source: draft-ietf-httpbis/latest/p1-messaging.xml @ 210

Last change on this file since 210 was 210, checked in by julian.reschke@…, 12 years ago

Move definition of "delta-seconds" into Part1; relates to #36.

  • Property svn:eol-style set to native
File size: 182.7 KB
Line 
1<?xml version="1.0" encoding="utf-8"?>
2<?xml-stylesheet type='text/xsl' href='../myxml2rfc.xslt'?>
3<!DOCTYPE rfc [
4  <!ENTITY MAY "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MAY</bcp14>">
5  <!ENTITY MUST "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST</bcp14>">
6  <!ENTITY MUST-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST NOT</bcp14>">
7  <!ENTITY OPTIONAL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>OPTIONAL</bcp14>">
8  <!ENTITY RECOMMENDED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>RECOMMENDED</bcp14>">
9  <!ENTITY REQUIRED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>REQUIRED</bcp14>">
10  <!ENTITY SHALL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL</bcp14>">
11  <!ENTITY SHALL-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL NOT</bcp14>">
12  <!ENTITY SHOULD "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD</bcp14>">
13  <!ENTITY SHOULD-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD NOT</bcp14>">
14  <!ENTITY ID-VERSION "latest">
15  <!ENTITY ID-MONTH "February">
16  <!ENTITY ID-YEAR "2008">
17  <!ENTITY caching                "<xref target='Part6' x:rel='#caching' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
18  <!ENTITY payload                "<xref target='Part3' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
19  <!ENTITY media-types            "<xref target='Part3' x:rel='#media.types' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
20  <!ENTITY content-codings        "<xref target='Part3' x:rel='#content.codings' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
21  <!ENTITY CONNECT                "<xref target='Part2' x:rel='#CONNECT' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
22  <!ENTITY content.negotiation    "<xref target='Part3' x:rel='#content.negotiation' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
23  <!ENTITY diff2045entity         "<xref target='Part3' x:rel='#differences.between.http.entities.and.rfc.2045.entities' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
24  <!ENTITY entity                 "<xref target='Part3' x:rel='#entity' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
25  <!ENTITY entity-body            "<xref target='Part3' x:rel='#entity.body' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
26  <!ENTITY entity-header-fields   "<xref target='Part3' x:rel='#entity.header.fields' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
27  <!ENTITY header-accept          "<xref target='Part3' x:rel='#header.accept' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
28  <!ENTITY header-cache-control   "<xref target='Part6' x:rel='#header.cache-control' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
29  <!ENTITY header-expect          "<xref target='Part2' x:rel='#header.expect' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
30  <!ENTITY header-pragma          "<xref target='Part6' x:rel='#header.pragma' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
31  <!ENTITY header-warning         "<xref target='Part6' x:rel='#header.warning' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
32  <!ENTITY idempotent-methods     "<xref target='Part2' x:rel='#idempotent.methods' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
33  <!ENTITY qvalue                 "<xref target='Part3' x:rel='#quality.values' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
34  <!ENTITY request-header-fields  "<xref target='Part2' x:rel='#request.header.fields' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
35  <!ENTITY response-header-fields "<xref target='Part2' x:rel='#response.header.fields' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
36  <!ENTITY method                 "<xref target='Part2' x:rel='#method' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
37  <!ENTITY status-codes           "<xref target='Part2' x:rel='#status.codes' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
38  <!ENTITY status-100             "<xref target='Part2' x:rel='#status.100' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
39  <!ENTITY status-1xx             "<xref target='Part2' x:rel='#status.1xx' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
40  <!ENTITY status-414             "<xref target='Part2' x:rel='#status.414' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
41]>
42<?rfc toc="yes" ?>
43<?rfc symrefs="yes" ?>
44<?rfc sortrefs="yes" ?>
45<?rfc compact="yes"?>
46<?rfc subcompact="no" ?>
47<?rfc linkmailto="no" ?>
48<?rfc editing="no" ?>
49<?rfc comments="yes"?>
50<?rfc inline="yes"?>
51<?rfc-ext allow-markup-in-artwork="yes" ?>
52<?rfc-ext include-references-in-index="yes" ?>
53<rfc obsoletes="2616" category="std"
54     ipr="full3978" docName="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"
55     xmlns:x='http://purl.org/net/xml2rfc/ext'>
56<front>
57
58  <title abbrev="HTTP/1.1, Part 1">HTTP/1.1, part 1: URIs, Connections, and Message Parsing</title>
59
60  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
61    <organization abbrev="Day Software">Day Software</organization>
62    <address>
63      <postal>
64        <street>23 Corporate Plaza DR, Suite 280</street>
65        <city>Newport Beach</city>
66        <region>CA</region>
67        <code>92660</code>
68        <country>USA</country>
69      </postal>
70      <phone>+1-949-706-5300</phone>
71      <facsimile>+1-949-706-5305</facsimile>
72      <email>fielding@gbiv.com</email>
73      <uri>http://roy.gbiv.com/</uri>
74    </address>
75  </author>
76
77  <author initials="J." surname="Gettys" fullname="Jim Gettys">
78    <organization>One Laptop per Child</organization>
79    <address>
80      <postal>
81        <street>21 Oak Knoll Road</street>
82        <city>Carlisle</city>
83        <region>MA</region>
84        <code>01741</code>
85        <country>USA</country>
86      </postal>
87      <email>jg@laptop.org</email>
88      <uri>http://www.laptop.org/</uri>
89    </address>
90  </author>
91 
92  <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
93    <organization abbrev="HP">Hewlett-Packard Company</organization>
94    <address>
95      <postal>
96        <street>HP Labs, Large Scale Systems Group</street>
97        <street>1501 Page Mill Road, MS 1177</street>
98        <city>Palo Alto</city>
99        <region>CA</region>
100        <code>94304</code>
101        <country>USA</country>
102      </postal>
103      <email>JeffMogul@acm.org</email>
104    </address>
105  </author>
106
107  <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
108    <organization abbrev="Microsoft">Microsoft Corporation</organization>
109    <address>
110      <postal>
111        <street>1 Microsoft Way</street>
112        <city>Redmond</city>
113        <region>WA</region>
114        <code>98052</code>
115        <country>USA</country>
116      </postal>
117      <email>henrikn@microsoft.com</email>
118    </address>
119  </author>
120
121  <author initials="L." surname="Masinter" fullname="Larry Masinter">
122    <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
123    <address>
124      <postal>
125        <street>345 Park Ave</street>
126        <city>San Jose</city>
127        <region>CA</region>
128        <code>95110</code>
129        <country>USA</country>
130      </postal>
131      <email>LMM@acm.org</email>
132      <uri>http://larry.masinter.net/</uri>
133    </address>
134  </author>
135 
136  <author initials="P." surname="Leach" fullname="Paul J. Leach">
137    <organization abbrev="Microsoft">Microsoft Corporation</organization>
138    <address>
139      <postal>
140        <street>1 Microsoft Way</street>
141        <city>Redmond</city>
142        <region>WA</region>
143        <code>98052</code>
144      </postal>
145      <email>paulle@microsoft.com</email>
146    </address>
147  </author>
148   
149  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
150    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
151    <address>
152      <postal>
153        <street>MIT Computer Science and Artificial Intelligence Laboratory</street>
154        <street>The Stata Center, Building 32</street>
155        <street>32 Vassar Street</street>
156        <city>Cambridge</city>
157        <region>MA</region>
158        <code>02139</code>
159        <country>USA</country>
160      </postal>
161      <email>timbl@w3.org</email>
162      <uri>http://www.w3.org/People/Berners-Lee/</uri>
163    </address>
164  </author>
165
166  <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
167    <organization abbrev="W3C">World Wide Web Consortium</organization>
168    <address>
169      <postal>
170        <street>W3C / ERCIM</street>
171        <street>2004, rte des Lucioles</street>
172        <city>Sophia-Antipolis</city>
173        <region>AM</region>
174        <code>06902</code>
175        <country>France</country>
176      </postal>
177      <email>ylafon@w3.org</email>
178      <uri>http://www.raubacapeu.net/people/yves/</uri>
179    </address>
180  </author>
181
182  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
183    <organization abbrev="greenbytes">greenbytes GmbH</organization>
184    <address>
185      <postal>
186        <street>Hafenweg 16</street>
187        <city>Muenster</city><region>NW</region><code>48155</code>
188        <country>Germany</country>
189      </postal>
190      <phone>+49 251 2807760</phone>   
191      <facsimile>+49 251 2807761</facsimile>   
192      <email>julian.reschke@greenbytes.de</email>       
193      <uri>http://greenbytes.de/tech/webdav/</uri>     
194    </address>
195  </author>
196
197  <date month="&ID-MONTH;" year="&ID-YEAR;"/>
198
199<abstract>
200<t>
201   The Hypertext Transfer Protocol (HTTP) is an application-level
202   protocol for distributed, collaborative, hypermedia information
203   systems. HTTP has been in use by the World Wide Web global information
204   initiative since 1990. This document is Part 1 of the seven-part specification
205   that defines the protocol referred to as "HTTP/1.1" and, taken together,
206   obsoletes RFC 2616.  Part 1 provides an overview of HTTP and
207   its associated terminology, defines the "http" and "https" Uniform
208   Resource Identifier (URI) schemes, defines the generic message syntax
209   and parsing requirements for HTTP message frames, and describes
210   general security concerns for implementations.
211</t>
212</abstract>
213
214<note title="Editorial Note (To be removed by RFC Editor)">
215  <t>
216    Discussion of this draft should take place on the HTTPBIS working group
217    mailing list (ietf-http-wg@w3.org). The current issues list is
218    at <eref target="http://www.tools.ietf.org/wg/httpbis/trac/report/11"/>
219    and related documents (including fancy diffs) can be found at
220    <eref target="http://www.tools.ietf.org/wg/httpbis/"/>.
221  </t>
222  <t>
223    This draft incorporates those issue resolutions that were either
224    collected in the original RFC2616 errata list (<eref target="http://purl.org/NET/http-errata"/>),
225    or which were agreed upon on the mailing list between October 2006 and
226    November 2007 (as published in "draft-lafon-rfc2616bis-03").
227  </t>
228</note>
229</front>
230<middle>
231<section title="Introduction" anchor="introduction">
232<t>
233   The Hypertext Transfer Protocol (HTTP) is an application-level
234   protocol for distributed, collaborative, hypermedia information
235   systems. HTTP has been in use by the World-Wide Web global
236   information initiative since 1990. The first version of HTTP, commonly
237   referred to as HTTP/0.9, was a simple protocol for raw data transfer
238   across the Internet with only a single method and no metadata.
239   HTTP/1.0, as defined by <xref target="RFC1945"/>, improved
240   the protocol by allowing messages to be in the format of MIME-like
241   messages, containing metadata about the data transferred and
242   modifiers on the request/response semantics. However, HTTP/1.0 did
243   not sufficiently take into consideration the effects of hierarchical
244   proxies, caching, the need for persistent connections, or name-based
245   virtual hosts. In addition, the proliferation of incompletely-implemented
246   applications calling themselves "HTTP/1.0" necessitated a
247   protocol version change in order for two communicating applications
248   to determine each other's true capabilities.
249</t>
250<t>
251   This document is Part 1 of the seven-part specification that defines
252   the protocol referred to as "HTTP/1.1", obsoleting <xref target="RFC2616"/>.
253   HTTP/1.1 remains compatible with HTTP/1.0 by including more stringent
254   requirements that enable reliable implementations and adding only
255   those new features that will either be safely ignored by an HTTP/1.0
256   recipient or only sent when communicating with a party advertising
257   compliance with HTTP/1.1.
258   Part 1 defines those aspects of HTTP/1.1 related to overall network
259   operation, message framing, interaction with transport protocols, and
260   URI schemes.
261</t>
262<t>
263   This document is currently disorganized in order to minimize the changes
264   between drafts and enable reviewers to see the smaller errata changes.
265   The next draft will reorganize the sections to better reflect the content.
266   In particular, the sections will be organized according to the typical
267   process of deciding when to use HTTP (URI schemes), overall network operation,
268   connection management, message framing, and generic message parsing.
269   The current mess reflects how widely dispersed these topics and associated
270   requirements had become in <xref target="RFC2616"/>.
271</t>
272
273<section title="Purpose" anchor="intro.purpose">
274<t>
275   Practical information systems require more functionality than simple
276   retrieval, including search, front-end update, and annotation. HTTP
277   allows an open-ended set of methods and headers that indicate the
278   purpose of a request <xref target="RFC2324"/>. It builds on the discipline of reference
279   provided by the Uniform Resource Identifier (URI) <xref target="RFC1630"/>, as a location
280   (URL) <xref target="RFC1738"/> or name (URN) <xref target="RFC1737"/>, for indicating the resource to which a
281   method is to be applied. Messages are passed in a format similar to
282   that used by Internet mail <xref target="RFC2822"/> as defined by the Multipurpose
283   Internet Mail Extensions (MIME) <xref target="RFC2045"/>.
284</t>
285<t>
286   HTTP is also used as a generic protocol for communication between
287   user agents and proxies/gateways to other Internet systems, including
288   those supported by the SMTP <xref target="RFC2821"/>, NNTP <xref target="RFC3977"/>, FTP <xref target="RFC959"/>, Gopher <xref target="RFC1436"/>,
289   and WAIS <xref target="WAIS"/> protocols. In this way, HTTP allows basic hypermedia
290   access to resources available from diverse applications.
291</t>
292</section>
293
294<section title="Requirements" anchor="intro.requirements">
295<t>
296   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
297   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
298   document are to be interpreted as described in <xref target="RFC2119"/>.
299</t>
300<t>
301   An implementation is not compliant if it fails to satisfy one or more
302   of the &MUST; or &REQUIRED; level requirements for the protocols it
303   implements. An implementation that satisfies all the &MUST; or &REQUIRED;
304   level and all the &SHOULD; level requirements for its protocols is said
305   to be "unconditionally compliant"; one that satisfies all the &MUST;
306   level requirements but not all the &SHOULD; level requirements for its
307   protocols is said to be "conditionally compliant."
308</t>
309</section>
310
311<section title="Terminology" anchor="intro.terminology">
312<t>
313   This specification uses a number of terms to refer to the roles
314   played by participants in, and objects of, the HTTP communication.
315</t>
316<t>
317  <iref item="connection"/>
318  <x:dfn>connection</x:dfn>
319  <list>
320    <t>
321      A transport layer virtual circuit established between two programs
322      for the purpose of communication.
323    </t>
324  </list>
325</t>
326<t>
327  <iref item="message"/>
328  <x:dfn>message</x:dfn>
329  <list>
330    <t>
331      The basic unit of HTTP communication, consisting of a structured
332      sequence of octets matching the syntax defined in <xref target="http.message"/> and
333      transmitted via the connection.
334    </t>
335  </list>
336</t>
337<t>
338  <iref item="request"/>
339  <x:dfn>request</x:dfn>
340  <list>
341    <t>
342      An HTTP request message, as defined in <xref target="request"/>.
343    </t>
344  </list>
345</t>
346<t>
347  <iref item="response"/>
348  <x:dfn>response</x:dfn>
349  <list>
350    <t>
351      An HTTP response message, as defined in <xref target="response"/>.
352    </t>
353  </list>
354</t>
355<t>
356  <iref item="resource"/>
357  <x:dfn>resource</x:dfn>
358  <list>
359    <t>
360      A network data object or service that can be identified by a URI,
361      as defined in <xref target="uri"/>. Resources may be available in multiple
362      representations (e.g. multiple languages, data formats, size, and
363      resolutions) or vary in other ways.
364    </t>
365  </list>
366</t>
367<t>
368  <iref item="entity"/>
369  <x:dfn>entity</x:dfn>
370  <list>
371    <t>
372      The information transferred as the payload of a request or
373      response. An entity consists of metainformation in the form of
374      entity-header fields and content in the form of an entity-body, as
375      described in &entity;.
376    </t>
377  </list>
378</t>
379<t>
380  <iref item="representation"/>
381  <x:dfn>representation</x:dfn>
382  <list>
383    <t>
384      An entity included with a response that is subject to content
385      negotiation, as described in &content.negotiation;. There may exist multiple
386      representations associated with a particular response status.
387    </t>
388  </list>
389</t>
390<t>
391  <iref item="content negotiation"/>
392  <x:dfn>content negotiation</x:dfn>
393  <list>
394    <t>
395      The mechanism for selecting the appropriate representation when
396      servicing a request, as described in &content.negotiation;. The
397      representation of entities in any response can be negotiated
398      (including error responses).
399    </t>
400  </list>
401</t>
402<t>
403  <iref item="variant"/>
404  <x:dfn>variant</x:dfn>
405  <list>
406    <t>
407      A resource may have one, or more than one, representation(s)
408      associated with it at any given instant. Each of these
409      representations is termed a `variant'.  Use of the term `variant'
410      does not necessarily imply that the resource is subject to content
411      negotiation.
412    </t>
413  </list>
414</t>
415<t>
416  <iref item="client"/>
417  <x:dfn>client</x:dfn>
418  <list>
419    <t>
420      A program that establishes connections for the purpose of sending
421      requests.
422    </t>
423  </list>
424</t>
425<t>
426  <iref item="user agent"/>
427  <x:dfn>user agent</x:dfn>
428  <list>
429    <t>
430      The client which initiates a request. These are often browsers,
431      editors, spiders (web-traversing robots), or other end user tools.
432    </t>
433  </list>
434</t>
435<t>
436  <iref item="server"/>
437  <x:dfn>server</x:dfn>
438  <list>
439    <t>
440      An application program that accepts connections in order to
441      service requests by sending back responses. Any given program may
442      be capable of being both a client and a server; our use of these
443      terms refers only to the role being performed by the program for a
444      particular connection, rather than to the program's capabilities
445      in general. Likewise, any server may act as an origin server,
446      proxy, gateway, or tunnel, switching behavior based on the nature
447      of each request.
448    </t>
449  </list>
450</t>
451<t>
452  <iref item="origin server"/>
453  <x:dfn>origin server</x:dfn>
454  <list>
455    <t>
456      The server on which a given resource resides or is to be created.
457    </t>
458  </list>
459</t>
460<t>
461  <iref item="proxy"/>
462  <x:dfn>proxy</x:dfn>
463  <list>
464    <t>
465      An intermediary program which acts as both a server and a client
466      for the purpose of making requests on behalf of other clients.
467      Requests are serviced internally or by passing them on, with
468      possible translation, to other servers. A proxy &MUST; implement
469      both the client and server requirements of this specification. A
470      "transparent proxy" is a proxy that does not modify the request or
471      response beyond what is required for proxy authentication and
472      identification. A "non-transparent proxy" is a proxy that modifies
473      the request or response in order to provide some added service to
474      the user agent, such as group annotation services, media type
475      transformation, protocol reduction, or anonymity filtering. Except
476      where either transparent or non-transparent behavior is explicitly
477      stated, the HTTP proxy requirements apply to both types of
478      proxies.
479    </t>
480  </list>
481</t>
482<t>
483  <iref item="gateway"/>
484  <x:dfn>gateway</x:dfn>
485  <list>
486    <t>
487      A server which acts as an intermediary for some other server.
488      Unlike a proxy, a gateway receives requests as if it were the
489      origin server for the requested resource; the requesting client
490      may not be aware that it is communicating with a gateway.
491    </t>
492  </list>
493</t>
494<t>
495  <iref item="tunnel"/>
496  <x:dfn>tunnel</x:dfn>
497  <list>
498    <t>
499      An intermediary program which is acting as a blind relay between
500      two connections. Once active, a tunnel is not considered a party
501      to the HTTP communication, though the tunnel may have been
502      initiated by an HTTP request. The tunnel ceases to exist when both
503      ends of the relayed connections are closed.
504    </t>
505  </list>
506</t>
507<t>
508  <iref item="cache"/>
509  <x:dfn>cache</x:dfn>
510  <list>
511    <t>
512      A program's local store of response messages and the subsystem
513      that controls its message storage, retrieval, and deletion. A
514      cache stores cacheable responses in order to reduce the response
515      time and network bandwidth consumption on future, equivalent
516      requests. Any client or server may include a cache, though a cache
517      cannot be used by a server that is acting as a tunnel.
518    </t>
519  </list>
520</t>
521<t>
522  <iref item="cacheable"/>
523  <x:dfn>cacheable</x:dfn>
524  <list>
525    <t>
526      A response is cacheable if a cache is allowed to store a copy of
527      the response message for use in answering subsequent requests. The
528      rules for determining the cacheability of HTTP responses are
529      defined in &caching;. Even if a resource is cacheable, there may
530      be additional constraints on whether a cache can use the cached
531      copy for a particular request.
532    </t>
533  </list>
534</t>
535<t>
536  <iref item="upstream"/>
537  <iref item="downstream"/>
538  <x:dfn>upstream</x:dfn>/<x:dfn>downstream</x:dfn>
539  <list>
540    <t>
541      Upstream and downstream describe the flow of a message: all
542      messages flow from upstream to downstream.
543    </t>
544  </list>
545</t>
546<t>
547  <iref item="inbound"/>
548  <iref item="outbound"/>
549  <x:dfn>inbound</x:dfn>/<x:dfn>outbound</x:dfn>
550  <list>
551    <t>
552      Inbound and outbound refer to the request and response paths for
553      messages: "inbound" means "traveling toward the origin server",
554      and "outbound" means "traveling toward the user agent"
555    </t>
556  </list>
557</t>
558</section>
559
560<section title="Overall Operation" anchor="intro.overall.operation">
561<t>
562   HTTP is a request/response protocol. A client sends a
563   request to the server in the form of a request method, URI, and
564   protocol version, followed by a MIME-like message containing request
565   modifiers, client information, and possible body content over a
566   connection with a server. The server responds with a status line,
567   including the message's protocol version and a success or error code,
568   followed by a MIME-like message containing server information, entity
569   metainformation, and possible entity-body content. The relationship
570   between HTTP and MIME is described in &diff2045entity;.
571</t>
572<t>
573   Most HTTP communication is initiated by a user agent and consists of
574   a request to be applied to a resource on some origin server. In the
575   simplest case, this may be accomplished via a single connection (v)
576   between the user agent (UA) and the origin server (O).
577</t>
578<figure><artwork type="drawing">
579       request chain ------------------------&gt;
580    UA -------------------v------------------- O
581       &lt;----------------------- response chain
582</artwork></figure>
583<t>
584   A more complicated situation occurs when one or more intermediaries
585   are present in the request/response chain. There are three common
586   forms of intermediary: proxy, gateway, and tunnel. A proxy is a
587   forwarding agent, receiving requests for a URI in its absolute form,
588   rewriting all or part of the message, and forwarding the reformatted
589   request toward the server identified by the URI. A gateway is a
590   receiving agent, acting as a layer above some other server(s) and, if
591   necessary, translating the requests to the underlying server's
592   protocol. A tunnel acts as a relay point between two connections
593   without changing the messages; tunnels are used when the
594   communication needs to pass through an intermediary (such as a
595   firewall) even when the intermediary cannot understand the contents
596   of the messages.
597</t>
598<figure><artwork type="drawing">
599       request chain --------------------------------------&gt;
600    UA -----v----- A -----v----- B -----v----- C -----v----- O
601       &lt;------------------------------------- response chain
602</artwork></figure>
603<t>
604   The figure above shows three intermediaries (A, B, and C) between the
605   user agent and origin server. A request or response message that
606   travels the whole chain will pass through four separate connections.
607   This distinction is important because some HTTP communication options
608   may apply only to the connection with the nearest, non-tunnel
609   neighbor, only to the end-points of the chain, or to all connections
610   along the chain. Although the diagram is linear, each participant may
611   be engaged in multiple, simultaneous communications. For example, B
612   may be receiving requests from many clients other than A, and/or
613   forwarding requests to servers other than C, at the same time that it
614   is handling A's request.
615</t>
616<t>
617   Any party to the communication which is not acting as a tunnel may
618   employ an internal cache for handling requests. The effect of a cache
619   is that the request/response chain is shortened if one of the
620   participants along the chain has a cached response applicable to that
621   request. The following illustrates the resulting chain if B has a
622   cached copy of an earlier response from O (via C) for a request which
623   has not been cached by UA or A.
624</t>
625<figure><artwork type="drawing">
626          request chain ----------&gt;
627       UA -----v----- A -----v----- B - - - - - - C - - - - - - O
628          &lt;--------- response chain
629</artwork></figure>
630<t>
631   Not all responses are usefully cacheable, and some requests may
632   contain modifiers which place special requirements on cache behavior.
633   HTTP requirements for cache behavior and cacheable responses are
634   defined in &caching;.
635</t>
636<t>
637   In fact, there are a wide variety of architectures and configurations
638   of caches and proxies currently being experimented with or deployed
639   across the World Wide Web. These systems include national hierarchies
640   of proxy caches to save transoceanic bandwidth, systems that
641   broadcast or multicast cache entries, organizations that distribute
642   subsets of cached data via CD-ROM, and so on. HTTP systems are used
643   in corporate intranets over high-bandwidth links, and for access via
644   PDAs with low-power radio links and intermittent connectivity. The
645   goal of HTTP/1.1 is to support the wide diversity of configurations
646   already deployed while introducing protocol constructs that meet the
647   needs of those who build web applications that require high
648   reliability and, failing that, at least reliable indications of
649   failure.
650</t>
651<t>
652   HTTP communication usually takes place over TCP/IP connections. The
653   default port is TCP 80 (<eref target="http://www.iana.org/assignments/port-numbers"/>), but other ports can be used. This does
654   not preclude HTTP from being implemented on top of any other protocol
655   on the Internet, or on other networks. HTTP only presumes a reliable
656   transport; any protocol that provides such guarantees can be used;
657   the mapping of the HTTP/1.1 request and response structures onto the
658   transport data units of the protocol in question is outside the scope
659   of this specification.
660</t>
661<t>
662   In HTTP/1.0, most implementations used a new connection for each
663   request/response exchange. In HTTP/1.1, a connection may be used for
664   one or more request/response exchanges, although connections may be
665   closed for a variety of reasons (see <xref target="persistent.connections"/>).
666</t>
667</section>
668</section>
669
670<section title="Notational Conventions and Generic Grammar" anchor="notation">
671
672<section title="Augmented BNF" anchor="notation.abnf">
673<t>
674   All of the mechanisms specified in this document are described in
675   both prose and an augmented Backus-Naur Form (BNF) similar to that
676   used by <xref target="RFC822ABNF"/>. Implementors will need to be familiar with the
677   notation in order to understand this specification. The augmented BNF
678   includes the following constructs:
679</t>
680<t>
681   name = definition
682  <list>
683    <t>
684      The name of a rule is simply the name itself (without any
685      enclosing "&lt;" and "&gt;") and is separated from its definition by the
686      equal "=" character. White space is only significant in that
687      indentation of continuation lines is used to indicate a rule
688      definition that spans more than one line. Certain basic rules are
689      in uppercase, such as SP, LWS, HTAB, CRLF, DIGIT, ALPHA, etc. Angle
690      brackets are used within definitions whenever their presence will
691      facilitate discerning the use of rule names.
692    </t>
693  </list>
694</t>
695<t>
696   "literal"
697  <list>
698    <t>
699      Quotation marks surround literal text. Unless stated otherwise,
700      the text is case-insensitive.
701    </t>
702  </list>
703</t>
704<t>
705   rule1 | rule2
706  <list>
707    <t>
708      Elements separated by a bar ("|") are alternatives, e.g., "yes |
709      no" will accept yes or no.
710    </t>
711  </list>
712</t>
713<t>
714   (rule1 rule2)
715  <list>
716    <t>
717      Elements enclosed in parentheses are treated as a single element.
718      Thus, "(elem (foo | bar) elem)" allows the token sequences "elem
719      foo elem" and "elem bar elem".
720    </t>
721  </list>
722</t>
723<t>
724   *rule
725  <list>
726    <t>
727      The character "*" preceding an element indicates repetition. The
728      full form is "&lt;n&gt;*&lt;m&gt;element" indicating at least &lt;n&gt; and at most
729      &lt;m&gt; occurrences of element. Default values are 0 and infinity so
730      that "*(element)" allows any number, including zero; "1*element"
731      requires at least one; and "1*2element" allows one or two.
732    </t>
733  </list>
734</t>
735<t>
736   [rule]
737  <list>
738    <t>
739      Square brackets enclose optional elements; "[foo bar]" is
740      equivalent to "*1(foo bar)".
741    </t>
742  </list>
743</t>
744<t>
745   N rule
746  <list>
747    <t>
748      Specific repetition: "&lt;n&gt;(element)" is equivalent to
749      "&lt;n&gt;*&lt;n&gt;(element)"; that is, exactly &lt;n&gt; occurrences of (element).
750      Thus 2DIGIT is a 2-digit number, and 3ALPHA is a string of three
751      alphabetic characters.
752    </t>
753  </list>
754</t>
755<t>
756   #rule
757  <list>
758    <t>
759      A construct "#" is defined, similar to "*", for defining lists of
760      elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating at least
761      &lt;n&gt; and at most &lt;m&gt; elements, each separated by one or more commas
762      (",") and &OPTIONAL; linear white space (LWS). This makes the usual
763      form of lists very easy; a rule such as
764    </t>
765    <t>
766         ( *LWS element *( *LWS "," *LWS element ))
767    </t>
768    <t>
769      can be shown as
770    </t>
771    <t>
772         1#element
773    </t>
774    <t>
775      Wherever this construct is used, null elements are allowed, but do
776      not contribute to the count of elements present. That is,
777      "(element), , (element) " is permitted, but counts as only two
778      elements. Therefore, where at least one element is required, at
779      least one non-null element &MUST; be present. Default values are 0
780      and infinity so that "#element" allows any number, including zero;
781      "1#element" requires at least one; and "1#2element" allows one or
782      two.
783    </t>
784  </list>
785</t>
786<t>
787   ; comment
788  <list>
789    <t>
790      A semi-colon, set off some distance to the right of rule text,
791      starts a comment that continues to the end of line. This is a
792      simple way of including useful notes in parallel with the
793      specifications.
794    </t>
795  </list>
796</t>
797<t>
798   implied *LWS
799  <list>
800    <t>
801      The grammar described by this specification is word-based. Except
802      where noted otherwise, linear white space (LWS) can be included
803      between any two adjacent words (token or quoted-string), and
804      between adjacent words and separators, without changing the
805      interpretation of a field. At least one delimiter (LWS and/or
806      separators) &MUST; exist between any two tokens (for the definition
807      of "token" below), since they would otherwise be interpreted as a
808      single token.
809    </t>
810  </list>
811</t>
812</section>
813
814<section title="Basic Rules" anchor="basic.rules">
815<x:anchor-alias value="OCTET"/>
816<x:anchor-alias value="CHAR"/>
817<x:anchor-alias value="ALPHA"/>
818<x:anchor-alias value="DIGIT"/>
819<x:anchor-alias value="CTL"/>
820<x:anchor-alias value="CR"/>
821<x:anchor-alias value="LF"/>
822<x:anchor-alias value="SP"/>
823<x:anchor-alias value="HTAB"/>
824<x:anchor-alias value="CRLF"/>
825<x:anchor-alias value="LWS"/>
826<x:anchor-alias value="TEXT"/>
827<x:anchor-alias value="HEX"/>
828<x:anchor-alias value="token"/>
829<x:anchor-alias value="separators"/>
830<x:anchor-alias value="comment"/>
831<x:anchor-alias value="ctext"/>
832<x:anchor-alias value="quoted-string"/>
833<x:anchor-alias value="qdtext"/>
834<x:anchor-alias value="quoted-pair"/>
835<t>
836   The following rules are used throughout this specification to
837   describe basic parsing constructs. The US-ASCII coded character set
838   is defined by ANSI X3.4-1986 <xref target="USASCII"/>.
839</t>
840<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="OCTET"/><iref primary="true" item="Grammar" subitem="CHAR"/><iref primary="true" item="Grammar" subitem="ALPHA"/><iref primary="true" item="Grammar" subitem="DIGIT"/><iref primary="true" item="Grammar" subitem="CTL"/><iref primary="true" item="Grammar" subitem="CR"/><iref primary="true" item="Grammar" subitem="LF"/><iref primary="true" item="Grammar" subitem="SP"/><iref primary="true" item="Grammar" subitem="HTAB"/><iref primary="true" item="Grammar" subitem="DQUOTE"/>
841  OCTET          = %x00-FF
842                   ; any 8-bit sequence of data
843  CHAR           = %x01-7F
844                   ; any US-ASCII character, excluding NUL
845  ALPHA          = %x41-5A | %x61-7A
846                   ; A-Z | a-z
847  DIGIT          = %x30-39
848                   ; any US-ASCII digit "0".."9"
849  CTL            = %x00-1F | %x7F
850                   ; (octets 0 - 31) and DEL (127)
851  CR             = %x0D
852                   ; US-ASCII CR, carriage return (13)
853  LF             = %x0A
854                   ; US-ASCII LF, linefeed (10)
855  SP             = %x20
856                   ; US-ASCII SP, space (32)
857  HTAB           = %x09
858                   ; US-ASCII HT, horizontal-tab (9)
859  DQUOTE         = %x22
860                   ; US-ASCII double-quote mark (34)
861</artwork></figure>
862<t>
863   HTTP/1.1 defines the sequence CR LF as the end-of-line marker for all
864   protocol elements except the entity-body (see <xref target="tolerant.applications"/> for
865   tolerant applications). The end-of-line marker within an entity-body
866   is defined by its associated media type, as described in &media-types;.
867</t>
868<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="CRLF"/>
869  CRLF           = CR LF
870</artwork></figure>
871<t>
872   HTTP/1.1 header field values can be folded onto multiple lines if the
873   continuation line begins with a space or horizontal tab. All linear
874   white space, including folding, has the same semantics as SP. A
875   recipient &MAY; replace any linear white space with a single SP before
876   interpreting the field value or forwarding the message downstream.
877</t>
878<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="LWS"/>
879  LWS            = [CRLF] 1*( SP | HTAB )
880</artwork></figure>
881<t>
882   The TEXT rule is only used for descriptive field contents and values
883   that are not intended to be interpreted by the message parser. Words
884   of *TEXT &MAY; contain characters from character sets other than ISO-8859-1
885   <xref target="ISO-8859-1"/> only when encoded according to the rules of
886   <xref target="RFC2047"/>.
887</t>
888<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TEXT"/>
889  TEXT           = &lt;any OCTET except CTLs, but including LWS&gt;
890</artwork></figure>
891<t>
892   A CRLF is allowed in the definition of TEXT only as part of a header
893   field continuation. It is expected that the folding LWS will be
894   replaced with a single SP before interpretation of the TEXT value.
895</t>
896<t>
897   Hexadecimal numeric characters are used in several protocol elements.
898</t>
899<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HEX"/>
900  HEX            = "A" | "B" | "C" | "D" | "E" | "F"
901                 | "a" | "b" | "c" | "d" | "e" | "f" | DIGIT
902</artwork></figure>
903<t>
904   Many HTTP/1.1 header field values consist of words separated by LWS
905   or special characters. These special characters &MUST; be in a quoted
906   string to be used within a parameter value (as defined in
907   <xref target="transfer.codings"/>).
908</t>
909<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="token"/><iref primary="true" item="Grammar" subitem="separators"/>
910  token          = 1*&lt;any CHAR except CTLs or separators&gt;
911  separators     = "(" | ")" | "&lt;" | "&gt;" | "@"
912                 | "," | ";" | ":" | "\" | DQUOTE
913                 | "/" | "[" | "]" | "?" | "="
914                 | "{" | "}" | SP | HTAB
915</artwork></figure>
916<t>
917   Comments can be included in some HTTP header fields by surrounding
918   the comment text with parentheses. Comments are only allowed in
919   fields containing "comment" as part of their field value definition.
920   In all other fields, parentheses are considered part of the field
921   value.
922</t>
923<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/>
924  comment        = "(" *( ctext | quoted-pair | comment ) ")"
925  ctext          = &lt;any TEXT excluding "(" and ")"&gt;
926</artwork></figure>
927<t>
928   A string of text is parsed as a single word if it is quoted using
929   double-quote marks.
930</t>
931<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-string"/><iref primary="true" item="Grammar" subitem="qdtext"/>
932  quoted-string  = ( DQUOTE *(qdtext | quoted-pair ) DQUOTE )
933  qdtext         = &lt;any TEXT excluding DQUOTE and "\">
934</artwork></figure>
935<t>
936   The backslash character ("\") &MAY; be used as a single-character
937   quoting mechanism only within quoted-string and comment constructs.
938</t>
939<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-pair"/>
940  quoted-pair    = "\" CHAR
941</artwork></figure>
942</section>
943
944<section title="ABNF Rules defined in other Parts of the Specification" anchor="abnf.dependencies">
945<t>
946  The ABNF rules below are defined in other parts:
947</t>
948<figure><!-- Part2--><artwork type="abnf2616">
949  request-header =  &lt;request-header, defined in &request-header-fields;&gt;
950  response-header = &lt;response-header, defined in &response-header-fields;&gt;
951</artwork></figure>
952<figure><!-- Part3--><artwork type="abnf2616">
953  accept-params   = &lt;accept-params, defined in &header-accept;&gt;
954  entity-body     = &lt;entity-body, defined in &entity-body;&gt;
955  entity-header   = &lt;entity-header, defined in &entity-header-fields;&gt;
956</artwork></figure>
957<figure><!-- Part6--><artwork type="abnf2616">
958  Cache-Control   = &lt;Cache-Control, defined in &header-pragma;&gt;
959  Pragma          = &lt;Pragma, defined in &header-pragma;&gt;
960  Warning         = &lt;Warning, defined in &header-warning;&gt;
961</artwork></figure>
962</section>
963
964</section>
965
966<section title="Protocol Parameters" anchor="protocol.parameters">
967
968<section title="HTTP Version" anchor="http.version">
969<t>
970   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate versions
971   of the protocol. The protocol versioning policy is intended to allow
972   the sender to indicate the format of a message and its capacity for
973   understanding further HTTP communication, rather than the features
974   obtained via that communication. No change is made to the version
975   number for the addition of message components which do not affect
976   communication behavior or which only add to extensible field values.
977   The &lt;minor&gt; number is incremented when the changes made to the
978   protocol add features which do not change the general message parsing
979   algorithm, but which may add to the message semantics and imply
980   additional capabilities of the sender. The &lt;major&gt; number is
981   incremented when the format of a message within the protocol is
982   changed. See <xref target="RFC2145"/> for a fuller explanation.
983</t>
984<t>
985   The version of an HTTP message is indicated by an HTTP-Version field
986   in the first line of the message. HTTP-Version is case-sensitive.
987</t>
988<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-Version"/>
989  HTTP-Version   = "HTTP" "/" 1*DIGIT "." 1*DIGIT
990</artwork></figure>
991<t>
992   Note that the major and minor numbers &MUST; be treated as separate
993   integers and that each &MAY; be incremented higher than a single digit.
994   Thus, HTTP/2.4 is a lower version than HTTP/2.13, which in turn is
995   lower than HTTP/12.3. Leading zeros &MUST; be ignored by recipients and
996   &MUST-NOT; be sent.
997</t>
998<t>
999   An application that sends a request or response message that includes
1000   HTTP-Version of "HTTP/1.1" &MUST; be at least conditionally compliant
1001   with this specification. Applications that are at least conditionally
1002   compliant with this specification &SHOULD; use an HTTP-Version of
1003   "HTTP/1.1" in their messages, and &MUST; do so for any message that is
1004   not compatible with HTTP/1.0. For more details on when to send
1005   specific HTTP-Version values, see <xref target="RFC2145"/>.
1006</t>
1007<t>
1008   The HTTP version of an application is the highest HTTP version for
1009   which the application is at least conditionally compliant.
1010</t>
1011<t>
1012   Proxy and gateway applications need to be careful when forwarding
1013   messages in protocol versions different from that of the application.
1014   Since the protocol version indicates the protocol capability of the
1015   sender, a proxy/gateway &MUST-NOT; send a message with a version
1016   indicator which is greater than its actual version. If a higher
1017   version request is received, the proxy/gateway &MUST; either downgrade
1018   the request version, or respond with an error, or switch to tunnel
1019   behavior.
1020</t>
1021<t>
1022   Due to interoperability problems with HTTP/1.0 proxies discovered
1023   since the publication of <xref target="RFC2068"/>, caching proxies &MUST;, gateways
1024   &MAY;, and tunnels &MUST-NOT; upgrade the request to the highest version
1025   they support. The proxy/gateway's response to that request &MUST; be in
1026   the same major version as the request.
1027</t>
1028<t>
1029  <list>
1030    <t>
1031      <x:h>Note:</x:h> Converting between versions of HTTP may involve modification
1032      of header fields required or forbidden by the versions involved.
1033    </t>
1034  </list>
1035</t>
1036</section>
1037
1038<section title="Uniform Resource Identifiers" anchor="uri">
1039<t>
1040   URIs have been known by many names: WWW addresses, Universal Document
1041   Identifiers, Universal Resource Identifiers <xref target="RFC1630"/>, and finally the
1042   combination of Uniform Resource Locators (URL) <xref target="RFC1738"/> and Names (URN)
1043   <xref target="RFC1737"/>. As far as HTTP is concerned, Uniform Resource Identifiers are
1044   simply formatted strings which identify--via name, location, or any
1045   other characteristic--a resource.
1046</t>
1047
1048<section title="General Syntax" anchor="general.syntax">
1049<t>
1050   URIs in HTTP can be represented in absolute form or relative to some
1051   known base URI <xref target="RFC1808"/>, depending upon the context of their use. The two
1052   forms are differentiated by the fact that absolute URIs always begin
1053   with a scheme name followed by a colon. For definitive information on
1054   URL syntax and semantics, see "Uniform Resource Identifiers (URI):
1055   Generic Syntax and Semantics," <xref target="RFC2396"/> (which replaces <xref target="RFC1738"/>
1056   and <xref target="RFC1808"/>). This specification adopts the
1057   definitions of "URI-reference", "absoluteURI", "fragment", "relativeURI", "port",
1058   "host", "abs_path", "query", and "authority" from that specification:
1059</t>
1060<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="absoluteURI"/><iref primary="true" item="Grammar" subitem="authority"/><iref primary="true" item="Grammar" subitem="path-absolute"/><iref primary="true" item="Grammar" subitem="port"/><iref primary="true" item="Grammar" subitem="query"/><iref primary="true" item="Grammar" subitem="relativeURI"/><iref primary="true" item="Grammar" subitem="uri-host"/>
1061  absoluteURI   = &lt;absoluteURI, defined in <xref target="RFC2396" x:fmt="," x:sec="3"/>>
1062  authority     = &lt;authority, defined in <xref target="RFC2396" x:fmt="," x:sec="3.2"/>>
1063  fragment      = &lt;fragment, defined in <xref target="RFC2396" x:fmt="," x:sec="4.1"/>>
1064  path-absolute = &lt;abs_path, defined in <xref target="RFC2396" x:fmt="," x:sec="3"/>>
1065  port          = &lt;port, defined in <xref target="RFC2396" x:fmt="," x:sec="3.2.2"/>>
1066  query         = &lt;query, defined in <xref target="RFC2396" x:fmt="," x:sec="3.4"/>>
1067  relativeURI   = &lt;relativeURI, defined in <xref target="RFC2396" x:fmt="," x:sec="5"/>>
1068  uri-host      = &lt;host, defined in <xref target="RFC2396" x:fmt="," x:sec="3.2.2"/>>
1069</artwork></figure>
1070<t>
1071   HTTP does not place any a priori limit on the length of
1072   a URI. Servers &MUST; be able to handle the URI of any resource they
1073   serve, and &SHOULD; be able to handle URIs of unbounded length if they
1074   provide GET-based forms that could generate such URIs. A server
1075   &SHOULD; return 414 (Request-URI Too Long) status if a URI is longer
1076   than the server can handle (see &status-414;).
1077</t>
1078<t>
1079  <list>
1080    <t>
1081      <x:h>Note:</x:h> Servers ought to be cautious about depending on URI lengths
1082      above 255 bytes, because some older client or proxy
1083      implementations might not properly support these lengths.
1084    </t>
1085  </list>
1086</t>
1087</section>
1088
1089<section title="http URL" anchor="http.url">
1090<t>
1091   The "http" scheme is used to locate network resources via the HTTP
1092   protocol. This section defines the scheme-specific syntax and
1093   semantics for http URLs.
1094</t>
1095<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="http-URL"/>
1096  http-URL = "http:" "//" uri-host [ ":" port ]
1097             [ path-absolute [ "?" query ]]
1098</artwork></figure>
1099<t>
1100   If the port is empty or not given, port 80 is assumed. The semantics
1101   are that the identified resource is located at the server listening
1102   for TCP connections on that port of that host, and the Request-URI
1103   for the resource is path-absolute (<xref target="request-uri"/>). The use of IP addresses
1104   in URLs &SHOULD; be avoided whenever possible (see <xref target="RFC1900"/>). If
1105   the path-absolute is not present in the URL, it &MUST; be given as "/" when
1106   used as a Request-URI for a resource (<xref target="request-uri"/>). If a proxy
1107   receives a host name which is not a fully qualified domain name, it
1108   &MAY; add its domain to the host name it received. If a proxy receives
1109   a fully qualified domain name, the proxy &MUST-NOT; change the host
1110   name.
1111</t>
1112</section>
1113
1114<section title="URI Comparison" anchor="uri.comparison">
1115<t>
1116   When comparing two URIs to decide if they match or not, a client
1117   &SHOULD; use a case-sensitive octet-by-octet comparison of the entire
1118   URIs, with these exceptions:
1119  <list style="symbols">
1120    <t>A port that is empty or not given is equivalent to the default
1121        port for that URI-reference;</t>
1122    <t>Comparisons of host names &MUST; be case-insensitive;</t>
1123    <t>Comparisons of scheme names &MUST; be case-insensitive;</t>
1124    <t>An empty path-absolute is equivalent to an path-absolute of "/".</t>
1125  </list>
1126</t>
1127<t>
1128   Characters other than those in the "reserved" set (see
1129   <xref target="RFC2396"/>) are equivalent to their ""%" HEX HEX" encoding.
1130</t>
1131<t>
1132   For example, the following three URIs are equivalent:
1133</t>
1134<figure><artwork type="example">
1135   http://example.com:80/~smith/home.html
1136   http://EXAMPLE.com/%7Esmith/home.html
1137   http://EXAMPLE.com:/%7esmith/home.html
1138</artwork></figure>
1139</section>
1140</section>
1141
1142<section title="Date/Time Formats" anchor="date.time.formats">
1143<section title="Full Date" anchor="full.date">
1144<t>
1145   HTTP applications have historically allowed three different formats
1146   for the representation of date/time stamps:
1147</t>
1148<figure><artwork type="example">
1149   Sun, 06 Nov 1994 08:49:37 GMT  ; RFC 822, updated by RFC 1123
1150   Sunday, 06-Nov-94 08:49:37 GMT ; obsolete RFC 850 format
1151   Sun Nov  6 08:49:37 1994       ; ANSI C's asctime() format
1152</artwork></figure>
1153<t>
1154   The first format is preferred as an Internet standard and represents
1155   a fixed-length subset of that defined by <xref target="RFC1123"/> (an update to
1156   <xref target="RFC822"/>). The other formats are described here only for
1157   compatibility with obsolete implementations.
1158   HTTP/1.1 clients and servers that parse the date value &MUST; accept
1159   all three formats (for compatibility with HTTP/1.0), though they &MUST;
1160   only generate the RFC 1123 format for representing HTTP-date values
1161   in header fields. See <xref target="tolerant.applications"/> for further information.
1162</t>
1163<t><list><t>
1164      <x:h>Note:</x:h> Recipients of date values are encouraged to be robust in
1165      accepting date values that may have been sent by non-HTTP
1166      applications, as is sometimes the case when retrieving or posting
1167      messages via proxies/gateways to SMTP or NNTP.
1168</t></list></t>
1169<t>
1170   All HTTP date/time stamps &MUST; be represented in Greenwich Mean Time
1171   (GMT), without exception. For the purposes of HTTP, GMT is exactly
1172   equal to UTC (Coordinated Universal Time). This is indicated in the
1173   first two formats by the inclusion of "GMT" as the three-letter
1174   abbreviation for time zone, and &MUST; be assumed when reading the
1175   asctime format. HTTP-date is case sensitive and &MUST-NOT; include
1176   additional LWS beyond that specifically included as SP in the
1177   grammar.
1178</t>
1179<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"/>
1180  HTTP-date    = rfc1123-date | rfc850-date | asctime-date
1181  rfc1123-date = wkday "," SP date1 SP time SP "GMT"
1182  rfc850-date  = weekday "," SP date2 SP time SP "GMT"
1183  asctime-date = wkday SP date3 SP time SP 4DIGIT
1184  date1        = 2DIGIT SP month SP 4DIGIT
1185                 ; day month year (e.g., 02 Jun 1982)
1186  date2        = 2DIGIT "-" month "-" 2DIGIT
1187                 ; day-month-year (e.g., 02-Jun-82)
1188  date3        = month SP ( 2DIGIT | ( SP 1DIGIT ))
1189                 ; month day (e.g., Jun  2)
1190  time         = 2DIGIT ":" 2DIGIT ":" 2DIGIT
1191                 ; 00:00:00 - 23:59:59
1192  wkday        = "Mon" | "Tue" | "Wed"
1193               | "Thu" | "Fri" | "Sat" | "Sun"
1194  weekday      = "Monday" | "Tuesday" | "Wednesday"
1195               | "Thursday" | "Friday" | "Saturday" | "Sunday"
1196  month        = "Jan" | "Feb" | "Mar" | "Apr"
1197               | "May" | "Jun" | "Jul" | "Aug"
1198               | "Sep" | "Oct" | "Nov" | "Dec"
1199</artwork></figure>
1200<t>
1201      <x:h>Note:</x:h> HTTP requirements for the date/time stamp format apply only
1202      to their usage within the protocol stream. Clients and servers are
1203      not required to use these formats for user presentation, request
1204      logging, etc.
1205</t>
1206</section>
1207
1208<section title="Time Spans" anchor="time.spans">
1209<t>
1210  Time spans are presented as non-negative decimal integers, representing time
1211  in seconds.
1212</t>
1213<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="delta-seconds"/>
1214  delta-seconds  = 1*DIGIT
1215</artwork></figure>
1216</section>
1217
1218</section>
1219
1220<section title="Transfer Codings" anchor="transfer.codings">
1221<t>
1222   Transfer-coding values are used to indicate an encoding
1223   transformation that has been, can be, or may need to be applied to an
1224   entity-body in order to ensure "safe transport" through the network.
1225   This differs from a content coding in that the transfer-coding is a
1226   property of the message, not of the original entity.
1227</t>
1228<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/>
1229  transfer-coding         = "chunked" | transfer-extension
1230  transfer-extension      = token *( ";" parameter )
1231</artwork></figure>
1232<t>
1233   Parameters are in  the form of attribute/value pairs.
1234</t>
1235<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"/>
1236  parameter               = attribute "=" value
1237  attribute               = token
1238  value                   = token | quoted-string
1239</artwork></figure>
1240<t>
1241   All transfer-coding values are case-insensitive. HTTP/1.1 uses
1242   transfer-coding values in the TE header field (<xref target="header.te"/>) and in
1243   the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1244</t>
1245<t>
1246   Whenever a transfer-coding is applied to a message-body, the set of
1247   transfer-codings &MUST; include "chunked", unless the message is
1248   terminated by closing the connection. When the "chunked" transfer-coding
1249   is used, it &MUST; be the last transfer-coding applied to the
1250   message-body. The "chunked" transfer-coding &MUST-NOT; be applied more
1251   than once to a message-body. These rules allow the recipient to
1252   determine the transfer-length of the message (<xref target="message.length"/>).
1253</t>
1254<t>
1255   Transfer-codings are analogous to the Content-Transfer-Encoding
1256   values of MIME <xref target="RFC2045"/>, which were designed to enable safe transport of
1257   binary data over a 7-bit transport service. However, safe transport
1258   has a different focus for an 8bit-clean transfer protocol. In HTTP,
1259   the only unsafe characteristic of message-bodies is the difficulty in
1260   determining the exact body length (<xref target="message.length"/>), or the desire to
1261   encrypt data over a shared transport.
1262</t>
1263<t>
1264   The Internet Assigned Numbers Authority (IANA) acts as a registry for
1265   transfer-coding value tokens. Initially, the registry contains the
1266   following tokens: "chunked" (<xref target="chunked.transfer.encoding"/>),
1267   "gzip", "compress", and "deflate" (&content-codings;).
1268</t>
1269<t>
1270   New transfer-coding value tokens &SHOULD; be registered in the same way
1271   as new content-coding value tokens (&content-codings;).
1272</t>
1273<t>
1274   A server which receives an entity-body with a transfer-coding it does
1275   not understand &SHOULD; return 501 (Not Implemented), and close the
1276   connection. A server &MUST-NOT; send transfer-codings to an HTTP/1.0
1277   client.
1278</t>
1279
1280<section title="Chunked Transfer Coding" anchor="chunked.transfer.encoding">
1281<t>
1282   The chunked encoding modifies the body of a message in order to
1283   transfer it as a series of chunks, each with its own size indicator,
1284   followed by an &OPTIONAL; trailer containing entity-header fields. This
1285   allows dynamically produced content to be transferred along with the
1286   information necessary for the recipient to verify that it has
1287   received the full message.
1288</t>
1289<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-part"/>
1290  Chunked-Body   = *chunk
1291                   last-chunk
1292                   trailer-part
1293                   CRLF
1294 
1295  chunk          = chunk-size [ chunk-extension ] CRLF
1296                   chunk-data CRLF
1297  chunk-size     = 1*HEX
1298  last-chunk     = 1*("0") [ chunk-extension ] CRLF
1299 
1300  chunk-extension= *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
1301  chunk-ext-name = token
1302  chunk-ext-val  = token | quoted-string
1303  chunk-data     = 1*OCTET ; a sequence of chunk-size octets
1304  trailer-part   = *(entity-header CRLF)
1305</artwork></figure>
1306<t>
1307   The chunk-size field is a string of hex digits indicating the size of
1308   the chunk-data in octets. The chunked encoding is ended by any chunk whose size is
1309   zero, followed by the trailer, which is terminated by an empty line.
1310</t>
1311<t>
1312   The trailer allows the sender to include additional HTTP header
1313   fields at the end of the message. The Trailer header field can be
1314   used to indicate which header fields are included in a trailer (see
1315   <xref target="header.trailer"/>).
1316</t>
1317<t>
1318   A server using chunked transfer-coding in a response &MUST-NOT; use the
1319   trailer for any header fields unless at least one of the following is
1320   true:
1321  <list style="numbers">
1322    <t>the request included a TE header field that indicates "trailers" is
1323     acceptable in the transfer-coding of the  response, as described in
1324     <xref target="header.te"/>; or,</t>
1325
1326    <t>the server is the origin server for the response, the trailer
1327     fields consist entirely of optional metadata, and the recipient
1328     could use the message (in a manner acceptable to the origin server)
1329     without receiving this metadata.  In other words, the origin server
1330     is willing to accept the possibility that the trailer fields might
1331     be silently discarded along the path to the client.</t>
1332  </list>
1333</t>
1334<t>
1335   This requirement prevents an interoperability failure when the
1336   message is being received by an HTTP/1.1 (or later) proxy and
1337   forwarded to an HTTP/1.0 recipient. It avoids a situation where
1338   compliance with the protocol would have necessitated a possibly
1339   infinite buffer on the proxy.
1340</t>
1341<t>
1342   A process for decoding the "chunked" transfer-coding
1343   can be represented in pseudo-code as:
1344</t>
1345<figure><artwork type="code">
1346    length := 0
1347    read chunk-size, chunk-extension (if any) and CRLF
1348    while (chunk-size &gt; 0) {
1349       read chunk-data and CRLF
1350       append chunk-data to entity-body
1351       length := length + chunk-size
1352       read chunk-size and CRLF
1353    }
1354    read entity-header
1355    while (entity-header not empty) {
1356       append entity-header to existing header fields
1357       read entity-header
1358    }
1359    Content-Length := length
1360    Remove "chunked" from Transfer-Encoding
1361</artwork></figure>
1362<t>
1363   All HTTP/1.1 applications &MUST; be able to receive and decode the
1364   "chunked" transfer-coding, and &MUST; ignore chunk-extension extensions
1365   they do not understand.
1366</t>
1367</section>
1368</section>
1369
1370<section title="Product Tokens" anchor="product.tokens">
1371<t>
1372   Product tokens are used to allow communicating applications to
1373   identify themselves by software name and version. Most fields using
1374   product tokens also allow sub-products which form a significant part
1375   of the application to be listed, separated by white space. By
1376   convention, the products are listed in order of their significance
1377   for identifying the application.
1378</t>
1379<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="product"/><iref primary="true" item="Grammar" subitem="product-version"/>
1380  product         = token ["/" product-version]
1381  product-version = token
1382</artwork></figure>
1383<t>
1384   Examples:
1385</t>
1386<figure><artwork type="example">
1387    User-Agent: CERN-LineMode/2.15 libwww/2.17b3
1388    Server: Apache/0.8.4
1389</artwork></figure>
1390<t>
1391   Product tokens &SHOULD; be short and to the point. They &MUST-NOT; be
1392   used for advertising or other non-essential information. Although any
1393   token character &MAY; appear in a product-version, this token &SHOULD;
1394   only be used for a version identifier (i.e., successive versions of
1395   the same product &SHOULD; only differ in the product-version portion of
1396   the product value).
1397</t>
1398</section>
1399
1400</section>
1401
1402<section title="HTTP Message" anchor="http.message">
1403
1404<section title="Message Types" anchor="message.types">
1405<t>
1406   HTTP messages consist of requests from client to server and responses
1407   from server to client.
1408</t>
1409<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-message"/>
1410  HTTP-message   = Request | Response     ; HTTP/1.1 messages
1411</artwork></figure>
1412<t>
1413   Request (<xref target="request"/>) and Response (<xref target="response"/>) messages use the generic
1414   message format of <xref target="RFC2822"/> for transferring entities (the payload
1415   of the message). Both types of message consist of a start-line, zero
1416   or more header fields (also known as "headers"), an empty line (i.e.,
1417   a line with nothing preceding the CRLF) indicating the end of the
1418   header fields, and possibly a message-body.
1419</t>
1420<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="generic-message"/><iref primary="true" item="Grammar" subitem="start-line"/>
1421  generic-message = start-line
1422                    *(message-header CRLF)
1423                    CRLF
1424                    [ message-body ]
1425  start-line      = Request-Line | Status-Line
1426</artwork></figure>
1427<t>
1428   In the interest of robustness, servers &SHOULD; ignore any empty
1429   line(s) received where a Request-Line is expected. In other words, if
1430   the server is reading the protocol stream at the beginning of a
1431   message and receives a CRLF first, it should ignore the CRLF.
1432</t>
1433<t>
1434   Certain buggy HTTP/1.0 client implementations generate extra CRLF's
1435   after a POST request. To restate what is explicitly forbidden by the
1436   BNF, an HTTP/1.1 client &MUST-NOT; preface or follow a request with an
1437   extra CRLF.
1438</t>
1439</section>
1440
1441<section title="Message Headers" anchor="message.headers">
1442<t>
1443   HTTP header fields, which include general-header (<xref target="general.header.fields"/>),
1444   request-header (&request-header-fields;), response-header (&response-header-fields;), and
1445   entity-header (&entity-header-fields;) fields, follow the same generic format as
1446   that given in <xref target="RFC2822" x:fmt="of" x:sec="2.1"/>. Each header field consists
1447   of a name followed by a colon (":") and the field value. Field names
1448   are case-insensitive. The field value &MAY; be preceded by any amount
1449   of LWS, though a single SP is preferred. Header fields can be
1450   extended over multiple lines by preceding each extra line with at
1451   least one SP or HTAB. Applications ought to follow "common form", where
1452   one is known or indicated, when generating HTTP constructs, since
1453   there might exist some implementations that fail to accept anything
1454   beyond the common forms.
1455</t>
1456<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"/>
1457  message-header = field-name ":" [ field-value ]
1458  field-name     = token
1459  field-value    = *( field-content | LWS )
1460  field-content  = &lt;field content&gt;
1461                   ; the OCTETs making up the field-value
1462                   ; and consisting of either *TEXT or combinations
1463                   ; of token, separators, and quoted-string
1464</artwork></figure>
1465<t>
1466   The field-content does not include any leading or trailing LWS:
1467   linear white space occurring before the first non-whitespace
1468   character of the field-value or after the last non-whitespace
1469   character of the field-value. Such leading or trailing LWS &MAY; be
1470   removed without changing the semantics of the field value. Any LWS
1471   that occurs between field-content &MAY; be replaced with a single SP
1472   before interpreting the field value or forwarding the message
1473   downstream.
1474</t>
1475<t>
1476   The order in which header fields with differing field names are
1477   received is not significant. However, it is "good practice" to send
1478   general-header fields first, followed by request-header or response-header
1479   fields, and ending with the entity-header fields.
1480</t>
1481<t>
1482   Multiple message-header fields with the same field-name &MAY; be
1483   present in a message if and only if the entire field-value for that
1484   header field is defined as a comma-separated list [i.e., #(values)].
1485   It &MUST; be possible to combine the multiple header fields into one
1486   "field-name: field-value" pair, without changing the semantics of the
1487   message, by appending each subsequent field-value to the first, each
1488   separated by a comma. The order in which header fields with the same
1489   field-name are received is therefore significant to the
1490   interpretation of the combined field value, and thus a proxy &MUST-NOT;
1491   change the order of these field values when a message is forwarded.
1492</t>
1493</section>
1494
1495<section title="Message Body" anchor="message.body">
1496<t>
1497   The message-body (if any) of an HTTP message is used to carry the
1498   entity-body associated with the request or response. The message-body
1499   differs from the entity-body only when a transfer-coding has been
1500   applied, as indicated by the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1501</t>
1502<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="message-body"/>
1503  message-body = entity-body
1504               | &lt;entity-body encoded as per Transfer-Encoding&gt;
1505</artwork></figure>
1506<t>
1507   Transfer-Encoding &MUST; be used to indicate any transfer-codings
1508   applied by an application to ensure safe and proper transfer of the
1509   message. Transfer-Encoding is a property of the message, not of the
1510   entity, and thus &MAY; be added or removed by any application along the
1511   request/response chain. (However, <xref target="transfer.codings"/> places restrictions on
1512   when certain transfer-codings may be used.)
1513</t>
1514<t>
1515   The rules for when a message-body is allowed in a message differ for
1516   requests and responses.
1517</t>
1518<t>
1519   The presence of a message-body in a request is signaled by the
1520   inclusion of a Content-Length or Transfer-Encoding header field in
1521   the request's message-headers. A message-body &MUST-NOT; be included in
1522   a request if the specification of the request method (&method;)
1523   explicitly disallows an entity-body in requests.
1524   When a request message contains both a message-body of non-zero
1525   length and a method that does not define any semantics for that
1526   request message-body, then an origin server &SHOULD; either ignore
1527   the message-body or respond with an appropriate error message
1528   (e.g., 413).  A proxy or gateway, when presented the same request,
1529   &SHOULD; either forward the request inbound with the message-body or
1530   ignore the message-body when determining a response.
1531</t>
1532<t>
1533   For response messages, whether or not a message-body is included with
1534   a message is dependent on both the request method and the response
1535   status code (<xref target="status.code.and.reason.phrase"/>). All responses to the HEAD request method
1536   &MUST-NOT; include a message-body, even though the presence of entity-header
1537   fields might lead one to believe they do. All 1xx
1538   (informational), 204 (No Content), and 304 (Not Modified) responses
1539   &MUST-NOT; include a message-body. All other responses do include a
1540   message-body, although it &MAY; be of zero length.
1541</t>
1542</section>
1543
1544<section title="Message Length" anchor="message.length">
1545<t>
1546   The transfer-length of a message is the length of the message-body as
1547   it appears in the message; that is, after any transfer-codings have
1548   been applied. When a message-body is included with a message, the
1549   transfer-length of that body is determined by one of the following
1550   (in order of precedence):
1551</t>
1552<t>
1553  <list style="numbers">
1554    <x:lt><t>
1555     Any response message which "&MUST-NOT;" include a message-body (such
1556     as the 1xx, 204, and 304 responses and any response to a HEAD
1557     request) is always terminated by the first empty line after the
1558     header fields, regardless of the entity-header fields present in
1559     the message.
1560    </t></x:lt>
1561    <x:lt><t>
1562     If a Transfer-Encoding header field (<xref target="header.transfer-encoding"/>)
1563     is present, then the transfer-length is
1564     defined by use of the "chunked" transfer-coding (<xref target="transfer.codings"/>),
1565     unless the message is terminated by closing the connection.
1566    </t></x:lt>
1567    <x:lt><t>
1568     If a Content-Length header field (<xref target="header.content-length"/>) is present, its
1569     decimal value in OCTETs represents both the entity-length and the
1570     transfer-length. The Content-Length header field &MUST-NOT; be sent
1571     if these two lengths are different (i.e., if a Transfer-Encoding
1572     header field is present). If a message is received with both a
1573     Transfer-Encoding header field and a Content-Length header field,
1574     the latter &MUST; be ignored.
1575    </t></x:lt>
1576    <x:lt><t>
1577     If the message uses the media type "multipart/byteranges", and the
1578     transfer-length is not otherwise specified, then this self-delimiting
1579     media type defines the transfer-length. This media type
1580     &MUST-NOT; be used unless the sender knows that the recipient can parse
1581     it; the presence in a request of a Range header with multiple byte-range
1582     specifiers from a 1.1 client implies that the client can parse
1583     multipart/byteranges responses.
1584    <list style="empty"><t>
1585       A range header might be forwarded by a 1.0 proxy that does not
1586       understand multipart/byteranges; in this case the server &MUST;
1587       delimit the message using methods defined in items 1, 3 or 5 of
1588       this section.
1589    </t></list>
1590    </t></x:lt>
1591    <x:lt><t>
1592     By the server closing the connection. (Closing the connection
1593     cannot be used to indicate the end of a request body, since that
1594     would leave no possibility for the server to send back a response.)
1595    </t></x:lt>
1596  </list>
1597</t>
1598<t>
1599   For compatibility with HTTP/1.0 applications, HTTP/1.1 requests
1600   containing a message-body &MUST; include a valid Content-Length header
1601   field unless the server is known to be HTTP/1.1 compliant. If a
1602   request contains a message-body and a Content-Length is not given,
1603   the server &SHOULD; respond with 400 (Bad Request) if it cannot
1604   determine the length of the message, or with 411 (Length Required) if
1605   it wishes to insist on receiving a valid Content-Length.
1606</t>
1607<t>
1608   All HTTP/1.1 applications that receive entities &MUST; accept the
1609   "chunked" transfer-coding (<xref target="transfer.codings"/>), thus allowing this mechanism
1610   to be used for messages when the message length cannot be determined
1611   in advance.
1612</t>
1613<t>
1614   Messages &MUST-NOT; include both a Content-Length header field and a
1615   transfer-coding. If the message does include a
1616   transfer-coding, the Content-Length &MUST; be ignored.
1617</t>
1618<t>
1619   When a Content-Length is given in a message where a message-body is
1620   allowed, its field value &MUST; exactly match the number of OCTETs in
1621   the message-body. HTTP/1.1 user agents &MUST; notify the user when an
1622   invalid length is received and detected.
1623</t>
1624</section>
1625
1626<section title="General Header Fields" anchor="general.header.fields">
1627<t>
1628   There are a few header fields which have general applicability for
1629   both request and response messages, but which do not apply to the
1630   entity being transferred. These header fields apply only to the
1631   message being transmitted.
1632</t>
1633<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="general-header"/>
1634  general-header = Cache-Control            ; &header-cache-control;
1635                 | Connection               ; <xref target="header.connection"/>
1636                 | Date                     ; <xref target="header.date"/>
1637                 | Pragma                   ; &header-pragma;
1638                 | Trailer                  ; <xref target="header.trailer"/>
1639                 | Transfer-Encoding        ; <xref target="header.transfer-encoding"/>
1640                 | Upgrade                  ; <xref target="header.upgrade"/>
1641                 | Via                      ; <xref target="header.via"/>
1642                 | Warning                  ; &header-warning;
1643</artwork></figure>
1644<t>
1645   General-header field names can be extended reliably only in
1646   combination with a change in the protocol version. However, new or
1647   experimental header fields may be given the semantics of general
1648   header fields if all parties in the communication recognize them to
1649   be general-header fields. Unrecognized header fields are treated as
1650   entity-header fields.
1651</t>
1652</section>
1653</section>
1654
1655<section title="Request" anchor="request">
1656<t>
1657   A request message from a client to a server includes, within the
1658   first line of that message, the method to be applied to the resource,
1659   the identifier of the resource, and the protocol version in use.
1660</t>
1661<!--                 Host                      ; should be moved here eventually -->
1662<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request"/>
1663  Request       = Request-Line              ; <xref target="request-line"/>
1664                  *(( general-header        ; <xref target="general.header.fields"/>
1665                   | request-header         ; &request-header-fields;
1666                   | entity-header ) CRLF)  ; &entity-header-fields;
1667                  CRLF
1668                  [ message-body ]          ; <xref target="message.body"/>
1669</artwork></figure>
1670
1671<section title="Request-Line" anchor="request-line">
1672<t>
1673   The Request-Line begins with a method token, followed by the
1674   Request-URI and the protocol version, and ending with CRLF. The
1675   elements are separated by SP characters. No CR or LF is allowed
1676   except in the final CRLF sequence.
1677</t>
1678<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request-Line"/>
1679  Request-Line   = Method SP Request-URI SP HTTP-Version CRLF
1680</artwork></figure>
1681
1682<section title="Method" anchor="method">
1683<t>
1684   The Method  token indicates the method to be performed on the
1685   resource identified by the Request-URI. The method is case-sensitive.
1686</t>
1687<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Method"/><iref primary="true" item="Grammar" subitem="extension-method"/>
1688  Method         = token
1689</artwork></figure>
1690</section>
1691
1692<section title="Request-URI" anchor="request-uri">
1693<t>
1694   The Request-URI is a Uniform Resource Identifier (<xref target="uri"/>) and
1695   identifies the resource upon which to apply the request.
1696</t>
1697<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request-URI"/>
1698  Request-URI    = "*"
1699                 | absoluteURI
1700                 | ( path-absolute [ "?" query ] )
1701                 | authority
1702</artwork></figure>
1703<t>
1704   The four options for Request-URI are dependent on the nature of the
1705   request. The asterisk "*" means that the request does not apply to a
1706   particular resource, but to the server itself, and is only allowed
1707   when the method used does not necessarily apply to a resource. One
1708   example would be
1709</t>
1710<figure><artwork type="example">
1711    OPTIONS * HTTP/1.1
1712</artwork></figure>
1713<t>
1714   The absoluteURI form is &REQUIRED; when the request is being made to a
1715   proxy. The proxy is requested to forward the request or service it
1716   from a valid cache, and return the response. Note that the proxy &MAY;
1717   forward the request on to another proxy or directly to the server
1718   specified by the absoluteURI. In order to avoid request loops, a
1719   proxy &MUST; be able to recognize all of its server names, including
1720   any aliases, local variations, and the numeric IP address. An example
1721   Request-Line would be:
1722</t>
1723<figure><artwork type="example">
1724    GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1
1725</artwork></figure>
1726<t>
1727   To allow for transition to absoluteURIs in all requests in future
1728   versions of HTTP, all HTTP/1.1 servers &MUST; accept the absoluteURI
1729   form in requests, even though HTTP/1.1 clients will only generate
1730   them in requests to proxies.
1731</t>
1732<t>
1733   The authority form is only used by the CONNECT method (&CONNECT;).
1734</t>
1735<t>
1736   The most common form of Request-URI is that used to identify a
1737   resource on an origin server or gateway. In this case the absolute
1738   path of the URI &MUST; be transmitted (see <xref target="general.syntax"/>, path-absolute) as
1739   the Request-URI, and the network location of the URI (authority) &MUST;
1740   be transmitted in a Host header field. For example, a client wishing
1741   to retrieve the resource above directly from the origin server would
1742   create a TCP connection to port 80 of the host "www.example.org" and send
1743   the lines:
1744</t>
1745<figure><artwork type="example">
1746    GET /pub/WWW/TheProject.html HTTP/1.1
1747    Host: www.example.org
1748</artwork></figure>
1749<t>
1750   followed by the remainder of the Request. Note that the absolute path
1751   cannot be empty; if none is present in the original URI, it &MUST; be
1752   given as "/" (the server root).
1753</t>
1754<t>
1755   The Request-URI is transmitted in the format specified in
1756   <xref target="general.syntax"/>. If the Request-URI is encoded using the "% HEX HEX" encoding
1757   <xref target="RFC2396"/>, the origin server &MUST; decode the Request-URI in order to
1758   properly interpret the request. Servers &SHOULD; respond to invalid
1759   Request-URIs with an appropriate status code.
1760</t>
1761<t>
1762   A transparent proxy &MUST-NOT; rewrite the "path-absolute" part of the
1763   received Request-URI when forwarding it to the next inbound server,
1764   except as noted above to replace a null path-absolute with "/".
1765</t>
1766<t>
1767  <list><t>
1768      <x:h>Note:</x:h> The "no rewrite" rule prevents the proxy from changing the
1769      meaning of the request when the origin server is improperly using
1770      a non-reserved URI character for a reserved purpose.  Implementors
1771      should be aware that some pre-HTTP/1.1 proxies have been known to
1772      rewrite the Request-URI.
1773  </t></list>
1774</t>
1775</section>
1776</section>
1777
1778<section title="The Resource Identified by a Request" anchor="the.resource.identified.by.a.request">
1779<t>
1780   The exact resource identified by an Internet request is determined by
1781   examining both the Request-URI and the Host header field.
1782</t>
1783<t>
1784   An origin server that does not allow resources to differ by the
1785   requested host &MAY; ignore the Host header field value when
1786   determining the resource identified by an HTTP/1.1 request. (But see
1787   <xref target="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses"/>
1788   for other requirements on Host support in HTTP/1.1.)
1789</t>
1790<t>
1791   An origin server that does differentiate resources based on the host
1792   requested (sometimes referred to as virtual hosts or vanity host
1793   names) &MUST; use the following rules for determining the requested
1794   resource on an HTTP/1.1 request:
1795  <list style="numbers">
1796    <t>If Request-URI is an absoluteURI, the host is part of the
1797     Request-URI. Any Host header field value in the request &MUST; be
1798     ignored.</t>
1799    <t>If the Request-URI is not an absoluteURI, and the request includes
1800     a Host header field, the host is determined by the Host header
1801     field value.</t>
1802    <t>If the host as determined by rule 1 or 2 is not a valid host on
1803     the server, the response &MUST; be a 400 (Bad Request) error message.</t>
1804  </list>
1805</t>
1806<t>
1807   Recipients of an HTTP/1.0 request that lacks a Host header field &MAY;
1808   attempt to use heuristics (e.g., examination of the URI path for
1809   something unique to a particular host) in order to determine what
1810   exact resource is being requested.
1811</t>
1812</section>
1813
1814</section>
1815
1816
1817<section title="Response" anchor="response">
1818<t>
1819   After receiving and interpreting a request message, a server responds
1820   with an HTTP response message.
1821</t>
1822<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Response"/>
1823  Response      = Status-Line               ; <xref target="status-line"/>
1824                  *(( general-header        ; <xref target="general.header.fields"/>
1825                   | response-header        ; &response-header-fields;
1826                   | entity-header ) CRLF)  ; &entity-header-fields;
1827                  CRLF
1828                  [ message-body ]          ; <xref target="message.body"/>
1829</artwork></figure>
1830
1831<section title="Status-Line" anchor="status-line">
1832<t>
1833   The first line of a Response message is the Status-Line, consisting
1834   of the protocol version followed by a numeric status code and its
1835   associated textual phrase, with each element separated by SP
1836   characters. No CR or LF is allowed except in the final CRLF sequence.
1837</t>
1838<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Status-Line"/>
1839  Status-Line = HTTP-Version SP Status-Code SP Reason-Phrase CRLF
1840</artwork></figure>
1841
1842<section title="Status Code and Reason Phrase" anchor="status.code.and.reason.phrase">
1843<t>
1844   The Status-Code element is a 3-digit integer result code of the
1845   attempt to understand and satisfy the request. These codes are fully
1846   defined in &status-codes;.  The Reason Phrase exists for the sole
1847   purpose of providing a textual description associated with the numeric
1848   status code, out of deference to earlier Internet application protocols
1849   that were more frequently used with interactive text clients.
1850   A client &SHOULD; ignore the content of the Reason Phrase.
1851</t>
1852<t>
1853   The first digit of the Status-Code defines the class of response. The
1854   last two digits do not have any categorization role. There are 5
1855   values for the first digit:
1856  <list style="symbols">
1857    <t>
1858      1xx: Informational - Request received, continuing process
1859    </t>
1860    <t>
1861      2xx: Success - The action was successfully received,
1862        understood, and accepted
1863    </t>
1864    <t>
1865      3xx: Redirection - Further action must be taken in order to
1866        complete the request
1867    </t>
1868    <t>
1869      4xx: Client Error - The request contains bad syntax or cannot
1870        be fulfilled
1871    </t>
1872    <t>
1873      5xx: Server Error - The server failed to fulfill an apparently
1874        valid request
1875    </t>
1876  </list>
1877</t>
1878<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"/>
1879  Status-Code    = 3DIGIT
1880  Reason-Phrase  = *&lt;TEXT, excluding CR, LF&gt;
1881</artwork></figure>
1882</section>
1883</section>
1884
1885</section>
1886
1887
1888<section title="Connections" anchor="connections">
1889
1890<section title="Persistent Connections" anchor="persistent.connections">
1891
1892<section title="Purpose" anchor="persistent.purpose">
1893<t>
1894   Prior to persistent connections, a separate TCP connection was
1895   established to fetch each URL, increasing the load on HTTP servers
1896   and causing congestion on the Internet. The use of inline images and
1897   other associated data often require a client to make multiple
1898   requests of the same server in a short amount of time. Analysis of
1899   these performance problems and results from a prototype
1900   implementation are available <xref target="Pad1995"/> <xref target="Spe"/>. Implementation experience and
1901   measurements of actual HTTP/1.1 (<xref target="RFC2068" x:fmt="none">RFC 2068</xref>) implementations show good
1902   results <xref target="Nie1997"/>. Alternatives have also been explored, for example,
1903   T/TCP <xref target="Tou1998"/>.
1904</t>
1905<t>
1906   Persistent HTTP connections have a number of advantages:
1907  <list style="symbols">
1908      <t>
1909        By opening and closing fewer TCP connections, CPU time is saved
1910        in routers and hosts (clients, servers, proxies, gateways,
1911        tunnels, or caches), and memory used for TCP protocol control
1912        blocks can be saved in hosts.
1913      </t>
1914      <t>
1915        HTTP requests and responses can be pipelined on a connection.
1916        Pipelining allows a client to make multiple requests without
1917        waiting for each response, allowing a single TCP connection to
1918        be used much more efficiently, with much lower elapsed time.
1919      </t>
1920      <t>
1921        Network congestion is reduced by reducing the number of packets
1922        caused by TCP opens, and by allowing TCP sufficient time to
1923        determine the congestion state of the network.
1924      </t>
1925      <t>
1926        Latency on subsequent requests is reduced since there is no time
1927        spent in TCP's connection opening handshake.
1928      </t>
1929      <t>
1930        HTTP can evolve more gracefully, since errors can be reported
1931        without the penalty of closing the TCP connection. Clients using
1932        future versions of HTTP might optimistically try a new feature,
1933        but if communicating with an older server, retry with old
1934        semantics after an error is reported.
1935      </t>
1936    </list>
1937</t>
1938<t>
1939   HTTP implementations &SHOULD; implement persistent connections.
1940</t>
1941</section>
1942
1943<section title="Overall Operation" anchor="persistent.overall">
1944<t>
1945   A significant difference between HTTP/1.1 and earlier versions of
1946   HTTP is that persistent connections are the default behavior of any
1947   HTTP connection. That is, unless otherwise indicated, the client
1948   &SHOULD; assume that the server will maintain a persistent connection,
1949   even after error responses from the server.
1950</t>
1951<t>
1952   Persistent connections provide a mechanism by which a client and a
1953   server can signal the close of a TCP connection. This signaling takes
1954   place using the Connection header field (<xref target="header.connection"/>). Once a close
1955   has been signaled, the client &MUST-NOT; send any more requests on that
1956   connection.
1957</t>
1958
1959<section title="Negotiation" anchor="persistent.negotiation">
1960<t>
1961   An HTTP/1.1 server &MAY; assume that a HTTP/1.1 client intends to
1962   maintain a persistent connection unless a Connection header including
1963   the connection-token "close" was sent in the request. If the server
1964   chooses to close the connection immediately after sending the
1965   response, it &SHOULD; send a Connection header including the
1966   connection-token close.
1967</t>
1968<t>
1969   An HTTP/1.1 client &MAY; expect a connection to remain open, but would
1970   decide to keep it open based on whether the response from a server
1971   contains a Connection header with the connection-token close. In case
1972   the client does not want to maintain a connection for more than that
1973   request, it &SHOULD; send a Connection header including the
1974   connection-token close.
1975</t>
1976<t>
1977   If either the client or the server sends the close token in the
1978   Connection header, that request becomes the last one for the
1979   connection.
1980</t>
1981<t>
1982   Clients and servers &SHOULD-NOT;  assume that a persistent connection is
1983   maintained for HTTP versions less than 1.1 unless it is explicitly
1984   signaled. See <xref target="compatibility.with.http.1.0.persistent.connections"/> for more information on backward
1985   compatibility with HTTP/1.0 clients.
1986</t>
1987<t>
1988   In order to remain persistent, all messages on the connection &MUST;
1989   have a self-defined message length (i.e., one not defined by closure
1990   of the connection), as described in <xref target="message.length"/>.
1991</t>
1992</section>
1993
1994<section title="Pipelining" anchor="pipelining">
1995<t>
1996   A client that supports persistent connections &MAY; "pipeline" its
1997   requests (i.e., send multiple requests without waiting for each
1998   response). A server &MUST; send its responses to those requests in the
1999   same order that the requests were received.
2000</t>
2001<t>
2002   Clients which assume persistent connections and pipeline immediately
2003   after connection establishment &SHOULD; be prepared to retry their
2004   connection if the first pipelined attempt fails. If a client does
2005   such a retry, it &MUST-NOT; pipeline before it knows the connection is
2006   persistent. Clients &MUST; also be prepared to resend their requests if
2007   the server closes the connection before sending all of the
2008   corresponding responses.
2009</t>
2010<t>
2011   Clients &SHOULD-NOT;  pipeline requests using non-idempotent methods or
2012   non-idempotent sequences of methods (see &idempotent-methods;). Otherwise, a
2013   premature termination of the transport connection could lead to
2014   indeterminate results. A client wishing to send a non-idempotent
2015   request &SHOULD; wait to send that request until it has received the
2016   response status for the previous request.
2017</t>
2018</section>
2019</section>
2020
2021<section title="Proxy Servers" anchor="persistent.proxy">
2022<t>
2023   It is especially important that proxies correctly implement the
2024   properties of the Connection header field as specified in <xref target="header.connection"/>.
2025</t>
2026<t>
2027   The proxy server &MUST; signal persistent connections separately with
2028   its clients and the origin servers (or other proxy servers) that it
2029   connects to. Each persistent connection applies to only one transport
2030   link.
2031</t>
2032<t>
2033   A proxy server &MUST-NOT; establish a HTTP/1.1 persistent connection
2034   with an HTTP/1.0 client (but see <xref target="RFC2068"/> for information and
2035   discussion of the problems with the Keep-Alive header implemented by
2036   many HTTP/1.0 clients).
2037</t>
2038</section>
2039
2040<section title="Practical Considerations" anchor="persistent.practical">
2041<t>
2042   Servers will usually have some time-out value beyond which they will
2043   no longer maintain an inactive connection. Proxy servers might make
2044   this a higher value since it is likely that the client will be making
2045   more connections through the same server. The use of persistent
2046   connections places no requirements on the length (or existence) of
2047   this time-out for either the client or the server.
2048</t>
2049<t>
2050   When a client or server wishes to time-out it &SHOULD; issue a graceful
2051   close on the transport connection. Clients and servers &SHOULD; both
2052   constantly watch for the other side of the transport close, and
2053   respond to it as appropriate. If a client or server does not detect
2054   the other side's close promptly it could cause unnecessary resource
2055   drain on the network.
2056</t>
2057<t>
2058   A client, server, or proxy &MAY; close the transport connection at any
2059   time. For example, a client might have started to send a new request
2060   at the same time that the server has decided to close the "idle"
2061   connection. From the server's point of view, the connection is being
2062   closed while it was idle, but from the client's point of view, a
2063   request is in progress.
2064</t>
2065<t>
2066   This means that clients, servers, and proxies &MUST; be able to recover
2067   from asynchronous close events. Client software &SHOULD; reopen the
2068   transport connection and retransmit the aborted sequence of requests
2069   without user interaction so long as the request sequence is
2070   idempotent (see &idempotent-methods;). Non-idempotent methods or sequences
2071   &MUST-NOT; be automatically retried, although user agents &MAY; offer a
2072   human operator the choice of retrying the request(s). Confirmation by
2073   user-agent software with semantic understanding of the application
2074   &MAY; substitute for user confirmation. The automatic retry &SHOULD-NOT;
2075   be repeated if the second sequence of requests fails.
2076</t>
2077<t>
2078   Servers &SHOULD; always respond to at least one request per connection,
2079   if at all possible. Servers &SHOULD-NOT;  close a connection in the
2080   middle of transmitting a response, unless a network or client failure
2081   is suspected.
2082</t>
2083<t>
2084   Clients that use persistent connections &SHOULD; limit the number of
2085   simultaneous connections that they maintain to a given server. A
2086   single-user client &SHOULD-NOT; maintain more than 2 connections with
2087   any server or proxy. A proxy &SHOULD; use up to 2*N connections to
2088   another server or proxy, where N is the number of simultaneously
2089   active users. These guidelines are intended to improve HTTP response
2090   times and avoid congestion.
2091</t>
2092</section>
2093</section>
2094
2095<section title="Message Transmission Requirements" anchor="message.transmission.requirements">
2096
2097<section title="Persistent Connections and Flow Control" anchor="persistent.flow">
2098<t>
2099   HTTP/1.1 servers &SHOULD; maintain persistent connections and use TCP's
2100   flow control mechanisms to resolve temporary overloads, rather than
2101   terminating connections with the expectation that clients will retry.
2102   The latter technique can exacerbate network congestion.
2103</t>
2104</section>
2105
2106<section title="Monitoring Connections for Error Status Messages" anchor="persistent.monitor">
2107<t>
2108   An HTTP/1.1 (or later) client sending a message-body &SHOULD; monitor
2109   the network connection for an error status while it is transmitting
2110   the request. If the client sees an error status, it &SHOULD;
2111   immediately cease transmitting the body. If the body is being sent
2112   using a "chunked" encoding (<xref target="transfer.codings"/>), a zero length chunk and
2113   empty trailer &MAY; be used to prematurely mark the end of the message.
2114   If the body was preceded by a Content-Length header, the client &MUST;
2115   close the connection.
2116</t>
2117</section>
2118
2119<section title="Use of the 100 (Continue) Status" anchor="use.of.the.100.status">
2120<t>
2121   The purpose of the 100 (Continue) status (see &status-100;) is to
2122   allow a client that is sending a request message with a request body
2123   to determine if the origin server is willing to accept the request
2124   (based on the request headers) before the client sends the request
2125   body. In some cases, it might either be inappropriate or highly
2126   inefficient for the client to send the body if the server will reject
2127   the message without looking at the body.
2128</t>
2129<t>
2130   Requirements for HTTP/1.1 clients:
2131  <list style="symbols">
2132    <t>
2133        If a client will wait for a 100 (Continue) response before
2134        sending the request body, it &MUST; send an Expect request-header
2135        field (&header-expect;) with the "100-continue" expectation.
2136    </t>
2137    <t>
2138        A client &MUST-NOT; send an Expect request-header field (&header-expect;)
2139        with the "100-continue" expectation if it does not intend
2140        to send a request body.
2141    </t>
2142  </list>
2143</t>
2144<t>
2145   Because of the presence of older implementations, the protocol allows
2146   ambiguous situations in which a client may send "Expect: 100-continue"
2147   without receiving either a 417 (Expectation Failed) status
2148   or a 100 (Continue) status. Therefore, when a client sends this
2149   header field to an origin server (possibly via a proxy) from which it
2150   has never seen a 100 (Continue) status, the client &SHOULD-NOT;  wait
2151   for an indefinite period before sending the request body.
2152</t>
2153<t>
2154   Requirements for HTTP/1.1 origin servers:
2155  <list style="symbols">
2156    <t> Upon receiving a request which includes an Expect request-header
2157        field with the "100-continue" expectation, an origin server &MUST;
2158        either respond with 100 (Continue) status and continue to read
2159        from the input stream, or respond with a final status code. The
2160        origin server &MUST-NOT; wait for the request body before sending
2161        the 100 (Continue) response. If it responds with a final status
2162        code, it &MAY; close the transport connection or it &MAY; continue
2163        to read and discard the rest of the request.  It &MUST-NOT;
2164        perform the requested method if it returns a final status code.
2165    </t>
2166    <t> An origin server &SHOULD-NOT;  send a 100 (Continue) response if
2167        the request message does not include an Expect request-header
2168        field with the "100-continue" expectation, and &MUST-NOT; send a
2169        100 (Continue) response if such a request comes from an HTTP/1.0
2170        (or earlier) client. There is an exception to this rule: for
2171        compatibility with <xref target="RFC2068"/>, a server &MAY; send a 100 (Continue)
2172        status in response to an HTTP/1.1 PUT or POST request that does
2173        not include an Expect request-header field with the "100-continue"
2174        expectation. This exception, the purpose of which is
2175        to minimize any client processing delays associated with an
2176        undeclared wait for 100 (Continue) status, applies only to
2177        HTTP/1.1 requests, and not to requests with any other HTTP-version
2178        value.
2179    </t>
2180    <t> An origin server &MAY; omit a 100 (Continue) response if it has
2181        already received some or all of the request body for the
2182        corresponding request.
2183    </t>
2184    <t> An origin server that sends a 100 (Continue) response &MUST;
2185    ultimately send a final status code, once the request body is
2186        received and processed, unless it terminates the transport
2187        connection prematurely.
2188    </t>
2189    <t> If an origin server receives a request that does not include an
2190        Expect request-header field with the "100-continue" expectation,
2191        the request includes a request body, and the server responds
2192        with a final status code before reading the entire request body
2193        from the transport connection, then the server &SHOULD-NOT;  close
2194        the transport connection until it has read the entire request,
2195        or until the client closes the connection. Otherwise, the client
2196        might not reliably receive the response message. However, this
2197        requirement is not be construed as preventing a server from
2198        defending itself against denial-of-service attacks, or from
2199        badly broken client implementations.
2200      </t>
2201    </list>
2202</t>
2203<t>
2204   Requirements for HTTP/1.1 proxies:
2205  <list style="symbols">
2206    <t> If a proxy receives a request that includes an Expect request-header
2207        field with the "100-continue" expectation, and the proxy
2208        either knows that the next-hop server complies with HTTP/1.1 or
2209        higher, or does not know the HTTP version of the next-hop
2210        server, it &MUST; forward the request, including the Expect header
2211        field.
2212    </t>
2213    <t> If the proxy knows that the version of the next-hop server is
2214        HTTP/1.0 or lower, it &MUST-NOT; forward the request, and it &MUST;
2215        respond with a 417 (Expectation Failed) status.
2216    </t>
2217    <t> Proxies &SHOULD; maintain a cache recording the HTTP version
2218        numbers received from recently-referenced next-hop servers.
2219    </t>
2220    <t> A proxy &MUST-NOT; forward a 100 (Continue) response if the
2221        request message was received from an HTTP/1.0 (or earlier)
2222        client and did not include an Expect request-header field with
2223        the "100-continue" expectation. This requirement overrides the
2224        general rule for forwarding of 1xx responses (see &status-1xx;).
2225    </t>
2226  </list>
2227</t>
2228</section>
2229
2230<section title="Client Behavior if Server Prematurely Closes Connection" anchor="connection.premature">
2231<t>
2232   If an HTTP/1.1 client sends a request which includes a request body,
2233   but which does not include an Expect request-header field with the
2234   "100-continue" expectation, and if the client is not directly
2235   connected to an HTTP/1.1 origin server, and if the client sees the
2236   connection close before receiving any status from the server, the
2237   client &SHOULD; retry the request.  If the client does retry this
2238   request, it &MAY; use the following "binary exponential backoff"
2239   algorithm to be assured of obtaining a reliable response:
2240  <list style="numbers">
2241    <t>
2242      Initiate a new connection to the server
2243    </t>
2244    <t>
2245      Transmit the request-headers
2246    </t>
2247    <t>
2248      Initialize a variable R to the estimated round-trip time to the
2249         server (e.g., based on the time it took to establish the
2250         connection), or to a constant value of 5 seconds if the round-trip
2251         time is not available.
2252    </t>
2253    <t>
2254       Compute T = R * (2**N), where N is the number of previous
2255         retries of this request.
2256    </t>
2257    <t>
2258       Wait either for an error response from the server, or for T
2259         seconds (whichever comes first)
2260    </t>
2261    <t>
2262       If no error response is received, after T seconds transmit the
2263         body of the request.
2264    </t>
2265    <t>
2266       If client sees that the connection is closed prematurely,
2267         repeat from step 1 until the request is accepted, an error
2268         response is received, or the user becomes impatient and
2269         terminates the retry process.
2270    </t>
2271  </list>
2272</t>
2273<t>
2274   If at any point an error status is received, the client
2275  <list style="symbols">
2276      <t>&SHOULD-NOT;  continue and</t>
2277
2278      <t>&SHOULD; close the connection if it has not completed sending the
2279        request message.</t>
2280    </list>
2281</t>
2282</section>
2283</section>
2284</section>
2285
2286
2287<section title="Header Field Definitions" anchor="header.fields">
2288<t>
2289   This section defines the syntax and semantics of HTTP/1.1 header fields
2290   related to message framing and transport protocols.
2291</t>
2292<t>
2293   For entity-header fields, both sender and recipient refer to either the
2294   client or the server, depending on who sends and who receives the entity.
2295</t>
2296
2297<section title="Connection" anchor="header.connection">
2298  <iref primary="true" item="Connection header" x:for-anchor=""/>
2299  <iref primary="true" item="Headers" subitem="Connection" x:for-anchor=""/>
2300<t>
2301   The Connection general-header field allows the sender to specify
2302   options that are desired for that particular connection and &MUST-NOT;
2303   be communicated by proxies over further connections.
2304</t>
2305<t>
2306   The Connection header has the following grammar:
2307</t>
2308<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="connection-token"/>
2309  Connection = "Connection" ":" 1#(connection-token)
2310  connection-token  = token
2311</artwork></figure>
2312<t>
2313   HTTP/1.1 proxies &MUST; parse the Connection header field before a
2314   message is forwarded and, for each connection-token in this field,
2315   remove any header field(s) from the message with the same name as the
2316   connection-token. Connection options are signaled by the presence of
2317   a connection-token in the Connection header field, not by any
2318   corresponding additional header field(s), since the additional header
2319   field may not be sent if there are no parameters associated with that
2320   connection option.
2321</t>
2322<t>
2323   Message headers listed in the Connection header &MUST-NOT; include
2324   end-to-end headers, such as Cache-Control.
2325</t>
2326<t>
2327   HTTP/1.1 defines the "close" connection option for the sender to
2328   signal that the connection will be closed after completion of the
2329   response. For example,
2330</t>
2331<figure><artwork type="example">
2332    Connection: close
2333</artwork></figure>
2334<t>
2335   in either the request or the response header fields indicates that
2336   the connection &SHOULD-NOT;  be considered `persistent' (<xref target="persistent.connections"/>)
2337   after the current request/response is complete.
2338</t>
2339<t>
2340   An HTTP/1.1 client that does not support persistent connections &MUST;
2341   include the "close" connection option in every request message.
2342</t>
2343<t>
2344   An HTTP/1.1 server that does not support persistent connections &MUST;
2345   include the "close" connection option in every response message that
2346   does not have a 1xx (informational) status code.
2347</t>
2348<t>
2349   A system receiving an HTTP/1.0 (or lower-version) message that
2350   includes a Connection header &MUST;, for each connection-token in this
2351   field, remove and ignore any header field(s) from the message with
2352   the same name as the connection-token. This protects against mistaken
2353   forwarding of such header fields by pre-HTTP/1.1 proxies. See <xref target="compatibility.with.http.1.0.persistent.connections"/>.
2354</t>
2355</section>
2356
2357<section title="Content-Length" anchor="header.content-length">
2358  <iref primary="true" item="Content-Length header" x:for-anchor=""/>
2359  <iref primary="true" item="Headers" subitem="Content-Length" x:for-anchor=""/>
2360<t>
2361   The Content-Length entity-header field indicates the size of the
2362   entity-body, in decimal number of OCTETs, sent to the recipient or,
2363   in the case of the HEAD method, the size of the entity-body that
2364   would have been sent had the request been a GET.
2365</t>
2366<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Length"/>
2367  Content-Length    = "Content-Length" ":" 1*DIGIT
2368</artwork></figure>
2369<t>
2370   An example is
2371</t>
2372<figure><artwork type="example">
2373    Content-Length: 3495
2374</artwork></figure>
2375<t>
2376   Applications &SHOULD; use this field to indicate the transfer-length of
2377   the message-body, unless this is prohibited by the rules in <xref target="message.length"/>.
2378</t>
2379<t>
2380   Any Content-Length greater than or equal to zero is a valid value.
2381   <xref target="message.length"/> describes how to determine the length of a message-body
2382   if a Content-Length is not given.
2383</t>
2384<t>
2385   Note that the meaning of this field is significantly different from
2386   the corresponding definition in MIME, where it is an optional field
2387   used within the "message/external-body" content-type. In HTTP, it
2388   &SHOULD; be sent whenever the message's length can be determined prior
2389   to being transferred, unless this is prohibited by the rules in
2390   <xref target="message.length"/>.
2391</t>
2392</section>
2393
2394<section title="Date" anchor="header.date">
2395  <iref primary="true" item="Date header" x:for-anchor=""/>
2396  <iref primary="true" item="Headers" subitem="Date" x:for-anchor=""/>
2397<t>
2398   The Date general-header field represents the date and time at which
2399   the message was originated, having the same semantics as orig-date in
2400   <xref target="RFC2822" x:fmt="of" x:sec="3.6.1"/>. The field value is an HTTP-date, as described in <xref target="full.date"/>;
2401   it &MUST; be sent in rfc1123-date format.
2402</t>
2403<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Date"/>
2404  Date  = "Date" ":" HTTP-date
2405</artwork></figure>
2406<t>
2407   An example is
2408</t>
2409<figure><artwork type="example">
2410    Date: Tue, 15 Nov 1994 08:12:31 GMT
2411</artwork></figure>
2412<t>
2413   Origin servers &MUST; include a Date header field in all responses,
2414   except in these cases:
2415  <list style="numbers">
2416      <t>If the response status code is 100 (Continue) or 101 (Switching
2417         Protocols), the response &MAY; include a Date header field, at
2418         the server's option.</t>
2419
2420      <t>If the response status code conveys a server error, e.g. 500
2421         (Internal Server Error) or 503 (Service Unavailable), and it is
2422         inconvenient or impossible to generate a valid Date.</t>
2423
2424      <t>If the server does not have a clock that can provide a
2425         reasonable approximation of the current time, its responses
2426         &MUST-NOT; include a Date header field. In this case, the rules
2427         in <xref target="clockless.origin.server.operation"/> &MUST; be followed.</t>
2428  </list>
2429</t>
2430<t>
2431   A received message that does not have a Date header field &MUST; be
2432   assigned one by the recipient if the message will be cached by that
2433   recipient or gatewayed via a protocol which requires a Date. An HTTP
2434   implementation without a clock &MUST-NOT; cache responses without
2435   revalidating them on every use. An HTTP cache, especially a shared
2436   cache, &SHOULD; use a mechanism, such as NTP <xref target="RFC1305"/>, to synchronize its
2437   clock with a reliable external standard.
2438</t>
2439<t>
2440   Clients &SHOULD; only send a Date header field in messages that include
2441   an entity-body, as in the case of the PUT and POST requests, and even
2442   then it is optional. A client without a clock &MUST-NOT; send a Date
2443   header field in a request.
2444</t>
2445<t>
2446   The HTTP-date sent in a Date header &SHOULD-NOT;  represent a date and
2447   time subsequent to the generation of the message. It &SHOULD; represent
2448   the best available approximation of the date and time of message
2449   generation, unless the implementation has no means of generating a
2450   reasonably accurate date and time. In theory, the date ought to
2451   represent the moment just before the entity is generated. In
2452   practice, the date can be generated at any time during the message
2453   origination without affecting its semantic value.
2454</t>
2455
2456<section title="Clockless Origin Server Operation" anchor="clockless.origin.server.operation">
2457<t>
2458   Some origin server implementations might not have a clock available.
2459   An origin server without a clock &MUST-NOT; assign Expires or Last-Modified
2460   values to a response, unless these values were associated
2461   with the resource by a system or user with a reliable clock. It &MAY;
2462   assign an Expires value that is known, at or before server
2463   configuration time, to be in the past (this allows "pre-expiration"
2464   of responses without storing separate Expires values for each
2465   resource).
2466</t>
2467</section>
2468</section>
2469
2470<section title="Host" anchor="header.host">
2471  <iref primary="true" item="Host header" x:for-anchor=""/>
2472  <iref primary="true" item="Headers" subitem="Host" x:for-anchor=""/>
2473<t>
2474   The Host request-header field specifies the Internet host and port
2475   number of the resource being requested, as obtained from the original
2476   URI given by the user or referring resource (generally an HTTP URL,
2477   as described in <xref target="http.url"/>). The Host field value &MUST; represent
2478   the naming authority of the origin server or gateway given by the
2479   original URL. This allows the origin server or gateway to
2480   differentiate between internally-ambiguous URLs, such as the root "/"
2481   URL of a server for multiple host names on a single IP address.
2482</t>
2483<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Host"/>
2484  Host = "Host" ":" uri-host [ ":" port ] ; <xref target="http.url"/>
2485</artwork></figure>
2486<t>
2487   A "host" without any trailing port information implies the default
2488   port for the service requested (e.g., "80" for an HTTP URL). For
2489   example, a request on the origin server for
2490   &lt;http://www.example.org/pub/WWW/&gt; would properly include:
2491</t>
2492<figure><artwork type="example">
2493    GET /pub/WWW/ HTTP/1.1
2494    Host: www.example.org
2495</artwork></figure>
2496<t>
2497   A client &MUST; include a Host header field in all HTTP/1.1 request
2498   messages. If the requested URI does not include an Internet host
2499   name for the service being requested, then the Host header field &MUST;
2500   be given with an empty value. An HTTP/1.1 proxy &MUST; ensure that any
2501   request message it forwards does contain an appropriate Host header
2502   field that identifies the service being requested by the proxy. All
2503   Internet-based HTTP/1.1 servers &MUST; respond with a 400 (Bad Request)
2504   status code to any HTTP/1.1 request message which lacks a Host header
2505   field.
2506</t>
2507<t>
2508   See Sections <xref target="the.resource.identified.by.a.request" format="counter"/>
2509   and <xref target="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses" format="counter"/>
2510   for other requirements relating to Host.
2511</t>
2512</section>
2513
2514<section title="TE" anchor="header.te">
2515  <iref primary="true" item="TE header" x:for-anchor=""/>
2516  <iref primary="true" item="Headers" subitem="TE" x:for-anchor=""/>
2517<t>
2518   The TE request-header field indicates what extension transfer-codings
2519   it is willing to accept in the response and whether or not it is
2520   willing to accept trailer fields in a chunked transfer-coding. Its
2521   value may consist of the keyword "trailers" and/or a comma-separated
2522   list of extension transfer-coding names with optional accept
2523   parameters (as described in <xref target="transfer.codings"/>).
2524</t>
2525<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="t-codings"/>
2526  TE        = "TE" ":" #( t-codings )
2527  t-codings = "trailers" | ( transfer-extension [ accept-params ] )
2528</artwork></figure>
2529<t>
2530   The presence of the keyword "trailers" indicates that the client is
2531   willing to accept trailer fields in a chunked transfer-coding, as
2532   defined in <xref target="chunked.transfer.encoding"/>. This keyword is reserved for use with
2533   transfer-coding values even though it does not itself represent a
2534   transfer-coding.
2535</t>
2536<t>
2537   Examples of its use are:
2538</t>
2539<figure><artwork type="example">
2540    TE: deflate
2541    TE:
2542    TE: trailers, deflate;q=0.5
2543</artwork></figure>
2544<t>
2545   The TE header field only applies to the immediate connection.
2546   Therefore, the keyword &MUST; be supplied within a Connection header
2547   field (<xref target="header.connection"/>) whenever TE is present in an HTTP/1.1 message.
2548</t>
2549<t>
2550   A server tests whether a transfer-coding is acceptable, according to
2551   a TE field, using these rules:
2552  <list style="numbers">
2553    <x:lt>
2554      <t>The "chunked" transfer-coding is always acceptable. If the
2555         keyword "trailers" is listed, the client indicates that it is
2556         willing to accept trailer fields in the chunked response on
2557         behalf of itself and any downstream clients. The implication is
2558         that, if given, the client is stating that either all
2559         downstream clients are willing to accept trailer fields in the
2560         forwarded response, or that it will attempt to buffer the
2561         response on behalf of downstream recipients.
2562      </t><t>
2563         <x:h>Note:</x:h> HTTP/1.1 does not define any means to limit the size of a
2564         chunked response such that a client can be assured of buffering
2565         the entire response.</t>
2566    </x:lt>
2567    <x:lt>
2568      <t>If the transfer-coding being tested is one of the transfer-codings
2569         listed in the TE field, then it is acceptable unless it
2570         is accompanied by a qvalue of 0. (As defined in &qvalue;, a
2571         qvalue of 0 means "not acceptable.")</t>
2572    </x:lt>
2573    <x:lt>
2574      <t>If multiple transfer-codings are acceptable, then the
2575         acceptable transfer-coding with the highest non-zero qvalue is
2576         preferred.  The "chunked" transfer-coding always has a qvalue
2577         of 1.</t>
2578    </x:lt>
2579  </list>
2580</t>
2581<t>
2582   If the TE field-value is empty or if no TE field is present, the only
2583   transfer-coding  is "chunked". A message with no transfer-coding is
2584   always acceptable.
2585</t>
2586</section>
2587
2588<section title="Trailer" anchor="header.trailer">
2589  <iref primary="true" item="Trailer header" x:for-anchor=""/>
2590  <iref primary="true" item="Headers" subitem="Trailer" x:for-anchor=""/>
2591<t>
2592   The Trailer general field value indicates that the given set of
2593   header fields is present in the trailer of a message encoded with
2594   chunked transfer-coding.
2595</t>
2596<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Trailer"/>
2597  Trailer  = "Trailer" ":" 1#field-name
2598</artwork></figure>
2599<t>
2600   An HTTP/1.1 message &SHOULD; include a Trailer header field in a
2601   message using chunked transfer-coding with a non-empty trailer. Doing
2602   so allows the recipient to know which header fields to expect in the
2603   trailer.
2604</t>
2605<t>
2606   If no Trailer header field is present, the trailer &SHOULD-NOT;  include
2607   any header fields. See <xref target="chunked.transfer.encoding"/> for restrictions on the use of
2608   trailer fields in a "chunked" transfer-coding.
2609</t>
2610<t>
2611   Message header fields listed in the Trailer header field &MUST-NOT;
2612   include the following header fields:
2613  <list style="symbols">
2614    <t>Transfer-Encoding</t>
2615    <t>Content-Length</t>
2616    <t>Trailer</t>
2617  </list>
2618</t>
2619</section>
2620
2621<section title="Transfer-Encoding" anchor="header.transfer-encoding">
2622  <iref primary="true" item="Transfer-Encoding header" x:for-anchor=""/>
2623  <iref primary="true" item="Headers" subitem="Transfer-Encoding" x:for-anchor=""/>
2624<t>
2625   The Transfer-Encoding general-header field indicates what (if any)
2626   type of transformation has been applied to the message body in order
2627   to safely transfer it between the sender and the recipient. This
2628   differs from the content-coding in that the transfer-coding is a
2629   property of the message, not of the entity.
2630</t>
2631<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/>
2632  Transfer-Encoding       = "Transfer-Encoding" ":" 1#transfer-coding
2633</artwork></figure>
2634<t>
2635   Transfer-codings are defined in <xref target="transfer.codings"/>. An example is:
2636</t>
2637<figure><artwork type="example">
2638  Transfer-Encoding: chunked
2639</artwork></figure>
2640<t>
2641   If multiple encodings have been applied to an entity, the transfer-codings
2642   &MUST; be listed in the order in which they were applied.
2643   Additional information about the encoding parameters &MAY; be provided
2644   by other entity-header fields not defined by this specification.
2645</t>
2646<t>
2647   Many older HTTP/1.0 applications do not understand the Transfer-Encoding
2648   header.
2649</t>
2650</section>
2651
2652<section title="Upgrade" anchor="header.upgrade">
2653  <iref primary="true" item="Upgrade header" x:for-anchor=""/>
2654  <iref primary="true" item="Headers" subitem="Upgrade" x:for-anchor=""/>
2655<t>
2656   The Upgrade general-header allows the client to specify what
2657   additional communication protocols it supports and would like to use
2658   if the server finds it appropriate to switch protocols. The server
2659   &MUST; use the Upgrade header field within a 101 (Switching Protocols)
2660   response to indicate which protocol(s) are being switched.
2661</t>
2662<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Upgrade"/>
2663  Upgrade        = "Upgrade" ":" 1#product
2664</artwork></figure>
2665<t>
2666   For example,
2667</t>
2668<figure><artwork type="example">
2669    Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
2670</artwork></figure>
2671<t>
2672   The Upgrade header field is intended to provide a simple mechanism
2673   for transition from HTTP/1.1 to some other, incompatible protocol. It
2674   does so by allowing the client to advertise its desire to use another
2675   protocol, such as a later version of HTTP with a higher major version
2676   number, even though the current request has been made using HTTP/1.1.
2677   This eases the difficult transition between incompatible protocols by
2678   allowing the client to initiate a request in the more commonly
2679   supported protocol while indicating to the server that it would like
2680   to use a "better" protocol if available (where "better" is determined
2681   by the server, possibly according to the nature of the method and/or
2682   resource being requested).
2683</t>
2684<t>
2685   The Upgrade header field only applies to switching application-layer
2686   protocols upon the existing transport-layer connection. Upgrade
2687   cannot be used to insist on a protocol change; its acceptance and use
2688   by the server is optional. The capabilities and nature of the
2689   application-layer communication after the protocol change is entirely
2690   dependent upon the new protocol chosen, although the first action
2691   after changing the protocol &MUST; be a response to the initial HTTP
2692   request containing the Upgrade header field.
2693</t>
2694<t>
2695   The Upgrade header field only applies to the immediate connection.
2696   Therefore, the upgrade keyword &MUST; be supplied within a Connection
2697   header field (<xref target="header.connection"/>) whenever Upgrade is present in an
2698   HTTP/1.1 message.
2699</t>
2700<t>
2701   The Upgrade header field cannot be used to indicate a switch to a
2702   protocol on a different connection. For that purpose, it is more
2703   appropriate to use a 301, 302, 303, or 305 redirection response.
2704</t>
2705<t>
2706   This specification only defines the protocol name "HTTP" for use by
2707   the family of Hypertext Transfer Protocols, as defined by the HTTP
2708   version rules of <xref target="http.version"/> and future updates to this
2709   specification. Any token can be used as a protocol name; however, it
2710   will only be useful if both the client and server associate the name
2711   with the same protocol.
2712</t>
2713</section>
2714
2715<section title="Via" anchor="header.via">
2716  <iref primary="true" item="Via header" x:for-anchor=""/>
2717  <iref primary="true" item="Headers" subitem="Via" x:for-anchor=""/>
2718<t>
2719   The Via general-header field &MUST; be used by gateways and proxies to
2720   indicate the intermediate protocols and recipients between the user
2721   agent and the server on requests, and between the origin server and
2722   the client on responses. It is analogous to the "Received" field of
2723   <xref target="RFC2822"/> and is intended to be used for tracking message forwards,
2724   avoiding request loops, and identifying the protocol capabilities of
2725   all senders along the request/response chain.
2726</t>
2727<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"/>
2728  Via =  "Via" ":" 1#( received-protocol received-by [ comment ] )
2729  received-protocol = [ protocol-name "/" ] protocol-version
2730  protocol-name     = token
2731  protocol-version  = token
2732  received-by       = ( uri-host [ ":" port ] ) | pseudonym
2733  pseudonym         = token
2734</artwork></figure>
2735<t>
2736   The received-protocol indicates the protocol version of the message
2737   received by the server or client along each segment of the
2738   request/response chain. The received-protocol version is appended to
2739   the Via field value when the message is forwarded so that information
2740   about the protocol capabilities of upstream applications remains
2741   visible to all recipients.
2742</t>
2743<t>
2744   The protocol-name is optional if and only if it would be "HTTP". The
2745   received-by field is normally the host and optional port number of a
2746   recipient server or client that subsequently forwarded the message.
2747   However, if the real host is considered to be sensitive information,
2748   it &MAY; be replaced by a pseudonym. If the port is not given, it &MAY;
2749   be assumed to be the default port of the received-protocol.
2750</t>
2751<t>
2752   Multiple Via field values represents each proxy or gateway that has
2753   forwarded the message. Each recipient &MUST; append its information
2754   such that the end result is ordered according to the sequence of
2755   forwarding applications.
2756</t>
2757<t>
2758   Comments &MAY; be used in the Via header field to identify the software
2759   of the recipient proxy or gateway, analogous to the User-Agent and
2760   Server header fields. However, all comments in the Via field are
2761   optional and &MAY; be removed by any recipient prior to forwarding the
2762   message.
2763</t>
2764<t>
2765   For example, a request message could be sent from an HTTP/1.0 user
2766   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
2767   forward the request to a public proxy at p.example.net, which completes
2768   the request by forwarding it to the origin server at www.example.com.
2769   The request received by www.example.com would then have the following
2770   Via header field:
2771</t>
2772<figure><artwork type="example">
2773    Via: 1.0 fred, 1.1 p.example.net (Apache/1.1)
2774</artwork></figure>
2775<t>
2776   Proxies and gateways used as a portal through a network firewall
2777   &SHOULD-NOT;, by default, forward the names and ports of hosts within
2778   the firewall region. This information &SHOULD; only be propagated if
2779   explicitly enabled. If not enabled, the received-by host of any host
2780   behind the firewall &SHOULD; be replaced by an appropriate pseudonym
2781   for that host.
2782</t>
2783<t>
2784   For organizations that have strong privacy requirements for hiding
2785   internal structures, a proxy &MAY; combine an ordered subsequence of
2786   Via header field entries with identical received-protocol values into
2787   a single such entry. For example,
2788</t>
2789<figure><artwork type="example">
2790    Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
2791</artwork></figure>
2792<t>
2793        could be collapsed to
2794</t>
2795<figure><artwork type="example">
2796    Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
2797</artwork></figure>
2798<t>
2799   Applications &SHOULD-NOT;  combine multiple entries unless they are all
2800   under the same organizational control and the hosts have already been
2801   replaced by pseudonyms. Applications &MUST-NOT; combine entries which
2802   have different received-protocol values.
2803</t>
2804</section>
2805
2806</section>
2807
2808<section title="IANA Considerations" anchor="IANA.considerations">
2809<t>
2810   <cref>TBD.</cref>
2811</t>
2812</section>
2813
2814<section title="Security Considerations" anchor="security.considerations">
2815<t>
2816   This section is meant to inform application developers, information
2817   providers, and users of the security limitations in HTTP/1.1 as
2818   described by this document. The discussion does not include
2819   definitive solutions to the problems revealed, though it does make
2820   some suggestions for reducing security risks.
2821</t>
2822
2823<section title="Personal Information" anchor="personal.information">
2824<t>
2825   HTTP clients are often privy to large amounts of personal information
2826   (e.g. the user's name, location, mail address, passwords, encryption
2827   keys, etc.), and &SHOULD; be very careful to prevent unintentional
2828   leakage of this information.
2829   We very strongly recommend that a convenient interface be provided
2830   for the user to control dissemination of such information, and that
2831   designers and implementors be particularly careful in this area.
2832   History shows that errors in this area often create serious security
2833   and/or privacy problems and generate highly adverse publicity for the
2834   implementor's company.
2835</t>
2836</section>
2837
2838<section title="Abuse of Server Log Information" anchor="abuse.of.server.log.information">
2839<t>
2840   A server is in the position to save personal data about a user's
2841   requests which might identify their reading patterns or subjects of
2842   interest. This information is clearly confidential in nature and its
2843   handling can be constrained by law in certain countries. People using
2844   HTTP to provide data are responsible for ensuring that
2845   such material is not distributed without the permission of any
2846   individuals that are identifiable by the published results.
2847</t>
2848</section>
2849
2850<section title="Attacks Based On File and Path Names" anchor="attack.pathname">
2851<t>
2852   Implementations of HTTP origin servers &SHOULD; be careful to restrict
2853   the documents returned by HTTP requests to be only those that were
2854   intended by the server administrators. If an HTTP server translates
2855   HTTP URIs directly into file system calls, the server &MUST; take
2856   special care not to serve files that were not intended to be
2857   delivered to HTTP clients. For example, UNIX, Microsoft Windows, and
2858   other operating systems use ".." as a path component to indicate a
2859   directory level above the current one. On such a system, an HTTP
2860   server &MUST; disallow any such construct in the Request-URI if it
2861   would otherwise allow access to a resource outside those intended to
2862   be accessible via the HTTP server. Similarly, files intended for
2863   reference only internally to the server (such as access control
2864   files, configuration files, and script code) &MUST; be protected from
2865   inappropriate retrieval, since they might contain sensitive
2866   information. Experience has shown that minor bugs in such HTTP server
2867   implementations have turned into security risks.
2868</t>
2869</section>
2870
2871<section title="DNS Spoofing" anchor="dns.spoofing">
2872<t>
2873   Clients using HTTP rely heavily on the Domain Name Service, and are
2874   thus generally prone to security attacks based on the deliberate
2875   mis-association of IP addresses and DNS names. Clients need to be
2876   cautious in assuming the continuing validity of an IP number/DNS name
2877   association.
2878</t>
2879<t>
2880   In particular, HTTP clients &SHOULD; rely on their name resolver for
2881   confirmation of an IP number/DNS name association, rather than
2882   caching the result of previous host name lookups. Many platforms
2883   already can cache host name lookups locally when appropriate, and
2884   they &SHOULD; be configured to do so. It is proper for these lookups to
2885   be cached, however, only when the TTL (Time To Live) information
2886   reported by the name server makes it likely that the cached
2887   information will remain useful.
2888</t>
2889<t>
2890   If HTTP clients cache the results of host name lookups in order to
2891   achieve a performance improvement, they &MUST; observe the TTL
2892   information reported by DNS.
2893</t>
2894<t>
2895   If HTTP clients do not observe this rule, they could be spoofed when
2896   a previously-accessed server's IP address changes. As network
2897   renumbering is expected to become increasingly common <xref target="RFC1900"/>, the
2898   possibility of this form of attack will grow. Observing this
2899   requirement thus reduces this potential security vulnerability.
2900</t>
2901<t>
2902   This requirement also improves the load-balancing behavior of clients
2903   for replicated servers using the same DNS name and reduces the
2904   likelihood of a user's experiencing failure in accessing sites which
2905   use that strategy.
2906</t>
2907</section>
2908
2909<section title="Proxies and Caching" anchor="attack.proxies">
2910<t>
2911   By their very nature, HTTP proxies are men-in-the-middle, and
2912   represent an opportunity for man-in-the-middle attacks. Compromise of
2913   the systems on which the proxies run can result in serious security
2914   and privacy problems. Proxies have access to security-related
2915   information, personal information about individual users and
2916   organizations, and proprietary information belonging to users and
2917   content providers. A compromised proxy, or a proxy implemented or
2918   configured without regard to security and privacy considerations,
2919   might be used in the commission of a wide range of potential attacks.
2920</t>
2921<t>
2922   Proxy operators should protect the systems on which proxies run as
2923   they would protect any system that contains or transports sensitive
2924   information. In particular, log information gathered at proxies often
2925   contains highly sensitive personal information, and/or information
2926   about organizations. Log information should be carefully guarded, and
2927   appropriate guidelines for use developed and followed. (<xref target="abuse.of.server.log.information"/>).
2928</t>
2929<t>
2930   Proxy implementors should consider the privacy and security
2931   implications of their design and coding decisions, and of the
2932   configuration options they provide to proxy operators (especially the
2933   default configuration).
2934</t>
2935<t>
2936   Users of a proxy need to be aware that they are no trustworthier than
2937   the people who run the proxy; HTTP itself cannot solve this problem.
2938</t>
2939<t>
2940   The judicious use of cryptography, when appropriate, may suffice to
2941   protect against a broad range of security and privacy attacks. Such
2942   cryptography is beyond the scope of the HTTP/1.1 specification.
2943</t>
2944</section>
2945
2946<section title="Denial of Service Attacks on Proxies" anchor="attack.DoS">
2947<t>
2948   They exist. They are hard to defend against. Research continues.
2949   Beware.
2950</t>
2951</section>
2952</section>
2953
2954<section title="Acknowledgments" anchor="ack">
2955<t>
2956   This specification makes heavy use of the augmented BNF and generic
2957   constructs defined by David H. Crocker for <xref target="RFC822ABNF"/>. Similarly, it
2958   reuses many of the definitions provided by Nathaniel Borenstein and
2959   Ned Freed for MIME <xref target="RFC2045"/>. We hope that their inclusion in this
2960   specification will help reduce past confusion over the relationship
2961   between HTTP and Internet mail message formats.
2962</t>
2963<t>
2964   HTTP has evolved considerably over the years. It has
2965   benefited from a large and active developer community--the many
2966   people who have participated on the www-talk mailing list--and it is
2967   that community which has been most responsible for the success of
2968   HTTP and of the World-Wide Web in general. Marc Andreessen, Robert
2969   Cailliau, Daniel W. Connolly, Bob Denny, John Franks, Jean-Francois
2970   Groff, Phillip M. Hallam-Baker, Hakon W. Lie, Ari Luotonen, Rob
2971   McCool, Lou Montulli, Dave Raggett, Tony Sanders, and Marc
2972   VanHeyningen deserve special recognition for their efforts in
2973   defining early aspects of the protocol.
2974</t>
2975<t>
2976   This document has benefited greatly from the comments of all those
2977   participating in the HTTP-WG. In addition to those already mentioned,
2978   the following individuals have contributed to this specification:
2979</t>
2980<t>
2981   Gary Adams, Harald Tveit Alvestrand, Keith Ball, Brian Behlendorf,
2982   Paul Burchard, Maurizio Codogno, Mike Cowlishaw, Roman Czyborra,
2983   Michael A. Dolan, Daniel DuBois, David J. Fiander, Alan Freier, Marc Hedlund, Greg Herlihy,
2984   Koen Holtman, Alex Hopmann, Bob Jernigan, Shel Kaphan, Rohit Khare,
2985   John Klensin, Martijn Koster, Alexei Kosut, David M. Kristol,
2986   Daniel LaLiberte, Ben Laurie, Paul J. Leach, Albert Lunde,
2987   John C. Mallery, Jean-Philippe Martin-Flatin, Mitra, David Morris,
2988   Gavin Nicol, Ross Patterson, Bill Perry, Jeffrey Perry, Scott Powers, Owen Rees,
2989   Luigi Rizzo, David Robinson, Marc Salomon, Rich Salz,
2990   Allan M. Schiffman, Jim Seidman, Chuck Shotton, Eric W. Sink,
2991   Simon E. Spero, Richard N. Taylor, Robert S. Thau,
2992   Bill (BearHeart) Weinman, Francois Yergeau, Mary Ellen Zurko,
2993   Josh Cohen.
2994</t>
2995<t>
2996   Thanks to the "cave men" of Palo Alto. You know who you are.
2997</t>
2998<t>
2999   Jim Gettys (the editor of <xref target="RFC2616"/>) wishes particularly
3000   to thank Roy Fielding, the editor of <xref target="RFC2068"/>, along
3001   with John Klensin, Jeff Mogul, Paul Leach, Dave Kristol, Koen
3002   Holtman, John Franks, Josh Cohen, Alex Hopmann, Scott Lawrence, and
3003   Larry Masinter for their help. And thanks go particularly to Jeff
3004   Mogul and Scott Lawrence for performing the "MUST/MAY/SHOULD" audit.
3005</t>
3006<t>
3007   The Apache Group, Anselm Baird-Smith, author of Jigsaw, and Henrik
3008   Frystyk implemented RFC 2068 early, and we wish to thank them for the
3009   discovery of many of the problems that this document attempts to
3010   rectify.
3011</t>
3012</section>
3013
3014</middle>
3015<back>
3016
3017<references title="Normative References">
3018
3019<reference anchor="ISO-8859-1">
3020  <front>
3021    <title>
3022     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
3023    </title>
3024    <author>
3025      <organization>International Organization for Standardization</organization>
3026    </author>
3027    <date year="1998"/>
3028  </front>
3029  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
3030</reference>
3031
3032<reference anchor="Part2">
3033  <front>
3034    <title abbrev="HTTP/1.1">HTTP/1.1, part 2: Message Semantics</title>
3035    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3036      <organization abbrev="Day Software">Day Software</organization>
3037      <address><email>fielding@gbiv.com</email></address>
3038    </author>
3039    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3040      <organization>One Laptop per Child</organization>
3041      <address><email>jg@laptop.org</email></address>
3042    </author>
3043    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3044      <organization abbrev="HP">Hewlett-Packard Company</organization>
3045      <address><email>JeffMogul@acm.org</email></address>
3046    </author>
3047    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
3048      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3049      <address><email>henrikn@microsoft.com</email></address>
3050    </author>
3051    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3052      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3053      <address><email>LMM@acm.org</email></address>
3054    </author>
3055    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3056      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3057      <address><email>paulle@microsoft.com</email></address>
3058    </author>
3059    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3060      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3061      <address><email>timbl@w3.org</email></address>
3062    </author>
3063    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
3064      <organization abbrev="W3C">World Wide Web Consortium</organization>
3065      <address><email>ylafon@w3.org</email></address>
3066    </author>
3067    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3068      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3069      <address><email>julian.reschke@greenbytes.de</email></address>
3070    </author>
3071    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3072  </front>
3073  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-&ID-VERSION;"/>
3074  <x:source href="p2-semantics.xml" basename="p2-semantics"/>
3075</reference>
3076
3077<reference anchor="Part3">
3078  <front>
3079    <title abbrev="HTTP/1.1">HTTP/1.1, part 3: Message Payload and Content Negotiation</title>
3080    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3081      <organization abbrev="Day Software">Day Software</organization>
3082      <address><email>fielding@gbiv.com</email></address>
3083    </author>
3084    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3085      <organization>One Laptop per Child</organization>
3086      <address><email>jg@laptop.org</email></address>
3087    </author>
3088    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3089      <organization abbrev="HP">Hewlett-Packard Company</organization>
3090      <address><email>JeffMogul@acm.org</email></address>
3091    </author>
3092    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
3093      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3094      <address><email>henrikn@microsoft.com</email></address>
3095    </author>
3096    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3097      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3098      <address><email>LMM@acm.org</email></address>
3099    </author>
3100    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3101      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3102      <address><email>paulle@microsoft.com</email></address>
3103    </author>
3104    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3105      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3106      <address><email>timbl@w3.org</email></address>
3107    </author>
3108    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
3109      <organization abbrev="W3C">World Wide Web Consortium</organization>
3110      <address><email>ylafon@w3.org</email></address>
3111    </author>
3112    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3113      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3114      <address><email>julian.reschke@greenbytes.de</email></address>
3115    </author>
3116    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3117  </front>
3118  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p3-payload-&ID-VERSION;"/>
3119  <x:source href="p3-payload.xml" basename="p3-payload"/>
3120</reference>
3121
3122<reference anchor="Part5">
3123  <front>
3124    <title abbrev="HTTP/1.1">HTTP/1.1, part 5: Range Requests and Partial Responses</title>
3125    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3126      <organization abbrev="Day Software">Day Software</organization>
3127      <address><email>fielding@gbiv.com</email></address>
3128    </author>
3129    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3130      <organization>One Laptop per Child</organization>
3131      <address><email>jg@laptop.org</email></address>
3132    </author>
3133    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3134      <organization abbrev="HP">Hewlett-Packard Company</organization>
3135      <address><email>JeffMogul@acm.org</email></address>
3136    </author>
3137    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
3138      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3139      <address><email>henrikn@microsoft.com</email></address>
3140    </author>
3141    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3142      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3143      <address><email>LMM@acm.org</email></address>
3144    </author>
3145    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3146      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3147      <address><email>paulle@microsoft.com</email></address>
3148    </author>
3149    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3150      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3151      <address><email>timbl@w3.org</email></address>
3152    </author>
3153    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
3154      <organization abbrev="W3C">World Wide Web Consortium</organization>
3155      <address><email>ylafon@w3.org</email></address>
3156    </author>
3157    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3158      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3159      <address><email>julian.reschke@greenbytes.de</email></address>
3160    </author>
3161    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3162  </front>
3163  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-&ID-VERSION;"/>
3164  <x:source href="p5-range.xml" basename="p5-range"/>
3165</reference>
3166
3167<reference anchor="Part6">
3168  <front>
3169    <title abbrev="HTTP/1.1">HTTP/1.1, part 6: Caching</title>
3170    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3171      <organization abbrev="Day Software">Day Software</organization>
3172      <address><email>fielding@gbiv.com</email></address>
3173    </author>
3174    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3175      <organization>One Laptop per Child</organization>
3176      <address><email>jg@laptop.org</email></address>
3177    </author>
3178    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3179      <organization abbrev="HP">Hewlett-Packard Company</organization>
3180      <address><email>JeffMogul@acm.org</email></address>
3181    </author>
3182    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
3183      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3184      <address><email>henrikn@microsoft.com</email></address>
3185    </author>
3186    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3187      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
3188      <address><email>LMM@acm.org</email></address>
3189    </author>
3190    <author initials="P." surname="Leach" fullname="Paul J. Leach">
3191      <organization abbrev="Microsoft">Microsoft Corporation</organization>
3192      <address><email>paulle@microsoft.com</email></address>
3193    </author>
3194    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3195      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
3196      <address><email>timbl@w3.org</email></address>
3197    </author>
3198    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
3199      <organization abbrev="W3C">World Wide Web Consortium</organization>
3200      <address><email>ylafon@w3.org</email></address>
3201    </author>
3202    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3203      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3204      <address><email>julian.reschke@greenbytes.de</email></address>
3205    </author>
3206    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3207  </front>
3208  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-&ID-VERSION;"/>
3209  <x:source href="p6-cache.xml" basename="p6-cache"/>
3210</reference>
3211
3212<reference anchor="RFC822ABNF">
3213  <front>
3214    <title abbrev="Standard for ARPA Internet Text Messages">Standard for the format of ARPA Internet text messages</title>
3215    <author initials="D.H." surname="Crocker" fullname="David H. Crocker">
3216      <organization>University of Delaware, Dept. of Electrical Engineering</organization>
3217      <address><email>DCrocker@UDel-Relay</email></address>
3218    </author>
3219    <date month="August" day="13" year="1982"/>
3220  </front>
3221  <seriesInfo name="STD" value="11"/>
3222  <seriesInfo name="RFC" value="822"/>
3223</reference>
3224
3225<reference anchor="RFC2045">
3226  <front>
3227    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
3228    <author initials="N." surname="Freed" fullname="Ned Freed">
3229      <organization>Innosoft International, Inc.</organization>
3230      <address><email>ned@innosoft.com</email></address>
3231    </author>
3232    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
3233      <organization>First Virtual Holdings</organization>
3234      <address><email>nsb@nsb.fv.com</email></address>
3235    </author>
3236    <date month="November" year="1996"/>
3237  </front>
3238  <seriesInfo name="RFC" value="2045"/>
3239</reference>
3240
3241<reference anchor="RFC2047">
3242  <front>
3243    <title abbrev="Message Header Extensions">MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text</title>
3244    <author initials="K." surname="Moore" fullname="Keith Moore">
3245      <organization>University of Tennessee</organization>
3246      <address><email>moore@cs.utk.edu</email></address>
3247    </author>
3248    <date month="November" year="1996"/>
3249  </front>
3250  <seriesInfo name="RFC" value="2047"/>
3251</reference>
3252
3253<reference anchor="RFC2119">
3254  <front>
3255    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
3256    <author initials="S." surname="Bradner" fullname="Scott Bradner">
3257      <organization>Harvard University</organization>
3258      <address><email>sob@harvard.edu</email></address>
3259    </author>
3260    <date month="March" year="1997"/>
3261  </front>
3262  <seriesInfo name="BCP" value="14"/>
3263  <seriesInfo name="RFC" value="2119"/>
3264</reference>
3265
3266<reference anchor="RFC2396">
3267  <front>
3268    <title abbrev="URI Generic Syntax">Uniform Resource Identifiers (URI): Generic Syntax</title>
3269    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3270      <organization abbrev="MIT/LCS">World Wide Web Consortium</organization>
3271      <address><email>timbl@w3.org</email></address>
3272    </author>
3273    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3274      <organization abbrev="U.C. Irvine">Department of Information and Computer Science</organization>
3275      <address><email>fielding@ics.uci.edu</email></address>
3276    </author>
3277    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3278      <organization abbrev="Xerox Corporation">Xerox PARC</organization>
3279      <address><email>masinter@parc.xerox.com</email></address>
3280    </author>
3281    <date month="August" year="1998"/>
3282  </front>
3283  <seriesInfo name="RFC" value="2396"/>
3284</reference>
3285
3286<reference anchor="USASCII">
3287  <front>
3288    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
3289    <author>
3290      <organization>American National Standards Institute</organization>
3291    </author>
3292    <date year="1986"/>
3293  </front>
3294  <seriesInfo name="ANSI" value="X3.4"/>
3295</reference>
3296
3297</references>
3298
3299<references title="Informative References">
3300
3301<reference anchor="Nie1997" target="http://doi.acm.org/10.1145/263105.263157">
3302  <front>
3303    <title>Network Performance Effects of HTTP/1.1, CSS1, and PNG</title>
3304    <author initials="H.F.." surname="Nielsen" fullname="H.F. Nielsen">
3305      <organization/>
3306    </author>
3307    <author initials="J." surname="Gettys" fullname="J. Gettys">
3308      <organization/>
3309    </author>
3310    <author initials="E." surname="Prud'hommeaux" fullname="E. Prud'hommeaux">
3311      <organization/>
3312    </author>
3313    <author initials="H." surname="Lie" fullname="H. Lie">
3314      <organization/>
3315    </author>
3316    <author initials="C." surname="Lilley" fullname="C. Lilley">
3317      <organization/>
3318    </author>
3319    <date year="1997" month="September"/>
3320  </front>
3321  <seriesInfo name="ACM" value="Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication SIGCOMM '97"/>
3322</reference>
3323
3324<reference anchor="Pad1995">
3325  <front>
3326    <title>Improving HTTP Latency</title>
3327    <author initials="V.N." surname="Padmanabhan" fullname="Venkata N. Padmanabhan">
3328      <organization/>
3329    </author>
3330    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
3331      <organization/>
3332    </author>
3333    <date year="1995" month="December"/>
3334  </front>
3335  <seriesInfo name="Computer Networks and ISDN Systems" value="v. 28, pp. 25-35"/>
3336  <annotation>
3337    Slightly revised version of paper in Proc. 2nd International WWW Conference '94: Mosaic and the Web, Oct. 1994,
3338    which is available at <eref target="http://www.ncsa.uiuc.edu/SDG/IT94/Proceedings/DDay/mogul/HTTPLatency.html"/>.
3339  </annotation>
3340</reference>
3341
3342<reference anchor="RFC822">
3343  <front>
3344    <title abbrev="Standard for ARPA Internet Text Messages">Standard for the format of ARPA Internet text messages</title>
3345    <author initials="D.H." surname="Crocker" fullname="David H. Crocker">
3346      <organization>University of Delaware, Dept. of Electrical Engineering</organization>
3347      <address><email>DCrocker@UDel-Relay</email></address>
3348    </author>
3349    <date month="August" day="13" year="1982"/>
3350  </front>
3351  <seriesInfo name="STD" value="11"/>
3352  <seriesInfo name="RFC" value="822"/>
3353</reference>
3354
3355<reference anchor="RFC959">
3356  <front>
3357    <title abbrev="File Transfer Protocol">File Transfer Protocol</title>
3358    <author initials="J." surname="Postel" fullname="J. Postel">
3359      <organization>Information Sciences Institute (ISI)</organization>
3360    </author>
3361    <author initials="J." surname="Reynolds" fullname="J. Reynolds">
3362      <organization/>
3363    </author>
3364    <date month="October" year="1985"/>
3365  </front>
3366  <seriesInfo name="STD" value="9"/>
3367  <seriesInfo name="RFC" value="959"/>
3368</reference>
3369
3370<reference anchor="RFC1123">
3371  <front>
3372    <title>Requirements for Internet Hosts - Application and Support</title>
3373    <author initials="R." surname="Braden" fullname="Robert Braden">
3374      <organization>University of Southern California (USC), Information Sciences Institute</organization>
3375      <address><email>Braden@ISI.EDU</email></address>
3376    </author>
3377    <date month="October" year="1989"/>
3378  </front>
3379  <seriesInfo name="STD" value="3"/>
3380  <seriesInfo name="RFC" value="1123"/>
3381</reference>
3382
3383<reference anchor="RFC1305">
3384  <front>
3385    <title>Network Time Protocol (Version 3) Specification, Implementation</title>
3386    <author initials="D." surname="Mills" fullname="David L. Mills">
3387      <organization>University of Delaware, Electrical Engineering Department</organization>
3388      <address><email>mills@udel.edu</email></address>
3389    </author>
3390    <date month="March" year="1992"/>
3391  </front>
3392  <seriesInfo name="RFC" value="1305"/>
3393</reference>
3394
3395<reference anchor="RFC1436">
3396  <front>
3397    <title abbrev="Gopher">The Internet Gopher Protocol (a distributed document search and retrieval protocol)</title>
3398    <author initials="F." surname="Anklesaria" fullname="Farhad Anklesaria">
3399      <organization>University of Minnesota, Computer and Information Services</organization>
3400      <address><email>fxa@boombox.micro.umn.edu</email></address>
3401    </author>
3402    <author initials="M." surname="McCahill" fullname="Mark McCahill">
3403      <organization>University of Minnesota, Computer and Information Services</organization>
3404      <address><email>mpm@boombox.micro.umn.edu</email></address>
3405    </author>
3406    <author initials="P." surname="Lindner" fullname="Paul Lindner">
3407      <organization>University of Minnesota, Computer and Information Services</organization>
3408      <address><email>lindner@boombox.micro.umn.edu</email></address>
3409    </author>
3410    <author initials="D." surname="Johnson" fullname="David Johnson">
3411      <organization>University of Minnesota, Computer and Information Services</organization>
3412      <address><email>dmj@boombox.micro.umn.edu</email></address>
3413    </author>
3414    <author initials="D." surname="Torrey" fullname="Daniel Torrey">
3415      <organization>University of Minnesota, Computer and Information Services</organization>
3416      <address><email>daniel@boombox.micro.umn.edu</email></address>
3417    </author>
3418    <author initials="B." surname="Alberti" fullname="Bob Alberti">
3419      <organization>University of Minnesota, Computer and Information Services</organization>
3420      <address><email>alberti@boombox.micro.umn.edu</email></address>
3421    </author>
3422    <date month="March" year="1993"/>
3423  </front>
3424  <seriesInfo name="RFC" value="1436"/>
3425</reference>
3426
3427<reference anchor="RFC1630">
3428  <front>
3429    <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>
3430    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3431      <organization>CERN, World-Wide Web project</organization>
3432      <address><email>timbl@info.cern.ch</email></address>
3433    </author>
3434    <date month="June" year="1994"/>
3435  </front>
3436  <seriesInfo name="RFC" value="1630"/>
3437</reference>
3438
3439<reference anchor="RFC1737">
3440  <front>
3441    <title abbrev="Requirements for Uniform Resource Names">Functional Requirements for Uniform Resource Names</title>
3442    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3443      <organization>Xerox Palo Alto Research Center</organization>
3444      <address><email>masinter@parc.xerox.com</email></address>
3445    </author>
3446    <author initials="K." surname="Sollins" fullname="Karen Sollins">
3447      <organization>MIT Laboratory for Computer Science</organization>
3448      <address><email>sollins@lcs.mit.edu</email></address>
3449    </author>
3450    <date month="December" year="1994"/>
3451  </front>
3452  <seriesInfo name="RFC" value="1737"/>
3453</reference>
3454
3455<reference anchor="RFC1738">
3456  <front>
3457    <title>Uniform Resource Locators (URL)</title>
3458    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3459      <organization>CERN, World-Wide Web project</organization>
3460      <address><email>timbl@info.cern.ch</email></address>
3461    </author>
3462    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3463      <organization>Xerox PARC</organization>
3464      <address><email>masinter@parc.xerox.com</email></address>
3465    </author>
3466    <author initials="M." surname="McCahill" fullname="Mark McCahill">
3467      <organization>University of Minnesota, Computer and Information Services</organization>
3468      <address><email>mpm@boombox.micro.umn.edu</email></address>
3469    </author>
3470    <date month="December" year="1994"/>
3471  </front>
3472  <seriesInfo name="RFC" value="1738"/>
3473</reference>
3474
3475<reference anchor="RFC1808">
3476  <front>
3477    <title>Relative Uniform Resource Locators</title>
3478    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
3479      <organization>University of California Irvine, Department of Information and Computer Science</organization>
3480      <address><email>fielding@ics.uci.edu</email></address>
3481    </author>
3482    <date month="June" year="1995"/>
3483  </front>
3484  <seriesInfo name="RFC" value="1808"/>
3485</reference>
3486
3487<reference anchor="RFC1900">
3488  <front>
3489    <title>Renumbering Needs Work</title>
3490    <author initials="B." surname="Carpenter" fullname="Brian E. Carpenter">
3491      <organization>CERN, Computing and Networks Division</organization>
3492      <address><email>brian@dxcoms.cern.ch</email></address>
3493    </author>
3494    <author initials="Y." surname="Rekhter" fullname="Yakov Rekhter">
3495      <organization>cisco Systems</organization>
3496      <address><email>yakov@cisco.com</email></address>
3497    </author>
3498    <date month="February" year="1996"/>
3499  </front>
3500  <seriesInfo name="RFC" value="1900"/>
3501</reference>
3502
3503<reference anchor="RFC1945">
3504  <front>
3505    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
3506    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3507      <organization>MIT, Laboratory for Computer Science</organization>
3508      <address><email>timbl@w3.org</email></address>
3509    </author>
3510    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3511      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
3512      <address><email>fielding@ics.uci.edu</email></address>
3513    </author>
3514    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3515      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
3516      <address><email>frystyk@w3.org</email></address>
3517    </author>
3518    <date month="May" year="1996"/>
3519  </front>
3520  <seriesInfo name="RFC" value="1945"/>
3521</reference>
3522
3523<reference anchor="RFC2068">
3524  <front>
3525    <title abbrev="HTTP/1.1">Hypertext Transfer Protocol -- HTTP/1.1</title>
3526    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
3527      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
3528      <address><email>fielding@ics.uci.edu</email></address>
3529    </author>
3530    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3531      <organization>MIT Laboratory for Computer Science</organization>
3532      <address><email>jg@w3.org</email></address>
3533    </author>
3534    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
3535      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
3536      <address><email>mogul@wrl.dec.com</email></address>
3537    </author>
3538    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3539      <organization>MIT Laboratory for Computer Science</organization>
3540      <address><email>frystyk@w3.org</email></address>
3541    </author>
3542    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
3543      <organization>MIT Laboratory for Computer Science</organization>
3544      <address><email>timbl@w3.org</email></address>
3545    </author>
3546    <date month="January" year="1997"/>
3547  </front>
3548  <seriesInfo name="RFC" value="2068"/>
3549</reference>
3550
3551<reference anchor="RFC2145">
3552  <front>
3553    <title abbrev="HTTP Version Numbers">Use and Interpretation of HTTP Version Numbers</title>
3554    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
3555      <organization>Western Research Laboratory</organization>
3556      <address><email>mogul@wrl.dec.com</email></address>
3557    </author>
3558    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
3559      <organization>Department of Information and Computer Science</organization>
3560      <address><email>fielding@ics.uci.edu</email></address>
3561    </author>
3562    <author initials="J." surname="Gettys" fullname="Jim Gettys">
3563      <organization>MIT Laboratory for Computer Science</organization>
3564      <address><email>jg@w3.org</email></address>
3565    </author>
3566    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
3567      <organization>W3 Consortium</organization>
3568      <address><email>frystyk@w3.org</email></address>
3569    </author>
3570    <date month="May" year="1997"/>
3571  </front>
3572  <seriesInfo name="RFC" value="2145"/>
3573</reference>
3574
3575<reference anchor="RFC2324">
3576  <front>
3577    <title abbrev="HTCPCP/1.0">Hyper Text Coffee Pot Control Protocol (HTCPCP/1.0)</title>
3578    <author initials="L." surname="Masinter" fullname="Larry Masinter">
3579      <organization>Xerox Palo Alto Research Center</organization>
3580      <address><email>masinter@parc.xerox.com</email></address>
3581    </author>
3582    <date month="April" day="1" year="1998"/>
3583  </front>
3584  <seriesInfo name="RFC" value="2324"/>
3585</reference>
3586
3587<reference anchor="RFC2616">
3588  <front>
3589    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
3590    <author initials="R." surname="Fielding" fullname="R. Fielding">
3591      <organization>University of California, Irvine</organization>
3592      <address><email>fielding@ics.uci.edu</email></address>
3593    </author>
3594    <author initials="J." surname="Gettys" fullname="J. Gettys">
3595      <organization>W3C</organization>
3596      <address><email>jg@w3.org</email></address>
3597    </author>
3598    <author initials="J." surname="Mogul" fullname="J. Mogul">
3599      <organization>Compaq Computer Corporation</organization>
3600      <address><email>mogul@wrl.dec.com</email></address>
3601    </author>
3602    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
3603      <organization>MIT Laboratory for Computer Science</organization>
3604      <address><email>frystyk@w3.org</email></address>
3605    </author>
3606    <author initials="L." surname="Masinter" fullname="L. Masinter">
3607      <organization>Xerox Corporation</organization>
3608      <address><email>masinter@parc.xerox.com</email></address>
3609    </author>
3610    <author initials="P." surname="Leach" fullname="P. Leach">
3611      <organization>Microsoft Corporation</organization>
3612      <address><email>paulle@microsoft.com</email></address>
3613    </author>
3614    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
3615      <organization>W3C</organization>
3616      <address><email>timbl@w3.org</email></address>
3617    </author>
3618    <date month="June" year="1999"/>
3619  </front>
3620  <seriesInfo name="RFC" value="2616"/>
3621</reference>
3622
3623<reference anchor="RFC2821">
3624  <front>
3625    <title>Simple Mail Transfer Protocol</title>
3626    <author initials="J." surname="Klensin" fullname="J. Klensin">
3627      <organization>AT&amp;T Laboratories</organization>
3628      <address><email>klensin@research.att.com</email></address>
3629    </author>
3630    <date year="2001" month="April"/>
3631  </front>
3632  <seriesInfo name="RFC" value="2821"/>
3633</reference>
3634
3635<reference anchor="RFC2822">
3636  <front>
3637    <title>Internet Message Format</title>
3638    <author initials="P." surname="Resnick" fullname="P. Resnick">
3639      <organization>QUALCOMM Incorporated</organization>
3640    </author>
3641    <date year="2001" month="April"/>
3642  </front>
3643  <seriesInfo name="RFC" value="2822"/>
3644</reference>
3645
3646<reference anchor='RFC3977'>
3647  <front>
3648    <title>Network News Transfer Protocol (NNTP)</title>
3649    <author initials='C.' surname='Feather' fullname='C. Feather'>
3650      <organization>THUS plc</organization>
3651      <address><email>clive@demon.net</email></address>
3652    </author>
3653    <date year='2006' month='October' />
3654  </front>
3655  <seriesInfo name="RFC" value="3977"/>
3656</reference>
3657
3658<reference anchor="RFC4288">
3659  <front>
3660    <title>Media Type Specifications and Registration Procedures</title>
3661    <author initials="N." surname="Freed" fullname="N. Freed">
3662      <organization>Sun Microsystems</organization>
3663      <address>
3664        <email>ned.freed@mrochek.com</email>
3665      </address>
3666    </author>
3667    <author initials="J." surname="Klensin" fullname="J. Klensin">
3668      <organization/>
3669      <address>
3670        <email>klensin+ietf@jck.com</email>
3671      </address>
3672    </author>
3673    <date year="2005" month="December"/>
3674  </front>
3675  <seriesInfo name="BCP" value="13"/>
3676  <seriesInfo name="RFC" value="4288"/>
3677</reference>
3678
3679<reference anchor="Spe" target="http://sunsite.unc.edu/mdma-release/http-prob.html">
3680  <front>
3681  <title>Analysis of HTTP Performance Problems</title>
3682  <author initials="S." surname="Spero" fullname="Simon E. Spero">
3683    <organization/>
3684  </author>
3685  <date/>
3686  </front>
3687</reference>
3688
3689<reference anchor="Tou1998" target="http://www.isi.edu/touch/pubs/http-perf96/">
3690  <front>
3691  <title>Analysis of HTTP Performance</title>
3692  <author initials="J." surname="Touch" fullname="Joe Touch">
3693    <organization>USC/Information Sciences Institute</organization>
3694    <address><email>touch@isi.edu</email></address>
3695  </author>
3696  <author initials="J." surname="Heidemann" fullname="John Heidemann">
3697    <organization>USC/Information Sciences Institute</organization>
3698    <address><email>johnh@isi.edu</email></address>
3699  </author>
3700  <author initials="K." surname="Obraczka" fullname="Katia Obraczka">
3701    <organization>USC/Information Sciences Institute</organization>
3702    <address><email>katia@isi.edu</email></address>
3703  </author>
3704  <date year="1998" month="Aug"/>
3705  </front>
3706  <seriesInfo name="ISI Research Report" value="ISI/RR-98-463"/>
3707  <annotation>(original report dated Aug. 1996)</annotation>
3708</reference>
3709
3710<reference anchor="WAIS">
3711  <front>
3712    <title>WAIS Interface Protocol Prototype Functional Specification (v1.5)</title>
3713    <author initials="F." surname="Davis" fullname="F. Davis">
3714      <organization>Thinking Machines Corporation</organization>
3715    </author>
3716    <author initials="B." surname="Kahle" fullname="B. Kahle">
3717      <organization>Thinking Machines Corporation</organization>
3718    </author>
3719    <author initials="H." surname="Morris" fullname="H. Morris">
3720      <organization>Thinking Machines Corporation</organization>
3721    </author>
3722    <author initials="J." surname="Salem" fullname="J. Salem">
3723      <organization>Thinking Machines Corporation</organization>
3724    </author>
3725    <author initials="T." surname="Shen" fullname="T. Shen">
3726      <organization>Thinking Machines Corporation</organization>
3727    </author>
3728    <author initials="R." surname="Wang" fullname="R. Wang">
3729      <organization>Thinking Machines Corporation</organization>
3730    </author>
3731    <author initials="J." surname="Sui" fullname="J. Sui">
3732      <organization>Thinking Machines Corporation</organization>
3733    </author>
3734    <author initials="M." surname="Grinbaum" fullname="M. Grinbaum">
3735      <organization>Thinking Machines Corporation</organization>
3736    </author>
3737    <date month="April" year="1990"/>
3738  </front>
3739  <seriesInfo name="Thinking Machines Corporation" value=""/>
3740</reference>
3741
3742</references>
3743
3744
3745<section title="Internet Media Types" anchor="internet.media.type.http">
3746<t>
3747   In addition to defining HTTP/1.1, this document serves
3748   as the specification for the Internet media type "message/http" and
3749   "application/http". The following is to be registered with IANA <xref target="RFC4288"/>.
3750</t>
3751<section title="Internet Media Type message/http" anchor="internet.media.type.message.http">
3752<iref item="Media Type" subitem="message/http" primary="true"/>
3753<iref item="message/http Media Type" primary="true"/>
3754<t>
3755   The message/http type can be used to enclose a single HTTP request or
3756   response message, provided that it obeys the MIME restrictions for all
3757   "message" types regarding line length and encodings.
3758</t>
3759<t>
3760  <list style="hanging" x:indent="12em">
3761    <t hangText="Type name:">
3762      message
3763    </t>
3764    <t hangText="Subtype name:">
3765      http
3766    </t>
3767    <t hangText="Required parameters:">
3768      none
3769    </t>
3770    <t hangText="Optional parameters:">
3771      version, msgtype
3772      <list style="hanging">
3773        <t hangText="version:">
3774          The HTTP-Version number of the enclosed message
3775          (e.g., "1.1"). If not present, the version can be
3776          determined from the first line of the body.
3777        </t>
3778        <t hangText="msgtype:">
3779          The message type -- "request" or "response". If not
3780          present, the type can be determined from the first
3781          line of the body.
3782        </t>
3783      </list>
3784    </t>
3785    <t hangText="Encoding considerations:">
3786      only "7bit", "8bit", or "binary" are permitted
3787    </t>
3788    <t hangText="Security considerations:">
3789      none
3790    </t>
3791    <t hangText="Interoperability considerations:">
3792      none
3793    </t>
3794    <t hangText="Published specification:">
3795      This specification (see <xref target="internet.media.type.message.http"/>).
3796    </t>
3797    <t hangText="Applications that use this media type:">
3798    </t>
3799    <t hangText="Additional information:">
3800      <list style="hanging">
3801        <t hangText="Magic number(s):">none</t>
3802        <t hangText="File extension(s):">none</t>
3803        <t hangText="Macintosh file type code(s):">none</t>
3804      </list>
3805    </t>
3806    <t hangText="Person and email address to contact for further information:">
3807      See Authors Section.
3808    </t>
3809                <t hangText="Intended usage:">
3810                  COMMON
3811    </t>
3812                <t hangText="Restrictions on usage:">
3813                  none
3814    </t>
3815    <t hangText="Author/Change controller:">
3816      IESG
3817    </t>
3818  </list>
3819</t>
3820</section>
3821<section title="Internet Media Type application/http" anchor="internet.media.type.application.http">
3822<iref item="Media Type" subitem="application/http" primary="true"/>
3823<iref item="application/http Media Type" primary="true"/>
3824<t>
3825   The application/http type can be used to enclose a pipeline of one or more
3826   HTTP request or response messages (not intermixed).
3827</t>
3828<t>
3829  <list style="hanging" x:indent="12em">
3830    <t hangText="Type name:">
3831      application
3832    </t>
3833    <t hangText="Subtype name:">
3834      http
3835    </t>
3836    <t hangText="Required parameters:">
3837      none
3838    </t>
3839    <t hangText="Optional parameters:">
3840      version, msgtype
3841      <list style="hanging">
3842        <t hangText="version:">
3843          The HTTP-Version number of the enclosed messages
3844          (e.g., "1.1"). If not present, the version can be
3845          determined from the first line of the body.
3846        </t>
3847        <t hangText="msgtype:">
3848          The message type -- "request" or "response". If not
3849          present, the type can be determined from the first
3850          line of the body.
3851        </t>
3852      </list>
3853    </t>
3854    <t hangText="Encoding considerations:">
3855      HTTP messages enclosed by this type
3856      are in "binary" format; use of an appropriate
3857      Content-Transfer-Encoding is required when
3858      transmitted via E-mail.
3859    </t>
3860    <t hangText="Security considerations:">
3861      none
3862    </t>
3863    <t hangText="Interoperability considerations:">
3864      none
3865    </t>
3866    <t hangText="Published specification:">
3867      This specification (see <xref target="internet.media.type.application.http"/>).
3868    </t>
3869    <t hangText="Applications that use this media type:">
3870    </t>
3871    <t hangText="Additional information:">
3872      <list style="hanging">
3873        <t hangText="Magic number(s):">none</t>
3874        <t hangText="File extension(s):">none</t>
3875        <t hangText="Macintosh file type code(s):">none</t>
3876      </list>
3877    </t>
3878    <t hangText="Person and email address to contact for further information:">
3879      See Authors Section.
3880    </t>
3881                <t hangText="Intended usage:">
3882                  COMMON
3883    </t>
3884                <t hangText="Restrictions on usage:">
3885                  none
3886    </t>
3887    <t hangText="Author/Change controller:">
3888      IESG
3889    </t>
3890  </list>
3891</t>
3892</section>
3893</section>
3894
3895<section title="Tolerant Applications" anchor="tolerant.applications">
3896<t>
3897   Although this document specifies the requirements for the generation
3898   of HTTP/1.1 messages, not all applications will be correct in their
3899   implementation. We therefore recommend that operational applications
3900   be tolerant of deviations whenever those deviations can be
3901   interpreted unambiguously.
3902</t>
3903<t>
3904   Clients &SHOULD; be tolerant in parsing the Status-Line and servers
3905   tolerant when parsing the Request-Line. In particular, they &SHOULD;
3906   accept any amount of SP or HTAB characters between fields, even though
3907   only a single SP is required.
3908</t>
3909<t>
3910   The line terminator for message-header fields is the sequence CRLF.
3911   However, we recommend that applications, when parsing such headers,
3912   recognize a single LF as a line terminator and ignore the leading CR.
3913</t>
3914<t>
3915   The character set of an entity-body &SHOULD; be labeled as the lowest
3916   common denominator of the character codes used within that body, with
3917   the exception that not labeling the entity is preferred over labeling
3918   the entity with the labels US-ASCII or ISO-8859-1. See &payload;.
3919</t>
3920<t>
3921   Additional rules for requirements on parsing and encoding of dates
3922   and other potential problems with date encodings include:
3923</t>
3924<t>
3925  <list style="symbols">
3926     <t>HTTP/1.1 clients and caches &SHOULD; assume that an RFC-850 date
3927        which appears to be more than 50 years in the future is in fact
3928        in the past (this helps solve the "year 2000" problem).</t>
3929
3930     <t>An HTTP/1.1 implementation &MAY; internally represent a parsed
3931        Expires date as earlier than the proper value, but &MUST-NOT;
3932        internally represent a parsed Expires date as later than the
3933        proper value.</t>
3934
3935     <t>All expiration-related calculations &MUST; be done in GMT. The
3936        local time zone &MUST-NOT; influence the calculation or comparison
3937        of an age or expiration time.</t>
3938
3939     <t>If an HTTP header incorrectly carries a date value with a time
3940        zone other than GMT, it &MUST; be converted into GMT using the
3941        most conservative possible conversion.</t>
3942  </list>
3943</t>
3944</section>
3945
3946<section title="Conversion of Date Formats" anchor="conversion.of.date.formats">
3947<t>
3948   HTTP/1.1 uses a restricted set of date formats (<xref target="full.date"/>) to
3949   simplify the process of date comparison. Proxies and gateways from
3950   other protocols &SHOULD; ensure that any Date header field present in a
3951   message conforms to one of the HTTP/1.1 formats and rewrite the date
3952   if necessary.
3953</t>
3954</section>
3955
3956<section title="Compatibility with Previous Versions" anchor="compatibility">
3957<t>
3958   It is beyond the scope of a protocol specification to mandate
3959   compliance with previous versions. HTTP/1.1 was deliberately
3960   designed, however, to make supporting previous versions easy. It is
3961   worth noting that, at the time of composing this specification
3962   (1996), we would expect commercial HTTP/1.1 servers to:
3963  <list style="symbols">
3964     <t>recognize the format of the Request-Line for HTTP/0.9, 1.0, and
3965        1.1 requests;</t>
3966
3967     <t>understand any valid request in the format of HTTP/0.9, 1.0, or
3968        1.1;</t>
3969
3970     <t>respond appropriately with a message in the same major version
3971        used by the client.</t>
3972  </list>
3973</t>
3974<t>
3975   And we would expect HTTP/1.1 clients to:
3976  <list style="symbols">
3977     <t>recognize the format of the Status-Line for HTTP/1.0 and 1.1
3978        responses;</t>
3979
3980     <t>understand any valid response in the format of HTTP/0.9, 1.0, or
3981        1.1.</t>
3982  </list>
3983</t>
3984<t>
3985   For most implementations of HTTP/1.0, each connection is established
3986   by the client prior to the request and closed by the server after
3987   sending the response. Some implementations implement the Keep-Alive
3988   version of persistent connections described in <xref x:sec="19.7.1" x:fmt="of" target="RFC2068"/>.
3989</t>
3990
3991<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
3992<t>
3993   This section summarizes major differences between versions HTTP/1.0
3994   and HTTP/1.1.
3995</t>
3996
3997<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">
3998<t>
3999   The requirements that clients and servers support the Host request-header,
4000   report an error if the Host request-header (<xref target="header.host"/>) is
4001   missing from an HTTP/1.1 request, and accept absolute URIs (<xref target="request-uri"/>)
4002   are among the most important changes defined by this
4003   specification.
4004</t>
4005<t>
4006   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
4007   addresses and servers; there was no other established mechanism for
4008   distinguishing the intended server of a request than the IP address
4009   to which that request was directed. The changes outlined above will
4010   allow the Internet, once older HTTP clients are no longer common, to
4011   support multiple Web sites from a single IP address, greatly
4012   simplifying large operational Web servers, where allocation of many
4013   IP addresses to a single host has created serious problems. The
4014   Internet will also be able to recover the IP addresses that have been
4015   allocated for the sole purpose of allowing special-purpose domain
4016   names to be used in root-level HTTP URLs. Given the rate of growth of
4017   the Web, and the number of servers already deployed, it is extremely
4018   important that all implementations of HTTP (including updates to
4019   existing HTTP/1.0 applications) correctly implement these
4020   requirements:
4021  <list style="symbols">
4022     <t>Both clients and servers &MUST; support the Host request-header.</t>
4023
4024     <t>A client that sends an HTTP/1.1 request &MUST; send a Host header.</t>
4025
4026     <t>Servers &MUST; report a 400 (Bad Request) error if an HTTP/1.1
4027        request does not include a Host request-header.</t>
4028
4029     <t>Servers &MUST; accept absolute URIs.</t>
4030  </list>
4031</t>
4032</section>
4033</section>
4034
4035<section title="Compatibility with HTTP/1.0 Persistent Connections" anchor="compatibility.with.http.1.0.persistent.connections">
4036<t>
4037   Some clients and servers might wish to be compatible with some
4038   previous implementations of persistent connections in HTTP/1.0
4039   clients and servers. Persistent connections in HTTP/1.0 are
4040   explicitly negotiated as they are not the default behavior. HTTP/1.0
4041   experimental implementations of persistent connections are faulty,
4042   and the new facilities in HTTP/1.1 are designed to rectify these
4043   problems. The problem was that some existing 1.0 clients may be
4044   sending Keep-Alive to a proxy server that doesn't understand
4045   Connection, which would then erroneously forward it to the next
4046   inbound server, which would establish the Keep-Alive connection and
4047   result in a hung HTTP/1.0 proxy waiting for the close on the
4048   response. The result is that HTTP/1.0 clients must be prevented from
4049   using Keep-Alive when talking to proxies.
4050</t>
4051<t>
4052   However, talking to proxies is the most important use of persistent
4053   connections, so that prohibition is clearly unacceptable. Therefore,
4054   we need some other mechanism for indicating a persistent connection
4055   is desired, which is safe to use even when talking to an old proxy
4056   that ignores Connection. Persistent connections are the default for
4057   HTTP/1.1 messages; we introduce a new keyword (Connection: close) for
4058   declaring non-persistence. See <xref target="header.connection"/>.
4059</t>
4060<t>
4061   The original HTTP/1.0 form of persistent connections (the Connection:
4062   Keep-Alive and Keep-Alive header) is documented in <xref target="RFC2068"/>.
4063</t>
4064</section>
4065
4066<section title="Changes from RFC 2068" anchor="changes.from.rfc.2068">
4067<t>
4068   This specification has been carefully audited to correct and
4069   disambiguate key word usage; RFC 2068 had many problems in respect to
4070   the conventions laid out in <xref target="RFC2119"/>.
4071</t>
4072<t>
4073   Transfer-coding and message lengths all interact in ways that
4074   required fixing exactly when chunked encoding is used (to allow for
4075   transfer encoding that may not be self delimiting); it was important
4076   to straighten out exactly how message lengths are computed. (Sections
4077   <xref target="transfer.codings" format="counter"/>, <xref target="message.length" format="counter"/>,
4078   <xref target="header.content-length" format="counter"/>,
4079   see also <xref target="Part3"/>, <xref target="Part5"/> and <xref target="Part6"/>)
4080</t>
4081<t>
4082   The use and interpretation of HTTP version numbers has been clarified
4083   by <xref target="RFC2145"/>. Require proxies to upgrade requests to highest protocol
4084   version they support to deal with problems discovered in HTTP/1.0
4085   implementations (<xref target="http.version"/>)
4086</t>
4087<t>
4088   Transfer-coding had significant problems, particularly with
4089   interactions with chunked encoding. The solution is that transfer-codings
4090   become as full fledged as content-codings. This involves
4091   adding an IANA registry for transfer-codings (separate from content
4092   codings), a new header field (TE) and enabling trailer headers in the
4093   future. Transfer encoding is a major performance benefit, so it was
4094   worth fixing <xref target="Nie1997"/>. TE also solves another, obscure, downward
4095   interoperability problem that could have occurred due to interactions
4096   between authentication trailers, chunked encoding and HTTP/1.0
4097   clients.(Section <xref target="transfer.codings" format="counter"/>, <xref target="chunked.transfer.encoding" format="counter"/>,
4098   and <xref target="header.te" format="counter"/>)
4099</t>
4100</section>
4101
4102<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
4103<t>
4104  The CHAR rule does not allow the NUL character anymore (this affects
4105  the comment and quoted-string rules).
4106  (<xref target="basic.rules"/>)
4107</t>
4108<t>
4109  Clarify that HTTP-Version is case sensitive.
4110  (<xref target="http.version"/>)
4111</t>
4112<t>
4113  Remove reference to non-existant identity transfer-coding value tokens.
4114  (Sections <xref format="counter" target="transfer.codings"/> and
4115  <xref format="counter" target="message.length"/>)
4116</t>
4117<t>
4118  Clarification that the chunk length does not include
4119  the count of the octets in the chunk header and trailer.
4120  (<xref target="chunked.transfer.encoding"/>)
4121</t>
4122<t>
4123  Fix BNF to add query, as the abs_path production in
4124  <xref x:sec="3" x:fmt="of" target="RFC2396"/> doesn't define it.
4125  (<xref target="request-uri"/>)
4126</t>
4127<t>
4128  Clarify exactly when close connection options must be sent.
4129  (<xref target="header.connection"/>)
4130</t>
4131</section>
4132</section>
4133
4134<section title="Change Log (to be removed by RFC Editor before publication)">
4135
4136<section title="Since RFC2616">
4137<t>
4138  Extracted relevant partitions from <xref target="RFC2616"/>.
4139</t>
4140</section>
4141
4142<section title="Since draft-ietf-httpbis-p1-messaging-00">
4143<t>
4144  Closed issues:
4145  <list style="symbols">
4146    <t>
4147      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/1"/>:
4148      "HTTP Version should be case sensitive"
4149      (<eref target="http://purl.org/NET/http-errata#verscase"/>)
4150    </t>
4151    <t>
4152      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/2"/>:
4153      "'unsafe' characters"
4154      (<eref target="http://purl.org/NET/http-errata#unsafe-uri"/>)
4155    </t>
4156    <t>
4157      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/3"/>:
4158      "Chunk Size Definition"
4159      (<eref target="http://purl.org/NET/http-errata#chunk-size"/>)
4160    </t>
4161    <t>
4162      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/4"/>:
4163      "Message Length"
4164      (<eref target="http://purl.org/NET/http-errata#msg-len-chars"/>)
4165    </t>
4166    <t>
4167      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/8"/>:
4168      "Media Type Registrations"
4169      (<eref target="http://purl.org/NET/http-errata#media-reg"/>)
4170    </t>
4171    <t>
4172      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/11"/>:
4173      "URI includes query"
4174      (<eref target="http://purl.org/NET/http-errata#uriquery"/>)
4175    </t>
4176    <t>
4177      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/15"/>:
4178      "No close on 1xx responses"
4179      (<eref target="http://purl.org/NET/http-errata#noclose1xx"/>)
4180    </t>
4181    <t>
4182      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/16"/>:
4183      "Remove 'identity' token references"
4184      (<eref target="http://purl.org/NET/http-errata#identity"/>)
4185    </t>
4186    <t>
4187      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/26"/>:
4188      "Import query BNF"
4189    </t>
4190    <t>
4191      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/31"/>:
4192      "qdtext BNF"
4193    </t>
4194    <t>
4195      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/35"/>:
4196      "Normative and Informative references"
4197    </t>
4198    <t>
4199      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/42"/>:
4200      "RFC2606 Compliance"
4201    </t>
4202    <t>
4203      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/45"/>:
4204      "RFC977 reference"
4205    </t>
4206    <t>
4207      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/46"/>:
4208      "RFC1700 references"
4209    </t>
4210    <t>
4211      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/47"/>:
4212      "inconsistency in date format explanation"
4213    </t>
4214    <t>
4215      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/48"/>:
4216      "Date reference typo"
4217    </t>
4218    <t>
4219      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/65"/>:
4220      "Informative references"
4221    </t>
4222    <t>
4223      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/66"/>:
4224      "ISO-8859-1 Reference"
4225    </t>
4226    <t>
4227      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/86"/>:
4228      "Normative up-to-date references"
4229    </t>
4230  </list>
4231</t>
4232<t>
4233  Other changes:
4234  <list style="symbols">
4235    <t>
4236      Update media type registrations to use RFC4288 template.
4237    </t>
4238    <t>
4239      Use names of RFC4234 core rules DQUOTE and HTAB,
4240      fix broken ABNF for chunk-data
4241      (work in progress on <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/36"/>)
4242    </t>
4243  </list>
4244</t>
4245</section>
4246
4247<section title="Since draft-ietf-httpbis-p1-messaging-01">
4248<t>
4249  Closed issues:
4250  <list style="symbols">
4251    <t>
4252      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/19"/>:
4253      "Bodies on GET (and other) requests"
4254    </t>
4255    <t>
4256      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/55"/>:
4257      "Updating to RFC4288"
4258    </t>
4259    <t>
4260      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/57"/>:
4261      "Status Code and Reason Phrase"
4262    </t>
4263    <t>
4264      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/82"/>:
4265      "rel_path not used"
4266    </t>
4267  </list>
4268</t>
4269<t>
4270  Ongoing work on ABNF conversion (<eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/36"/>):
4271  <list style="symbols">
4272    <t>
4273      Get rid of duplicate BNF rule names ("host" -> "uri-host", "trailer" ->
4274      "trailer-part").
4275    </t>
4276    <t>
4277      Avoid underscore character in rule names ("http_URL" ->
4278      "http-URL", "abs_path" -> "path-absolute").
4279    </t>
4280    <t>
4281      Add rules for terms imported from URI spec ("absoluteURI", "authority",
4282      "path-absolute", "port", "query", "relativeURI", "host) -- these will
4283      have to be updated when switching over to RFC3986.
4284    </t>
4285    <t>
4286      Synchronize core rules with RFC5234 (this includes a change to CHAR
4287      which now excludes NUL).
4288    </t>
4289    <t>
4290      Get rid of prose rules that span multiple lines.
4291    </t>
4292    <t>
4293      Get rid of unused rules LOALPHA and UPALPHA.
4294    </t>
4295    <t>
4296      Move "Product Tokens" section (back) into Part 1, as "token" is used
4297      in the definition of the Upgrade header.
4298    </t>
4299    <t>
4300      Add explicit references to BNF syntax and rules imported from other parts of the specification.
4301    </t>
4302    <t>
4303      Move definition of "delta-seconds" into Part1 as a common BNF element.
4304    </t>
4305  </list>
4306</t>
4307</section>
4308
4309</section>
4310
4311</back>
4312</rfc>
Note: See TracBrowser for help on using the repository browser.