N rule
678  <list>
679    <t>
680      Specific repetition: "&lt;n&gt;(element)" is equivalent to
681      "&lt;n&gt;*&lt;n&gt;(element)"; that is, exactly &lt;n&gt; occurrences of (element).
682      Thus 2DIGIT is a 2-digit number, and 3ALPHA is a string of three
683      alphabetic characters.
684    </t>
685  </list>
686</t>
687<t>
688   #rule
689  <list>
690    <t>
691      A construct "#" is defined, similar to "*", for defining lists of
692      elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating at least
693      &lt;n&gt; and at most &lt;m&gt; elements, each separated by one or more commas
694      (",") and &OPTIONAL; linear white space (LWS). This makes the usual
695      form of lists very easy; a rule such as
696    </t>
697    <t>
698         ( *LWS element *( *LWS "," *LWS element ))
699    </t>
700    <t>
701      can be shown as
702    </t>
703    <t>
704         1#element
705    </t>
706    <t>
707      Wherever this construct is used, null elements are allowed, but do
708      not contribute to the count of elements present. That is,
709      "(element), , (element) " is permitted, but counts as only two
710      elements. Therefore, where at least one element is required, at
711      least one non-null element &MUST; be present. Default values are 0
712      and infinity so that "#element" allows any number, including zero;
713      "1#element" requires at least one; and "1#2element" allows one or
714      two.
715    </t>
716  </list>
717</t>
718<t>
719   ; comment
720  <list>
721    <t>
722      A semi-colon, set off some distance to the right of rule text,
723      starts a comment that continues to the end of line. This is a
724      simple way of including useful notes in parallel with the
725      specifications.
726    </t>
727  </list>
728</t>
729<t>
730   implied *LWS
731  <list>
732    <t>
733      The grammar described by this specification is word-based. Except
734      where noted otherwise, linear white space (LWS) can be included
735      between any two adjacent words (token or quoted-string), and
736      between adjacent words and separators, without changing the
737      interpretation of a field. At least one delimiter (LWS and/or
738      separators) &MUST; exist between any two tokens (for the definition
739      of "token" below), since they would otherwise be interpreted as a
740      single token.
741    </t>
742  </list>
743</t>
744</section>
745
746<section title="Basic Rules" anchor="basic.rules">
747<x:anchor-alias value="OCTET"/>
748<x:anchor-alias value="CHAR"/>
749<x:anchor-alias value="UPALPHA"/>
750<x:anchor-alias value="LOALPHA"/>
751<x:anchor-alias value="ALPHA"/>
752<x:anchor-alias value="DIGIT"/>
753<x:anchor-alias value="CTL"/>
754<x:anchor-alias value="CR"/>
755<x:anchor-alias value="LF"/>
756<x:anchor-alias value="SP"/>
757<x:anchor-alias value="HT"/>
758<x:anchor-alias value="CRLF"/>
759<x:anchor-alias value="LWS"/>
760<x:anchor-alias value="TEXT"/>
761<x:anchor-alias value="HEX"/>
762<x:anchor-alias value="token"/>
763<x:anchor-alias value="separators"/>
764<x:anchor-alias value="comment"/>
765<x:anchor-alias value="ctext"/>
766<x:anchor-alias value="quoted-string"/>
767<x:anchor-alias value="qdtext"/>
768<x:anchor-alias value="quoted-pair"/>
769<t>
770   The following rules are used throughout this specification to
771   describe basic parsing constructs. The US-ASCII coded character set
772   is defined by ANSI X3.4-1986 <xref target="USASCII"/>.
773</t>
774<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="HT"/>
775    OCTET          = &lt;any 8-bit sequence of data&gt;
776    CHAR           = &lt;any US-ASCII character (octets 0 - 127)&gt;
777    UPALPHA        = &lt;any US-ASCII uppercase letter "A".."Z"&gt;
778    LOALPHA        = &lt;any US-ASCII lowercase letter "a".."z"&gt;
779    ALPHA          = UPALPHA | LOALPHA
780    DIGIT          = &lt;any US-ASCII digit "0".."9"&gt;
781    CTL            = &lt;any US-ASCII control character
782                     (octets 0 - 31) and DEL (127)&gt;
783    CR             = &lt;US-ASCII CR, carriage return (13)&gt;
784    LF             = &lt;US-ASCII LF, linefeed (10)&gt;
785    SP             = &lt;US-ASCII SP, space (32)&gt;
786    HT             = &lt;US-ASCII HT, horizontal-tab (9)&gt;
787    &lt;"&gt;            = &lt;US-ASCII double-quote mark (34)&gt;
788</artwork></figure>
789<t>
790   HTTP/1.1 defines the sequence CR LF as the end-of-line marker for all
791   protocol elements except the entity-body (see <xref target="tolerant.applications"/> for
792   tolerant applications). The end-of-line marker within an entity-body
793   is defined by its associated media type, as described in &payload;.
794</t>
795<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="CRLF"/>
796    CRLF           = CR LF
797</artwork></figure>
798<t>
799   HTTP/1.1 header field values can be folded onto multiple lines if the
800   continuation line begins with a space or horizontal tab. All linear
801   white space, including folding, has the same semantics as SP. A
802   recipient &MAY; replace any linear white space with a single SP before
803   interpreting the field value or forwarding the message downstream.
804</t>
805<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="LWS"/>
806    LWS            = [CRLF] 1*( SP | HT )
807</artwork></figure>
808<t>
809   The TEXT rule is only used for descriptive field contents and values
810   that are not intended to be interpreted by the message parser. Words
811   of *TEXT &MAY; contain characters from character sets other than ISO-8859-1
812   <xref target="ISO-8859"/> only when encoded according to the rules of RFC 2047
813   <xref target="RFC2047"/>.
814</t>
815<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TEXT"/>
816    TEXT           = &lt;any OCTET except CTLs,
817                     but including LWS&gt;
818</artwork></figure>
819<t>
820   A CRLF is allowed in the definition of TEXT only as part of a header
821   field continuation. It is expected that the folding LWS will be
822   replaced with a single SP before interpretation of the TEXT value.
823</t>
824<t>
825   Hexadecimal numeric characters are used in several protocol elements.
826</t>
827<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HEX"/>
828    HEX            = "A" | "B" | "C" | "D" | "E" | "F"
829                   | "a" | "b" | "c" | "d" | "e" | "f" | DIGIT
830</artwork></figure>
831<t>
832   Many HTTP/1.1 header field values consist of words separated by LWS
833   or special characters. These special characters &MUST; be in a quoted
834   string to be used within a parameter value (as defined in
835   <xref target="transfer.codings"/>).
836</t>
837<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="token"/><iref primary="true" item="Grammar" subitem="separators"/>
838    token          = 1*&lt;any CHAR except CTLs or separators&gt;
839    separators     = "(" | ")" | "&lt;" | "&gt;" | "@"
840                   | "," | ";" | ":" | "\" | &lt;"&gt;
841                   | "/" | "[" | "]" | "?" | "="
842                   | "{" | "}" | SP | HT
843</artwork></figure>
844<t>
845   Comments can be included in some HTTP header fields by surrounding
846   the comment text with parentheses. Comments are only allowed in
847   fields containing "comment" as part of their field value definition.
848   In all other fields, parentheses are considered part of the field
849   value.
850</t>
851<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/>
852    comment        = "(" *( ctext | quoted-pair | comment ) ")"
853    ctext          = &lt;any TEXT excluding "(" and ")"&gt;
854</artwork></figure>
855<t>
856   A string of text is parsed as a single word if it is quoted using
857   double-quote marks.
858</t>
859<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-string"/><iref primary="true" item="Grammar" subitem="qdtext"/>
860    quoted-string  = ( &lt;"&gt; *(qdtext | quoted-pair ) &lt;"&gt; )
861    qdtext         = &lt;any TEXT except &lt;"&gt;&gt;
862</artwork></figure>
863<t>
864   The backslash character ("\") &MAY; be used as a single-character
865   quoting mechanism only within quoted-string and comment constructs.
866</t>
867<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-pair"/>
868    quoted-pair    = "\" CHAR
869</artwork></figure>
870</section>
871</section>
872
873<section title="Protocol Parameters" anchor="protocol.parameters">
874
875<section title="HTTP Version" anchor="http.version">
876<t>
877   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate versions
878   of the protocol. The protocol versioning policy is intended to allow
879   the sender to indicate the format of a message and its capacity for
880   understanding further HTTP communication, rather than the features
881   obtained via that communication. No change is made to the version
882   number for the addition of message components which do not affect
883   communication behavior or which only add to extensible field values.
884   The &lt;minor&gt; number is incremented when the changes made to the
885   protocol add features which do not change the general message parsing
886   algorithm, but which may add to the message semantics and imply
887   additional capabilities of the sender. The &lt;major&gt; number is
888   incremented when the format of a message within the protocol is
889   changed. See RFC 2145 <xref target="RFC2145"/> for a fuller explanation.
890</t>
891<t>
892   The version of an HTTP message is indicated by an HTTP-Version field
893   in the first line of the message.
894</t>
895<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-Version"/>
896       HTTP-Version   = "HTTP" "/" 1*DIGIT "." 1*DIGIT
897</artwork></figure>
898<t>
899   Note that the major and minor numbers &MUST; be treated as separate
900   integers and that each &MAY; be incremented higher than a single digit.
901   Thus, HTTP/2.4 is a lower version than HTTP/2.13, which in turn is
902   lower than HTTP/12.3. Leading zeros &MUST; be ignored by recipients and
903   &MUST-NOT; be sent.
904</t>
905<t>
906   An application that sends a request or response message that includes
907   HTTP-Version of "HTTP/1.1" &MUST; be at least conditionally compliant
908   with this specification. Applications that are at least conditionally
909   compliant with this specification &SHOULD; use an HTTP-Version of
910   "HTTP/1.1" in their messages, and &MUST; do so for any message that is
911   not compatible with HTTP/1.0. For more details on when to send
912   specific HTTP-Version values, see RFC 2145 <xref target="RFC2145"/>.
913</t>
914<t>
915   The HTTP version of an application is the highest HTTP version for
916   which the application is at least conditionally compliant.
917</t>
918<t>
919   Proxy and gateway applications need to be careful when forwarding
920   messages in protocol versions different from that of the application.
921   Since the protocol version indicates the protocol capability of the
922   sender, a proxy/gateway &MUST-NOT; send a message with a version
923   indicator which is greater than its actual version. If a higher
924   version request is received, the proxy/gateway &MUST; either downgrade
925   the request version, or respond with an error, or switch to tunnel
926   behavior.
927</t>
928<t>
929   Due to interoperability problems with HTTP/1.0 proxies discovered
930   since the publication of RFC 2068 <xref target="RFC2068"/>, caching proxies MUST, gateways
931   &MAY;, and tunnels &MUST-NOT; upgrade the request to the highest version
932   they support. The proxy/gateway's response to that request &MUST; be in
933   the same major version as the request.
934</t>
935<t>
936  <list>
937    <t>
938      <x:h>Note:</x:h> Converting between versions of HTTP may involve modification
939      of header fields required or forbidden by the versions involved.
940    </t>
941  </list>
942</t>
943</section>
944
945<section title="Uniform Resource Identifiers" anchor="uri">
946<t>
947   URIs have been known by many names: WWW addresses, Universal Document
948   Identifiers, Universal Resource Identifiers <xref target="RFC1630"/>, and finally the
949   combination of Uniform Resource Locators (URL) <xref target="RFC1738"/> and Names (URN)
950   <xref target="RFC1737"/>. As far as HTTP is concerned, Uniform Resource Identifiers are
951   simply formatted strings which identify--via name, location, or any
952   other characteristic--a resource.
953</t>
954
955<section title="General Syntax" anchor="general.syntax">
956<t>
957   URIs in HTTP can be represented in absolute form or relative to some
958   known base URI <xref target="RFC1808"/>, depending upon the context of their use. The two
959   forms are differentiated by the fact that absolute URIs always begin
960   with a scheme name followed by a colon. For definitive information on
961   URL syntax and semantics, see "Uniform Resource Identifiers (URI):
962   Generic Syntax and Semantics," RFC 2396 <xref target="RFC2396"/> (which replaces RFCs
963   1738 <xref target="RFC1738"/> and RFC 1808 <xref target="RFC1808"/>). This specification adopts the
964   definitions of "URI-reference", "absoluteURI", "relativeURI", "port",
965   "host","abs_path", "rel_path", and "authority" from that
966   specification.
967</t>
968<t>
969   The HTTP protocol does not place any a priori limit on the length of
970   a URI. Servers &MUST; be able to handle the URI of any resource they
971   serve, and &SHOULD; be able to handle URIs of unbounded length if they
972   provide GET-based forms that could generate such URIs. A server
973   &SHOULD; return 414 (Request-URI Too Long) status if a URI is longer
974   than the server can handle (see &status-414;).
975</t>
976<t>
977  <list>
978    <t>
979      <x:h>Note:</x:h> Servers ought to be cautious about depending on URI lengths
980      above 255 bytes, because some older client or proxy
981      implementations might not properly support these lengths.
982    </t>
983  </list>
984</t>
985</section>
986
987<section title="http URL" anchor="http.url">
988<t>
989   The "http" scheme is used to locate network resources via the HTTP
990   protocol. This section defines the scheme-specific syntax and
991   semantics for http URLs.
992</t>
993<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="http_URL"/>
994http_URL = "http:" "//" host [ ":" port ] [ abs_path [ "?" query ]]
995</artwork></figure>
996<t>
997   If the port is empty or not given, port 80 is assumed. The semantics
998   are that the identified resource is located at the server listening
999   for TCP connections on that port of that host, and the Request-URI
1000   for the resource is abs_path (<xref target="request-uri"/>). The use of IP addresses
1001   in URLs &SHOULD; be avoided whenever possible (see RFC 1900 <xref target="RFC1900"/>). If
1002   the abs_path is not present in the URL, it &MUST; be given as "/" when
1003   used as a Request-URI for a resource (<xref target="request-uri"/>). If a proxy
1004
1005   receives a host name which is not a fully qualified domain name, it
1006   &MAY; add its domain to the host name it received. If a proxy receives
1007   a fully qualified domain name, the proxy &MUST-NOT; change the host
1008   name.
1009</t>
1010</section>
1011
1012<section title="URI Comparison" anchor="uri.comparison">
1013<t>
1014   When comparing two URIs to decide if they match or not, a client
1015   &SHOULD; use a case-sensitive octet-by-octet comparison of the entire
1016   URIs, with these exceptions:
1017  <list style="symbols">
1018    <t>A port that is empty or not given is equivalent to the default
1019        port for that URI-reference;</t>
1020    <t>Comparisons of host names &MUST; be case-insensitive;</t>
1021    <t>Comparisons of scheme names &MUST; be case-insensitive;</t>
1022    <t>An empty abs_path is equivalent to an abs_path of "/".</t>
1023  </list>
1024</t>
1025<t>
1026   Characters other than those in the "reserved" and "unsafe" sets (see
1027   RFC 2396 <xref target="RFC2396"/>) are equivalent to their ""%" HEX HEX" encoding.
1028</t>
1029<t>
1030   For example, the following three URIs are equivalent:
1031</t>
1032<figure><artwork type="example">
1033   http://abc.com:80/~smith/home.html
1034   http://ABC.com/%7Esmith/home.html
1035   http://ABC.com:/%7esmith/home.html
1036</artwork></figure>
1037</section>
1038</section>
1039
1040<section title="Date/Time Formats" anchor="date.time.formats">
1041<section title="Full Date" anchor="full.date">
1042<t>
1043   HTTP applications have historically allowed three different formats
1044   for the representation of date/time stamps:
1045</t>
1046<figure><artwork type="example">
1047   Sun, 06 Nov 1994 08:49:37 GMT  ; RFC 822, updated by RFC 1123
1048   Sunday, 06-Nov-94 08:49:37 GMT ; RFC 850, obsoleted by RFC 1036
1049   Sun Nov  6 08:49:37 1994       ; ANSI C's asctime() format
1050</artwork></figure>
1051<t>
1052   The first format is preferred as an Internet standard and represents
1053   a fixed-length subset of that defined by RFC 1123 <xref target="RFC1123"/> (an update to
1054   RFC 822 <xref target="RFC822"/>). The second format is in common use, but is based on the
1055   obsolete RFC 850 <xref target="RFC1036"/> date format and lacks a four-digit year.
1056   HTTP/1.1 clients and servers that parse the date value &MUST; accept
1057   all three formats (for compatibility with HTTP/1.0), though they &MUST;
1058   only generate the RFC 1123 format for representing HTTP-date values
1059   in header fields. See <xref target="tolerant.applications"/> for further information.
1060</t>
1061<t><list><t>
1062      <x:h>Note:</x:h> Recipients of date values are encouraged to be robust in
1063      accepting date values that may have been sent by non-HTTP
1064      applications, as is sometimes the case when retrieving or posting
1065      messages via proxies/gateways to SMTP or NNTP.
1066</t></list></t>
1067<t>
1068   All HTTP date/time stamps &MUST; be represented in Greenwich Mean Time
1069   (GMT), without exception. For the purposes of HTTP, GMT is exactly
1070   equal to UTC (Coordinated Universal Time). This is indicated in the
1071   first two formats by the inclusion of "GMT" as the three-letter
1072   abbreviation for time zone, and &MUST; be assumed when reading the
1073   asctime format. HTTP-date is case sensitive and &MUST-NOT; include
1074   additional LWS beyond that specifically included as SP in the
1075   grammar.
1076</t>
1077<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"/>
1078    HTTP-date    = rfc1123-date | rfc850-date | asctime-date
1079    rfc1123-date = wkday "," SP date1 SP time SP "GMT"
1080    rfc850-date  = weekday "," SP date2 SP time SP "GMT"
1081    asctime-date = wkday SP date3 SP time SP 4DIGIT
1082    date1        = 2DIGIT SP month SP 4DIGIT
1083                   ; day month year (e.g., 02 Jun 1982)
1084    date2        = 2DIGIT "-" month "-" 2DIGIT
1085                   ; day-month-year (e.g., 02-Jun-82)
1086    date3        = month SP ( 2DIGIT | ( SP 1DIGIT ))
1087                   ; month day (e.g., Jun  2)
1088    time         = 2DIGIT ":" 2DIGIT ":" 2DIGIT
1089                   ; 00:00:00 - 23:59:59
1090    wkday        = "Mon" | "Tue" | "Wed"
1091                 | "Thu" | "Fri" | "Sat" | "Sun"
1092    weekday      = "Monday" | "Tuesday" | "Wednesday"
1093                 | "Thursday" | "Friday" | "Saturday" | "Sunday"
1094    month        = "Jan" | "Feb" | "Mar" | "Apr"
1095                 | "May" | "Jun" | "Jul" | "Aug"
1096                 | "Sep" | "Oct" | "Nov" | "Dec"
1097</artwork></figure>
1098<t>
1099      <x:h>Note:</x:h> HTTP requirements for the date/time stamp format apply only
1100      to their usage within the protocol stream. Clients and servers are
1101      not required to use these formats for user presentation, request
1102      logging, etc.
1103</t>
1104</section>
1105</section>
1106
1107<section title="Transfer Codings" anchor="transfer.codings">
1108<t>
1109   Transfer-coding values are used to indicate an encoding
1110   transformation that has been, can be, or may need to be applied to an
1111   entity-body in order to ensure "safe transport" through the network.
1112   This differs from a content coding in that the transfer-coding is a
1113   property of the message, not of the original entity.
1114</t>
1115<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/>
1116    transfer-coding         = "chunked" | transfer-extension
1117    transfer-extension      = token *( ";" parameter )
1118</artwork></figure>
1119<t>
1120   Parameters are in  the form of attribute/value pairs.
1121</t>
1122<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"/>
1123    parameter               = attribute "=" value
1124    attribute               = token
1125    value                   = token | quoted-string
1126</artwork></figure>
1127<t>
1128   All transfer-coding values are case-insensitive. HTTP/1.1 uses
1129   transfer-coding values in the TE header field (<xref target="header.te"/>) and in
1130   the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1131</t>
1132<t>
1133   Whenever a transfer-coding is applied to a message-body, the set of
1134   transfer-codings &MUST; include "chunked", unless the message is
1135   terminated by closing the connection. When the "chunked" transfer-coding
1136   is used, it &MUST; be the last transfer-coding applied to the
1137   message-body. The "chunked" transfer-coding &MUST-NOT; be applied more
1138   than once to a message-body. These rules allow the recipient to
1139   determine the transfer-length of the message (<xref target="message.length"/>).
1140</t>
1141<t>
1142   Transfer-codings are analogous to the Content-Transfer-Encoding
1143   values of MIME <xref target="RFC2045"/>, which were designed to enable safe transport of
1144   binary data over a 7-bit transport service. However, safe transport
1145   has a different focus for an 8bit-clean transfer protocol. In HTTP,
1146   the only unsafe characteristic of message-bodies is the difficulty in
1147   determining the exact body length (<xref target="message.length"/>), or the desire to
1148   encrypt data over a shared transport.
1149</t>
1150<t>
1151   The Internet Assigned Numbers Authority (IANA) acts as a registry for
1152   transfer-coding value tokens. Initially, the registry contains the
1153   following tokens: "chunked" (<xref target="chunked.transfer.encoding"/>), "identity" (section
1154   3.6.2), "gzip" (&payload;), "compress" (&payload;), and "deflate"
1155   (&payload;).
1156</t>
1157<t>
1158   New transfer-coding value tokens &SHOULD; be registered in the same way
1159   as new content-coding value tokens (&payload;).
1160</t>
1161<t>
1162   A server which receives an entity-body with a transfer-coding it does
1163   not understand &SHOULD; return 501 (Unimplemented), and close the
1164   connection. A server &MUST-NOT; send transfer-codings to an HTTP/1.0
1165   client.
1166</t>
1167
1168<section title="Chunked Transfer Coding" anchor="chunked.transfer.encoding">
1169<t>
1170   The chunked encoding modifies the body of a message in order to
1171   transfer it as a series of chunks, each with its own size indicator,
1172   followed by an &OPTIONAL; trailer containing entity-header fields. This
1173   allows dynamically produced content to be transferred along with the
1174   information necessary for the recipient to verify that it has
1175   received the full message.
1176</t>
1177<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"/>
1178    Chunked-Body   = *chunk
1179                     last-chunk
1180                     trailer
1181                     CRLF
1182
1183    chunk          = chunk-size [ chunk-extension ] CRLF
1184                     chunk-data CRLF
1185    chunk-size     = 1*HEX
1186    last-chunk     = 1*("0") [ chunk-extension ] CRLF
1187
1188    chunk-extension= *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
1189    chunk-ext-name = token
1190    chunk-ext-val  = token | quoted-string
1191    chunk-data     = chunk-size(OCTET)
1192    trailer        = *(entity-header CRLF)
1193</artwork></figure>
1194<t>
1195   The chunk-size field is a string of hex digits indicating the size of
1196   the chunk. The chunked encoding is ended by any chunk whose size is
1197   zero, followed by the trailer, which is terminated by an empty line.
1198</t>
1199<t>
1200   The trailer allows the sender to include additional HTTP header
1201   fields at the end of the message. The Trailer header field can be
1202   used to indicate which header fields are included in a trailer (see
1203   <xref target="header.trailer"/>).
1204</t>
1205<t>
1206   A server using chunked transfer-coding in a response &MUST-NOT; use the
1207   trailer for any header fields unless at least one of the following is
1208   true:
1209  <list style="numbers">
1210    <t>the request included a TE header field that indicates "trailers" is
1211     acceptable in the transfer-coding of the  response, as described in
1212     <xref target="header.te"/>; or,</t>
1213
1214    <t>the server is the origin server for the response, the trailer
1215     fields consist entirely of optional metadata, and the recipient
1216     could use the message (in a manner acceptable to the origin server)
1217     without receiving this metadata.  In other words, the origin server
1218     is willing to accept the possibility that the trailer fields might
1219     be silently discarded along the path to the client.</t>
1220  </list>
1221</t>
1222<t>
1223   This requirement prevents an interoperability failure when the
1224   message is being received by an HTTP/1.1 (or later) proxy and
1225   forwarded to an HTTP/1.0 recipient. It avoids a situation where
1226   compliance with the protocol would have necessitated a possibly
1227   infinite buffer on the proxy.
1228</t>
1229<t>
1230   A process for decoding the "chunked" transfer-coding
1231   can be represented in pseudo-code as:
1232</t>
1233<figure><artwork type="code">
1234    length := 0
1235    read chunk-size, chunk-extension (if any) and CRLF
1236    while (chunk-size &gt; 0) {
1237       read chunk-data and CRLF
1238       append chunk-data to entity-body
1239       length := length + chunk-size
1240       read chunk-size and CRLF
1241    }
1242    read entity-header
1243    while (entity-header not empty) {
1244       append entity-header to existing header fields
1245       read entity-header
1246    }
1247    Content-Length := length
1248    Remove "chunked" from Transfer-Encoding
1249</artwork></figure>
1250<t>
1251   All HTTP/1.1 applications &MUST; be able to receive and decode the
1252   "chunked" transfer-coding, and &MUST; ignore chunk-extension extensions
1253   they do not understand.
1254</t>
1255</section>
1256</section>
1257
1258</section>
1259
1260<section title="HTTP Message" anchor="http.message">
1261
1262<section title="Message Types" anchor="message.types">
1263<t>
1264   HTTP messages consist of requests from client to server and responses
1265   from server to client.
1266</t>
1267<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-message"/>
1268    HTTP-message   = Request | Response     ; HTTP/1.1 messages
1269</artwork></figure>
1270<t>
1271   Request (<xref target="request"/>) and Response (<xref target="response"/>) messages use the generic
1272   message format of RFC 822 <xref target="RFC822"/> for transferring entities (the payload
1273   of the message). Both types of message consist of a start-line, zero
1274   or more header fields (also known as "headers"), an empty line (i.e.,
1275   a line with nothing preceding the CRLF) indicating the end of the
1276   header fields, and possibly a message-body.
1277</t>
1278<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="generic-message"/><iref primary="true" item="Grammar" subitem="start-line"/>
1279     generic-message = start-line
1280                       *(message-header CRLF)
1281                       CRLF
1282                       [ message-body ]
1283     start-line      = Request-Line | Status-Line
1284</artwork></figure>
1285<t>
1286   In the interest of robustness, servers &SHOULD; ignore any empty
1287   line(s) received where a Request-Line is expected. In other words, if
1288   the server is reading the protocol stream at the beginning of a
1289   message and receives a CRLF first, it should ignore the CRLF.
1290</t>
1291<t>
1292   Certain buggy HTTP/1.0 client implementations generate extra CRLF's
1293   after a POST request. To restate what is explicitly forbidden by the
1294   BNF, an HTTP/1.1 client &MUST-NOT; preface or follow a request with an
1295   extra CRLF.
1296</t>
1297</section>
1298
1299<section title="Message Headers" anchor="message.headers">
1300<t>
1301   HTTP header fields, which include general-header (<xref target="general.header.fields"/>),
1302   request-header (&request-header-fields;), response-header (&response-header-fields;), and
1303   entity-header (&entity-header-fields;) fields, follow the same generic format as
1304   that given in <xref target="RFC822" x:fmt="sec" x:sec="3.1"/> of RFC 822 <xref target="RFC822"/>. Each header field consists
1305   of a name followed by a colon (":") and the field value. Field names
1306   are case-insensitive. The field value &MAY; be preceded by any amount
1307   of LWS, though a single SP is preferred. Header fields can be
1308   extended over multiple lines by preceding each extra line with at
1309   least one SP or HT. Applications ought to follow "common form", where
1310   one is known or indicated, when generating HTTP constructs, since
1311   there might exist some implementations that fail to accept anything
1312   beyond the common forms.
1313</t>
1314<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"/>
1315    message-header = field-name ":" [ field-value ]
1316    field-name     = token
1317    field-value    = *( field-content | LWS )
1318    field-content  = &lt;the OCTETs making up the field-value
1319                     and consisting of either *TEXT or combinations
1320                     of token, separators, and quoted-string&gt;
1321</artwork></figure>
1322<t>
1323   The field-content does not include any leading or trailing LWS:
1324   linear white space occurring before the first non-whitespace
1325   character of the field-value or after the last non-whitespace
1326   character of the field-value. Such leading or trailing LWS &MAY; be
1327   removed without changing the semantics of the field value. Any LWS
1328   that occurs between field-content &MAY; be replaced with a single SP
1329   before interpreting the field value or forwarding the message
1330   downstream.
1331</t>
1332<t>
1333   The order in which header fields with differing field names are
1334   received is not significant. However, it is "good practice" to send
1335   general-header fields first, followed by request-header or response-header
1336   fields, and ending with the entity-header fields.
1337</t>
1338<t>
1339   Multiple message-header fields with the same field-name &MAY; be
1340   present in a message if and only if the entire field-value for that
1341   header field is defined as a comma-separated list [i.e., #(values)].
1342   It &MUST; be possible to combine the multiple header fields into one
1343   "field-name: field-value" pair, without changing the semantics of the
1344   message, by appending each subsequent field-value to the first, each
1345   separated by a comma. The order in which header fields with the same
1346   field-name are received is therefore significant to the
1347   interpretation of the combined field value, and thus a proxy &MUST-NOT;
1348   change the order of these field values when a message is forwarded.
1349</t>
1350</section>
1351
1352<section title="Message Body" anchor="message.body">
1353<t>
1354   The message-body (if any) of an HTTP message is used to carry the
1355   entity-body associated with the request or response. The message-body
1356   differs from the entity-body only when a transfer-coding has been
1357   applied, as indicated by the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1358</t>
1359<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="message-body"/>
1360    message-body = entity-body
1361                 | &lt;entity-body encoded as per Transfer-Encoding&gt;
1362</artwork></figure>
1363<t>
1364   Transfer-Encoding &MUST; be used to indicate any transfer-codings
1365   applied by an application to ensure safe and proper transfer of the
1366   message. Transfer-Encoding is a property of the message, not of the
1367   entity, and thus &MAY; be added or removed by any application along the
1368   request/response chain. (However, <xref target="transfer.codings"/> places restrictions on
1369   when certain transfer-codings may be used.)
1370</t>
1371<t>
1372   The rules for when a message-body is allowed in a message differ for
1373   requests and responses.
1374</t>
1375<t>
1376   The presence of a message-body in a request is signaled by the
1377   inclusion of a Content-Length or Transfer-Encoding header field in
1378   the request's message-headers. A message-body &MUST-NOT; be included in
1379   a request if the specification of the request method (&method;)
1380   does not allow sending an entity-body in requests. A server &SHOULD;
1381   read and forward a message-body on any request; if the request method
1382   does not include defined semantics for an entity-body, then the
1383   message-body &SHOULD; be ignored when handling the request.
1384</t>
1385<t>
1386   For response messages, whether or not a message-body is included with
1387   a message is dependent on both the request method and the response
1388   status code (<xref target="status.code.and.reason.phrase"/>). All responses to the HEAD request method
1389   &MUST-NOT; include a message-body, even though the presence of entity-header
1390   fields might lead one to believe they do. All 1xx
1391   (informational), 204 (no content), and 304 (not modified) responses
1392   &MUST-NOT; include a message-body. All other responses do include a
1393   message-body, although it &MAY; be of zero length.
1394</t>
1395</section>
1396
1397<section title="Message Length" anchor="message.length">
1398<t>
1399   The transfer-length of a message is the length of the message-body as
1400   it appears in the message; that is, after any transfer-codings have
1401   been applied. When a message-body is included with a message, the
1402   transfer-length of that body is determined by one of the following
1403   (in order of precedence):
1404</t>
1405<t>
1406  <list style="numbers">
1407    <x:lt><t>
1408     Any response message which "&MUST-NOT;" include a message-body (such
1409     as the 1xx, 204, and 304 responses and any response to a HEAD
1410     request) is always terminated by the first empty line after the
1411     header fields, regardless of the entity-header fields present in
1412     the message.
1413    </t></x:lt>
1414    <x:lt><t>
1415     If a Transfer-Encoding header field (<xref target="header.transfer-encoding"/>) is present and
1416     has any value other than "identity", then the transfer-length is
1417     defined by use of the "chunked" transfer-coding (<xref target="transfer.codings"/>),
1418     unless the message is terminated by closing the connection.
1419    </t></x:lt>
1420    <x:lt><t>
1421     If a Content-Length header field (<xref target="header.content-length"/>) is present, its
1422     decimal value in OCTETs represents both the entity-length and the
1423     transfer-length. The Content-Length header field &MUST-NOT; be sent
1424     if these two lengths are different (i.e., if a Transfer-Encoding
1425     header field is present). If a message is received with both a
1426     Transfer-Encoding header field and a Content-Length header field,
1427     the latter &MUST; be ignored.
1428    </t></x:lt>
1429    <x:lt><t>
1430     If the message uses the media type "multipart/byteranges", and the
1431     ransfer-length is not otherwise specified, then this self-elimiting
1432     media type defines the transfer-length. This media type
1433     UST NOT be used unless the sender knows that the recipient can arse
1434     it; the presence in a request of a Range header with ultiple byte-range
1435     specifiers from a 1.1 client implies that the lient can parse
1436     multipart/byteranges responses.
1437    <list style="empty"><t>
1438       A range header might be forwarded by a 1.0 proxy that does not
1439       understand multipart/byteranges; in this case the server &MUST;
1440       delimit the message using methods defined in items 1, 3 or 5 of
1441       this section.
1442    </t></list>
1443    </t></x:lt>
1444    <x:lt><t>
1445     By the server closing the connection. (Closing the connection
1446     cannot be used to indicate the end of a request body, since that
1447     would leave no possibility for the server to send back a response.)
1448    </t></x:lt>
1449  </list>
1450</t>
1451<t>
1452   For compatibility with HTTP/1.0 applications, HTTP/1.1 requests
1453   containing a message-body &MUST; include a valid Content-Length header
1454   field unless the server is known to be HTTP/1.1 compliant. If a
1455   request contains a message-body and a Content-Length is not given,
1456   the server &SHOULD; respond with 400 (bad request) if it cannot
1457   determine the length of the message, or with 411 (length required) if
1458   it wishes to insist on receiving a valid Content-Length.
1459</t>
1460<t>
1461   All HTTP/1.1 applications that receive entities &MUST; accept the
1462   "chunked" transfer-coding (<xref target="transfer.codings"/>), thus allowing this mechanism
1463   to be used for messages when the message length cannot be determined
1464   in advance.
1465</t>
1466<t>
1467   Messages &MUST-NOT; include both a Content-Length header field and a
1468   non-identity transfer-coding. If the message does include a non-identity
1469   transfer-coding, the Content-Length &MUST; be ignored.
1470</t>
1471<t>
1472   When a Content-Length is given in a message where a message-body is
1473   allowed, its field value &MUST; exactly match the number of OCTETs in
1474   the message-body. HTTP/1.1 user agents &MUST; notify the user when an
1475   invalid length is received and detected.
1476</t>
1477</section>
1478
1479<section title="General Header Fields" anchor="general.header.fields">
1480<t>
1481   There are a few header fields which have general applicability for
1482   both request and response messages, but which do not apply to the
1483   entity being transferred. These header fields apply only to the
1484   message being transmitted.
1485</t>
1486<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="general-header"/>
1487    general-header = Cache-Control            ; &header-cache-control;
1488                   | Connection               ; <xref target="header.connection"/>
1489                   | Date                     ; <xref target="header.date"/>
1490                   | Pragma                   ; &header-pragma;
1491                   | Trailer                  ; <xref target="header.trailer"/>
1492                   | Transfer-Encoding        ; <xref target="header.transfer-encoding"/>
1493                   | Upgrade                  ; <xref target="header.upgrade"/>
1494                   | Via                      ; <xref target="header.via"/>
1495                   | Warning                  ; &header-warning;
1496</artwork></figure>
1497<t>
1498   General-header field names can be extended reliably only in
1499   combination with a change in the protocol version. However, new or
1500   experimental header fields may be given the semantics of general
1501   header fields if all parties in the communication recognize them to
1502   be general-header fields. Unrecognized header fields are treated as
1503   entity-header fields.
1504</t>
1505</section>
1506</section>
1507
1508<section title="Request" anchor="request">
1509<t>
1510   A request message from a client to a server includes, within the
1511   first line of that message, the method to be applied to the resource,
1512   the identifier of the resource, and the protocol version in use.
1513</t>
1514<!--                 Host                      ; should be moved here eventually -->
1515<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request"/>
1516     Request       = Request-Line              ; <xref target="request-line"/>
1517                     *(( general-header        ; <xref target="general.header.fields"/>
1518                      | request-header         ; &request-header-fields;
1519                      | entity-header ) CRLF)  ; &entity-header-fields;
1520                     CRLF
1521                     [ message-body ]          ; <xref target="message.body"/>
1522</artwork></figure>
1523
1524<section title="Request-Line" anchor="request-line">
1525<t>
1526   The Request-Line begins with a method token, followed by the
1527   Request-URI and the protocol version, and ending with CRLF. The
1528   elements are separated by SP characters. No CR or LF is allowed
1529   except in the final CRLF sequence.
1530</t>
1531<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request-Line"/>
1532     Request-Line   = Method SP Request-URI SP HTTP-Version CRLF
1533</artwork></figure>
1534
1535<section title="Method" anchor="method">
1536<t>
1537   The Method  token indicates the method to be performed on the
1538   resource identified by the Request-URI. The method is case-sensitive.
1539</t>
1540<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Method"/><iref primary="true" item="Grammar" subitem="extension-method"/>
1541    Method         = token
1542</artwork></figure>
1543</section>
1544
1545<section title="Request-URI" anchor="request-uri">
1546<t>
1547   The Request-URI is a Uniform Resource Identifier (<xref target="uri"/>) and
1548   identifies the resource upon which to apply the request.
1549</t>
1550<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request-URI"/>
1551    Request-URI    = "*" | absoluteURI | abs_path | authority
1552</artwork></figure>
1553<t>
1554   The four options for Request-URI are dependent on the nature of the
1555   request. The asterisk "*" means that the request does not apply to a
1556   particular resource, but to the server itself, and is only allowed
1557   when the method used does not necessarily apply to a resource. One
1558   example would be
1559</t>
1560<figure><artwork type="example">
1561    OPTIONS * HTTP/1.1
1562</artwork></figure>
1563<t>
1564   The absoluteURI form is &REQUIRED; when the request is being made to a
1565   proxy. The proxy is requested to forward the request or service it
1566   from a valid cache, and return the response. Note that the proxy &MAY;
1567   forward the request on to another proxy or directly to the server
1568   specified by the absoluteURI. In order to avoid request loops, a
1569   proxy &MUST; be able to recognize all of its server names, including
1570   any aliases, local variations, and the numeric IP address. An example
1571   Request-Line would be:
1572</t>
1573<figure><artwork type="example">
1574    GET http://www.w3.org/pub/WWW/TheProject.html HTTP/1.1
1575</artwork></figure>
1576<t>
1577   To allow for transition to absoluteURIs in all requests in future
1578   versions of HTTP, all HTTP/1.1 servers &MUST; accept the absoluteURI
1579   form in requests, even though HTTP/1.1 clients will only generate
1580   them in requests to proxies.
1581</t>
1582<t>
1583   The authority form is only used by the CONNECT method (&CONNECT;).
1584</t>
1585<t>
1586   The most common form of Request-URI is that used to identify a
1587   resource on an origin server or gateway. In this case the absolute
1588   path of the URI &MUST; be transmitted (see <xref target="general.syntax"/>, abs_path) as
1589   the Request-URI, and the network location of the URI (authority) &MUST;
1590   be transmitted in a Host header field. For example, a client wishing
1591   to retrieve the resource above directly from the origin server would
1592   create a TCP connection to port 80 of the host "www.w3.org" and send
1593   the lines:
1594</t>
1595<figure><artwork type="example">
1596    GET /pub/WWW/TheProject.html HTTP/1.1
1597    Host: www.w3.org
1598</artwork></figure>
1599<t>
1600   followed by the remainder of the Request. Note that the absolute path
1601   cannot be empty; if none is present in the original URI, it &MUST; be
1602   given as "/" (the server root).
1603</t>
1604<t>
1605   The Request-URI is transmitted in the format specified in
1606   <xref target="general.syntax"/>. If the Request-URI is encoded using the "% HEX HEX" encoding
1607   <xref target="RFC2396"/>, the origin server &MUST; decode the Request-URI in order to
1608   properly interpret the request. Servers &SHOULD; respond to invalid
1609   Request-URIs with an appropriate status code.
1610</t>
1611<t>
1612   A transparent proxy &MUST-NOT; rewrite the "abs_path" part of the
1613   received Request-URI when forwarding it to the next inbound server,
1614   except as noted above to replace a null abs_path with "/".
1615</t>
1616<t>
1617  <list><t>
1618      <x:h>Note:</x:h> The "no rewrite" rule prevents the proxy from changing the
1619      meaning of the request when the origin server is improperly using
1620      a non-reserved URI character for a reserved purpose.  Implementors
1621      should be aware that some pre-HTTP/1.1 proxies have been known to
1622      rewrite the Request-URI.
1623  </t></list>
1624</t>
1625</section>
1626</section>
1627
1628<section title="The Resource Identified by a Request" anchor="the.resource.identified.by.a.request">
1629<t>
1630   The exact resource identified by an Internet request is determined by
1631   examining both the Request-URI and the Host header field.
1632</t>
1633<t>
1634   An origin server that does not allow resources to differ by the
1635   requested host &MAY; ignore the Host header field value when
1636   determining the resource identified by an HTTP/1.1 request. (But see
1637   <xref target="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses"/>
1638   for other requirements on Host support in HTTP/1.1.)
1639</t>
1640<t>
1641   An origin server that does differentiate resources based on the host
1642   requested (sometimes referred to as virtual hosts or vanity host
1643   names) &MUST; use the following rules for determining the requested
1644   resource on an HTTP/1.1 request:
1645  <list style="numbers">
1646    <t>If Request-URI is an absoluteURI, the host is part of the
1647     Request-URI. Any Host header field value in the request &MUST; be
1648     ignored.</t>
1649    <t>If the Request-URI is not an absoluteURI, and the request includes
1650     a Host header field, the host is determined by the Host header
1651     field value.</t>
1652    <t>If the host as determined by rule 1 or 2 is not a valid host on
1653     the server, the response &MUST; be a 400 (Bad Request) error message.</t>
1654  </list>
1655</t>
1656<t>
1657   Recipients of an HTTP/1.0 request that lacks a Host header field &MAY;
1658   attempt to use heuristics (e.g., examination of the URI path for
1659   something unique to a particular host) in order to determine what
1660   exact resource is being requested.
1661</t>
1662</section>
1663
1664</section>
1665
1666
1667<section title="Response" anchor="response">
1668<t>
1669   After receiving and interpreting a request message, a server responds
1670   with an HTTP response message.
1671</t>
1672<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Response"/>
1673    Response      = Status-Line               ; <xref target="status-line"/>
1674                    *(( general-header        ; <xref target="general.header.fields"/>
1675                     | response-header        ; &response-header-fields;
1676                     | entity-header ) CRLF)  ; &entity-header-fields;
1677                    CRLF
1678                    [ message-body ]          ; <xref target="message.body"/>
1679</artwork></figure>
1680
1681<section title="Status-Line" anchor="status-line">
1682<t>
1683   The first line of a Response message is the Status-Line, consisting
1684   of the protocol version followed by a numeric status code and its
1685   associated textual phrase, with each element separated by SP
1686   characters. No CR or LF is allowed except in the final CRLF sequence.
1687</t>
1688<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Status-Line"/>
1689    Status-Line = HTTP-Version SP Status-Code SP Reason-Phrase CRLF
1690</artwork></figure>
1691
1692<section title="Status Code and Reason Phrase" anchor="status.code.and.reason.phrase">
1693<t>
1694   The Status-Code element is a 3-digit integer result code of the
1695   attempt to understand and satisfy the request. These codes are fully
1696   defined in &status-codes;. The Reason-Phrase is intended to give a short
1697   textual description of the Status-Code. The Status-Code is intended
1698   for use by automata and the Reason-Phrase is intended for the human
1699   user. The client is not required to examine or display the Reason-Phrase.
1700</t>
1701<t>
1702   The first digit of the Status-Code defines the class of response. The
1703   last two digits do not have any categorization role. There are 5
1704   values for the first digit:
1705  <list style="symbols">
1706    <t>
1707      1xx: Informational - Request received, continuing process
1708    </t>
1709    <t>
1710      2xx: Success - The action was successfully received,
1711        understood, and accepted
1712    </t>
1713    <t>
1714      3xx: Redirection - Further action must be taken in order to
1715        complete the request
1716    </t>
1717    <t>
1718      4xx: Client Error - The request contains bad syntax or cannot
1719        be fulfilled
1720    </t>
1721    <t>
1722      5xx: Server Error - The server failed to fulfill an apparently
1723        valid request
1724    </t>
1725  </list>
1726</t>
1727<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"/>
1728   Status-Code    = 3DIGIT
1729   Reason-Phrase  = *&lt;TEXT, excluding CR, LF&gt;
1730</artwork></figure>
1731</section>
1732</section>
1733
1734</section>
1735
1736
1737<section title="Connections" anchor="connections">
1738
1739<section title="Persistent Connections" anchor="persistent.connections">
1740
1741<section title="Purpose" anchor="persistent.purpose">
1742<t>
1743   Prior to persistent connections, a separate TCP connection was
1744   established to fetch each URL, increasing the load on HTTP servers
1745   and causing congestion on the Internet. The use of inline images and
1746   other associated data often require a client to make multiple
1747   requests of the same server in a short amount of time. Analysis of
1748   these performance problems and results from a prototype
1749   implementation are available <xref target="Pad1995"/> <xref target="Spe"/>. Implementation experience and
1750   measurements of actual HTTP/1.1 (RFC 2068) implementations show good
1751   results <xref target="Nie1997"/>. Alternatives have also been explored, for example,
1752   T/TCP <xref target="Tou1998"/>.
1753</t>
1754<t>
1755   Persistent HTTP connections have a number of advantages:
1756  <list style="symbols">
1757      <t>
1758        By opening and closing fewer TCP connections, CPU time is saved
1759        in routers and hosts (clients, servers, proxies, gateways,
1760        tunnels, or caches), and memory used for TCP protocol control
1761        blocks can be saved in hosts.
1762      </t>
1763      <t>
1764        HTTP requests and responses can be pipelined on a connection.
1765        Pipelining allows a client to make multiple requests without
1766        waiting for each response, allowing a single TCP connection to
1767        be used much more efficiently, with much lower elapsed time.
1768      </t>
1769      <t>
1770        Network congestion is reduced by reducing the number of packets
1771        caused by TCP opens, and by allowing TCP sufficient time to
1772        determine the congestion state of the network.
1773      </t>
1774      <t>
1775        Latency on subsequent requests is reduced since there is no time
1776        spent in TCP's connection opening handshake.
1777      </t>
1778      <t>
1779        HTTP can evolve more gracefully, since errors can be reported
1780        without the penalty of closing the TCP connection. Clients using
1781        future versions of HTTP might optimistically try a new feature,
1782        but if communicating with an older server, retry with old
1783        semantics after an error is reported.
1784      </t>
1785    </list>
1786</t>
1787<t>
1788   HTTP implementations &SHOULD; implement persistent connections.
1789</t>
1790</section>
1791
1792<section title="Overall Operation" anchor="persistent.overall">
1793<t>
1794   A significant difference between HTTP/1.1 and earlier versions of
1795   HTTP is that persistent connections are the default behavior of any
1796   HTTP connection. That is, unless otherwise indicated, the client
1797   &SHOULD; assume that the server will maintain a persistent connection,
1798   even after error responses from the server.
1799</t>
1800<t>
1801   Persistent connections provide a mechanism by which a client and a
1802   server can signal the close of a TCP connection. This signaling takes
1803   place using the Connection header field (<xref target="header.connection"/>). Once a close
1804   has been signaled, the client &MUST-NOT; send any more requests on that
1805   connection.
1806</t>
1807
1808<section title="Negotiation" anchor="persistent.negotiation">
1809<t>
1810   An HTTP/1.1 server &MAY; assume that a HTTP/1.1 client intends to
1811   maintain a persistent connection unless a Connection header including
1812   the connection-token "close" was sent in the request. If the server
1813   chooses to close the connection immediately after sending the
1814   response, it &SHOULD; send a Connection header including the
1815   connection-token close.
1816</t>
1817<t>
1818   An HTTP/1.1 client &MAY; expect a connection to remain open, but would
1819   decide to keep it open based on whether the response from a server
1820   contains a Connection header with the connection-token close. In case
1821   the client does not want to maintain a connection for more than that
1822   request, it &SHOULD; send a Connection header including the
1823   connection-token close.
1824</t>
1825<t>
1826   If either the client or the server sends the close token in the
1827   Connection header, that request becomes the last one for the
1828   connection.
1829</t>
1830<t>
1831   Clients and servers &SHOULD-NOT;  assume that a persistent connection is
1832   maintained for HTTP versions less than 1.1 unless it is explicitly
1833   signaled. See <xref target="compatibility.with.http.1.0.persistent.connections"/> for more information on backward
1834   compatibility with HTTP/1.0 clients.
1835</t>
1836<t>
1837   In order to remain persistent, all messages on the connection &MUST;
1838   have a self-defined message length (i.e., one not defined by closure
1839   of the connection), as described in <xref target="message.length"/>.
1840</t>
1841</section>
1842
1843<section title="Pipelining" anchor="pipelining">
1844<t>
1845   A client that supports persistent connections &MAY; "pipeline" its
1846   requests (i.e., send multiple requests without waiting for each
1847   response). A server &MUST; send its responses to those requests in the
1848   same order that the requests were received.
1849</t>
1850<t>
1851   Clients which assume persistent connections and pipeline immediately
1852   after connection establishment &SHOULD; be prepared to retry their
1853   connection if the first pipelined attempt fails. If a client does
1854   such a retry, it &MUST-NOT; pipeline before it knows the connection is
1855   persistent. Clients &MUST; also be prepared to resend their requests if
1856   the server closes the connection before sending all of the
1857   corresponding responses.
1858</t>
1859<t>
1860   Clients &SHOULD-NOT;  pipeline requests using non-idempotent methods or
1861   non-idempotent sequences of methods (see &idempotent-methods;). Otherwise, a
1862   premature termination of the transport connection could lead to
1863   indeterminate results. A client wishing to send a non-idempotent
1864   request &SHOULD; wait to send that request until it has received the
1865   response status for the previous request.
1866</t>
1867</section>
1868</section>
1869
1870<section title="Proxy Servers" anchor="persistent.proxy">
1871<t>
1872   It is especially important that proxies correctly implement the
1873   properties of the Connection header field as specified in <xref target="header.connection"/>.
1874</t>
1875<t>
1876   The proxy server &MUST; signal persistent connections separately with
1877   its clients and the origin servers (or other proxy servers) that it
1878   connects to. Each persistent connection applies to only one transport
1879   link.
1880</t>
1881<t>
1882   A proxy server &MUST-NOT; establish a HTTP/1.1 persistent connection
1883   with an HTTP/1.0 client (but see RFC 2068 <xref target="RFC2068"/> for information and
1884   discussion of the problems with the Keep-Alive header implemented by
1885   many HTTP/1.0 clients).
1886</t>
1887</section>
1888
1889<section title="Practical Considerations" anchor="persistent.practical">
1890<t>
1891   Servers will usually have some time-out value beyond which they will
1892   no longer maintain an inactive connection. Proxy servers might make
1893   this a higher value since it is likely that the client will be making
1894   more connections through the same server. The use of persistent
1895   connections places no requirements on the length (or existence) of
1896   this time-out for either the client or the server.
1897</t>
1898<t>
1899   When a client or server wishes to time-out it &SHOULD; issue a graceful
1900   close on the transport connection. Clients and servers &SHOULD; both
1901   constantly watch for the other side of the transport close, and
1902   respond to it as appropriate. If a client or server does not detect
1903   the other side's close promptly it could cause unnecessary resource
1904   drain on the network.
1905</t>
1906<t>
1907   A client, server, or proxy &MAY; close the transport connection at any
1908   time. For example, a client might have started to send a new request
1909   at the same time that the server has decided to close the "idle"
1910   connection. From the server's point of view, the connection is being
1911   closed while it was idle, but from the client's point of view, a
1912   request is in progress.
1913</t>
1914<t>
1915   This means that clients, servers, and proxies &MUST; be able to recover
1916   from asynchronous close events. Client software &SHOULD; reopen the
1917   transport connection and retransmit the aborted sequence of requests
1918   without user interaction so long as the request sequence is
1919   idempotent (see &idempotent-methods;). Non-idempotent methods or sequences
1920   &MUST-NOT; be automatically retried, although user agents &MAY; offer a
1921   human operator the choice of retrying the request(s). Confirmation by
1922   user-agent software with semantic understanding of the application
1923   &MAY; substitute for user confirmation. The automatic retry &SHOULD-NOT; 
1924   be repeated if the second sequence of requests fails.
1925</t>
1926<t>
1927   Servers &SHOULD; always respond to at least one request per connection,
1928   if at all possible. Servers &SHOULD-NOT;  close a connection in the
1929   middle of transmitting a response, unless a network or client failure
1930   is suspected.
1931</t>
1932<t>
1933   Clients that use persistent connections &SHOULD; limit the number of
1934   simultaneous connections that they maintain to a given server. A
1935   single-user client &SHOULD-NOT; maintain more than 2 connections with
1936   any server or proxy. A proxy &SHOULD; use up to 2*N connections to
1937   another server or proxy, where N is the number of simultaneously
1938   active users. These guidelines are intended to improve HTTP response
1939   times and avoid congestion.
1940</t>
1941</section>
1942</section>
1943
1944<section title="Message Transmission Requirements" anchor="message.transmission.requirements">
1945
1946<section title="Persistent Connections and Flow Control" anchor="persistent.flow">
1947<t>
1948   HTTP/1.1 servers &SHOULD; maintain persistent connections and use TCP's
1949   flow control mechanisms to resolve temporary overloads, rather than
1950   terminating connections with the expectation that clients will retry.
1951   The latter technique can exacerbate network congestion.
1952</t>
1953</section>
1954
1955<section title="Monitoring Connections for Error Status Messages" anchor="persistent.monitor">
1956<t>
1957   An HTTP/1.1 (or later) client sending a message-body &SHOULD; monitor
1958   the network connection for an error status while it is transmitting
1959   the request. If the client sees an error status, it &SHOULD;
1960   immediately cease transmitting the body. If the body is being sent
1961   using a "chunked" encoding (<xref target="transfer.codings"/>), a zero length chunk and
1962   empty trailer &MAY; be used to prematurely mark the end of the message.
1963   If the body was preceded by a Content-Length header, the client &MUST;
1964   close the connection.
1965</t>
1966</section>
1967
1968<section title="Use of the 100 (Continue) Status" anchor="use.of.the.100.status">
1969<t>
1970   The purpose of the 100 (Continue) status (see &status-100;) is to
1971   allow a client that is sending a request message with a request body
1972   to determine if the origin server is willing to accept the request
1973   (based on the request headers) before the client sends the request
1974   body. In some cases, it might either be inappropriate or highly
1975   inefficient for the client to send the body if the server will reject
1976   the message without looking at the body.
1977</t>
1978<t>
1979   Requirements for HTTP/1.1 clients:
1980  <list style="symbols">
1981    <t>
1982        If a client will wait for a 100 (Continue) response before
1983        sending the request body, it &MUST; send an Expect request-header
1984        field (&header-expect;) with the "100-continue" expectation.
1985    </t>
1986    <t>
1987        A client &MUST-NOT; send an Expect request-header field (&header-expect;)
1988        with the "100-continue" expectation if it does not intend
1989        to send a request body.
1990    </t>
1991  </list>
1992</t>
1993<t>
1994   Because of the presence of older implementations, the protocol allows
1995   ambiguous situations in which a client may send "Expect: 100-continue"
1996   without receiving either a 417 (Expectation Failed) status
1997   or a 100 (Continue) status. Therefore, when a client sends this
1998   header field to an origin server (possibly via a proxy) from which it
1999   has never seen a 100 (Continue) status, the client &SHOULD-NOT;  wait
2000   for an indefinite period before sending the request body.
2001</t>
2002<t>
2003   Requirements for HTTP/1.1 origin servers:
2004  <list style="symbols">
2005    <t> Upon receiving a request which includes an Expect request-header
2006        field with the "100-continue" expectation, an origin server &MUST;
2007        either respond with 100 (Continue) status and continue to read
2008        from the input stream, or respond with a final status code. The
2009        origin server &MUST-NOT; wait for the request body before sending
2010        the 100 (Continue) response. If it responds with a final status
2011        code, it &MAY; close the transport connection or it &MAY; continue
2012        to read and discard the rest of the request.  It &MUST-NOT;
2013        perform the requested method if it returns a final status code.
2014    </t>
2015    <t> An origin server &SHOULD-NOT;  send a 100 (Continue) response if
2016        the request message does not include an Expect request-header
2017        field with the "100-continue" expectation, and &MUST-NOT; send a
2018        100 (Continue) response if such a request comes from an HTTP/1.0
2019        (or earlier) client. There is an exception to this rule: for
2020        compatibility with RFC 2068, a server &MAY; send a 100 (Continue)
2021        status in response to an HTTP/1.1 PUT or POST request that does
2022        not include an Expect request-header field with the "100-continue"
2023        expectation. This exception, the purpose of which is
2024        to minimize any client processing delays associated with an
2025        undeclared wait for 100 (Continue) status, applies only to
2026        HTTP/1.1 requests, and not to requests with any other HTTP-version
2027        value.
2028    </t>
2029    <t> An origin server &MAY; omit a 100 (Continue) response if it has
2030        already received some or all of the request body for the
2031        corresponding request.
2032    </t>
2033    <t> An origin server that sends a 100 (Continue) response &MUST;
2034    ultimately send a final status code, once the request body is
2035        received and processed, unless it terminates the transport
2036        connection prematurely.
2037    </t>
2038    <t> If an origin server receives a request that does not include an
2039        Expect request-header field with the "100-continue" expectation,
2040        the request includes a request body, and the server responds
2041        with a final status code before reading the entire request body
2042        from the transport connection, then the server &SHOULD-NOT;  close
2043        the transport connection until it has read the entire request,
2044        or until the client closes the connection. Otherwise, the client
2045        might not reliably receive the response message. However, this
2046        requirement is not be construed as preventing a server from
2047        defending itself against denial-of-service attacks, or from
2048        badly broken client implementations.
2049      </t>
2050    </list>
2051</t>
2052<t>
2053   Requirements for HTTP/1.1 proxies:
2054  <list style="symbols">
2055    <t> If a proxy receives a request that includes an Expect request-header
2056        field with the "100-continue" expectation, and the proxy
2057        either knows that the next-hop server complies with HTTP/1.1 or
2058        higher, or does not know the HTTP version of the next-hop
2059        server, it &MUST; forward the request, including the Expect header
2060        field.
2061    </t>
2062    <t> If the proxy knows that the version of the next-hop server is
2063        HTTP/1.0 or lower, it &MUST-NOT; forward the request, and it &MUST;
2064        respond with a 417 (Expectation Failed) status.
2065    </t>
2066    <t> Proxies &SHOULD; maintain a cache recording the HTTP version
2067        numbers received from recently-referenced next-hop servers.
2068    </t>
2069    <t> A proxy &MUST-NOT; forward a 100 (Continue) response if the
2070        request message was received from an HTTP/1.0 (or earlier)
2071        client and did not include an Expect request-header field with
2072        the "100-continue" expectation. This requirement overrides the
2073        general rule for forwarding of 1xx responses (see &status-1xx;).
2074    </t>
2075  </list>
2076</t>
2077</section>
2078
2079<section title="Client Behavior if Server Prematurely Closes Connection" anchor="connection.premature">
2080<t>
2081   If an HTTP/1.1 client sends a request which includes a request body,
2082   but which does not include an Expect request-header field with the
2083   "100-continue" expectation, and if the client is not directly
2084   connected to an HTTP/1.1 origin server, and if the client sees the
2085   connection close before receiving any status from the server, the
2086   client &SHOULD; retry the request.  If the client does retry this
2087   request, it &MAY; use the following "binary exponential backoff"
2088   algorithm to be assured of obtaining a reliable response:
2089  <list style="numbers">
2090    <t>
2091      Initiate a new connection to the server
2092    </t>
2093    <t>
2094      Transmit the request-headers
2095    </t>
2096    <t>
2097      Initialize a variable R to the estimated round-trip time to the
2098         server (e.g., based on the time it took to establish the
2099         connection), or to a constant value of 5 seconds if the round-trip
2100         time is not available.
2101    </t>
2102    <t>
2103       Compute T = R * (2**N), where N is the number of previous
2104         retries of this request.
2105    </t>
2106    <t>
2107       Wait either for an error response from the server, or for T
2108         seconds (whichever comes first)
2109    </t>
2110    <t>
2111       If no error response is received, after T seconds transmit the
2112         body of the request.
2113    </t>
2114    <t>
2115       If client sees that the connection is closed prematurely,
2116         repeat from step 1 until the request is accepted, an error
2117         response is received, or the user becomes impatient and
2118         terminates the retry process.
2119    </t>
2120  </list>
2121</t>
2122<t>
2123   If at any point an error status is received, the client
2124  <list style="symbols">
2125      <t>&SHOULD-NOT;  continue and</t>
2126
2127      <t>&SHOULD; close the connection if it has not completed sending the
2128        request message.</t>
2129    </list>
2130</t>
2131</section>
2132</section>
2133</section>
2134
2135
2136<section title="Header Field Definitions" anchor="header.fields">
2137<t>
2138   This section defines the syntax and semantics of all standard
2139   HTTP/1.1 header fields. For entity-header fields, both sender and
2140   recipient refer to either the client or the server, depending on who
2141   sends and who receives the entity.
2142</t>
2143
2144<section title="Connection" anchor="header.connection">
2145  <iref primary="true" item="Connection header" x:for-anchor=""/>
2146  <iref primary="true" item="Headers" subitem="Connection" x:for-anchor=""/>
2147<t>
2148   The Connection general-header field allows the sender to specify
2149   options that are desired for that particular connection and &MUST-NOT;
2150   be communicated by proxies over further connections.
2151</t>
2152<t>
2153   The Connection header has the following grammar:
2154</t>
2155<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="connection-token"/>
2156    Connection = "Connection" ":" 1#(connection-token)
2157    connection-token  = token
2158</artwork></figure>
2159<t>
2160   HTTP/1.1 proxies &MUST; parse the Connection header field before a
2161   message is forwarded and, for each connection-token in this field,
2162   remove any header field(s) from the message with the same name as the
2163   connection-token. Connection options are signaled by the presence of
2164   a connection-token in the Connection header field, not by any
2165   corresponding additional header field(s), since the additional header
2166   field may not be sent if there are no parameters associated with that
2167   connection option.
2168</t>
2169<t>
2170   Message headers listed in the Connection header &MUST-NOT; include
2171   end-to-end headers, such as Cache-Control.
2172</t>
2173<t>
2174   HTTP/1.1 defines the "close" connection option for the sender to
2175   signal that the connection will be closed after completion of the
2176   response. For example,
2177</t>
2178<figure><artwork type="example">
2179    Connection: close
2180</artwork></figure>
2181<t>
2182   in either the request or the response header fields indicates that
2183   the connection &SHOULD-NOT;  be considered `persistent' (<xref target="persistent.connections"/>)
2184   after the current request/response is complete.
2185</t>
2186<t>
2187   HTTP/1.1 applications that do not support persistent connections &MUST;
2188   include the "close" connection option in every message.
2189</t>
2190<t>
2191   A system receiving an HTTP/1.0 (or lower-version) message that
2192   includes a Connection header MUST, for each connection-token in this
2193   field, remove and ignore any header field(s) from the message with
2194   the same name as the connection-token. This protects against mistaken
2195   forwarding of such header fields by pre-HTTP/1.1 proxies. See <xref target="compatibility.with.http.1.0.persistent.connections"/>.
2196</t>
2197</section>
2198
2199<section title="Content-Length" anchor="header.content-length">
2200  <iref primary="true" item="Content-Length header" x:for-anchor=""/>
2201  <iref primary="true" item="Headers" subitem="Content-Length" x:for-anchor=""/>
2202<t>
2203   The Content-Length entity-header field indicates the size of the
2204   entity-body, in decimal number of OCTETs, sent to the recipient or,
2205   in the case of the HEAD method, the size of the entity-body that
2206   would have been sent had the request been a GET.
2207</t>
2208<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Length"/>
2209    Content-Length    = "Content-Length" ":" 1*DIGIT
2210</artwork></figure>
2211<t>
2212   An example is
2213</t>
2214<figure><artwork type="example">
2215    Content-Length: 3495
2216</artwork></figure>
2217<t>
2218   Applications &SHOULD; use this field to indicate the transfer-length of
2219   the message-body, unless this is prohibited by the rules in <xref target="message.length"/>.
2220</t>
2221<t>
2222   Any Content-Length greater than or equal to zero is a valid value.
2223   <xref target="message.length"/> describes how to determine the length of a message-body
2224   if a Content-Length is not given.
2225</t>
2226<t>
2227   Note that the meaning of this field is significantly different from
2228   the corresponding definition in MIME, where it is an optional field
2229   used within the "message/external-body" content-type. In HTTP, it
2230   &SHOULD; be sent whenever the message's length can be determined prior
2231   to being transferred, unless this is prohibited by the rules in
2232   <xref target="message.length"/>.
2233</t>
2234</section>
2235
2236<section title="Date" anchor="header.date">
2237  <iref primary="true" item="Date header" x:for-anchor=""/>
2238  <iref primary="true" item="Headers" subitem="Date" x:for-anchor=""/>
2239<t>
2240   The Date general-header field represents the date and time at which
2241   the message was originated, having the same semantics as orig-date in
2242   RFC 822. The field value is an HTTP-date, as described in <xref target="full.date"/>;
2243   it &MUST; be sent in RFC 1123 <xref target="RFC1123"/>-date format.
2244</t>
2245<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Date"/>
2246    Date  = "Date" ":" HTTP-date
2247</artwork></figure>
2248<t>
2249   An example is
2250</t>
2251<figure><artwork type="example">
2252    Date: Tue, 15 Nov 1994 08:12:31 GMT
2253</artwork></figure>
2254<t>
2255   Origin servers &MUST; include a Date header field in all responses,
2256   except in these cases:
2257  <list style="numbers">
2258      <t>If the response status code is 100 (Continue) or 101 (Switching
2259         Protocols), the response &MAY; include a Date header field, at
2260         the server's option.</t>
2261
2262      <t>If the response status code conveys a server error, e.g. 500
2263         (Internal Server Error) or 503 (Service Unavailable), and it is
2264         inconvenient or impossible to generate a valid Date.</t>
2265
2266      <t>If the server does not have a clock that can provide a
2267         reasonable approximation of the current time, its responses
2268         &MUST-NOT; include a Date header field. In this case, the rules
2269         in <xref target="clockless.origin.server.operation"/> &MUST; be followed.</t>
2270  </list>
2271</t>
2272<t>
2273   A received message that does not have a Date header field &MUST; be
2274   assigned one by the recipient if the message will be cached by that
2275   recipient or gatewayed via a protocol which requires a Date. An HTTP
2276   implementation without a clock &MUST-NOT; cache responses without
2277   revalidating them on every use. An HTTP cache, especially a shared
2278   cache, &SHOULD; use a mechanism, such as NTP <xref target="RFC1305"/>, to synchronize its
2279   clock with a reliable external standard.
2280</t>
2281<t>
2282   Clients &SHOULD; only send a Date header field in messages that include
2283   an entity-body, as in the case of the PUT and POST requests, and even
2284   then it is optional. A client without a clock &MUST-NOT; send a Date
2285   header field in a request.
2286</t>
2287<t>
2288   The HTTP-date sent in a Date header &SHOULD-NOT;  represent a date and
2289   time subsequent to the generation of the message. It &SHOULD; represent
2290   the best available approximation of the date and time of message
2291   generation, unless the implementation has no means of generating a
2292   reasonably accurate date and time. In theory, the date ought to
2293   represent the moment just before the entity is generated. In
2294   practice, the date can be generated at any time during the message
2295   origination without affecting its semantic value.
2296</t>
2297
2298<section title="Clockless Origin Server Operation" anchor="clockless.origin.server.operation">
2299<t>
2300   Some origin server implementations might not have a clock available.
2301   An origin server without a clock &MUST-NOT; assign Expires or Last-Modified
2302   values to a response, unless these values were associated
2303   with the resource by a system or user with a reliable clock. It &MAY;
2304   assign an Expires value that is known, at or before server
2305   configuration time, to be in the past (this allows "pre-expiration"
2306   of responses without storing separate Expires values for each
2307   resource).
2308</t>
2309</section>
2310</section>
2311
2312<section title="Host" anchor="header.host">
2313  <iref primary="true" item="Host header" x:for-anchor=""/>
2314  <iref primary="true" item="Headers" subitem="Host" x:for-anchor=""/>
2315<t>
2316   The Host request-header field specifies the Internet host and port
2317   number of the resource being requested, as obtained from the original
2318   URI given by the user or referring resource (generally an HTTP URL,
2319   as described in <xref target="http.url"/>). The Host field value &MUST; represent
2320   the naming authority of the origin server or gateway given by the
2321   original URL. This allows the origin server or gateway to
2322   differentiate between internally-ambiguous URLs, such as the root "/"
2323   URL of a server for multiple host names on a single IP address.
2324</t>
2325<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Host"/>
2326    Host = "Host" ":" host [ ":" port ] ; <xref target="http.url"/>
2327</artwork></figure>
2328<t>
2329   A "host" without any trailing port information implies the default
2330   port for the service requested (e.g., "80" for an HTTP URL). For
2331   example, a request on the origin server for
2332   &lt;http://www.w3.org/pub/WWW/&gt; would properly include:
2333</t>
2334<figure><artwork type="example">
2335    GET /pub/WWW/ HTTP/1.1
2336    Host: www.w3.org
2337</artwork></figure>
2338<t>
2339   A client &MUST; include a Host header field in all HTTP/1.1 request
2340   messages . If the requested URI does not include an Internet host
2341   name for the service being requested, then the Host header field &MUST;
2342   be given with an empty value. An HTTP/1.1 proxy &MUST; ensure that any
2343   request message it forwards does contain an appropriate Host header
2344   field that identifies the service being requested by the proxy. All
2345   Internet-based HTTP/1.1 servers &MUST; respond with a 400 (Bad Request)
2346   status code to any HTTP/1.1 request message which lacks a Host header
2347   field.
2348</t>
2349<t>
2350   See sections <xref target="the.resource.identified.by.a.request" format="counter"/>
2351   and <xref target="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses" format="counter"/>
2352   for other requirements relating to Host.
2353</t>
2354</section>
2355
2356<section title="TE" anchor="header.te">
2357  <iref primary="true" item="TE header" x:for-anchor=""/>
2358  <iref primary="true" item="Headers" subitem="TE" x:for-anchor=""/>
2359<t>
2360   The TE request-header field indicates what extension transfer-codings
2361   it is willing to accept in the response and whether or not it is
2362   willing to accept trailer fields in a chunked transfer-coding. Its
2363   value may consist of the keyword "trailers" and/or a comma-separated
2364   list of extension transfer-coding names with optional accept
2365   parameters (as described in <xref target="transfer.codings"/>).
2366</t>
2367<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="t-codings"/>
2368    TE        = "TE" ":" #( t-codings )
2369    t-codings = "trailers" | ( transfer-extension [ accept-params ] )
2370</artwork></figure>
2371<t>
2372   The presence of the keyword "trailers" indicates that the client is
2373   willing to accept trailer fields in a chunked transfer-coding, as
2374   defined in <xref target="chunked.transfer.encoding"/>. This keyword is reserved for use with
2375   transfer-coding values even though it does not itself represent a
2376   transfer-coding.
2377</t>
2378<t>
2379   Examples of its use are:
2380</t>
2381<figure><artwork type="example">
2382    TE: deflate
2383    TE:
2384    TE: trailers, deflate;q=0.5
2385</artwork></figure>
2386<t>
2387   The TE header field only applies to the immediate connection.
2388   Therefore, the keyword &MUST; be supplied within a Connection header
2389   field (<xref target="header.connection"/>) whenever TE is present in an HTTP/1.1 message.
2390</t>
2391<t>
2392   A server tests whether a transfer-coding is acceptable, according to
2393   a TE field, using these rules:
2394  <list style="numbers">
2395    <x:lt>
2396      <t>The "chunked" transfer-coding is always acceptable. If the
2397         keyword "trailers" is listed, the client indicates that it is
2398         willing to accept trailer fields in the chunked response on
2399         behalf of itself and any downstream clients. The implication is
2400         that, if given, the client is stating that either all
2401         downstream clients are willing to accept trailer fields in the
2402         forwarded response, or that it will attempt to buffer the
2403         response on behalf of downstream recipients.
2404      </t><t>
2405         <x:h>Note:</x:h> HTTP/1.1 does not define any means to limit the size of a
2406         chunked response such that a client can be assured of buffering
2407         the entire response.</t>
2408    </x:lt>
2409    <x:lt>
2410      <t>If the transfer-coding being tested is one of the transfer-codings
2411         listed in the TE field, then it is acceptable unless it
2412         is accompanied by a qvalue of 0. (As defined in &qvalue;, a
2413         qvalue of 0 means "not acceptable.")</t>
2414    </x:lt>
2415    <x:lt>
2416      <t>If multiple transfer-codings are acceptable, then the
2417         acceptable transfer-coding with the highest non-zero qvalue is
2418         preferred.  The "chunked" transfer-coding always has a qvalue
2419         of 1.</t>
2420    </x:lt>
2421  </list>
2422</t>
2423<t>
2424   If the TE field-value is empty or if no TE field is present, the only
2425   transfer-coding  is "chunked". A message with no transfer-coding is
2426   always acceptable.
2427</t>
2428</section>
2429
2430<section title="Trailer" anchor="header.trailer">
2431  <iref primary="true" item="Trailer header" x:for-anchor=""/>
2432  <iref primary="true" item="Headers" subitem="Trailer" x:for-anchor=""/>
2433<t>
2434   The Trailer general field value indicates that the given set of
2435   header fields is present in the trailer of a message encoded with
2436   chunked transfer-coding.
2437</t>
2438<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Trailer"/>
2439    Trailer  = "Trailer" ":" 1#field-name
2440</artwork></figure>
2441<t>
2442   An HTTP/1.1 message &SHOULD; include a Trailer header field in a
2443   message using chunked transfer-coding with a non-empty trailer. Doing
2444   so allows the recipient to know which header fields to expect in the
2445   trailer.
2446</t>
2447<t>
2448   If no Trailer header field is present, the trailer &SHOULD-NOT;  include
2449   any header fields. See <xref target="chunked.transfer.encoding"/> for restrictions on the use of
2450   trailer fields in a "chunked" transfer-coding.
2451</t>
2452<t>
2453   Message header fields listed in the Trailer header field &MUST-NOT;
2454   include the following header fields:
2455  <list style="symbols">
2456    <t>Transfer-Encoding</t>
2457    <t>Content-Length</t>
2458    <t>Trailer</t>
2459  </list>
2460</t>
2461</section>
2462
2463<section title="Transfer-Encoding" anchor="header.transfer-encoding">
2464  <iref primary="true" item="Transfer-Encoding header" x:for-anchor=""/>
2465  <iref primary="true" item="Headers" subitem="Transfer-Encoding" x:for-anchor=""/>
2466<t>
2467   The Transfer-Encoding general-header field indicates what (if any)
2468   type of transformation has been applied to the message body in order
2469   to safely transfer it between the sender and the recipient. This
2470   differs from the content-coding in that the transfer-coding is a
2471   property of the message, not of the entity.
2472</t>
2473<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/>
2474  Transfer-Encoding       = "Transfer-Encoding" ":" 1#transfer-coding
2475</artwork></figure>
2476<t>
2477   Transfer-codings are defined in <xref target="transfer.codings"/>. An example is:
2478</t>
2479<figure><artwork type="example">
2480  Transfer-Encoding: chunked
2481</artwork></figure>
2482<t>
2483   If multiple encodings have been applied to an entity, the transfer-codings
2484   &MUST; be listed in the order in which they were applied.
2485   Additional information about the encoding parameters &MAY; be provided
2486   by other entity-header fields not defined by this specification.
2487</t>
2488<t>
2489   Many older HTTP/1.0 applications do not understand the Transfer-Encoding
2490   header.
2491</t>
2492</section>
2493
2494<section title="Upgrade" anchor="header.upgrade">
2495  <iref primary="true" item="Upgrade header" x:for-anchor=""/>
2496  <iref primary="true" item="Headers" subitem="Upgrade" x:for-anchor=""/>
2497<t>
2498   The Upgrade general-header allows the client to specify what
2499   additional communication protocols it supports and would like to use
2500   if the server finds it appropriate to switch protocols. The server
2501   &MUST; use the Upgrade header field within a 101 (Switching Protocols)
2502   response to indicate which protocol(s) are being switched.
2503</t>
2504<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Upgrade"/>
2505    Upgrade        = "Upgrade" ":" 1#product
2506</artwork></figure>
2507<t>
2508   For example,
2509</t>
2510<figure><artwork type="example">
2511    Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
2512</artwork></figure>
2513<t>
2514   The Upgrade header field is intended to provide a simple mechanism
2515   for transition from HTTP/1.1 to some other, incompatible protocol. It
2516   does so by allowing the client to advertise its desire to use another
2517   protocol, such as a later version of HTTP with a higher major version
2518   number, even though the current request has been made using HTTP/1.1.
2519   This eases the difficult transition between incompatible protocols by
2520   allowing the client to initiate a request in the more commonly
2521   supported protocol while indicating to the server that it would like
2522   to use a "better" protocol if available (where "better" is determined
2523   by the server, possibly according to the nature of the method and/or
2524   resource being requested).
2525</t>
2526<t>
2527   The Upgrade header field only applies to switching application-layer
2528   protocols upon the existing transport-layer connection. Upgrade
2529   cannot be used to insist on a protocol change; its acceptance and use
2530   by the server is optional. The capabilities and nature of the
2531   application-layer communication after the protocol change is entirely
2532   dependent upon the new protocol chosen, although the first action
2533   after changing the protocol &MUST; be a response to the initial HTTP
2534   request containing the Upgrade header field.
2535</t>
2536<t>
2537   The Upgrade header field only applies to the immediate connection.
2538   Therefore, the upgrade keyword &MUST; be supplied within a Connection
2539   header field (<xref target="header.connection"/>) whenever Upgrade is present in an
2540   HTTP/1.1 message.
2541</t>
2542<t>
2543   The Upgrade header field cannot be used to indicate a switch to a
2544   protocol on a different connection. For that purpose, it is more
2545   appropriate to use a 301, 302, 303, or 305 redirection response.
2546</t>
2547<t>
2548   This specification only defines the protocol name "HTTP" for use by
2549   the family of Hypertext Transfer Protocols, as defined by the HTTP
2550   version rules of <xref target="http.version"/> and future updates to this
2551   specification. Any token can be used as a protocol name; however, it
2552   will only be useful if both the client and server associate the name
2553   with the same protocol.
2554</t>
2555</section>
2556
2557<section title="Via" anchor="header.via">
2558  <iref primary="true" item="Via header" x:for-anchor=""/>
2559  <iref primary="true" item="Headers" subitem="Via" x:for-anchor=""/>
2560<t>
2561   The Via general-header field &MUST; be used by gateways and proxies to
2562   indicate the intermediate protocols and recipients between the user
2563   agent and the server on requests, and between the origin server and
2564   the client on responses. It is analogous to the "Received" field of
2565   RFC 822 <xref target="RFC822"/> and is intended to be used for tracking message forwards,
2566   avoiding request loops, and identifying the protocol capabilities of
2567   all senders along the request/response chain.
2568</t>
2569<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"/>
2570   Via =  "Via" ":" 1#( received-protocol received-by [ comment ] )
2571   received-protocol = [ protocol-name "/" ] protocol-version
2572   protocol-name     = token
2573   protocol-version  = token
2574   received-by       = ( host [ ":" port ] ) | pseudonym
2575   pseudonym         = token
2576</artwork></figure>
2577<t>
2578   The received-protocol indicates the protocol version of the message
2579   received by the server or client along each segment of the
2580   request/response chain. The received-protocol version is appended to
2581   the Via field value when the message is forwarded so that information
2582   about the protocol capabilities of upstream applications remains
2583   visible to all recipients.
2584</t>
2585<t>
2586   The protocol-name is optional if and only if it would be "HTTP". The
2587   received-by field is normally the host and optional port number of a
2588   recipient server or client that subsequently forwarded the message.
2589   However, if the real host is considered to be sensitive information,
2590   it &MAY; be replaced by a pseudonym. If the port is not given, it &MAY;
2591   be assumed to be the default port of the received-protocol.
2592</t>
2593<t>
2594   Multiple Via field values represents each proxy or gateway that has
2595   forwarded the message. Each recipient &MUST; append its information
2596   such that the end result is ordered according to the sequence of
2597   forwarding applications.
2598</t>
2599<t>
2600   Comments &MAY; be used in the Via header field to identify the software
2601   of the recipient proxy or gateway, analogous to the User-Agent and
2602   Server header fields. However, all comments in the Via field are
2603   optional and &MAY; be removed by any recipient prior to forwarding the
2604   message.
2605</t>
2606<t>
2607   For example, a request message could be sent from an HTTP/1.0 user
2608   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
2609   forward the request to a public proxy at nowhere.com, which completes
2610   the request by forwarding it to the origin server at www.ics.uci.edu.
2611   The request received by www.ics.uci.edu would then have the following
2612   Via header field:
2613</t>
2614<figure><artwork type="example">
2615    Via: 1.0 fred, 1.1 nowhere.com (Apache/1.1)
2616</artwork></figure>
2617<t>
2618   Proxies and gateways used as a portal through a network firewall
2619   &SHOULD-NOT;, by default, forward the names and ports of hosts within
2620   the firewall region. This information &SHOULD; only be propagated if
2621   explicitly enabled. If not enabled, the received-by host of any host
2622   behind the firewall &SHOULD; be replaced by an appropriate pseudonym
2623   for that host.
2624</t>
2625<t>
2626   For organizations that have strong privacy requirements for hiding
2627   internal structures, a proxy &MAY; combine an ordered subsequence of
2628   Via header field entries with identical received-protocol values into
2629   a single such entry. For example,
2630</t>
2631<figure><artwork type="example">
2632    Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
2633</artwork></figure>
2634<t>
2635        could be collapsed to
2636</t>
2637<figure><artwork type="example">
2638    Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
2639</artwork></figure>
2640<t>
2641   Applications &SHOULD-NOT;  combine multiple entries unless they are all
2642   under the same organizational control and the hosts have already been
2643   replaced by pseudonyms. Applications &MUST-NOT; combine entries which
2644   have different received-protocol values.
2645</t>
2646</section>
2647
2648</section>
2649
2650<section title="IANA Considerations" anchor="IANA.considerations">
2651<t>
2652   TBD.
2653</t>
2654</section>
2655
2656<section title="Security Considerations" anchor="security.considerations">
2657<t>
2658   This section is meant to inform application developers, information
2659   providers, and users of the security limitations in HTTP/1.1 as
2660   described by this document. The discussion does not include
2661   definitive solutions to the problems revealed, though it does make
2662   some suggestions for reducing security risks.
2663</t>
2664
2665<section title="Personal Information" anchor="personal.information">
2666<t>
2667   HTTP clients are often privy to large amounts of personal information
2668   (e.g. the user's name, location, mail address, passwords, encryption
2669   keys, etc.), and &SHOULD; be very careful to prevent unintentional
2670   leakage of this information via the HTTP protocol to other sources.
2671   We very strongly recommend that a convenient interface be provided
2672   for the user to control dissemination of such information, and that
2673   designers and implementors be particularly careful in this area.
2674   History shows that errors in this area often create serious security
2675   and/or privacy problems and generate highly adverse publicity for the
2676   implementor's company.
2677</t>
2678</section>
2679
2680<section title="Abuse of Server Log Information" anchor="abuse.of.server.log.information">
2681<t>
2682   A server is in the position to save personal data about a user's
2683   requests which might identify their reading patterns or subjects of
2684   interest. This information is clearly confidential in nature and its
2685   handling can be constrained by law in certain countries. People using
2686   the HTTP protocol to provide data are responsible for ensuring that
2687   such material is not distributed without the permission of any
2688   individuals that are identifiable by the published results.
2689</t>
2690</section>
2691
2692<section title="Attacks Based On File and Path Names" anchor="attack.pathname">
2693<t>
2694   Implementations of HTTP origin servers &SHOULD; be careful to restrict
2695   the documents returned by HTTP requests to be only those that were
2696   intended by the server administrators. If an HTTP server translates
2697   HTTP URIs directly into file system calls, the server &MUST; take
2698   special care not to serve files that were not intended to be
2699   delivered to HTTP clients. For example, UNIX, Microsoft Windows, and
2700   other operating systems use ".." as a path component to indicate a
2701   directory level above the current one. On such a system, an HTTP
2702   server &MUST; disallow any such construct in the Request-URI if it
2703   would otherwise allow access to a resource outside those intended to
2704   be accessible via the HTTP server. Similarly, files intended for
2705   reference only internally to the server (such as access control
2706   files, configuration files, and script code) &MUST; be protected from
2707   inappropriate retrieval, since they might contain sensitive
2708   information. Experience has shown that minor bugs in such HTTP server
2709   implementations have turned into security risks.
2710</t>
2711</section>
2712
2713<section title="DNS Spoofing" anchor="dns.spoofing">
2714<t>
2715   Clients using HTTP rely heavily on the Domain Name Service, and are
2716   thus generally prone to security attacks based on the deliberate
2717   mis-association of IP addresses and DNS names. Clients need to be
2718   cautious in assuming the continuing validity of an IP number/DNS name
2719   association.
2720</t>
2721<t>
2722   In particular, HTTP clients &SHOULD; rely on their name resolver for
2723   confirmation of an IP number/DNS name association, rather than
2724   caching the result of previous host name lookups. Many platforms
2725   already can cache host name lookups locally when appropriate, and
2726   they &SHOULD; be configured to do so. It is proper for these lookups to
2727   be cached, however, only when the TTL (Time To Live) information
2728   reported by the name server makes it likely that the cached
2729   information will remain useful.
2730</t>
2731<t>
2732   If HTTP clients cache the results of host name lookups in order to
2733   achieve a performance improvement, they &MUST; observe the TTL
2734   information reported by DNS.
2735</t>
2736<t>
2737   If HTTP clients do not observe this rule, they could be spoofed when
2738   a previously-accessed server's IP address changes. As network
2739   renumbering is expected to become increasingly common <xref target="RFC1900"/>, the
2740   possibility of this form of attack will grow. Observing this
2741   requirement thus reduces this potential security vulnerability.
2742</t>
2743<t>
2744   This requirement also improves the load-balancing behavior of clients
2745   for replicated servers using the same DNS name and reduces the
2746   likelihood of a user's experiencing failure in accessing sites which
2747   use that strategy.
2748</t>
2749</section>
2750
2751<section title="Proxies and Caching" anchor="attack.proxies">
2752<t>
2753   By their very nature, HTTP proxies are men-in-the-middle, and
2754   represent an opportunity for man-in-the-middle attacks. Compromise of
2755   the systems on which the proxies run can result in serious security
2756   and privacy problems. Proxies have access to security-related
2757   information, personal information about individual users and
2758   organizations, and proprietary information belonging to users and
2759   content providers. A compromised proxy, or a proxy implemented or
2760   configured without regard to security and privacy considerations,
2761   might be used in the commission of a wide range of potential attacks.
2762</t>
2763<t>
2764   Proxy operators should protect the systems on which proxies run as
2765   they would protect any system that contains or transports sensitive
2766   information. In particular, log information gathered at proxies often
2767   contains highly sensitive personal information, and/or information
2768   about organizations. Log information should be carefully guarded, and
2769   appropriate guidelines for use developed and followed. (<xref target="abuse.of.server.log.information"/>).
2770</t>
2771<t>
2772   Proxy implementors should consider the privacy and security
2773   implications of their design and coding decisions, and of the
2774   configuration options they provide to proxy operators (especially the
2775   default configuration).
2776</t>
2777<t>
2778   Users of a proxy need to be aware that they are no trustworthier than
2779   the people who run the proxy; HTTP itself cannot solve this problem.
2780</t>
2781<t>
2782   The judicious use of cryptography, when appropriate, may suffice to
2783   protect against a broad range of security and privacy attacks. Such
2784   cryptography is beyond the scope of the HTTP/1.1 specification.
2785</t>
2786</section>
2787
2788<section title="Denial of Service Attacks on Proxies" anchor="attack.DoS">
2789<t>
2790   They exist. They are hard to defend against. Research continues.
2791   Beware.
2792</t>
2793</section>
2794</section>
2795
2796<section title="Acknowledgments" anchor="ack">
2797<t>
2798   This specification makes heavy use of the augmented BNF and generic
2799   constructs defined by David H. Crocker for RFC 822 <xref target="RFC822"/>. Similarly, it
2800   reuses many of the definitions provided by Nathaniel Borenstein and
2801   Ned Freed for MIME <xref target="RFC2045"/>. We hope that their inclusion in this
2802   specification will help reduce past confusion over the relationship
2803   between HTTP and Internet mail message formats.
2804</t>
2805<t>
2806   The HTTP protocol has evolved considerably over the years. It has
2807   benefited from a large and active developer community--the many
2808   people who have participated on the www-talk mailing list--and it is
2809   that community which has been most responsible for the success of
2810   HTTP and of the World-Wide Web in general. Marc Andreessen, Robert
2811   Cailliau, Daniel W. Connolly, Bob Denny, John Franks, Jean-Francois
2812   Groff, Phillip M. Hallam-Baker, Hakon W. Lie, Ari Luotonen, Rob
2813   McCool, Lou Montulli, Dave Raggett, Tony Sanders, and Marc
2814   VanHeyningen deserve special recognition for their efforts in
2815   defining early aspects of the protocol.
2816</t>
2817<t>
2818   This document has benefited greatly from the comments of all those
2819   participating in the HTTP-WG. In addition to those already mentioned,
2820   the following individuals have contributed to this specification:
2821</t>
2822<figure><artwork type="text/plain">
2823    Gary Adams                  Ross Patterson
2824    Harald Tveit Alvestrand     Albert Lunde
2825    Keith Ball                  John C. Mallery
2826    Brian Behlendorf            Jean-Philippe Martin-Flatin
2827    Paul Burchard               Mitra
2828    Maurizio Codogno            David Morris
2829    Mike Cowlishaw              Gavin Nicol
2830    Roman Czyborra              Bill Perry
2831    Michael A. Dolan            Jeffrey Perry
2832    David J. Fiander            Scott Powers
2833    Alan Freier                 Owen Rees
2834    Marc Hedlund                Luigi Rizzo
2835    Greg Herlihy                David Robinson
2836    Koen Holtman                Marc Salomon
2837    Alex Hopmann                Rich Salz
2838    Bob Jernigan                Allan M. Schiffman
2839    Shel Kaphan                 Jim Seidman
2840    Rohit Khare                 Chuck Shotton
2841    John Klensin                Eric W. Sink
2842    Martijn Koster              Simon E. Spero
2843    Alexei Kosut                Richard N. Taylor
2844    David M. Kristol            Robert S. Thau
2845    Daniel LaLiberte            Bill (BearHeart) Weinman
2846    Ben Laurie                  Francois Yergeau
2847    Paul J. Leach               Mary Ellen Zurko
2848    Daniel DuBois               Josh Cohen
2849</artwork></figure>
2850<t>
2851   Based on an XML translation of RFC 2616 by Julian Reschke.
2852</t>
2853</section>
2854
2855</middle>
2856<back>
2857<references>
2858
2859<reference anchor="Part2">
2860  <front>
2861    <title abbrev="HTTP/1.1">HTTP/1.1, part 2: Message Semantics</title>
2862 
2863    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
2864      <organization abbrev="Day Software">Day Software</organization>
2865      <address>
2866        <email>fielding@gbiv.com</email>
2867      </address>
2868    </author>
2869 
2870    <author initials="J." surname="Gettys" fullname="Jim Gettys">
2871      <organization>One Laptop per Child</organization>
2872      <address>
2873        <email>jg@laptop.org</email>
2874      </address>
2875    </author>
2876   
2877    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
2878      <organization abbrev="HP">Hewlett-Packard Company</organization>
2879      <address>
2880        <email>JeffMogul@acm.org</email>
2881      </address>
2882    </author>
2883 
2884    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
2885      <organization abbrev="Microsoft">Microsoft Corporation</organization>
2886      <address>
2887        <email>henrikn@microsoft.com</email>
2888      </address>
2889    </author>
2890 
2891    <author initials="L." surname="Masinter" fullname="Larry Masinter">
2892      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
2893      <address>
2894        <email>LMM@acm.org</email>
2895      </address>
2896    </author>
2897   
2898    <author initials="P." surname="Leach" fullname="Paul J. Leach">
2899      <organization abbrev="Microsoft">Microsoft Corporation</organization>
2900      <address>
2901        <email>paulle@microsoft.com</email>
2902      </address>
2903    </author>
2904     
2905    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
2906      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
2907      <address>
2908        <email>timbl@w3.org</email>
2909      </address>
2910    </author>
2911 
2912    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
2913  </front>
2914  <seriesInfo name="Internet-Draft" value="draft-fielding-p2-semantics-&ID-VERSION;"/>
2915  <x:source href="p2-semantics.xml" basename="p2-semantics"/>
2916</reference>
2917
2918<reference anchor="Part3">
2919  <front>
2920    <title abbrev="HTTP/1.1">HTTP/1.1, part 3: Message Payload and Content Negotiation</title>
2921 
2922    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
2923      <organization abbrev="Day Software">Day Software</organization>
2924      <address>
2925        <email>fielding@gbiv.com</email>
2926      </address>
2927    </author>
2928 
2929    <author initials="J." surname="Gettys" fullname="Jim Gettys">
2930      <organization>One Laptop per Child</organization>
2931      <address>
2932        <email>jg@laptop.org</email>
2933      </address>
2934    </author>
2935   
2936    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
2937      <organization abbrev="HP">Hewlett-Packard Company</organization>
2938      <address>
2939        <email>JeffMogul@acm.org</email>
2940      </address>
2941    </author>
2942 
2943    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
2944      <organization abbrev="Microsoft">Microsoft Corporation</organization>
2945      <address>
2946        <email>henrikn@microsoft.com</email>
2947      </address>
2948    </author>
2949 
2950    <author initials="L." surname="Masinter" fullname="Larry Masinter">
2951      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
2952      <address>
2953        <email>LMM@acm.org</email>
2954      </address>
2955    </author>
2956   
2957    <author initials="P." surname="Leach" fullname="Paul J. Leach">
2958      <organization abbrev="Microsoft">Microsoft Corporation</organization>
2959      <address>
2960        <email>paulle@microsoft.com</email>
2961      </address>
2962    </author>
2963     
2964    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
2965      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
2966      <address>
2967        <email>timbl@w3.org</email>
2968      </address>
2969    </author>
2970 
2971    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
2972  </front>
2973  <seriesInfo name="Internet-Draft" value="draft-fielding-p3-payload-&ID-VERSION;"/>
2974  <x:source href="p3-payload.xml" basename="p3-payload"/>
2975</reference>
2976
2977<reference anchor="Part6">
2978  <front>
2979    <title abbrev="HTTP/1.1">HTTP/1.1, part 6: Caching</title>
2980 
2981    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
2982      <organization abbrev="Day Software">Day Software</organization>
2983      <address>
2984        <email>fielding@gbiv.com</email>
2985      </address>
2986    </author>
2987 
2988    <author initials="J." surname="Gettys" fullname="Jim Gettys">
2989      <organization>One Laptop per Child</organization>
2990      <address>
2991        <email>jg@laptop.org</email>
2992      </address>
2993    </author>
2994   
2995    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
2996      <organization abbrev="HP">Hewlett-Packard Company</organization>
2997      <address>
2998        <email>JeffMogul@acm.org</email>
2999      </address>
3000    </author>
3001 
3002    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
3003      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3004      <address>
3005        <email>henrikn@microsoft.com</email>
3006      </address>
3007    </author>
3008 
3009    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3010      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3011      <address>
3012        <email>LMM@acm.org</email>
3013      </address>
3014    </author>
3015   
3016    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3017      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3018      <address>
3019        <email>paulle@microsoft.com</email>
3020      </address>
3021    </author>
3022     
3023    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3024      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3025      <address>
3026        <email>timbl@w3.org</email>
3027      </address>
3028    </author>
3029 
3030    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3031  </front>
3032  <seriesInfo name="Internet-Draft" value="draft-fielding-p6-cache-&ID-VERSION;"/>
3033  <x:source href="p6-cache.xml" basename="p6-cache"/>
3034</reference>
3035
3036<reference anchor="RFC1436">
3037<front>
3038<title abbrev="Gopher">The Internet Gopher Protocol (a distributed document search and retrieval protocol)</title>
3039<author initials="F." surname="Anklesaria" fullname="Farhad Anklesaria">
3040<organization>University of Minnesota, Computer and Information Services</organization>
3041<address>
3042<postal>
3043<street>100 Union Street SE</street>
3044<street>Room 152 Shepherd Labs</street>
3045<city>Minneapolis</city>
3046<region>MN</region>
3047<code>55455</code>
3048<country>US</country></postal>
3049<phone>+1 612 625 1300</phone>
3050<email>fxa@boombox.micro.umn.edu</email></address></author>
3051<author initials="M." surname="McCahill" fullname="Mark McCahill">
3052<organization>University of Minnesota, Computer and Information Services</organization>
3053<address>
3054<postal>
3055<street>100 Union Street SE</street>
3056<street>Room 152 Shepherd Labs</street>
3057<city>Minneapolis</city>
3058<region>MN</region>
3059<code>55455</code>
3060<country>US</country></postal>
3061<phone>+1 612 625 1300</phone>
3062<email>mpm@boombox.micro.umn.edu</email></address></author>
3063<author initials="P." surname="Lindner" fullname="Paul Lindner">
3064<organization>University of Minnesota, Computer and Information Services</organization>
3065<address>
3066<postal>
3067<street>100 Union Street SE</street>
3068<street>Room 152 Shepherd Labs</street>
3069<city>Minneapolis</city>
3070<region>MN</region>
3071<code>55455</code>
3072<country>US</country></postal>
3073<phone>+1 612 625 1300</phone>
3074<email>lindner@boombox.micro.umn.edu</email></address></author>
3075<author initials="D." surname="Johnson" fullname="David Johnson">
3076<organization>University of Minnesota, Computer and Information Services</organization>
3077<address>
3078<postal>
3079<street>100 Union Street SE</street>
3080<street>Room 152 Shepherd Labs</street>
3081<city>Minneapolis</city>
3082<region>MN</region>
3083<code>55455</code>
3084<country>US</country></postal>
3085<phone>+1 612 625 1300</phone>
3086<email>dmj@boombox.micro.umn.edu</email></address></author>
3087<author initials="D." surname="Torrey" fullname="Daniel Torrey">
3088<organization>University of Minnesota, Computer and Information Services</organization>
3089<address>
3090<postal>
3091<street>100 Union Street SE</street>
3092<street>Room 152 Shepherd Labs</street>
3093<city>Minneapolis</city>
3094<region>MN</region>
3095<code>55455</code>
3096<country>US</country></postal>
3097<phone>+1 612 625 1300</phone>
3098<email>daniel@boombox.micro.umn.edu</email></address></author>
3099<author initials="B." surname="Alberti" fullname="Bob Alberti">
3100<organization>University of Minnesota, Computer and Information Services</organization>
3101<address>
3102<postal>
3103<street>100 Union Street SE</street>
3104<street>Room 152 Shepherd Labs</street>
3105<city>Minneapolis</city>
3106<region>MN</region>
3107<code>55455</code>
3108<country>US</country></postal>
3109<phone>+1 612 625 1300</phone>
3110<email>alberti@boombox.micro.umn.edu</email></address></author>
3111<date month="March" year="1993"/>
3112</front>
3113<seriesInfo name="RFC" value="1436"/>
3114</reference>
3115
3116<reference anchor="RFC1630">
3117<front>
3118<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>
3119<author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3120<organization>CERN, World-Wide Web project</organization>
3121<address>
3122<postal>
3123<street>1211 Geneva 23</street>
3124<city/>
3125<region/>
3126<code/>
3127<country>CH</country></postal>
3128<phone>+41 22 7673755</phone>
3129<facsimile>+41 22 7677155</facsimile>
3130<email>timbl@info.cern.ch</email></address></author>
3131<date month="June" year="1994"/></front>
3132<seriesInfo name="RFC" value="1630"/>
3133</reference>
3134
3135<reference anchor="RFC1738">
3136<front>
3137<title>Uniform Resource Locators (URL)</title>
3138<author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3139<organization>CERN, World-Wide Web project</organization>
3140<address>
3141<postal>
3142<street>1211 Geneva 23</street>
3143<city/>
3144<region/>
3145<code/>
3146<country>CH</country></postal>
3147<phone>+41 22 7673755</phone>
3148<facsimile>+41 22 7677155</facsimile>
3149<email>timbl@info.cern.ch</email></address></author>
3150<author initials="L." surname="Masinter" fullname="Larry Masinter">
3151<organization>Xerox PARC</organization>
3152<address>
3153<postal>
3154<street>3333 Coyote Hill Road</street>
3155<city>Palo Alto</city>
3156<region>CA</region>
3157<code>94034</code>
3158<country>US</country></postal>
3159<phone>+1 415 812 4365</phone>
3160<facsimile>+1 415 812 4333</facsimile>
3161<email>masinter@parc.xerox.com</email></address></author>
3162<author initials="M." surname="McCahill" fullname="Mark McCahill">
3163<organization>University of Minnesota, Computer and Information Services</organization>
3164<address>
3165<postal>
3166<street>100 Union Street SE, Shepherd Labs</street>
3167<street>Room 152</street>
3168<city>Minneapolis</city>
3169<region>MN</region>
3170<code>55455</code>
3171<country>US</country></postal>
3172<phone>+1 612 625 1300</phone>
3173<email>mpm@boombox.micro.umn.edu</email></address></author>
3174<date month="December" year="1994"/>
3175</front>
3176<seriesInfo name="RFC" value="1738"/>
3177</reference>
3178
3179<reference anchor="RFC1945">
3180<front>
3181<title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
3182<author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3183<organization>MIT, Laboratory for Computer Science</organization>
3184<address>
3185<postal>
3186<street>545 Technology Square</street>
3187<city>Cambridge</city>
3188<region>MA</region>
3189<code>02139</code>
3190<country>US</country></postal>
3191<phone/>
3192<facsimile>+1 617 258 8682</facsimile>
3193<email>timbl@w3.org</email></address></author>
3194<author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3195<organization>University of California, Irvine, Department of Information and Computer Science</organization>
3196<address>
3197<postal>
3198<street/>
3199<city>Irvine</city>
3200<region>CA</region>
3201<code>92717-3425</code>
3202<country>US</country></postal>
3203<phone/>
3204<facsimile>+1 714 824 4056</facsimile>
3205<email>fielding@ics.uci.edu</email></address></author>
3206<author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3207<organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
3208<address>
3209<postal>
3210<street>545 Technology Square</street>
3211<city>Cambridge</city>
3212<region>MA</region>
3213<code>02139</code>
3214<country>US</country></postal>
3215<phone/>
3216<facsimile>+1 617 258 8682</facsimile>
3217<email>frystyk@w3.org</email></address></author>
3218<date month="May" year="1996"/>
3219</front>
3220<seriesInfo name="RFC" value="1945"/>
3221</reference>
3222
3223<reference anchor="RFC2045">
3224<front>
3225<title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
3226<author initials="N." surname="Freed" fullname="Ned Freed">
3227<organization>Innosoft International, Inc.</organization>
3228<address>
3229<postal>
3230<street>1050 East Garvey Avenue South</street>
3231<city>West Covina</city>
3232<region>CA</region>
3233<code>91790</code>
3234<country>US</country></postal>
3235<phone>+1 818 919 3600</phone>
3236<facsimile>+1 818 919 3614</facsimile>
3237<email>ned@innosoft.com</email></address></author>
3238<author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
3239<organization>First Virtual Holdings</organization>
3240<address>
3241<postal>
3242<street>25 Washington Avenue</street>
3243<city>Morristown</city>
3244<region>NJ</region>
3245<code>07960</code>
3246<country>US</country></postal>
3247<phone>+1 201 540 8967</phone>
3248<facsimile>+1 201 993 3032</facsimile>
3249<email>nsb@nsb.fv.com</email></address></author>
3250<date month="November" year="1996"/>
3251</front>
3252<seriesInfo name="RFC" value="2045"/>
3253</reference>
3254
3255<reference anchor="RFC1123">
3256<front>
3257<title>Requirements for Internet Hosts - Application and Support</title>
3258<author initials="R." surname="Braden" fullname="Robert Braden">
3259<organization>University of Southern California (USC), Information Sciences Institute</organization>
3260<address>
3261<postal>
3262<street>4676 Admiralty Way</street>
3263<city>Marina del Rey</city>
3264<region>CA</region>
3265<code>90292-6695</code>
3266<country>US</country></postal>
3267<phone>+1 213 822 1511</phone>
3268<email>Braden@ISI.EDU</email></address></author>
3269<date month="October" year="1989"/></front>
3270<seriesInfo name="STD" value="3"/>
3271<seriesInfo name="RFC" value="1123"/>
3272</reference>
3273
3274<reference anchor="RFC822">
3275<front>
3276<title abbrev="Standard for ARPA Internet Text Messages">Standard for the format of ARPA Internet text messages</title>
3277<author initials="D.H." surname="Crocker" fullname="David H. Crocker">
3278<organization>University of Delaware, Dept. of Electrical Engineering</organization>
3279<address>
3280<postal>
3281<street/>
3282<city>Newark</city>
3283<region>DE</region>
3284<code>19711</code>
3285<country>US</country></postal>
3286<email>DCrocker@UDel-Relay</email></address></author>
3287<date month="August" day="13" year="1982"/></front>
3288<seriesInfo name="STD" value="11"/>
3289<seriesInfo name="RFC" value="822"/>
3290</reference>
3291
3292<reference anchor="WAIS">
3293<front>
3294<title>WAIS Interface Protocol Prototype Functional Specification (v1.5)</title>
3295<author initials="F." surname="Davis" fullname="F. Davis">
3296<organization>Thinking Machines Corporation</organization>
3297</author>
3298<author initials="B." surname="Kahle" fullname="B. Kahle">
3299<organization>Thinking Machines Corporation</organization>
3300</author>
3301<author initials="H." surname="Morris" fullname="H. Morris">
3302<organization>Thinking Machines Corporation</organization>
3303</author>
3304<author initials="J." surname="Salem" fullname="J. Salem">
3305<organization>Thinking Machines Corporation</organization>
3306</author>
3307<author initials="T." surname="Shen" fullname="T. Shen">
3308<organization>Thinking Machines Corporation</organization>
3309</author>
3310<author initials="R." surname="Wang" fullname="R. Wang">
3311<organization>Thinking Machines Corporation</organization>
3312</author>
3313<author initials="J." surname="Sui" fullname="J. Sui">
3314<organization>Thinking Machines Corporation</organization>
3315</author>
3316<author initials="M." surname="Grinbaum" fullname="M. Grinbaum">
3317<organization>Thinking Machines Corporation</organization>
3318</author>
3319<date month="April" year="1990"/></front>
3320<seriesInfo name="Thinking Machines Corporation" value=""/>
3321</reference>
3322
3323<reference anchor="RFC1808">
3324<front>
3325<title>Relative Uniform Resource Locators</title>
3326<author initials="R." surname="Fielding" fullname="Roy T. Fielding">
3327<organization>University of California Irvine, Department of Information and Computer Science</organization>
3328<address>
3329<postal>
3330<street/>
3331<city>Irvine</city>
3332<region>CA</region>
3333<code>92717-3425</code>
3334<country>US</country></postal>
3335<phone>+1 714 824 4049</phone>
3336<facsimile>+1 714 824 4056</facsimile>
3337<email>fielding@ics.uci.edu</email></address></author>
3338<date month="June" year="1995"/>
3339</front>
3340<seriesInfo name="RFC" value="1808"/>
3341</reference>
3342
3343<reference anchor="RFC1036">
3344<front>
3345<title abbrev="Standard for USENET Messages">Standard for interchange of USENET messages</title>
3346<author initials="M." surname="Horton" fullname="M. Horton">
3347<organization>AT&amp;amp;T Bell Laboratories</organization></author>
3348<author initials="R." surname="Adams" fullname="R. Adams">
3349<organization>Center for Seismic Studies</organization></author>
3350<date month="December" year="1987"/></front>
3351<seriesInfo name="RFC" value="1036"/>
3352</reference>
3353
3354<reference anchor="RFC977">
3355<front>
3356<title abbrev="Network News Transfer Protocol">Network News Transfer Protocol</title>
3357<author initials="B." surname="Kantor" fullname="Brian Kantor">
3358<organization>University of California, San Diego (UCSD)</organization></author>
3359<author initials="P." surname="Lapsley" fullname="Phil Lapsley">
3360<organization>University of California, Berkeley (UCB)</organization></author>
3361<date month="February" year="1986"/></front>
3362<seriesInfo name="RFC" value="977"/>
3363</reference>
3364
3365<reference anchor="RFC2047">
3366<front>
3367<title abbrev="Message Header Extensions">MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text</title>
3368<author initials="K." surname="Moore" fullname="Keith Moore">
3369<organization>University of Tennessee</organization>
3370<address>
3371<postal>
3372<street>107 Ayres Hall</street>
3373<street>Knoxville TN 37996-1301</street></postal>
3374<email>moore@cs.utk.edu</email></address></author>
3375<date month="November" year="1996"/>
3376<area>Applications</area>
3377<keyword>Amercian Standard Code for Information Interchange</keyword>
3378<keyword>mail</keyword>
3379<keyword>multipurpose internet mail extensions</keyword>
3380</front>
3381<seriesInfo name="RFC" value="2047"/>
3382</reference>
3383
3384<reference anchor="RFC821">
3385<front>
3386<title>Simple Mail Transfer Protocol</title>
3387<author initials="J.B." surname="Postel" fullname="Jonathan B. Postel">
3388<organization>University of Southern California (USC)/Information Sciences Institute</organization>
3389<address>
3390<postal>
3391<street>4676 Admiralty Way</street>
3392<city>Marina del Rey</city>
3393<region>CA</region>
3394<code>90291</code>
3395<country>US</country></postal>
3396<phone>+1 213 822 1511</phone></address></author>
3397<date month="August" year="1982"/></front>
3398<seriesInfo name="STD" value="10"/>
3399<seriesInfo name="RFC" value="821"/>
3400</reference>
3401
3402<reference anchor="RFC1590">
3403<front>
3404<title>Media Type Registration Procedure</title>
3405<author initials="J." surname="Postel" fullname="Jon Postel">
3406<organization>USC/Information Sciences Institute</organization>
3407<address>
3408<postal>
3409<street>4676 Admiralty Way</street>
3410<city>Marina del Rey</city>
3411<region>CA</region>
3412<code>90292</code>
3413<country>US</country></postal>
3414<phone>+1 310 822 1511</phone>
3415<facsimile>+1 310 823 6714</facsimile>
3416<email>Postel@ISI.EDU</email></address></author>
3417<date month="November" year="1996"/>
3418</front>
3419<seriesInfo name="RFC" value="1590"/>
3420</reference>
3421
3422<reference anchor="RFC959">
3423<front>
3424<title abbrev="File Transfer Protocol">File Transfer Protocol</title>
3425<author initials="J." surname="Postel" fullname="J. Postel">
3426<organization>Information Sciences Institute (ISI)</organization></author>
3427<author initials="J." surname="Reynolds" fullname="J. Reynolds">
3428<organization/></author>
3429<date month="October" year="1985"/></front>
3430<seriesInfo name="STD" value="9"/>
3431<seriesInfo name="RFC" value="959"/>
3432</reference>
3433
3434<reference anchor="RFC1700">
3435<front>
3436<title abbrev="Assigned Numbers">Assigned Numbers</title>
3437<author initials="J." surname="Reynolds" fullname="Joyce K. Reynolds">
3438<organization>USC/Information Sciences Institute</organization>
3439<address>
3440<postal>
3441<street>4676 Admiralty Way</street>
3442<city>Marina del Rey</city>
3443<region>CA</region>
3444<code>90292-6695</code>
3445<country>US</country></postal>
3446<phone>+1 310 822 1511</phone>
3447<email>jkrey@isi.edu</email></address></author>
3448<author initials="J." surname="Postel" fullname="Jon Postel">
3449<organization>USC/Information Sciences Institute</organization>
3450<address>
3451<postal>
3452<street>4676 Admiralty Way</street>
3453<city>Marina del Rey</city>
3454<region>CA</region>
3455<code>90292-6695</code>
3456<country>US</country></postal>
3457<phone>+1 310 822 1511</phone>
3458<email>postel@isi.edu</email></address></author>
3459<date month="October" year="1994"/>
3460<abstract>
3461<t/></abstract></front>
3462<seriesInfo name="STD" value="2"/>
3463<seriesInfo name="RFC" value="1700"/>
3464</reference>
3465
3466<reference anchor="RFC1737">
3467<front>
3468<title abbrev="Requirements for Uniform Resource Names">Functional Requirements for Uniform Resource Names</title>
3469<author initials="L." surname="Masinter" fullname="Larry Masinter">
3470<organization>Xerox Palo Alto Research Center</organization>
3471<address>
3472<postal>
3473<street>3333 Coyote Hill Road</street>
3474<city>Palo Alto</city>
3475<region>CA</region>
3476<code>94304</code>
3477<country>US</country></postal>
3478<phone>+1 415 812 4365</phone>
3479<facsimile>+1 415 812 4333</facsimile>
3480<email>masinter@parc.xerox.com</email></address></author>
3481<author initials="K." surname="Sollins" fullname="Karen Sollins">
3482<organization>MIT Laboratory for Computer Science</organization>
3483<address>
3484<postal>
3485<street>545 Technology Square</street>
3486<city>Cambridge</city>
3487<region>MA</region>
3488<code>02139</code>
3489<country>US</country></postal>
3490<phone>+1 617 253 2673</phone>
3491<email>sollins@lcs.mit.edu</email></address></author>
3492<date month="December" year="1994"/></front>
3493<seriesInfo name="RFC" value="1737"/>
3494</reference>
3495
3496<reference anchor="USASCII">
3497  <front>
3498    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
3499    <author>
3500      <organization>American National Standards Institute</organization>
3501    </author>
3502    <date year="1986"/>
3503  </front>
3504  <seriesInfo name="ANSI" value="X3.4"/>
3505</reference>
3506
3507<reference anchor="ISO-8859">
3508<front>
3509<title>Information technology - 8-bit single byte coded graphic - character sets</title>
3510<author>
3511<organization>International Organization for Standardization</organization>
3512</author>
3513<date year="1987-1990"/>
3514</front>
3515<annotation>
3516  Part 1: Latin alphabet No. 1, ISO-8859-1:1987.
3517  Part 2: Latin alphabet No. 2, ISO-8859-2, 1987.
3518  Part 3: Latin alphabet No. 3, ISO-8859-3, 1988.
3519  Part 4: Latin alphabet No. 4, ISO-8859-4, 1988.
3520  Part 5: Latin/Cyrillic alphabet, ISO-8859-5, 1988.
3521  Part 6: Latin/Arabic alphabet, ISO-8859-6, 1987.
3522  Part 7: Latin/Greek alphabet, ISO-8859-7, 1987.
3523  Part 8: Latin/Hebrew alphabet, ISO-8859-8, 1988.
3524  Part 9: Latin alphabet No. 5, ISO-8859-9, 1990.
3525</annotation>
3526</reference>
3527
3528<reference anchor="RFC1900">
3529<front>
3530<title>Renumbering Needs Work</title>
3531<author initials="B." surname="Carpenter" fullname="Brian E. Carpenter">
3532<organization>CERN, Computing and Networks Division</organization>
3533<address>
3534<postal>
3535<street>1211 Geneva 23</street>
3536<country>CH</country></postal>
3537<phone>+41 22 7674967</phone>
3538<facsimile>+41 22 7677155</facsimile>
3539<email>brian@dxcoms.cern.ch</email></address></author>
3540<author initials="Y." surname="Rekhter" fullname="Yakov Rekhter">
3541<organization>cisco Systems</organization>
3542<address>
3543<postal>
3544<street>170 West Tasman Drive</street>
3545<city>San Jose</city>
3546<region>CA</region>
3547<code>95134</code>
3548<country>US</country></postal>
3549<phone>+1 914 528 0090</phone>
3550<email>yakov@cisco.com</email></address></author>
3551<date month="February" year="1996"/>
3552</front>
3553<seriesInfo name="RFC" value="1900"/>
3554</reference>
3555
3556<reference anchor="Pad1995">
3557<front>
3558<title>Improving HTTP Latency</title>
3559<author initials="V.N." surname="Padmanabhan" fullname="Venkata N. Padmanabhan"><organization/></author>
3560<author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul"><organization/></author>
3561<date year="1995" month="Dec"/>
3562</front>
3563<seriesInfo name="Computer Networks and ISDN Systems" value="v. 28, pp. 25-35"/>
3564<annotation>
3565  Slightly revised version of paper in Proc. 2nd International WWW Conference '94: Mosaic and the Web, Oct. 1994,
3566  which is available at <eref target="http://www.ncsa.uiuc.edu/SDG/IT94/Proceedings/DDay/mogul/HTTPLatency.html"/>.
3567</annotation>
3568</reference>
3569
3570<reference anchor="Tou1998" target="http://www.isi.edu/touch/pubs/http-perf96/">
3571<front>
3572<title>Analysis of HTTP Performance</title>
3573<author initials="J." surname="Touch" fullname="Joe Touch">
3574  <organization>USC/Information Sciences Institute</organization>
3575  <address><email>touch@isi.edu</email></address>
3576</author>
3577<author initials="J." surname="Heidemann" fullname="John Heidemann">
3578  <organization>USC/Information Sciences Institute</organization>
3579  <address><email>johnh@isi.edu</email></address>
3580</author>
3581<author initials="K." surname="Obraczka" fullname="Katia Obraczka">
3582  <organization>USC/Information Sciences Institute</organization>
3583  <address><email>katia@isi.edu</email></address>
3584</author>
3585<date year="1998" month="Aug"/>
3586</front>
3587<seriesInfo name="ISI Research Report" value="ISI/RR-98-463 (original report dated Aug.1996)"/>
3588</reference>
3589
3590<reference anchor="RFC1305">
3591<front>
3592<title>Network Time Protocol (Version 3) Specification, Implementation</title>
3593<author initials="D." surname="Mills" fullname="David L. Mills">
3594<organization>University of Delaware, Electrical Engineering Department</organization>
3595<address>
3596<postal>
3597<street/>
3598<city>Newark</city>
3599<region>DE</region>
3600<code>19716</code>
3601<country>US</country></postal>
3602<phone>+1 302 451 8247</phone>
3603<email>mills@udel.edu</email></address></author>
3604<date month="March" year="1992"/>
3605<abstract>
3606<t>This document describes the Network Time Protocol (NTP), specifies its normal structure and summarizes information useful for its implementation. NTP provides the mechanisms to synchronize time and coordinate time distribution in a large, diverse internet operating at rates from mundane to lightwave. It uses a returnable-time design in which a distributed subnet of time servers operating in a self-organizing, hierarchical-master-slave configuration synchronizes local clocks within the subnet and to national time standards via wire or radio. The servers can also redistribute reference time via local routing algorithms and time daemons.</t></abstract></front>
3607<seriesInfo name="RFC" value="1305"/>
3608</reference>
3609
3610<reference anchor="Spe" target="http://sunsite.unc.edu/mdma-release/http-prob.html">
3611<front>
3612<title>Analysis of HTTP Performance Problems</title>
3613<author initials="S." surname="Spero" fullname="Simon E. Spero">
3614  <organization/>
3615</author>
3616<date/>
3617</front>
3618</reference>
3619
3620<reference anchor="RFC2068">
3621<front>
3622<title abbrev="HTTP/1.1">Hypertext Transfer Protocol -- HTTP/1.1</title>
3623<author initials="R." surname="Fielding" fullname="Roy T. Fielding">
3624<organization>University of California, Irvine, Department of Information and Computer Science</organization>
3625<address>
3626<postal>
3627<street/>
3628<city>Irvine</city>
3629<region>CA</region>
3630<code>92717-3425</code>
3631<country>US</country></postal>
3632<facsimile>+1 714 824 4056</facsimile>
3633<email>fielding@ics.uci.edu</email></address></author>
3634<author initials="J." surname="Gettys" fullname="Jim Gettys">
3635<organization>MIT Laboratory for Computer Science</organization>
3636<address>
3637<postal>
3638<street>545 Technology Square</street>
3639<city>Cambridge</city>
3640<region>MA</region>
3641<code>02139</code>
3642<country>US</country></postal>
3643<facsimile>+1 617 258 8682</facsimile>
3644<email>jg@w3.org</email></address></author>
3645<author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3646<organization>Digital Equipment Corporation, Western Research Laboratory</organization>
3647<address>
3648<postal>
3649<street>250 University Avenue</street>
3650<city>Palo Alto</city>
3651<region>CA</region>
3652<code>94301</code>
3653<country>US</country></postal>
3654<email>mogul@wrl.dec.com</email></address></author>
3655<author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3656<organization>MIT Laboratory for Computer Science</organization>
3657<address>
3658<postal>
3659<street>545 Technology Square</street>
3660<city>Cambridge</city>
3661<region>MA</region>
3662<code>02139</code>
3663<country>US</country></postal>
3664<facsimile>+1 617 258 8682</facsimile>
3665<email>frystyk@w3.org</email></address></author>
3666<author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3667<organization>MIT Laboratory for Computer Science</organization>
3668<address>
3669<postal>
3670<street>545 Technology Square</street>
3671<city>Cambridge</city>
3672<region>MA</region>
3673<code>02139</code>
3674<country>US</country></postal>
3675<facsimile>+1 617 258 8682</facsimile>
3676<email>timbl@w3.org</email></address></author>
3677<date month="January" year="1997"/>
3678<abstract>
3679<t>The Hypertext Transfer Protocol (HTTP) is an application-level protocol for distributed, collaborative, hypermedia information systems. It is a generic, stateless, object-oriented protocol which can be used for many tasks, such as name servers and distributed object management systems, through extension of its request methods. A feature of HTTP is the typing and negotiation of data representation, allowing systems to be built independently of the data being transferred.</t>
3680<t>HTTP has been in use by the World-Wide Web global information initiative since 1990. This specification defines the protocol referred to as "HTTP/1.1".</t></abstract></front>
3681<seriesInfo name="RFC" value="2068"/>
3682</reference>
3683
3684<reference anchor="RFC2119">
3685<front>
3686<title abbrev="RFC Key Words">Key words for use in RFCs to Indicate Requirement Levels</title>
3687<author initials="S." surname="Bradner" fullname="Scott Bradner">
3688<organization>Harvard University</organization>
3689<address>
3690<postal>
3691<street>1350 Mass. Ave.</street>
3692<street>Cambridge</street>
3693<street>MA 02138</street></postal>
3694<phone>- +1 617 495 3864</phone>
3695<email>-</email></address></author>
3696<date month="March" year="1997"/>
3697<area>General</area>
3698<keyword>keyword</keyword>
3699</front>
3700
3701<seriesInfo name="BCP" value="14"/>
3702<seriesInfo name="RFC" value="2119"/>
3703</reference>
3704
3705<reference anchor="RFC2145">
3706<front>
3707<title abbrev="HTTP Version Numbers">Use and Interpretation of HTTP Version Numbers</title>
3708<author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
3709<organization>Western Research Laboratory</organization>
3710<address>
3711<postal>
3712<street>Digital Equipment Corporation</street>
3713<street>250 University Avenue</street>
3714<street>Palo Alto</street>
3715<street>California</street>
3716<street>94305</street>
3717<country>USA</country></postal>
3718<email>mogul@wrl.dec.com</email></address></author>
3719<author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3720<organization>Department of Information and Computer Science</organization>
3721<address>
3722<postal>
3723<street>University of California</street>
3724<street>Irvine</street>
3725<street>CA 92717-3425</street>
3726<country>USA</country></postal>
3727<facsimile>+1 (714) 824-4056</facsimile>
3728<email>fielding@ics.uci.edu</email></address></author>
3729<author initials="J." surname="Gettys" fullname="Jim Gettys">
3730<organization>MIT Laboratory for Computer Science</organization>
3731<address>
3732<postal>
3733<street>545 Technology Square</street>
3734<street>Cambridge</street>
3735<street>MA 02139</street>
3736<country>USA</country></postal>
3737<facsimile>+1 (617) 258 8682</facsimile>
3738<email>jg@w3.org</email></address></author>
3739<author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3740<organization>W3 Consortium</organization>
3741<address>
3742<postal>
3743<street>MIT Laboratory for Computer Science</street>
3744<street>545 Technology Square</street>
3745<street>Cambridge</street>
3746<street>MA 02139</street>
3747<country>USA</country></postal>
3748<facsimile>+1 (617) 258 8682</facsimile>
3749<email>frystyk@w3.org</email></address></author>
3750<date month="May" year="1997"/>
3751<area>Applications</area>
3752<keyword>HTTP</keyword>
3753<keyword>hypertext transfer protocol</keyword>
3754<abstract>
3755<t>
3756   HTTP request and response messages include an HTTP protocol version
3757   number.  Some confusion exists concerning the proper use and
3758   interpretation of HTTP version numbers, and concerning
3759   interoperability of HTTP implementations of different protocol
3760   versions.  This document is an attempt to clarify the situation.  It
3761   is not a modification of the intended meaning of the existing
3762   HTTP/1.0 and HTTP/1.1 documents, but it does describe the intention
3763   of the authors of those documents, and can be considered definitive
3764   when there is any ambiguity in those documents concerning HTTP
3765   version numbers, for all versions of HTTP.
3766</t></abstract></front>
3767<seriesInfo name="RFC" value="2145"/>
3768</reference>
3769
3770<reference anchor="Nie1997">
3771<front>
3772<title>Network Performance Effects of HTTP/1.1, CSS1, and PNG</title>
3773<author initials="H.F.." surname="Nielsen" fullname="H.F. Nielsen">
3774  <organization/>
3775</author>
3776<author initials="J." surname="Gettys" fullname="J. Gettys">
3777  <organization/>
3778</author>
3779<author initials="E." surname="Prud'hommeaux" fullname="E. Prud'hommeaux">
3780  <organization/>
3781</author>
3782<author initials="H." surname="Lie" fullname="H. Lie">
3783  <organization/>
3784</author>
3785<author initials="C." surname="Lilley" fullname="C. Lilley">
3786  <organization/>
3787</author>
3788<date year="1997" month="Sep"/>
3789</front>
3790<seriesInfo name="Proceedings of ACM SIGCOMM '97, Cannes France" value=""/>
3791</reference>
3792
3793<reference anchor="RFC2396">
3794<front>
3795<title abbrev="URI Generic Syntax">Uniform Resource Identifiers (URI): Generic Syntax</title>
3796<author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3797<organization abbrev="MIT/LCS">World Wide Web Consortium</organization>
3798<address>
3799<postal>
3800<street>MIT Laboratory for Computer Science, NE43-356</street>
3801<street>545 Technology Square</street>
3802<city>Cambridge</city>
3803<region>MA</region>
3804<code>02139</code></postal>
3805<facsimile>+1(617)258-8682</facsimile>
3806<email>timbl@w3.org</email></address></author>
3807<author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3808<organization abbrev="U.C. Irvine">Department of Information and Computer Science</organization>
3809<address>
3810<postal>
3811<street>University of California, Irvine</street>
3812<city>Irvine</city>
3813<region>CA</region>
3814<code>92697-3425</code></postal>
3815<facsimile>+1(949)824-1715</facsimile>
3816<email>fielding@ics.uci.edu</email></address></author>
3817<author initials="L." surname="Masinter" fullname="Larry Masinter">
3818<organization abbrev="Xerox Corporation">Xerox PARC</organization>
3819<address>
3820<postal>
3821<street>3333 Coyote Hill Road</street>
3822<city>Palo Alto</city>
3823<region>CA</region>
3824<code>94034</code></postal>
3825<facsimile>+1(415)812-4333</facsimile>
3826<email>masinter@parc.xerox.com</email></address></author>
3827<date month="August" year="1998"/>
3828<area>Applications</area>
3829<keyword>uniform resource</keyword>
3830<keyword>URI</keyword>
3831</front>
3832<seriesInfo name="RFC" value="2396"/>
3833</reference>
3834
3835<reference anchor="RFC2616">
3836  <front>
3837    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
3838    <author initials="R." surname="Fielding" fullname="R. Fielding">
3839      <organization>University of California, Irvine</organization>
3840      <address><email>fielding@ics.uci.edu</email></address>
3841    </author>
3842    <author initials="J." surname="Gettys" fullname="J. Gettys">
3843      <organization>W3C</organization>
3844      <address><email>jg@w3.org</email></address>
3845    </author>
3846    <author initials="J." surname="Mogul" fullname="J. Mogul">
3847      <organization>Compaq Computer Corporation</organization>
3848      <address><email>mogul@wrl.dec.com</email></address>
3849    </author>
3850    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
3851      <organization>MIT Laboratory for Computer Science</organization>
3852      <address><email>frystyk@w3.org</email></address>
3853    </author>
3854    <author initials="L." surname="Masinter" fullname="L. Masinter">
3855      <organization>Xerox Corporation</organization>
3856      <address><email>masinter@parc.xerox.com</email></address>
3857    </author>
3858    <author initials="P." surname="Leach" fullname="P. Leach">
3859      <organization>Microsoft Corporation</organization>
3860      <address><email>paulle@microsoft.com</email></address>
3861    </author>
3862    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
3863      <organization>W3C</organization>
3864      <address><email>timbl@w3.org</email></address>
3865    </author>
3866    <date month="June" year="1999"/>
3867  </front>
3868  <seriesInfo name="RFC" value="2616"/>
3869</reference>
3870
3871<reference anchor="RFC2617">
3872<front>
3873<title abbrev="HTTP Authentication">HTTP Authentication: Basic and Digest Access Authentication</title>
3874<author initials="J." surname="Franks" fullname="John Franks">
3875<organization>Northwestern University, Department of Mathematics</organization>
3876<address>
3877<postal>
3878<street/>
3879<city>Evanston</city>
3880<region>IL</region>
3881<code>60208-2730</code>
3882<country>US</country></postal>
3883<email>john@math.nwu.edu</email></address></author>
3884<author initials="P.M." surname="Hallam-Baker" fullname="Phillip M. Hallam-Baker">
3885<organization>Verisign Inc.</organization>
3886<address>
3887<postal>
3888<street>301 Edgewater Place</street>
3889<street>Suite 210</street>
3890<city>Wakefield</city>
3891<region>MA</region>
3892<code>01880</code>
3893<country>US</country></postal>
3894<email>pbaker@verisign.com</email></address></author>
3895<author initials="J.L." surname="Hostetler" fullname="Jeffery L. Hostetler">
3896<organization>AbiSource, Inc.</organization>
3897<address>
3898<postal>
3899<street>6 Dunlap Court</street>
3900<city>Savoy</city>
3901<region>IL</region>
3902<code>61874</code>
3903<country>US</country></postal>
3904<email>jeff@AbiSource.com</email></address></author>
3905<author initials="S.D." surname="Lawrence" fullname="Scott D. Lawrence">
3906<organization>Agranat Systems, Inc.</organization>
3907<address>
3908<postal>
3909<street>5 Clocktower Place</street>
3910<street>Suite 400</street>
3911<city>Maynard</city>
3912<region>MA</region>
3913<code>01754</code>
3914<country>US</country></postal>
3915<email>lawrence@agranat.com</email></address></author>
3916<author initials="P.J." surname="Leach" fullname="Paul J. Leach">
3917<organization>Microsoft Corporation</organization>
3918<address>
3919<postal>
3920<street>1 Microsoft Way</street>
3921<city>Redmond</city>
3922<region>WA</region>
3923<code>98052</code>
3924<country>US</country></postal>
3925<email>paulle@microsoft.com</email></address></author>
3926<author initials="A." surname="Luotonen" fullname="Ari Luotonen">
3927<organization>Netscape Communications Corporation</organization>
3928<address>
3929<postal>
3930<street>501 East Middlefield Road</street>
3931<city>Mountain View</city>
3932<region>CA</region>
3933<code>94043</code>
3934<country>US</country></postal></address></author>
3935<author initials="L." surname="Stewart" fullname="Lawrence C. Stewart">
3936<organization>Open Market, Inc.</organization>
3937<address>
3938<postal>
3939<street>215 First Street</street>
3940<city>Cambridge</city>
3941<region>MA</region>
3942<code>02142</code>
3943<country>US</country></postal>
3944<email>stewart@OpenMarket.com</email></address></author>
3945<date month="June" year="1999"/>
3946</front>
3947<seriesInfo name="RFC" value="2617"/>
3948</reference>
3949
3950<reference anchor="RFC2324">
3951<front>
3952<title abbrev="HTCPCP/1.0">Hyper Text Coffee Pot Control Protocol (HTCPCP/1.0)</title>
3953<author initials="L." surname="Masinter" fullname="Larry Masinter">
3954<organization>Xerox Palo Alto Research Center</organization>
3955<address>
3956<postal>
3957<street>3333 Coyote Hill Road</street>
3958<city>Palo Alto</city>
3959<region>CA</region>
3960<code>94304</code></postal>
3961<email>masinter@parc.xerox.com</email></address></author>
3962<date month="April" day="1" year="1998"/>
3963<area>General</area>
3964<keyword>control protocol</keyword>
3965<keyword>coffee</keyword>
3966</front>
3967<seriesInfo name="RFC" value="2324"/>
3968</reference>
3969
3970</references>
3971
3972<section title="Internet Media Type message/http and application/http" anchor="internet.media.type.http">
3973<iref item="Media Type" subitem="message/http" primary="true"/>
3974<iref item="message/http Media Type" primary="true"/>
3975<iref item="Media Type" subitem="application/http" primary="true"/>
3976<iref item="application/http Media Type" primary="true"/>
3977<t>
3978   In addition to defining the HTTP/1.1 protocol, this document serves
3979   as the specification for the Internet media type "message/http" and
3980   "application/http". The message/http type can be used to enclose a
3981   single HTTP request or response message, provided that it obeys the
3982   MIME restrictions for all "message" types regarding line length and
3983   encodings. The application/http type can be used to enclose a
3984   pipeline of one or more HTTP request or response messages (not
3985   intermixed). The following is to be registered with IANA <xref target="RFC1590"/>.
3986</t>
3987<t>
3988  <list style="hanging" x:indent="12em">
3989    <t hangText="Media Type name:">
3990      message
3991    </t>
3992    <t hangText="Media subtype name:">
3993      http
3994    </t>
3995    <t hangText="Required parameters:">
3996      none
3997    </t>
3998    <t hangText="Optional parameters:">
3999      version, msgtype
4000      <list style="hanging">
4001        <t hangText="version:">
4002          The HTTP-Version number of the enclosed message
4003          (e.g., "1.1"). If not present, the version can be
4004          determined from the first line of the body.
4005        </t>
4006        <t hangText="msgtype:">
4007          The message type -- "request" or "response". If not
4008          present, the type can be determined from the first
4009          line of the body.
4010        </t>
4011      </list>
4012    </t>
4013    <t hangText="Encoding considerations:">
4014      only "7bit", "8bit", or "binary" are permitted
4015    </t>
4016    <t hangText="Security considerations:">
4017      none
4018    </t>
4019  </list>
4020</t>
4021<t>
4022  <list style="hanging" x:indent="12em">
4023    <t hangText="Media Type name:">
4024      application
4025    </t>
4026    <t hangText="Media subtype name:">
4027      http
4028    </t>
4029    <t hangText="Required parameters:">
4030      none
4031    </t>
4032    <t hangText="Optional parameters:">
4033      version, msgtype
4034      <list style="hanging">
4035        <t hangText="version:">
4036          The HTTP-Version number of the enclosed messages
4037          (e.g., "1.1"). If not present, the version can be
4038          determined from the first line of the body.
4039        </t>
4040        <t hangText="msgtype:">
4041          The message type -- "request" or "response". If not
4042          present, the type can be determined from the first
4043          line of the body.
4044        </t>
4045      </list>
4046    </t>
4047    <t hangText="Encoding considerations:">
4048      HTTP messages enclosed by this type
4049      are in "binary" format; use of an appropriate
4050      Content-Transfer-Encoding is required when
4051      transmitted via E-mail.
4052    </t>
4053    <t hangText="Security considerations:">
4054      none
4055    </t>
4056  </list>
4057</t>
4058</section>
4059
4060<section title="Tolerant Applications" anchor="tolerant.applications">
4061<t>
4062   Although this document specifies the requirements for the generation
4063   of HTTP/1.1 messages, not all applications will be correct in their
4064   implementation. We therefore recommend that operational applications
4065   be tolerant of deviations whenever those deviations can be
4066   interpreted unambiguously.
4067</t>
4068<t>
4069   Clients &SHOULD; be tolerant in parsing the Status-Line and servers
4070   tolerant when parsing the Request-Line. In particular, they &SHOULD;
4071   accept any amount of SP or HT characters between fields, even though
4072   only a single SP is required.
4073</t>
4074<t>
4075   The line terminator for message-header fields is the sequence CRLF.
4076   However, we recommend that applications, when parsing such headers,
4077   recognize a single LF as a line terminator and ignore the leading CR.
4078</t>
4079<t>
4080   The character set of an entity-body &SHOULD; be labeled as the lowest
4081   common denominator of the character codes used within that body, with
4082   the exception that not labeling the entity is preferred over labeling
4083   the entity with the labels US-ASCII or ISO-8859-1. See &payload;.
4084</t>
4085<t>
4086   Additional rules for requirements on parsing and encoding of dates
4087   and other potential problems with date encodings include:
4088</t>
4089<t>
4090  <list style="symbols">
4091     <t>HTTP/1.1 clients and caches &SHOULD; assume that an RFC-850 date
4092        which appears to be more than 50 years in the future is in fact
4093        in the past (this helps solve the "year 2000" problem).</t>
4094
4095     <t>An HTTP/1.1 implementation &MAY; internally represent a parsed
4096        Expires date as earlier than the proper value, but &MUST-NOT;
4097        internally represent a parsed Expires date as later than the
4098        proper value.</t>
4099
4100     <t>All expiration-related calculations &MUST; be done in GMT. The
4101        local time zone &MUST-NOT; influence the calculation or comparison
4102        of an age or expiration time.</t>
4103
4104     <t>If an HTTP header incorrectly carries a date value with a time
4105        zone other than GMT, it &MUST; be converted into GMT using the
4106        most conservative possible conversion.</t>
4107  </list>
4108</t>
4109</section>
4110
4111<section title="Conversion of Date Formats" anchor="conversion.of.date.formats">
4112<t>
4113   HTTP/1.1 uses a restricted set of date formats (<xref target="full.date"/>) to
4114   simplify the process of date comparison. Proxies and gateways from
4115   other protocols &SHOULD; ensure that any Date header field present in a
4116   message conforms to one of the HTTP/1.1 formats and rewrite the date
4117   if necessary.
4118</t>
4119</section>
4120
4121<section title="Compatibility with Previous Versions" anchor="compatibility">
4122<t>
4123   It is beyond the scope of a protocol specification to mandate
4124   compliance with previous versions. HTTP/1.1 was deliberately
4125   designed, however, to make supporting previous versions easy. It is
4126   worth noting that, at the time of composing this specification
4127   (1996), we would expect commercial HTTP/1.1 servers to:
4128  <list style="symbols">
4129     <t>recognize the format of the Request-Line for HTTP/0.9, 1.0, and
4130        1.1 requests;</t>
4131
4132     <t>understand any valid request in the format of HTTP/0.9, 1.0, or
4133        1.1;</t>
4134
4135     <t>respond appropriately with a message in the same major version
4136        used by the client.</t>
4137  </list>
4138</t>
4139<t>
4140   And we would expect HTTP/1.1 clients to:
4141  <list style="symbols">
4142     <t>recognize the format of the Status-Line for HTTP/1.0 and 1.1
4143        responses;</t>
4144
4145     <t>understand any valid response in the format of HTTP/0.9, 1.0, or
4146        1.1.</t>
4147  </list>
4148</t>
4149<t>
4150   For most implementations of HTTP/1.0, each connection is established
4151   by the client prior to the request and closed by the server after
4152   sending the response. Some implementations implement the Keep-Alive
4153   version of persistent connections described in <xref x:sec="19.7.1" x:fmt="sec" target="RFC2068"/> of RFC
4154   2068 <xref target="RFC2068"/>.
4155</t>
4156
4157<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
4158<t>
4159   This section summarizes major differences between versions HTTP/1.0
4160   and HTTP/1.1.
4161</t>
4162
4163<section title="Changes to Simplify Multi-homed Web Servers and Conserve IP Addresses" anchor="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses">
4164<t>
4165   The requirements that clients and servers support the Host request-header,
4166   report an error if the Host request-header (<xref target="header.host"/>) is
4167   missing from an HTTP/1.1 request, and accept absolute URIs (<xref target="request-uri"/>)
4168   are among the most important changes defined by this
4169   specification.
4170</t>
4171<t>
4172   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
4173   addresses and servers; there was no other established mechanism for
4174   distinguishing the intended server of a request than the IP address
4175   to which that request was directed. The changes outlined above will
4176   allow the Internet, once older HTTP clients are no longer common, to
4177   support multiple Web sites from a single IP address, greatly
4178   simplifying large operational Web servers, where allocation of many
4179   IP addresses to a single host has created serious problems. The
4180   Internet will also be able to recover the IP addresses that have been
4181   allocated for the sole purpose of allowing special-purpose domain
4182   names to be used in root-level HTTP URLs. Given the rate of growth of
4183   the Web, and the number of servers already deployed, it is extremely
4184   important that all implementations of HTTP (including updates to
4185   existing HTTP/1.0 applications) correctly implement these
4186   requirements:
4187  <list style="symbols">
4188     <t>Both clients and servers &MUST; support the Host request-header.</t>
4189
4190     <t>A client that sends an HTTP/1.1 request &MUST; send a Host header.</t>
4191
4192     <t>Servers &MUST; report a 400 (Bad Request) error if an HTTP/1.1
4193        request does not include a Host request-header.</t>
4194
4195     <t>Servers &MUST; accept absolute URIs.</t>
4196  </list>
4197</t>
4198</section>
4199</section>
4200
4201<section title="Compatibility with HTTP/1.0 Persistent Connections" anchor="compatibility.with.http.1.0.persistent.connections">
4202<t>
4203   Some clients and servers might wish to be compatible with some
4204   previous implementations of persistent connections in HTTP/1.0
4205   clients and servers. Persistent connections in HTTP/1.0 are
4206   explicitly negotiated as they are not the default behavior. HTTP/1.0
4207   experimental implementations of persistent connections are faulty,
4208   and the new facilities in HTTP/1.1 are designed to rectify these
4209   problems. The problem was that some existing 1.0 clients may be
4210   sending Keep-Alive to a proxy server that doesn't understand
4211   Connection, which would then erroneously forward it to the next
4212   inbound server, which would establish the Keep-Alive connection and
4213   result in a hung HTTP/1.0 proxy waiting for the close on the
4214   response. The result is that HTTP/1.0 clients must be prevented from
4215   using Keep-Alive when talking to proxies.
4216</t>
4217<t>
4218   However, talking to proxies is the most important use of persistent
4219   connections, so that prohibition is clearly unacceptable. Therefore,
4220   we need some other mechanism for indicating a persistent connection
4221   is desired, which is safe to use even when talking to an old proxy
4222   that ignores Connection. Persistent connections are the default for
4223   HTTP/1.1 messages; we introduce a new keyword (Connection: close) for
4224   declaring non-persistence. See <xref target="header.connection"/>.
4225</t>
4226<t>
4227   The original HTTP/1.0 form of persistent connections (the Connection:
4228   Keep-Alive and Keep-Alive header) is documented in RFC 2068. <xref target="RFC2068"/>
4229</t>
4230</section>
4231
4232<section title="Changes from RFC 2068" anchor="changes.from.rfc.2068">
4233<t>
4234   This specification has been carefully audited to correct and
4235   disambiguate key word usage; RFC 2068 had many problems in respect to
4236   the conventions laid out in RFC 2119 <xref target="RFC2119"/>.
4237</t>
4238<t>
4239   Transfer-coding and message lengths all interact in ways that
4240   required fixing exactly when chunked encoding is used (to allow for
4241   transfer encoding that may not be self delimiting); it was important
4242   to straighten out exactly how message lengths are computed.
4243</t>
4244<t>
4245   The use and interpretation of HTTP version numbers has been clarified
4246   by RFC 2145. Require proxies to upgrade requests to highest protocol
4247   version they support to deal with problems discovered in HTTP/1.0
4248   implementations (<xref target="http.version"/>)
4249</t>
4250<t>
4251   Proxies should be able to add Content-Length when appropriate.
4252</t>
4253<t>
4254   Transfer-coding had significant problems, particularly with
4255   interactions with chunked encoding. The solution is that transfer-codings
4256   become as full fledged as content-codings. This involves
4257   adding an IANA registry for transfer-codings (separate from content
4258   codings), a new header field (TE) and enabling trailer headers in the
4259   future. Transfer encoding is a major performance benefit, so it was
4260   worth fixing <xref target="Nie1997"/>. TE also solves another, obscure, downward
4261   interoperability problem that could have occurred due to interactions
4262   between authentication trailers, chunked encoding and HTTP/1.0
4263   clients.(Section <xref target="transfer.codings" format="counter"/>, <xref target="chunked.transfer.encoding" format="counter"/>,
4264   and <xref target="header.te" format="counter"/>)
4265</t>
4266</section>
4267</section>
4268</back>
4269</rfc>
Note: See TracBrowser for help on using the repository browser.