source: draft-ietf-httpbis/23/draft-ietf-httpbis-p1-messaging-23.xml @ 2568

Last change on this file since 2568 was 2304, checked in by julian.reschke@…, 7 years ago

prepare release of -23

  • Property svn:mime-type set to text/xml
File size: 216.9 KB
Line 
1<?xml version="1.0" encoding="UTF-8"?>
2<!--
3    This XML document is the output of clean-for-DTD.xslt; a tool that strips
4    extensions to RFC2629(bis) from documents for processing with xml2rfc.
5-->
6<?xml-stylesheet type='text/xsl' href='../myxml2rfc.xslt'?>
7<?rfc toc="yes" ?>
8<?rfc symrefs="yes" ?>
9<?rfc sortrefs="yes" ?>
10<?rfc compact="yes"?>
11<?rfc subcompact="no" ?>
12<?rfc linkmailto="no" ?>
13<?rfc editing="no" ?>
14<?rfc comments="yes"?>
15<?rfc inline="yes"?>
16<?rfc rfcedstyle="yes"?>
17<!DOCTYPE rfc
18  PUBLIC "" "rfc2629.dtd">
19<rfc obsoletes="2145,2616" updates="2817,2818" category="std" ipr="pre5378Trust200902" docName="draft-ietf-httpbis-p1-messaging-23">
20
21
22<front>
23
24  <title abbrev="HTTP/1.1 Message Syntax and Routing">Hypertext Transfer Protocol (HTTP/1.1): Message Syntax and Routing</title>
25
26  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
27    <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
28    <address>
29      <postal>
30        <street>345 Park Ave</street>
31        <city>San Jose</city>
32        <region>CA</region>
33        <code>95110</code>
34        <country>USA</country>
35      </postal>
36      <email>fielding@gbiv.com</email>
37      <uri>http://roy.gbiv.com/</uri>
38    </address>
39  </author>
40
41  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
42    <organization abbrev="greenbytes">greenbytes GmbH</organization>
43    <address>
44      <postal>
45        <street>Hafenweg 16</street>
46        <city>Muenster</city><region>NW</region><code>48155</code>
47        <country>Germany</country>
48      </postal>
49      <email>julian.reschke@greenbytes.de</email>
50      <uri>http://greenbytes.de/tech/webdav/</uri>
51    </address>
52  </author>
53
54  <date month="July" year="2013" day="15"/>
55  <workgroup>HTTPbis Working Group</workgroup>
56
57<abstract>
58<t>
59   The Hypertext Transfer Protocol (HTTP) is an application-level protocol for
60   distributed, collaborative, hypertext information systems. HTTP has been in
61   use by the World Wide Web global information initiative since 1990.
62   This document provides an overview of HTTP architecture and its associated
63   terminology, defines the "http" and "https" Uniform Resource Identifier
64   (URI) schemes, defines the HTTP/1.1 message syntax and parsing requirements,
65   and describes general security concerns for implementations.
66</t>   
67</abstract>
68
69<note title="Editorial Note (To be removed by RFC Editor)">
70  <t>
71    Discussion of this draft takes place on the HTTPBIS working group
72    mailing list (ietf-http-wg@w3.org), which is archived at
73    <eref target="http://lists.w3.org/Archives/Public/ietf-http-wg/"/>.
74  </t>
75  <t>
76    The current issues list is at
77    <eref target="http://tools.ietf.org/wg/httpbis/trac/report/3"/> and related
78    documents (including fancy diffs) can be found at
79    <eref target="http://tools.ietf.org/wg/httpbis/"/>.
80  </t>
81  <t>
82    The changes in this draft are summarized in <xref target="changes.since.22"/>.
83  </t>
84</note>
85</front>
86<middle>
87<section title="Introduction" anchor="introduction">
88<t>
89   The Hypertext Transfer Protocol (HTTP) is an application-level
90   request/response protocol that uses extensible semantics and self-descriptive
91   message payloads for flexible interaction with network-based hypertext
92   information systems. This document is the first in a series of documents
93   that collectively form the HTTP/1.1 specification:
94   <list style="empty">
95    <t>RFC xxx1: Message Syntax and Routing</t>
96    <t>RFC xxx2: Semantics and Content</t>
97    <t>RFC xxx3: Conditional Requests</t>
98    <t>RFC xxx4: Range Requests</t>
99    <t>RFC xxx5: Caching</t>
100    <t>RFC xxx6: Authentication</t>
101   </list>
102</t>
103<t>
104   This HTTP/1.1 specification obsoletes and moves to historic status
105   RFC 2616, its predecessor
106   RFC 2068, and
107   RFC 2145 (on HTTP versioning).
108   This specification also updates the use of CONNECT to establish a tunnel,
109   previously defined in RFC 2817,
110   and defines the "https" URI scheme that was described informally in
111   RFC 2818.
112</t>
113<t>
114   HTTP is a generic interface protocol for information systems. It is
115   designed to hide the details of how a service is implemented by presenting
116   a uniform interface to clients that is independent of the types of
117   resources provided. Likewise, servers do not need to be aware of each
118   client's purpose: an HTTP request can be considered in isolation rather
119   than being associated with a specific type of client or a predetermined
120   sequence of application steps. The result is a protocol that can be used
121   effectively in many different contexts and for which implementations can
122   evolve independently over time.
123</t>
124<t>
125   HTTP is also designed for use as an intermediation protocol for translating
126   communication to and from non-HTTP information systems.
127   HTTP proxies and gateways can provide access to alternative information
128   services by translating their diverse protocols into a hypertext
129   format that can be viewed and manipulated by clients in the same way
130   as HTTP services.
131</t>
132<t>
133   One consequence of this flexibility is that the protocol cannot be
134   defined in terms of what occurs behind the interface. Instead, we
135   are limited to defining the syntax of communication, the intent
136   of received communication, and the expected behavior of recipients.
137   If the communication is considered in isolation, then successful
138   actions ought to be reflected in corresponding changes to the
139   observable interface provided by servers. However, since multiple
140   clients might act in parallel and perhaps at cross-purposes, we
141   cannot require that such changes be observable beyond the scope
142   of a single response.
143</t>
144<t>
145   This document describes the architectural elements that are used or
146   referred to in HTTP, defines the "http" and "https" URI schemes,
147   describes overall network operation and connection management,
148   and defines HTTP message framing and forwarding requirements.
149   Our goal is to define all of the mechanisms necessary for HTTP message
150   handling that are independent of message semantics, thereby defining the
151   complete set of requirements for message parsers and
152   message-forwarding intermediaries.
153</t>
154
155
156<section title="Requirement Notation" anchor="intro.requirements">
157<t>
158   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
159   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
160   document are to be interpreted as described in <xref target="RFC2119"/>.
161</t>
162<t>
163   Conformance criteria and considerations regarding error handling
164   are defined in <xref target="conformance"/>.
165</t>
166</section>
167
168<section title="Syntax Notation" anchor="notation">
169<iref primary="true" item="Grammar" subitem="ALPHA"/>
170<iref primary="true" item="Grammar" subitem="CR"/>
171<iref primary="true" item="Grammar" subitem="CRLF"/>
172<iref primary="true" item="Grammar" subitem="CTL"/>
173<iref primary="true" item="Grammar" subitem="DIGIT"/>
174<iref primary="true" item="Grammar" subitem="DQUOTE"/>
175<iref primary="true" item="Grammar" subitem="HEXDIG"/>
176<iref primary="true" item="Grammar" subitem="HTAB"/>
177<iref primary="true" item="Grammar" subitem="LF"/>
178<iref primary="true" item="Grammar" subitem="OCTET"/>
179<iref primary="true" item="Grammar" subitem="SP"/>
180<iref primary="true" item="Grammar" subitem="VCHAR"/>
181<t>
182   This specification uses the Augmented Backus-Naur Form (ABNF) notation
183   of <xref target="RFC5234"/> with the list rule extension defined in
184   <xref target="abnf.extension"/>.  <xref target="collected.abnf"/> shows
185   the collected ABNF with the list rule expanded.
186</t>
187<t anchor="core.rules">
188 
189 
190 
191 
192 
193 
194 
195 
196 
197 
198 
199 
200   The following core rules are included by
201   reference, as defined in <xref target="RFC5234"/>, Appendix B.1:
202   ALPHA (letters), CR (carriage return), CRLF (CR LF), CTL (controls),
203   DIGIT (decimal 0-9), DQUOTE (double quote),
204   HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line feed),
205   OCTET (any 8-bit sequence of data), SP (space), and
206   VCHAR (any visible <xref target="USASCII"/> character).
207</t>
208<t>
209   As a convention, ABNF rule names prefixed with "obs-" denote
210   "obsolete" grammar rules that appear for historical reasons.
211</t>
212</section>
213</section>
214
215<section title="Architecture" anchor="architecture">
216<t>
217   HTTP was created for the World Wide Web architecture
218   and has evolved over time to support the scalability needs of a worldwide
219   hypertext system. Much of that architecture is reflected in the terminology
220   and syntax productions used to define HTTP.
221</t>
222
223<section title="Client/Server Messaging" anchor="operation">
224<iref primary="true" item="client"/>
225<iref primary="true" item="server"/>
226<iref primary="true" item="connection"/>
227<t>
228   HTTP is a stateless request/response protocol that operates by exchanging
229   messages (<xref target="http.message"/>) across a reliable
230   transport or session-layer
231   "connection" (<xref target="connection.management"/>).
232   An HTTP "client" is a program that establishes a connection
233   to a server for the purpose of sending one or more HTTP requests.
234   An HTTP "server" is a program that accepts connections
235   in order to service HTTP requests by sending HTTP responses.
236</t>
237<iref primary="true" item="user agent"/>
238<iref primary="true" item="origin server"/>
239<iref primary="true" item="browser"/>
240<iref primary="true" item="spider"/>
241<iref primary="true" item="sender"/>
242<iref primary="true" item="recipient"/>
243<t>
244   The terms client and server refer only to the roles that
245   these programs perform for a particular connection.  The same program
246   might act as a client on some connections and a server on others.
247   We use the term "user agent" to refer to any of the various
248   client programs that initiate a request, including (but not limited to)
249   browsers, spiders (web-based robots), command-line tools, native
250   applications, and mobile apps.  The term "origin server" is
251   used to refer to the program that can originate authoritative responses to
252   a request. For general requirements, we use the terms
253   "sender" and "recipient" to refer to any
254   component that sends or receives, respectively, a given message.
255</t>
256<t>
257   HTTP relies upon the Uniform Resource Identifier (URI)
258   standard <xref target="RFC3986"/> to indicate the target resource
259   (<xref target="target-resource"/>) and relationships between resources.
260   Messages are passed in a format similar to that used by Internet mail
261   <xref target="RFC5322"/> and the Multipurpose Internet Mail Extensions
262   (MIME) <xref target="RFC2045"/> (see Appendix A of <xref target="Part2"/> for the differences
263   between HTTP and MIME messages).
264</t>
265<t>
266   Most HTTP communication consists of a retrieval request (GET) for
267   a representation of some resource identified by a URI.  In the
268   simplest case, this might be accomplished via a single bidirectional
269   connection (===) between the user agent (UA) and the origin server (O).
270</t>
271<figure><artwork type="drawing"><![CDATA[
272         request   >
273    UA ======================================= O
274                                <   response
275]]></artwork></figure>
276<iref primary="true" item="message"/>
277<iref primary="true" item="request"/>
278<iref primary="true" item="response"/>
279<t>
280   A client sends an HTTP request to a server in the form of a request
281   message, beginning with a request-line that includes a method, URI, and
282   protocol version (<xref target="request.line"/>),
283   followed by header fields containing
284   request modifiers, client information, and representation metadata
285   (<xref target="header.fields"/>),
286   an empty line to indicate the end of the header section, and finally
287   a message body containing the payload body (if any,
288   <xref target="message.body"/>).
289</t>
290<t>
291   A server responds to a client's request by sending one or more HTTP
292   response
293   messages, each beginning with a status line that
294   includes the protocol version, a success or error code, and textual
295   reason phrase (<xref target="status.line"/>),
296   possibly followed by header fields containing server
297   information, resource metadata, and representation metadata
298   (<xref target="header.fields"/>),
299   an empty line to indicate the end of the header section, and finally
300   a message body containing the payload body (if any,
301   <xref target="message.body"/>).
302</t>
303<t>
304   A connection might be used for multiple request/response exchanges,
305   as defined in <xref target="persistent.connections"/>.
306</t>
307<t>
308   The following example illustrates a typical message exchange for a
309   GET request on the URI "http://www.example.com/hello.txt":
310</t>
311<figure><preamble>
312client request:
313</preamble><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
314  GET /hello.txt HTTP/1.1
315  User-Agent: curl/7.16.3 libcurl/7.16.3 OpenSSL/0.9.7l zlib/1.2.3
316  Host: www.example.com
317  Accept-Language: en, mi
318 
319  ]]></artwork></figure>
320<figure><preamble>
321server response:
322</preamble><artwork type="message/http; msgtype=&#34;response&#34;"><![CDATA[
323  HTTP/1.1 200 OK
324  Date: Mon, 27 Jul 2009 12:28:53 GMT
325  Server: Apache
326  Last-Modified: Wed, 22 Jul 2009 19:15:56 GMT
327  ETag: "34aa387-d-1568eb00"
328  Accept-Ranges: bytes
329  Content-Length: 51
330  Vary: Accept-Encoding
331  Content-Type: text/plain
332 
333  Hello World! My payload includes a trailing CRLF.
334  ]]></artwork>
335</figure>
336</section>
337
338<section title="Implementation Diversity" anchor="implementation-diversity">
339<t>
340   When considering the design of HTTP, it is easy to fall into a trap of
341   thinking that all user agents are general-purpose browsers and all origin
342   servers are large public websites. That is not the case in practice.
343   Common HTTP user agents include household appliances, stereos, scales,
344   firmware update scripts, command-line programs, mobile apps,
345   and communication devices in a multitude of shapes and sizes.  Likewise,
346   common HTTP origin servers include home automation units, configurable
347   networking components, office machines, autonomous robots, news feeds,
348   traffic cameras, ad selectors, and video delivery platforms.
349</t>
350<t>
351   The term "user agent" does not imply that there is a human user directly
352   interacting with the software agent at the time of a request. In many
353   cases, a user agent is installed or configured to run in the background
354   and save its results for later inspection (or save only a subset of those
355   results that might be interesting or erroneous). Spiders, for example, are
356   typically given a start URI and configured to follow certain behavior while
357   crawling the Web as a hypertext graph.
358</t>
359<t>
360   The implementation diversity of HTTP means that we cannot assume the
361   user agent can make interactive suggestions to a user or provide adequate
362   warning for security or privacy options.  In the few cases where this
363   specification requires reporting of errors to the user, it is acceptable
364   for such reporting to only be observable in an error console or log file.
365   Likewise, requirements that an automated action be confirmed by the user
366   before proceeding might be met via advance configuration choices,
367   run-time options, or simple avoidance of the unsafe action; confirmation
368   does not imply any specific user interface or interruption of normal
369   processing if the user has already made that choice.
370</t>
371</section>
372
373<section title="Intermediaries" anchor="intermediaries">
374<iref primary="true" item="intermediary"/>
375<t>
376   HTTP enables the use of intermediaries to satisfy requests through
377   a chain of connections.  There are three common forms of HTTP
378   intermediary: proxy, gateway, and tunnel.  In some cases,
379   a single intermediary might act as an origin server, proxy, gateway,
380   or tunnel, switching behavior based on the nature of each request.
381</t>
382<figure><artwork type="drawing"><![CDATA[
383         >             >             >             >
384    UA =========== A =========== B =========== C =========== O
385               <             <             <             <
386]]></artwork></figure>
387<t>
388   The figure above shows three intermediaries (A, B, and C) between the
389   user agent and origin server. A request or response message that
390   travels the whole chain will pass through four separate connections.
391   Some HTTP communication options
392   might apply only to the connection with the nearest, non-tunnel
393   neighbor, only to the end-points of the chain, or to all connections
394   along the chain. Although the diagram is linear, each participant might
395   be engaged in multiple, simultaneous communications. For example, B
396   might be receiving requests from many clients other than A, and/or
397   forwarding requests to servers other than C, at the same time that it
398   is handling A's request. Likewise, later requests might be sent through a
399   different path of connections, often based on dynamic configuration for
400   load balancing.   
401</t>
402<t>
403<iref primary="true" item="upstream"/><iref primary="true" item="downstream"/>
404<iref primary="true" item="inbound"/><iref primary="true" item="outbound"/>
405   We use the terms "upstream" and "downstream"
406   to describe various requirements in relation to the directional flow of a
407   message: all messages flow from upstream to downstream.
408   Likewise, we use the terms inbound and outbound to refer to
409   directions in relation to the request path:
410   "inbound" means toward the origin server and
411   "outbound" means toward the user agent.
412</t>
413<t><iref primary="true" item="proxy"/>
414   A "proxy" is a message forwarding agent that is selected by the
415   client, usually via local configuration rules, to receive requests
416   for some type(s) of absolute URI and attempt to satisfy those
417   requests via translation through the HTTP interface.  Some translations
418   are minimal, such as for proxy requests for "http" URIs, whereas
419   other requests might require translation to and from entirely different
420   application-level protocols. Proxies are often used to group an
421   organization's HTTP requests through a common intermediary for the
422   sake of security, annotation services, or shared caching.
423</t>
424<t>
425<iref primary="true" item="transforming proxy"/>
426<iref primary="true" item="non-transforming proxy"/>
427   An HTTP-to-HTTP proxy is called a "transforming proxy" if it is designed
428   or configured to modify request or response messages in a semantically
429   meaningful way (i.e., modifications, beyond those required by normal
430   HTTP processing, that change the message in a way that would be
431   significant to the original sender or potentially significant to
432   downstream recipients).  For example, a transforming proxy might be
433   acting as a shared annotation server (modifying responses to include
434   references to a local annotation database), a malware filter, a
435   format transcoder, or an intranet-to-Internet privacy filter.  Such
436   transformations are presumed to be desired by the client (or client
437   organization) that selected the proxy and are beyond the scope of
438   this specification.  However, when a proxy is not intended to transform
439   a given message, we use the term "non-transforming proxy" to target
440   requirements that preserve HTTP message semantics. See Section 6.3.4 of <xref target="Part2"/> and
441   Section 7.5 of <xref target="Part6"/> for status and warning codes related to transformations.
442</t>
443<t><iref primary="true" item="gateway"/><iref primary="true" item="reverse proxy"/>
444<iref primary="true" item="accelerator"/>
445   A "gateway" (a.k.a., "reverse proxy") is an
446   intermediary that acts as an origin server for the outbound connection, but
447   translates received requests and forwards them inbound to another server or
448   servers. Gateways are often used to encapsulate legacy or untrusted
449   information services, to improve server performance through
450   "accelerator" caching, and to enable partitioning or load
451   balancing of HTTP services across multiple machines.
452</t>
453<t>
454   All HTTP requirements applicable to an origin server
455   also apply to the outbound communication of a gateway.
456   A gateway communicates with inbound servers using any protocol that
457   it desires, including private extensions to HTTP that are outside
458   the scope of this specification.  However, an HTTP-to-HTTP gateway
459   that wishes to interoperate with third-party HTTP servers ought to conform
460   to user agent requirements on the gateway's inbound connection.
461</t>
462<t><iref primary="true" item="tunnel"/>
463   A "tunnel" acts as a blind relay between two connections
464   without changing the messages. Once active, a tunnel is not
465   considered a party to the HTTP communication, though the tunnel might
466   have been initiated by an HTTP request. A tunnel ceases to exist when
467   both ends of the relayed connection are closed. Tunnels are used to
468   extend a virtual connection through an intermediary, such as when
469   Transport Layer Security (TLS, <xref target="RFC5246"/>) is used to
470   establish confidential communication through a shared firewall proxy.
471</t>
472<t><iref primary="true" item="interception proxy"/>
473<iref primary="true" item="transparent proxy"/>
474<iref primary="true" item="captive portal"/>
475   The above categories for intermediary only consider those acting as
476   participants in the HTTP communication.  There are also intermediaries
477   that can act on lower layers of the network protocol stack, filtering or
478   redirecting HTTP traffic without the knowledge or permission of message
479   senders. Network intermediaries often introduce security flaws or
480   interoperability problems by violating HTTP semantics.  For example, an
481   "interception proxy" <xref target="RFC3040"/> (also commonly
482   known as a "transparent proxy" <xref target="RFC1919"/> or
483   "captive portal")
484   differs from an HTTP proxy because it is not selected by the client.
485   Instead, an interception proxy filters or redirects outgoing TCP port 80
486   packets (and occasionally other common port traffic).
487   Interception proxies are commonly found on public network access points,
488   as a means of enforcing account subscription prior to allowing use of
489   non-local Internet services, and within corporate firewalls to enforce
490   network usage policies.
491   They are indistinguishable from a man-in-the-middle attack.
492</t>
493<t>
494   HTTP is defined as a stateless protocol, meaning that each request message
495   can be understood in isolation.  Many implementations depend on HTTP's
496   stateless design in order to reuse proxied connections or dynamically
497   load-balance requests across multiple servers.  Hence, servers MUST NOT
498   assume that two requests on the same connection are from the same user
499   agent unless the connection is secured and specific to that agent.
500   Some non-standard HTTP extensions (e.g., <xref target="RFC4559"/>) have
501   been known to violate this requirement, resulting in security and
502   interoperability problems.
503</t>
504</section>
505
506<section title="Caches" anchor="caches">
507<iref primary="true" item="cache"/>
508<t>
509   A "cache" is a local store of previous response messages and the
510   subsystem that controls its message storage, retrieval, and deletion.
511   A cache stores cacheable responses in order to reduce the response
512   time and network bandwidth consumption on future, equivalent
513   requests. Any client or server MAY employ a cache, though a cache
514   cannot be used by a server while it is acting as a tunnel.
515</t>
516<t>
517   The effect of a cache is that the request/response chain is shortened
518   if one of the participants along the chain has a cached response
519   applicable to that request. The following illustrates the resulting
520   chain if B has a cached copy of an earlier response from O (via C)
521   for a request that has not been cached by UA or A.
522</t>
523<figure><artwork type="drawing"><![CDATA[
524            >             >
525       UA =========== A =========== B - - - - - - C - - - - - - O
526                  <             <
527]]></artwork></figure>
528<t><iref primary="true" item="cacheable"/>
529   A response is "cacheable" if a cache is allowed to store a copy of
530   the response message for use in answering subsequent requests.
531   Even when a response is cacheable, there might be additional
532   constraints placed by the client or by the origin server on when
533   that cached response can be used for a particular request. HTTP
534   requirements for cache behavior and cacheable responses are
535   defined in Section 2 of <xref target="Part6"/>. 
536</t>
537<t>
538   There are a wide variety of architectures and configurations
539   of caches deployed across the World Wide Web and
540   inside large organizations. These include national hierarchies
541   of proxy caches to save transoceanic bandwidth, collaborative systems that
542   broadcast or multicast cache entries, archives of pre-fetched cache
543   entries for use in off-line or high-latency environments, and so on.
544</t>
545</section>
546
547<section title="Conformance and Error Handling" anchor="conformance">
548<t>
549   This specification targets conformance criteria according to the role of
550   a participant in HTTP communication.  Hence, HTTP requirements are placed
551   on senders, recipients, clients, servers, user agents, intermediaries,
552   origin servers, proxies, gateways, or caches, depending on what behavior
553   is being constrained by the requirement. Additional (social) requirements
554   are placed on implementations, resource owners, and protocol element
555   registrations when they apply beyond the scope of a single communication.
556</t>
557<t>
558   The verb "generate" is used instead of "send" where a requirement
559   differentiates between creating a protocol element and merely forwarding a
560   received element downstream.
561</t>
562<t>
563   An implementation is considered conformant if it complies with all of the
564   requirements associated with the roles it partakes in HTTP.
565</t>
566<t>
567   Conformance applies to both the syntax and semantics of HTTP protocol
568   elements. A sender MUST NOT generate protocol elements that convey a
569   meaning that is known by that sender to be false. A sender MUST NOT
570   generate protocol elements that do not match the grammar defined by the
571   ABNF rules for those protocol elements that are applicable to the sender's
572   role. If a received protocol element is processed, the recipient MUST be
573   able to parse any value that would match the ABNF rules for that protocol
574   element, excluding only those rules not applicable to the recipient's role.
575</t>
576<t>
577   Unless noted otherwise, a recipient MAY attempt to recover a usable
578   protocol element from an invalid construct.  HTTP does not define
579   specific error handling mechanisms except when they have a direct impact
580   on security, since different applications of the protocol require
581   different error handling strategies.  For example, a Web browser might
582   wish to transparently recover from a response where the
583   Location header field doesn't parse according to the ABNF,
584   whereas a systems control client might consider any form of error recovery
585   to be dangerous.
586</t>
587</section>
588
589<section title="Protocol Versioning" anchor="http.version">
590 
591 
592<t>
593   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate
594   versions of the protocol. This specification defines version "1.1".
595   The protocol version as a whole indicates the sender's conformance
596   with the set of requirements laid out in that version's corresponding
597   specification of HTTP.
598</t>
599<t>
600   The version of an HTTP message is indicated by an HTTP-version field
601   in the first line of the message. HTTP-version is case-sensitive.
602</t>
603<figure><iref primary="true" item="Grammar" subitem="HTTP-version"/><iref primary="true" item="Grammar" subitem="HTTP-name"/><artwork type="abnf2616"><![CDATA[
604  HTTP-version  = HTTP-name "/" DIGIT "." DIGIT
605  HTTP-name     = %x48.54.54.50 ; "HTTP", case-sensitive
606]]></artwork></figure>
607<t>
608   The HTTP version number consists of two decimal digits separated by a "."
609   (period or decimal point).  The first digit ("major version") indicates the
610   HTTP messaging syntax, whereas the second digit ("minor version") indicates
611   the highest minor version within that major version to which the sender is
612   conformant and able to understand for future communication.  The minor
613   version advertises the sender's communication capabilities even when the
614   sender is only using a backwards-compatible subset of the protocol,
615   thereby letting the recipient know that more advanced features can
616   be used in response (by servers) or in future requests (by clients).
617</t>
618<t>
619   When an HTTP/1.1 message is sent to an HTTP/1.0 recipient
620   <xref target="RFC1945"/> or a recipient whose version is unknown,
621   the HTTP/1.1 message is constructed such that it can be interpreted
622   as a valid HTTP/1.0 message if all of the newer features are ignored.
623   This specification places recipient-version requirements on some
624   new features so that a conformant sender will only use compatible
625   features until it has determined, through configuration or the
626   receipt of a message, that the recipient supports HTTP/1.1.
627</t>
628<t>
629   The interpretation of a header field does not change between minor
630   versions of the same major HTTP version, though the default
631   behavior of a recipient in the absence of such a field can change.
632   Unless specified otherwise, header fields defined in HTTP/1.1 are
633   defined for all versions of HTTP/1.x.  In particular, the <xref target="header.host" format="none">Host</xref>
634   and <xref target="header.connection" format="none">Connection</xref> header fields ought to be implemented by all
635   HTTP/1.x implementations whether or not they advertise conformance with
636   HTTP/1.1.
637</t>
638<t>
639   New header fields can be defined such that, when they are
640   understood by a recipient, they might override or enhance the
641   interpretation of previously defined header fields.  When an
642   implementation receives an unrecognized header field, the recipient
643   MUST ignore that header field for local processing regardless of
644   the message's HTTP version.  An unrecognized header field received
645   by a proxy MUST be forwarded downstream unless the header field's
646   field-name is listed in the message's <xref target="header.connection" format="none">Connection</xref> header field
647   (see <xref target="header.connection"/>).
648   These requirements allow HTTP's functionality to be enhanced without
649   requiring prior update of deployed intermediaries.
650</t>
651<t>
652   Intermediaries that process HTTP messages (i.e., all intermediaries
653   other than those acting as tunnels) MUST send their own HTTP-version
654   in forwarded messages.  In other words, they MUST NOT blindly
655   forward the first line of an HTTP message without ensuring that the
656   protocol version in that message matches a version to which that
657   intermediary is conformant for both the receiving and
658   sending of messages.  Forwarding an HTTP message without rewriting
659   the HTTP-version might result in communication errors when downstream
660   recipients use the message sender's version to determine what features
661   are safe to use for later communication with that sender.
662</t>
663<t>
664   A client SHOULD send a request version equal to the highest
665   version to which the client is conformant and
666   whose major version is no higher than the highest version supported
667   by the server, if this is known.  A client MUST NOT send a
668   version to which it is not conformant.
669</t>
670<t>
671   A client MAY send a lower request version if it is known that
672   the server incorrectly implements the HTTP specification, but only
673   after the client has attempted at least one normal request and determined
674   from the response status or header fields (e.g., Server) that
675   the server improperly handles higher request versions.
676</t>
677<t>
678   A server SHOULD send a response version equal to the highest
679   version to which the server is conformant and
680   whose major version is less than or equal to the one received in the
681   request.  A server MUST NOT send a version to which it is not
682   conformant.  A server MAY send a 505 (HTTP Version Not
683   Supported) response if it cannot send a response using the
684   major version used in the client's request.
685</t>
686<t>
687   A server MAY send an HTTP/1.0 response to a request
688   if it is known or suspected that the client incorrectly implements the
689   HTTP specification and is incapable of correctly processing later
690   version responses, such as when a client fails to parse the version
691   number correctly or when an intermediary is known to blindly forward
692   the HTTP-version even when it doesn't conform to the given minor
693   version of the protocol. Such protocol downgrades SHOULD NOT be
694   performed unless triggered by specific client attributes, such as when
695   one or more of the request header fields (e.g., User-Agent)
696   uniquely match the values sent by a client known to be in error.
697</t>
698<t>
699   The intention of HTTP's versioning design is that the major number
700   will only be incremented if an incompatible message syntax is
701   introduced, and that the minor number will only be incremented when
702   changes made to the protocol have the effect of adding to the message
703   semantics or implying additional capabilities of the sender.  However,
704   the minor version was not incremented for the changes introduced between
705   <xref target="RFC2068"/> and <xref target="RFC2616"/>, and this revision
706   has specifically avoided any such changes to the protocol.
707</t>
708<t>
709   When an HTTP message is received with a major version number that the
710   recipient implements, but a higher minor version number than what the
711   recipient implements, the recipient SHOULD process the message as if it
712   were in the highest minor version within that major version to which the
713   recipient is conformant. A recipient can assume that a message with a
714   higher minor version, when sent to a recipient that has not yet indicated
715   support for that higher version, is sufficiently backwards-compatible to be
716   safely processed by any implementation of the same major version.
717</t>
718</section>
719
720<section title="Uniform Resource Identifiers" anchor="uri">
721<iref primary="true" item="resource"/>
722<t>
723   Uniform Resource Identifiers (URIs) <xref target="RFC3986"/> are used
724   throughout HTTP as the means for identifying resources (Section 2 of <xref target="Part2"/>).
725   URI references are used to target requests, indicate redirects, and define
726   relationships.
727</t>
728 
729 
730 
731 
732 
733 
734 
735 
736 
737 
738 
739 
740<t>
741   This specification adopts the definitions of "URI-reference",
742   "absolute-URI", "relative-part", "authority", "port", "host",
743   "path-abempty", "segment", "query", and "fragment" from the
744   URI generic syntax.
745   In addition, we define an "absolute-path" rule (that differs from
746   RFC 3986's "path-absolute" in that it allows a leading "//")
747   and a "partial-URI" rule for protocol elements
748   that allow a relative URI but not a fragment.
749</t>
750<figure><iref primary="true" item="Grammar" subitem="URI-reference"><!--exported production--></iref><iref primary="true" item="Grammar" subitem="absolute-URI"/><iref primary="true" item="Grammar" subitem="authority"/><iref primary="true" item="Grammar" subitem="absolute-path"/><iref primary="true" item="Grammar" subitem="port"/><iref primary="true" item="Grammar" subitem="query"/><iref primary="true" item="Grammar" subitem="fragment"/><iref primary="true" item="Grammar" subitem="segment"/><iref primary="true" item="Grammar" subitem="uri-host"/><iref primary="true" item="Grammar" subitem="partial-URI"><!--exported production--></iref><artwork type="abnf2616"><![CDATA[
751  URI-reference = <URI-reference, defined in [RFC3986], Section 4.1>
752  absolute-URI  = <absolute-URI, defined in [RFC3986], Section 4.3>
753  relative-part = <relative-part, defined in [RFC3986], Section 4.2>
754  authority     = <authority, defined in [RFC3986], Section 3.2>
755  uri-host      = <host, defined in [RFC3986], Section 3.2.2>
756  port          = <port, defined in [RFC3986], Section 3.2.3>
757  path-abempty  = <path-abempty, defined in [RFC3986], Section 3.3>
758  segment       = <segment, defined in [RFC3986], Section 3.3>
759  query         = <query, defined in [RFC3986], Section 3.4>
760  fragment      = <fragment, defined in [RFC3986], Section 3.5>
761 
762  absolute-path = 1*( "/" segment )
763  partial-URI   = relative-part [ "?" query ]
764]]></artwork></figure>
765<t>
766   Each protocol element in HTTP that allows a URI reference will indicate
767   in its ABNF production whether the element allows any form of reference
768   (URI-reference), only a URI in absolute form (absolute-URI), only the
769   path and optional query components, or some combination of the above.
770   Unless otherwise indicated, URI references are parsed
771   relative to the effective request URI
772   (<xref target="effective.request.uri"/>).
773</t>
774
775<section title="http URI scheme" anchor="http.uri">
776 
777  <iref item="http URI scheme" primary="true"/>
778  <iref item="URI scheme" subitem="http" primary="true"/>
779<t>
780   The "http" URI scheme is hereby defined for the purpose of minting
781   identifiers according to their association with the hierarchical
782   namespace governed by a potential HTTP origin server listening for
783   TCP (<xref target="RFC0793"/>) connections on a given port.
784</t>
785<figure><iref primary="true" item="Grammar" subitem="http-URI"><!--terminal production--></iref><artwork type="abnf2616"><![CDATA[
786  http-URI = "http:" "//" authority path-abempty [ "?" query ]
787             [ "#" fragment ]
788]]></artwork></figure>
789<t>
790   The HTTP origin server is identified by the generic syntax's
791   <xref target="uri" format="none">authority</xref> component, which includes a host identifier
792   and optional TCP port (<xref target="RFC3986"/>, Section 3.2.2).
793   The remainder of the URI, consisting of both the hierarchical path
794   component and optional query component, serves as an identifier for
795   a potential resource within that origin server's name space.
796</t>
797<t>
798   If the host identifier is provided as an IP address,
799   then the origin server is any listener on the indicated TCP port at
800   that IP address. If host is a registered name, then that name is
801   considered an indirect identifier and the recipient might use a name
802   resolution service, such as DNS, to find the address of a listener
803   for that host.
804   The host MUST NOT be empty; if an "http" URI is received with an
805   empty host, then it MUST be rejected as invalid.
806   If the port subcomponent is empty or not given, then TCP port 80 is
807   assumed (the default reserved port for WWW services).
808</t>
809<t>
810   Regardless of the form of host identifier, access to that host is not
811   implied by the mere presence of its name or address. The host might or might
812   not exist and, even when it does exist, might or might not be running an
813   HTTP server or listening to the indicated port. The "http" URI scheme
814   makes use of the delegated nature of Internet names and addresses to
815   establish a naming authority (whatever entity has the ability to place
816   an HTTP server at that Internet name or address) and allows that
817   authority to determine which names are valid and how they might be used.
818</t>
819<t>
820   When an "http" URI is used within a context that calls for access to the
821   indicated resource, a client MAY attempt access by resolving
822   the host to an IP address, establishing a TCP connection to that address
823   on the indicated port, and sending an HTTP request message
824   (<xref target="http.message"/>) containing the URI's identifying data
825   (<xref target="message.routing"/>) to the server.
826   If the server responds to that request with a non-interim HTTP response
827   message, as described in Section 6 of <xref target="Part2"/>, then that response
828   is considered an authoritative answer to the client's request.
829</t>
830<t>
831   Although HTTP is independent of the transport protocol, the "http"
832   scheme is specific to TCP-based services because the name delegation
833   process depends on TCP for establishing authority.
834   An HTTP service based on some other underlying connection protocol
835   would presumably be identified using a different URI scheme, just as
836   the "https" scheme (below) is used for resources that require an
837   end-to-end secured connection. Other protocols might also be used to
838   provide access to "http" identified resources — it is only the
839   authoritative interface that is specific to TCP.
840</t>
841<t>
842   The URI generic syntax for authority also includes a deprecated
843   userinfo subcomponent (<xref target="RFC3986"/>, Section 3.2.1)
844   for including user authentication information in the URI.  Some
845   implementations make use of the userinfo component for internal
846   configuration of authentication information, such as within command
847   invocation options, configuration files, or bookmark lists, even
848   though such usage might expose a user identifier or password.
849   Senders MUST exclude the userinfo subcomponent (and its "@"
850   delimiter) when an "http" URI is transmitted within a message as a
851   request target or header field value.
852   Recipients of an "http" URI reference SHOULD parse for userinfo and
853   treat its presence as an error, since it is likely being used to obscure
854   the authority for the sake of phishing attacks.
855</t>
856</section>
857
858<section title="https URI scheme" anchor="https.uri">
859   
860   <iref item="https URI scheme"/>
861   <iref item="URI scheme" subitem="https"/>
862<t>
863   The "https" URI scheme is hereby defined for the purpose of minting
864   identifiers according to their association with the hierarchical
865   namespace governed by a potential HTTP origin server listening to a
866   given TCP port for TLS-secured connections
867   (<xref target="RFC0793"/>, <xref target="RFC5246"/>).
868</t>
869<t>
870   All of the requirements listed above for the "http" scheme are also
871   requirements for the "https" scheme, except that a default TCP port
872   of 443 is assumed if the port subcomponent is empty or not given,
873   and the TCP connection MUST be secured, end-to-end, through the
874   use of strong encryption prior to sending the first HTTP request.
875</t>
876<figure><iref primary="true" item="Grammar" subitem="https-URI"><!--terminal production--></iref><artwork type="abnf2616"><![CDATA[
877  https-URI = "https:" "//" authority path-abempty [ "?" query ]
878              [ "#" fragment ]
879]]></artwork></figure>
880<t>
881   Note that the "https" URI scheme depends on both TLS and TCP for
882   establishing authority.
883   Resources made available via the "https" scheme have no shared
884   identity with the "http" scheme even if their resource identifiers
885   indicate the same authority (the same host listening to the same
886   TCP port).  They are distinct name spaces and are considered to be
887   distinct origin servers.  However, an extension to HTTP that is
888   defined to apply to entire host domains, such as the Cookie protocol
889   <xref target="RFC6265"/>, can allow information
890   set by one service to impact communication with other services
891   within a matching group of host domains.
892</t>
893<t>
894   The process for authoritative access to an "https" identified
895   resource is defined in <xref target="RFC2818"/>.
896</t>
897</section>
898
899<section title="http and https URI Normalization and Comparison" anchor="uri.comparison">
900<t>
901   Since the "http" and "https" schemes conform to the URI generic syntax,
902   such URIs are normalized and compared according to the algorithm defined
903   in <xref target="RFC3986"/>, Section 6, using the defaults
904   described above for each scheme.
905</t>
906<t>
907   If the port is equal to the default port for a scheme, the normal form is
908   to omit the port subcomponent. When not being used in absolute form as the
909   request target of an OPTIONS request, an empty path component is equivalent
910   to an absolute path of "/", so the normal form is to provide a path of "/"
911   instead. The scheme and host are case-insensitive and normally provided in
912   lowercase; all other components are compared in a case-sensitive manner.
913   Characters other than those in the "reserved" set are equivalent to their
914   percent-encoded octets (see <xref target="RFC3986"/>, Section 2.1): the normal form is to not encode them.
915</t>
916<t>
917   For example, the following three URIs are equivalent:
918</t>
919<figure><artwork type="example"><![CDATA[
920   http://example.com:80/~smith/home.html
921   http://EXAMPLE.com/%7Esmith/home.html
922   http://EXAMPLE.com:/%7esmith/home.html
923]]></artwork></figure>
924</section>
925</section>
926</section>
927
928<section title="Message Format" anchor="http.message">
929
930
931
932
933<iref item="header section"/>
934<iref item="headers"/>
935<iref item="header field"/>
936<t>
937   All HTTP/1.1 messages consist of a start-line followed by a sequence of
938   octets in a format similar to the Internet Message Format
939   <xref target="RFC5322"/>: zero or more header fields (collectively
940   referred to as the "headers" or the "header section"), an empty line
941   indicating the end of the header section, and an optional message body.
942</t>
943<figure><iref primary="true" item="Grammar" subitem="HTTP-message"><!--terminal production--></iref><artwork type="abnf2616"><![CDATA[
944  HTTP-message   = start-line
945                   *( header-field CRLF )
946                   CRLF
947                   [ message-body ]
948]]></artwork></figure>
949<t>
950   The normal procedure for parsing an HTTP message is to read the
951   start-line into a structure, read each header field into a hash
952   table by field name until the empty line, and then use the parsed
953   data to determine if a message body is expected.  If a message body
954   has been indicated, then it is read as a stream until an amount
955   of octets equal to the message body length is read or the connection
956   is closed.
957</t>
958<t>
959   Recipients MUST parse an HTTP message as a sequence of octets in an
960   encoding that is a superset of US-ASCII <xref target="USASCII"/>.
961   Parsing an HTTP message as a stream of Unicode characters, without regard
962   for the specific encoding, creates security vulnerabilities due to the
963   varying ways that string processing libraries handle invalid multibyte
964   character sequences that contain the octet LF (%x0A).  String-based
965   parsers can only be safely used within protocol elements after the element
966   has been extracted from the message, such as within a header field-value
967   after message parsing has delineated the individual fields.
968</t>
969<t>
970   An HTTP message can be parsed as a stream for incremental processing or
971   forwarding downstream.  However, recipients cannot rely on incremental
972   delivery of partial messages, since some implementations will buffer or
973   delay message forwarding for the sake of network efficiency, security
974   checks, or payload transformations.
975</t>
976<t>
977   A sender MUST NOT send whitespace between the start-line and
978   the first header field.
979   A recipient that receives whitespace between the start-line and
980   the first header field MUST either reject the message as invalid or
981   consume each whitespace-preceded line without further processing of it
982   (i.e., ignore the entire line, along with any subsequent lines preceded
983   by whitespace, until a properly formed header field is received or the
984   header block is terminated).
985</t>
986<t>
987   The presence of such whitespace in a request
988   might be an attempt to trick a server into ignoring that field or
989   processing the line after it as a new request, either of which might
990   result in a security vulnerability if other implementations within
991   the request chain interpret the same message differently.
992   Likewise, the presence of such whitespace in a response might be
993   ignored by some clients or cause others to cease parsing.
994</t>
995
996<section title="Start Line" anchor="start.line">
997 
998<t>
999   An HTTP message can either be a request from client to server or a
1000   response from server to client.  Syntactically, the two types of message
1001   differ only in the start-line, which is either a request-line (for requests)
1002   or a status-line (for responses), and in the algorithm for determining
1003   the length of the message body (<xref target="message.body"/>).
1004</t>
1005<t>
1006   In theory, a client could receive requests and a server could receive
1007   responses, distinguishing them by their different start-line formats,
1008   but in practice servers are implemented to only expect a request
1009   (a response is interpreted as an unknown or invalid request method)
1010   and clients are implemented to only expect a response.
1011</t>
1012<figure><iref primary="true" item="Grammar" subitem="start-line"/><artwork type="abnf2616"><![CDATA[
1013  start-line     = request-line / status-line
1014]]></artwork></figure>
1015
1016<section title="Request Line" anchor="request.line">
1017 
1018 
1019<t>
1020   A request-line begins with a method token, followed by a single
1021   space (SP), the request-target, another single space (SP), the
1022   protocol version, and ending with CRLF.
1023</t>
1024<figure><iref primary="true" item="Grammar" subitem="request-line"/><artwork type="abnf2616"><![CDATA[
1025  request-line   = method SP request-target SP HTTP-version CRLF
1026]]></artwork></figure>
1027<iref primary="true" item="method"/>
1028<t anchor="method">
1029   The method token indicates the request method to be performed on the
1030   target resource. The request method is case-sensitive.
1031</t>
1032<figure><iref primary="true" item="Grammar" subitem="method"/><artwork type="abnf2616"><![CDATA[
1033  method         = token
1034]]></artwork></figure>
1035<t>
1036   The methods defined by this specification can be found in
1037   Section 4 of <xref target="Part2"/>, along with information regarding the HTTP method registry
1038   and considerations for defining new methods.
1039</t>
1040<iref item="request-target"/>
1041<t>
1042   The request-target identifies the target resource upon which to apply
1043   the request, as defined in <xref target="request-target"/>.
1044</t>
1045<t>
1046   Recipients typically parse the request-line into its component parts by
1047   splitting on whitespace (see <xref target="message.robustness"/>), since
1048   no whitespace is allowed in the three components.
1049   Unfortunately, some user agents fail to properly encode or exclude
1050   whitespace found in hypertext references, resulting in those disallowed
1051   characters being sent in a request-target.
1052</t>
1053<t>
1054   Recipients of an invalid request-line SHOULD respond with either a
1055   400 (Bad Request) error or a 301 (Moved Permanently)
1056   redirect with the request-target properly encoded.  Recipients SHOULD NOT
1057   attempt to autocorrect and then process the request without a redirect,
1058   since the invalid request-line might be deliberately crafted to bypass
1059   security filters along the request chain.
1060</t>
1061<t>
1062   HTTP does not place a pre-defined limit on the length of a request-line.
1063   A server that receives a method longer than any that it implements
1064   SHOULD respond with a 501 (Not Implemented) status code.
1065   A server MUST be prepared to receive URIs of unbounded length and
1066   respond with the 414 (URI Too Long) status code if the received
1067   request-target would be longer than the server wishes to handle
1068   (see Section 6.5.12 of <xref target="Part2"/>).
1069</t>
1070<t>
1071   Various ad-hoc limitations on request-line length are found in practice.
1072   It is RECOMMENDED that all HTTP senders and recipients support, at a
1073   minimum, request-line lengths of 8000 octets.
1074</t>
1075</section>
1076
1077<section title="Status Line" anchor="status.line">
1078 
1079 
1080 
1081 
1082<t>
1083   The first line of a response message is the status-line, consisting
1084   of the protocol version, a space (SP), the status code, another space,
1085   a possibly-empty textual phrase describing the status code, and
1086   ending with CRLF.
1087</t>
1088<figure><iref primary="true" item="Grammar" subitem="status-line"/><artwork type="abnf2616"><![CDATA[
1089  status-line = HTTP-version SP status-code SP reason-phrase CRLF
1090]]></artwork></figure>
1091<t>
1092   The status-code element is a 3-digit integer code describing the
1093   result of the server's attempt to understand and satisfy the client's
1094   corresponding request. The rest of the response message is to be
1095   interpreted in light of the semantics defined for that status code.
1096   See Section 6 of <xref target="Part2"/> for information about the semantics of status codes,
1097   including the classes of status code (indicated by the first digit),
1098   the status codes defined by this specification, considerations for the
1099   definition of new status codes, and the IANA registry.
1100</t>
1101<figure><iref primary="true" item="Grammar" subitem="status-code"/><artwork type="abnf2616"><![CDATA[
1102  status-code    = 3DIGIT
1103]]></artwork></figure>
1104<t>  
1105   The reason-phrase element exists for the sole purpose of providing a
1106   textual description associated with the numeric status code, mostly
1107   out of deference to earlier Internet application protocols that were more
1108   frequently used with interactive text clients. A client SHOULD ignore
1109   the reason-phrase content.
1110</t>
1111<figure><iref primary="true" item="Grammar" subitem="reason-phrase"/><artwork type="abnf2616"><![CDATA[
1112  reason-phrase  = *( HTAB / SP / VCHAR / obs-text )
1113]]></artwork></figure>
1114</section>
1115</section>
1116
1117<section title="Header Fields" anchor="header.fields">
1118 
1119 
1120 
1121 
1122 
1123<t>
1124   Each HTTP header field consists of a case-insensitive field name
1125   followed by a colon (":"), optional leading whitespace, the field value,
1126   and optional trailing whitespace.
1127</t>
1128<figure><iref primary="true" item="Grammar" subitem="header-field"/><iref primary="true" item="Grammar" subitem="field-name"/><iref primary="true" item="Grammar" subitem="field-value"/><iref primary="true" item="Grammar" subitem="field-content"/><iref primary="true" item="Grammar" subitem="obs-fold"/><artwork type="abnf2616"><![CDATA[
1129  header-field   = field-name ":" OWS field-value OWS
1130  field-name     = token
1131  field-value    = *( field-content / obs-fold )
1132  field-content  = *( HTAB / SP / VCHAR / obs-text )
1133  obs-fold       = CRLF ( SP / HTAB )
1134                 ; obsolete line folding
1135                 ; see Section 3.2.4
1136]]></artwork></figure>
1137<t>
1138   The field-name token labels the corresponding field-value as having the
1139   semantics defined by that header field.  For example, the Date
1140   header field is defined in Section 7.1.1.2 of <xref target="Part2"/> as containing the origination
1141   timestamp for the message in which it appears.
1142</t>
1143
1144<section title="Field Extensibility" anchor="field.extensibility">
1145<t>
1146   HTTP header fields are fully extensible: there is no limit on the
1147   introduction of new field names, each presumably defining new semantics,
1148   nor on the number of header fields used in a given message.  Existing
1149   fields are defined in each part of this specification and in many other
1150   specifications outside the core standard.
1151   New header fields can be introduced without changing the protocol version
1152   if their defined semantics allow them to be safely ignored by recipients
1153   that do not recognize them.
1154</t>
1155<t>
1156   New HTTP header fields ought to be registered with IANA in the
1157   Message Header Field Registry, as described in Section 8.3 of <xref target="Part2"/>.
1158   A proxy MUST forward unrecognized header fields unless the
1159   field-name is listed in the <xref target="header.connection" format="none">Connection</xref> header field
1160   (<xref target="header.connection"/>) or the proxy is specifically
1161   configured to block, or otherwise transform, such fields.
1162   Other recipients SHOULD ignore unrecognized header fields.
1163</t>
1164</section>
1165
1166<section title="Field Order" anchor="field.order">
1167<t>
1168   The order in which header fields with differing field names are
1169   received is not significant. However, it is "good practice" to send
1170   header fields that contain control data first, such as <xref target="header.host" format="none">Host</xref>
1171   on requests and Date on responses, so that implementations
1172   can decide when not to handle a message as early as possible.  A server
1173   MUST wait until the entire header section is received before interpreting
1174   a request message, since later header fields might include conditionals,
1175   authentication credentials, or deliberately misleading duplicate
1176   header fields that would impact request processing.
1177</t>
1178<t>
1179   A sender MUST NOT generate multiple header fields with the same field
1180   name in a message unless either the entire field value for that
1181   header field is defined as a comma-separated list [i.e., #(values)]
1182   or the header field is a well-known exception (as noted below).
1183</t>
1184<t>
1185   Multiple header fields with the same field name can be combined into
1186   one "field-name: field-value" pair, without changing the semantics of the
1187   message, by appending each subsequent field value to the combined
1188   field value in order, separated by a comma. The order in which
1189   header fields with the same field name are received is therefore
1190   significant to the interpretation of the combined field value;
1191   a proxy MUST NOT change the order of these field values when
1192   forwarding a message.
1193</t>
1194<t><list>
1195  <t>
1196   Note: In practice, the "Set-Cookie" header field (<xref target="RFC6265"/>)
1197   often appears multiple times in a response message and does not use the
1198   list syntax, violating the above requirements on multiple header fields
1199   with the same name. Since it cannot be combined into a single field-value,
1200   recipients ought to handle "Set-Cookie" as a special case while processing
1201   header fields. (See Appendix A.2.3 of <xref target="Kri2001"/> for details.)
1202  </t>
1203</list></t>
1204</section>
1205
1206<section title="Whitespace" anchor="whitespace">
1207<t anchor="rule.LWS">
1208   This specification uses three rules to denote the use of linear
1209   whitespace: OWS (optional whitespace), RWS (required whitespace), and
1210   BWS ("bad" whitespace).
1211</t>
1212<t anchor="rule.OWS">
1213   The OWS rule is used where zero or more linear whitespace octets might
1214   appear. For protocol elements where optional whitespace is preferred to
1215   improve readability, a sender SHOULD generate the optional whitespace
1216   as a single SP; otherwise, a sender SHOULD NOT generate optional
1217   whitespace except as needed to white-out invalid or unwanted protocol
1218   elements during in-place message filtering.
1219</t>
1220<t anchor="rule.RWS">
1221   The RWS rule is used when at least one linear whitespace octet is required
1222   to separate field tokens. A sender SHOULD generate RWS as a single SP.
1223</t>
1224<t anchor="rule.BWS">
1225   The BWS rule is used where the grammar allows optional whitespace only for
1226   historical reasons. A sender MUST NOT generate BWS in messages.
1227   A recipient MUST parse for such bad whitespace and remove it before
1228   interpreting the protocol element.
1229</t>
1230<t anchor="rule.whitespace">
1231 
1232 
1233 
1234</t>
1235<figure><iref primary="true" item="Grammar" subitem="OWS"/><iref primary="true" item="Grammar" subitem="RWS"/><iref primary="true" item="Grammar" subitem="BWS"/><artwork type="abnf2616"><![CDATA[
1236  OWS            = *( SP / HTAB )
1237                 ; optional whitespace
1238  RWS            = 1*( SP / HTAB )
1239                 ; required whitespace
1240  BWS            = OWS
1241                 ; "bad" whitespace
1242]]></artwork></figure>
1243</section>
1244
1245<section title="Field Parsing" anchor="field.parsing">
1246<t>
1247   No whitespace is allowed between the header field-name and colon.
1248   In the past, differences in the handling of such whitespace have led to
1249   security vulnerabilities in request routing and response handling.
1250   A server MUST reject any received request message that contains
1251   whitespace between a header field-name and colon with a response code of
1252   400 (Bad Request). A proxy MUST remove any such whitespace
1253   from a response message before forwarding the message downstream.
1254</t>
1255<t>
1256   A field value is preceded by optional whitespace (OWS); a single SP is
1257   preferred. The field value does not include any leading or trailing white
1258   space: OWS occurring before the first non-whitespace octet of the field
1259   value or after the last non-whitespace octet of the field value ought to be
1260   excluded by parsers when extracting the field value from a header field.
1261</t>
1262<t>
1263   A recipient of field-content containing multiple sequential octets of
1264   optional (OWS) or required (RWS) whitespace SHOULD either replace the
1265   sequence with a single SP or transform any non-SP octets in the sequence to
1266   SP octets before interpreting the field value or forwarding the message
1267   downstream.
1268</t>
1269<t>
1270   Historically, HTTP header field values could be extended over multiple
1271   lines by preceding each extra line with at least one space or horizontal
1272   tab (obs-fold). This specification deprecates such line folding except
1273   within the message/http media type
1274   (<xref target="internet.media.type.message.http"/>).
1275   Senders MUST NOT generate messages that include line folding
1276   (i.e., that contain any field-value that contains a match to the
1277   <xref target="header.fields" format="none">obs-fold</xref> rule) unless the message is intended for packaging
1278   within the message/http media type.
1279</t>
1280<t>
1281   A server that receives an <xref target="header.fields" format="none">obs-fold</xref> in a request message that
1282   is not within a message/http container MUST either reject the message by
1283   sending a 400 (Bad Request), preferably with a
1284   representation explaining that obsolete line folding is unacceptable, or
1285   replace each received <xref target="header.fields" format="none">obs-fold</xref> with one or more
1286   <xref target="core.rules" format="none">SP</xref> octets prior to interpreting the field value or
1287   forwarding the message downstream.
1288</t>
1289<t>
1290   A proxy or gateway that receives an <xref target="header.fields" format="none">obs-fold</xref> in a response
1291   message that is not within a message/http container MUST either discard
1292   the message and replace it with a 502 (Bad Gateway)
1293   response, preferably with a representation explaining that unacceptable
1294   line folding was received, or replace each received <xref target="header.fields" format="none">obs-fold</xref>
1295   with one or more <xref target="core.rules" format="none">SP</xref> octets prior to interpreting the field
1296   value or forwarding the message downstream.
1297</t>
1298<t>
1299   A user agent that receives an <xref target="header.fields" format="none">obs-fold</xref> in a response message
1300   that is not within a message/http container MUST replace each received
1301   <xref target="header.fields" format="none">obs-fold</xref> with one or more <xref target="core.rules" format="none">SP</xref> octets prior to
1302   interpreting the field value.
1303</t>
1304<t>
1305   Historically, HTTP has allowed field content with text in the ISO-8859-1
1306   <xref target="ISO-8859-1"/> charset, supporting other charsets only
1307   through use of <xref target="RFC2047"/> encoding.
1308   In practice, most HTTP header field values use only a subset of the
1309   US-ASCII charset <xref target="USASCII"/>. Newly defined
1310   header fields SHOULD limit their field values to US-ASCII octets.
1311   Recipients SHOULD treat other octets in field content (obs-text) as
1312   opaque data.
1313</t>
1314</section>
1315
1316<section title="Field Limits" anchor="field.limits">
1317<t>
1318   HTTP does not place a pre-defined limit on the length of each header field
1319   or on the length of the header block as a whole.  Various ad-hoc
1320   limitations on individual header field length are found in practice,
1321   often depending on the specific field semantics.
1322</t>
1323<t>
1324   A server MUST be prepared to receive request header fields of unbounded
1325   length and respond with an appropriate 4xx (Client Error)
1326   status code if the received header field(s) are larger than the server
1327   wishes to process.
1328</t>
1329<t>
1330   A client MUST be prepared to receive response header fields of unbounded
1331   length. A client MAY discard or truncate received header fields that are
1332   larger than the client wishes to process if the field semantics are such
1333   that the dropped value(s) can be safely ignored without changing the
1334   response semantics.
1335</t>
1336</section>
1337
1338<section title="Field value components" anchor="field.components">
1339<t anchor="rule.token.separators">
1340 
1341 
1342 
1343 
1344   Many HTTP header field values consist of words (token or quoted-string)
1345   separated by whitespace or special characters. These special characters
1346   MUST be in a quoted string to be used within a parameter value (as defined
1347   in <xref target="transfer.codings"/>).
1348</t>
1349<figure><iref primary="true" item="Grammar" subitem="word"/><iref primary="true" item="Grammar" subitem="token"/><iref primary="true" item="Grammar" subitem="tchar"/><iref primary="true" item="Grammar" subitem="special"><!--unused production--></iref><artwork type="abnf2616"><![CDATA[
1350  word           = token / quoted-string
1351
1352  token          = 1*tchar
1353
1354  tchar          = "!" / "#" / "$" / "%" / "&" / "'" / "*"
1355                 / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
1356                 / DIGIT / ALPHA
1357                 ; any VCHAR, except special
1358
1359  special        = "(" / ")" / "<" / ">" / "@" / ","
1360                 / ";" / ":" / "\" / DQUOTE / "/" / "["
1361                 / "]" / "?" / "=" / "{" / "}"
1362]]></artwork></figure>
1363<t anchor="rule.quoted-string">
1364 
1365 
1366 
1367   A string of text is parsed as a single word if it is quoted using
1368   double-quote marks.
1369</t>
1370<figure><iref primary="true" item="Grammar" subitem="quoted-string"/><iref primary="true" item="Grammar" subitem="qdtext"/><iref primary="true" item="Grammar" subitem="obs-text"/><artwork type="abnf2616"><![CDATA[
1371  quoted-string  = DQUOTE *( qdtext / quoted-pair ) DQUOTE
1372  qdtext         = HTAB / SP /%x21 / %x23-5B / %x5D-7E / obs-text
1373  obs-text       = %x80-FF
1374]]></artwork></figure>
1375<t anchor="rule.quoted-pair">
1376 
1377   The backslash octet ("\") can be used as a single-octet
1378   quoting mechanism within quoted-string constructs:
1379</t>
1380<figure><iref primary="true" item="Grammar" subitem="quoted-pair"/><artwork type="abnf2616"><![CDATA[
1381  quoted-pair    = "\" ( HTAB / SP / VCHAR / obs-text )
1382]]></artwork></figure>
1383<t>
1384   Recipients that process the value of a quoted-string MUST handle a
1385   quoted-pair as if it were replaced by the octet following the backslash.
1386</t>
1387<t>
1388   Senders SHOULD NOT generate a quoted-pair in a quoted-string except where
1389   necessary to quote DQUOTE and backslash octets occurring within that string.
1390</t>
1391<t anchor="rule.comment">
1392 
1393 
1394   Comments can be included in some HTTP header fields by surrounding
1395   the comment text with parentheses. Comments are only allowed in
1396   fields containing "comment" as part of their field value definition.
1397</t>
1398<figure><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/><artwork type="abnf2616"><![CDATA[
1399  comment        = "(" *( ctext / quoted-cpair / comment ) ")"
1400  ctext          = HTAB / SP / %x21-27 / %x2A-5B / %x5D-7E / obs-text
1401]]></artwork></figure>
1402<t anchor="rule.quoted-cpair">
1403 
1404   The backslash octet ("\") can be used as a single-octet
1405   quoting mechanism within comment constructs:
1406</t>
1407<figure><iref primary="true" item="Grammar" subitem="quoted-cpair"/><artwork type="abnf2616"><![CDATA[
1408  quoted-cpair   = "\" ( HTAB / SP / VCHAR / obs-text )
1409]]></artwork></figure>
1410<t>
1411   Senders SHOULD NOT escape octets in comments that do not require escaping
1412   (i.e., other than the backslash octet "\" and the parentheses "(" and ")").
1413</t>
1414</section>
1415
1416</section>
1417
1418<section title="Message Body" anchor="message.body">
1419 
1420<t>
1421   The message body (if any) of an HTTP message is used to carry the
1422   payload body of that request or response.  The message body is
1423   identical to the payload body unless a transfer coding has been
1424   applied, as described in <xref target="header.transfer-encoding"/>.
1425</t>
1426<figure><iref primary="true" item="Grammar" subitem="message-body"/><artwork type="abnf2616"><![CDATA[
1427  message-body = *OCTET
1428]]></artwork></figure>
1429<t>
1430   The rules for when a message body is allowed in a message differ for
1431   requests and responses.
1432</t>
1433<t>
1434   The presence of a message body in a request is signaled by a
1435   <xref target="header.content-length" format="none">Content-Length</xref> or <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header
1436   field. Request message framing is independent of method semantics,
1437   even if the method does not define any use for a message body.
1438</t>
1439<t>
1440   The presence of a message body in a response depends on both
1441   the request method to which it is responding and the response
1442   status code (<xref target="status.line"/>).
1443   Responses to the HEAD request method never include a message body
1444   because the associated response header fields (e.g.,
1445   <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>, <xref target="header.content-length" format="none">Content-Length</xref>, etc.),
1446   if present, indicate only what their values would have been if the request
1447   method had been GET (Section 4.3.2 of <xref target="Part2"/>).
1448   2xx (Successful) responses to CONNECT switch to tunnel
1449   mode instead of having a message body (Section 4.3.6 of <xref target="Part2"/>).
1450   All 1xx (Informational), 204 (No Content), and
1451   304 (Not Modified) responses do not include a message body.
1452   All other responses do include a message body, although the body
1453   might be of zero length.
1454</t>
1455
1456<section title="Transfer-Encoding" anchor="header.transfer-encoding">
1457  <iref primary="true" item="Transfer-Encoding header field"/>
1458  <iref item="chunked (Coding Format)"/>
1459 
1460<t>
1461   The Transfer-Encoding header field lists the transfer coding names
1462   corresponding to the sequence of transfer codings that have been
1463   (or will be) applied to the payload body in order to form the message body.
1464   Transfer codings are defined in <xref target="transfer.codings"/>.
1465</t>
1466<figure><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/><artwork type="abnf2616"><![CDATA[
1467  Transfer-Encoding = 1#transfer-coding
1468]]></artwork></figure>
1469<t>
1470   Transfer-Encoding is analogous to the Content-Transfer-Encoding field of
1471   MIME, which was designed to enable safe transport of binary data over a
1472   7-bit transport service (<xref target="RFC2045"/>, Section 6).
1473   However, safe transport has a different focus for an 8bit-clean transfer
1474   protocol. In HTTP's case, Transfer-Encoding is primarily intended to
1475   accurately delimit a dynamically generated payload and to distinguish
1476   payload encodings that are only applied for transport efficiency or
1477   security from those that are characteristics of the selected resource.
1478</t>
1479<t>
1480   All HTTP/1.1 recipients MUST implement the chunked transfer coding
1481   (<xref target="chunked.encoding"/>) because it plays a crucial role in
1482   framing messages when the payload body size is not known in advance.
1483   If chunked is applied to a payload body, the sender MUST NOT apply
1484   chunked more than once (i.e., chunking an already chunked message is not
1485   allowed).
1486   If any transfer coding is applied to a request payload body, the
1487   sender MUST apply chunked as the final transfer coding to ensure that
1488   the message is properly framed.
1489   If any transfer coding is applied to a response payload body, the
1490   sender MUST either apply chunked as the final transfer coding or
1491   terminate the message by closing the connection.
1492</t>
1493<figure><preamble>
1494   For example,
1495</preamble><artwork type="example"><![CDATA[
1496  Transfer-Encoding: gzip, chunked
1497]]></artwork><postamble>
1498   indicates that the payload body has been compressed using the gzip
1499   coding and then chunked using the chunked coding while forming the
1500   message body.
1501</postamble></figure>
1502<t>
1503   Unlike Content-Encoding (Section 3.1.2.1 of <xref target="Part2"/>),
1504   Transfer-Encoding is a property of the message, not of the representation, and
1505   any recipient along the request/response chain MAY decode the received
1506   transfer coding(s) or apply additional transfer coding(s) to the message
1507   body, assuming that corresponding changes are made to the Transfer-Encoding
1508   field-value. Additional information about the encoding parameters MAY be
1509   provided by other header fields not defined by this specification.
1510</t>
1511<t>
1512   Transfer-Encoding MAY be sent in a response to a HEAD request or in a
1513   304 (Not Modified) response (Section 4.1 of <xref target="Part4"/>) to a GET request,
1514   neither of which includes a message body,
1515   to indicate that the origin server would have applied a transfer coding
1516   to the message body if the request had been an unconditional GET.
1517   This indication is not required, however, because any recipient on
1518   the response chain (including the origin server) can remove transfer
1519   codings when they are not needed.
1520</t>
1521<t>
1522   Transfer-Encoding was added in HTTP/1.1.  It is generally assumed that
1523   implementations advertising only HTTP/1.0 support will not understand
1524   how to process a transfer-encoded payload.
1525   A client MUST NOT send a request containing Transfer-Encoding unless it
1526   knows the server will handle HTTP/1.1 (or later) requests; such knowledge
1527   might be in the form of specific user configuration or by remembering the
1528   version of a prior received response.
1529   A server MUST NOT send a response containing Transfer-Encoding unless
1530   the corresponding request indicates HTTP/1.1 (or later).
1531</t>
1532<t>
1533   A server that receives a request message with a transfer coding it does
1534   not understand SHOULD respond with 501 (Not Implemented).
1535</t>
1536</section>
1537
1538<section title="Content-Length" anchor="header.content-length">
1539  <iref primary="true" item="Content-Length header field"/>
1540 
1541<t>
1542   When a message does not have a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header
1543   field, a Content-Length header field can provide the anticipated size,
1544   as a decimal number of octets, for a potential payload body.
1545   For messages that do include a payload body, the Content-Length field-value
1546   provides the framing information necessary for determining where the body
1547   (and message) ends.  For messages that do not include a payload body, the
1548   Content-Length indicates the size of the selected representation
1549   (Section 3 of <xref target="Part2"/>).
1550</t>
1551<figure><iref primary="true" item="Grammar" subitem="Content-Length"/><artwork type="abnf2616"><![CDATA[
1552  Content-Length = 1*DIGIT
1553]]></artwork></figure>
1554<t>
1555   An example is
1556</t>
1557<figure><artwork type="example"><![CDATA[
1558  Content-Length: 3495
1559]]></artwork></figure>
1560<t>
1561   A sender MUST NOT send a Content-Length header field in any message that
1562   contains a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field.
1563</t>
1564<t>
1565   A user agent SHOULD send a Content-Length in a request message when no
1566   <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> is sent and the request method defines
1567   a meaning for an enclosed payload body. For example, a Content-Length
1568   header field is normally sent in a POST request even when the value is
1569   0 (indicating an empty payload body).  A user agent SHOULD NOT send a
1570   Content-Length header field when the request message does not contain a
1571   payload body and the method semantics do not anticipate such a body.
1572</t>
1573<t>
1574   A server MAY send a Content-Length header field in a response to a HEAD
1575   request (Section 4.3.2 of <xref target="Part2"/>); a server MUST NOT send Content-Length in such a
1576   response unless its field-value equals the decimal number of octets that
1577   would have been sent in the payload body of a response if the same
1578   request had used the GET method.
1579</t>
1580<t>
1581   A server MAY send a Content-Length header field in a
1582   304 (Not Modified) response to a conditional GET request
1583   (Section 4.1 of <xref target="Part4"/>); a server MUST NOT send Content-Length in such a
1584   response unless its field-value equals the decimal number of octets that
1585   would have been sent in the payload body of a 200 (OK)
1586   response to the same request.
1587</t>
1588<t>
1589   A server MUST NOT send a Content-Length header field in any response
1590   with a status code of
1591   1xx (Informational) or 204 (No Content).
1592   A server SHOULD NOT send a Content-Length header field in any
1593   2xx (Successful) response to a CONNECT request (Section 4.3.6 of <xref target="Part2"/>).
1594</t>
1595<t>
1596   Aside from the cases defined above, in the absence of Transfer-Encoding,
1597   an origin server SHOULD send a Content-Length header field when the
1598   payload body size is known prior to sending the complete header block.
1599   This will allow downstream recipients to measure transfer progress,
1600   know when a received message is complete, and potentially reuse the
1601   connection for additional requests.
1602</t>
1603<t>
1604   Any Content-Length field value greater than or equal to zero is valid.
1605   Since there is no predefined limit to the length of a payload,
1606   recipients SHOULD anticipate potentially large decimal numerals and
1607   prevent parsing errors due to integer conversion overflows
1608   (<xref target="attack.protocol.element.size.overflows"/>).
1609</t>
1610<t>
1611   If a message is received that has multiple Content-Length header fields
1612   with field-values consisting of the same decimal value, or a single
1613   Content-Length header field with a field value containing a list of
1614   identical decimal values (e.g., "Content-Length: 42, 42"), indicating that
1615   duplicate Content-Length header fields have been generated or combined by an
1616   upstream message processor, then the recipient MUST either reject the
1617   message as invalid or replace the duplicated field-values with a single
1618   valid Content-Length field containing that decimal value prior to
1619   determining the message body length.
1620</t>
1621<t><list>
1622  <t>
1623   Note: HTTP's use of Content-Length for message framing differs
1624   significantly from the same field's use in MIME, where it is an optional
1625   field used only within the "message/external-body" media-type.
1626  </t>
1627</list></t>
1628</section>
1629
1630<section title="Message Body Length" anchor="message.body.length">
1631  <iref item="chunked (Coding Format)"/>
1632<t>
1633   The length of a message body is determined by one of the following
1634   (in order of precedence):
1635</t>
1636<t>
1637  <list style="numbers">
1638    <t>
1639     Any response to a HEAD request and any response with a
1640     1xx (Informational), 204 (No Content), or
1641     304 (Not Modified) status code is always
1642     terminated by the first empty line after the header fields, regardless of
1643     the header fields present in the message, and thus cannot contain a
1644     message body.
1645    </t>
1646    <t>
1647     Any 2xx (Successful) response to a CONNECT request implies that the
1648     connection will become a tunnel immediately after the empty line that
1649     concludes the header fields.  A client MUST ignore any
1650     <xref target="header.content-length" format="none">Content-Length</xref> or <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header
1651     fields received in such a message.
1652    </t>
1653    <t>
1654     If a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field is present
1655     and the chunked transfer coding (<xref target="chunked.encoding"/>)
1656     is the final encoding, the message body length is determined by reading
1657     and decoding the chunked data until the transfer coding indicates the
1658     data is complete.
1659    <vspace blankLines="1"/>
1660     If a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field is present in a
1661     response and the chunked transfer coding is not the final encoding, the
1662     message body length is determined by reading the connection until it is
1663     closed by the server.
1664     If a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field is present in a request and the
1665     chunked transfer coding is not the final encoding, the message body
1666     length cannot be determined reliably; the server MUST respond with
1667     the 400 (Bad Request) status code and then close the connection.
1668    <vspace blankLines="1"/>
1669     If a message is received with both a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>
1670     and a <xref target="header.content-length" format="none">Content-Length</xref> header field, the Transfer-Encoding
1671     overrides the Content-Length. Such a message might indicate an attempt
1672     to perform request or response smuggling (bypass of security-related
1673     checks on message routing or content) and thus ought to be handled as
1674     an error.  A sender MUST remove the received Content-Length field
1675     prior to forwarding such a message downstream.
1676    </t>
1677    <t>
1678     If a message is received without <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> and with
1679     either multiple <xref target="header.content-length" format="none">Content-Length</xref> header fields having
1680     differing field-values or a single Content-Length header field having an
1681     invalid value, then the message framing is invalid and MUST be treated
1682     as an error to prevent request or response smuggling.
1683     If this is a request message, the server MUST respond with
1684     a 400 (Bad Request) status code and then close the connection.
1685     If this is a response message received by a proxy, the proxy
1686     MUST close the connection to the server, discard the received response,
1687     and send a 502 (Bad Gateway) response to the client.
1688     If this is a response message received by a user agent, it MUST be
1689     treated as an error by discarding the message and closing the connection.
1690    </t>
1691    <t>
1692     If a valid <xref target="header.content-length" format="none">Content-Length</xref> header field is present without
1693     <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>, its decimal value defines the
1694     expected message body length in octets.
1695     If the sender closes the connection or the recipient times out before the
1696     indicated number of octets are received, the recipient MUST consider
1697     the message to be incomplete and close the connection.
1698    </t>
1699    <t>
1700     If this is a request message and none of the above are true, then the
1701     message body length is zero (no message body is present).
1702    </t>
1703    <t>
1704     Otherwise, this is a response message without a declared message body
1705     length, so the message body length is determined by the number of octets
1706     received prior to the server closing the connection.
1707    </t>
1708  </list>
1709</t>
1710<t>
1711   Since there is no way to distinguish a successfully completed,
1712   close-delimited message from a partially-received message interrupted
1713   by network failure, a server SHOULD use encoding or
1714   length-delimited messages whenever possible.  The close-delimiting
1715   feature exists primarily for backwards compatibility with HTTP/1.0.
1716</t>
1717<t>
1718   A server MAY reject a request that contains a message body but
1719   not a <xref target="header.content-length" format="none">Content-Length</xref> by responding with
1720   411 (Length Required).
1721</t>
1722<t>
1723   Unless a transfer coding other than chunked has been applied,
1724   a client that sends a request containing a message body SHOULD
1725   use a valid <xref target="header.content-length" format="none">Content-Length</xref> header field if the message body
1726   length is known in advance, rather than the chunked transfer coding, since some
1727   existing services respond to chunked with a 411 (Length Required)
1728   status code even though they understand the chunked transfer coding.  This
1729   is typically because such services are implemented via a gateway that
1730   requires a content-length in advance of being called and the server
1731   is unable or unwilling to buffer the entire request before processing.
1732</t>
1733<t>
1734   A user agent that sends a request containing a message body MUST send a
1735   valid <xref target="header.content-length" format="none">Content-Length</xref> header field if it does not know the
1736   server will handle HTTP/1.1 (or later) requests; such knowledge can be in
1737   the form of specific user configuration or by remembering the version of a
1738   prior received response.
1739</t>
1740<t>
1741   If the final response to the last request on a connection has been
1742   completely received and there remains additional data to read, a user agent
1743   MAY discard the remaining data or attempt to determine if that data
1744   belongs as part of the prior response body, which might be the case if the
1745   prior message's Content-Length value is incorrect. A client MUST NOT
1746   process, cache, or forward such extra data as a separate response, since
1747   such behavior would be vulnerable to cache poisoning.
1748</t>
1749</section>
1750</section>
1751
1752<section anchor="incomplete.messages" title="Handling Incomplete Messages">
1753<t>
1754   A server that receives an incomplete request message, usually due to a
1755   canceled request or a triggered time-out exception, MAY send an error
1756   response prior to closing the connection.
1757</t>
1758<t>
1759   A client that receives an incomplete response message, which can occur
1760   when a connection is closed prematurely or when decoding a supposedly
1761   chunked transfer coding fails, MUST record the message as incomplete.
1762   Cache requirements for incomplete responses are defined in
1763   Section 3 of <xref target="Part6"/>.
1764</t>
1765<t>
1766   If a response terminates in the middle of the header block (before the
1767   empty line is received) and the status code might rely on header fields to
1768   convey the full meaning of the response, then the client cannot assume
1769   that meaning has been conveyed; the client might need to repeat the
1770   request in order to determine what action to take next.
1771</t>
1772<t>
1773   A message body that uses the chunked transfer coding is
1774   incomplete if the zero-sized chunk that terminates the encoding has not
1775   been received.  A message that uses a valid <xref target="header.content-length" format="none">Content-Length</xref> is
1776   incomplete if the size of the message body received (in octets) is less than
1777   the value given by Content-Length.  A response that has neither chunked
1778   transfer coding nor Content-Length is terminated by closure of the
1779   connection, and thus is considered complete regardless of the number of
1780   message body octets received, provided that the header block was received
1781   intact.
1782</t>
1783</section>
1784
1785<section title="Message Parsing Robustness" anchor="message.robustness">
1786<t>
1787   Older HTTP/1.0 user agent implementations might send an extra CRLF
1788   after a POST request as a workaround for some early server
1789   applications that failed to read message body content that was
1790   not terminated by a line-ending. An HTTP/1.1 user agent MUST NOT
1791   preface or follow a request with an extra CRLF.  If terminating
1792   the request message body with a line-ending is desired, then the
1793   user agent MUST count the terminating CRLF octets as part of the
1794   message body length.
1795</t>
1796<t>
1797   In the interest of robustness, servers SHOULD ignore at least one
1798   empty line received where a request-line is expected. In other words, if
1799   a server is reading the protocol stream at the beginning of a
1800   message and receives a CRLF first, the server SHOULD ignore the CRLF.
1801</t>
1802<t>
1803   Although the line terminator for the start-line and header
1804   fields is the sequence CRLF, recipients MAY recognize a
1805   single LF as a line terminator and ignore any preceding CR.
1806</t>
1807<t>
1808   Although the request-line and status-line grammar rules require that each
1809   of the component elements be separated by a single SP octet, recipients
1810   MAY instead parse on whitespace-delimited word boundaries and, aside
1811   from the CRLF terminator, treat any form of whitespace as the SP separator
1812   while ignoring preceding or trailing whitespace;
1813   such whitespace includes one or more of the following octets:
1814   SP, HTAB, VT (%x0B), FF (%x0C), or bare CR.
1815</t>
1816<t>
1817   When a server listening only for HTTP request messages, or processing
1818   what appears from the start-line to be an HTTP request message,
1819   receives a sequence of octets that does not match the HTTP-message
1820   grammar aside from the robustness exceptions listed above, the
1821   server SHOULD respond with a 400 (Bad Request) response. 
1822</t>
1823</section>
1824</section>
1825
1826<section title="Transfer Codings" anchor="transfer.codings">
1827 
1828 
1829<t>
1830   Transfer coding names are used to indicate an encoding
1831   transformation that has been, can be, or might need to be applied to a
1832   payload body in order to ensure "safe transport" through the network.
1833   This differs from a content coding in that the transfer coding is a
1834   property of the message rather than a property of the representation
1835   that is being transferred.
1836</t>
1837<figure><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/><artwork type="abnf2616"><![CDATA[
1838  transfer-coding    = "chunked" ; Section 4.1
1839                     / "compress" ; Section 4.2.1
1840                     / "deflate" ; Section 4.2.2
1841                     / "gzip" ; Section 4.2.3
1842                     / transfer-extension
1843  transfer-extension = token *( OWS ";" OWS transfer-parameter )
1844]]></artwork></figure>
1845<t anchor="rule.parameter">
1846 
1847 
1848 
1849   Parameters are in the form of attribute/value pairs.
1850</t>
1851<figure><iref primary="true" item="Grammar" subitem="transfer-parameter"/><iref primary="true" item="Grammar" subitem="attribute"/><iref primary="true" item="Grammar" subitem="value"/><iref primary="true" item="Grammar" subitem="date2"/><iref primary="true" item="Grammar" subitem="date3"/><artwork type="abnf2616"><![CDATA[
1852  transfer-parameter = attribute BWS "=" BWS value
1853  attribute          = token
1854  value              = word
1855]]></artwork></figure>
1856<t>
1857   All transfer-coding names are case-insensitive and ought to be registered
1858   within the HTTP Transfer Coding registry, as defined in
1859   <xref target="transfer.coding.registry"/>.
1860   They are used in the <xref target="header.te" format="none">TE</xref> (<xref target="header.te"/>) and
1861   <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> (<xref target="header.transfer-encoding"/>)
1862   header fields.
1863</t>
1864
1865<section title="Chunked Transfer Coding" anchor="chunked.encoding">
1866  <iref primary="true" item="chunked (Coding Format)"/>
1867 
1868 
1869 
1870 
1871 
1872 
1873 
1874 
1875 
1876 
1877 
1878<t>
1879   The chunked transfer coding modifies the body of a message in order to
1880   transfer it as a series of chunks, each with its own size indicator,
1881   followed by an OPTIONAL trailer containing header fields. This
1882   allows dynamically generated content to be transferred along with the
1883   information necessary for the recipient to verify that it has
1884   received the full message.
1885</t>
1886<figure><iref primary="true" item="Grammar" subitem="chunked-body"><!--terminal production--></iref><iref primary="true" item="Grammar" subitem="chunk"/><iref primary="true" item="Grammar" subitem="chunk-size"/><iref primary="true" item="Grammar" subitem="last-chunk"/><iref primary="true" item="Grammar" subitem="chunk-ext"/><iref primary="true" item="Grammar" subitem="chunk-ext-name"/><iref primary="true" item="Grammar" subitem="chunk-ext-val"/><iref primary="true" item="Grammar" subitem="chunk-data"/><iref primary="true" item="Grammar" subitem="trailer-part"/><iref primary="true" item="Grammar" subitem="quoted-str-nf"/><iref primary="true" item="Grammar" subitem="qdtext-nf"/><artwork type="abnf2616"><![CDATA[
1887  chunked-body   = *chunk
1888                   last-chunk
1889                   trailer-part
1890                   CRLF
1891 
1892  chunk          = chunk-size [ chunk-ext ] CRLF
1893                   chunk-data CRLF
1894  chunk-size     = 1*HEXDIG
1895  last-chunk     = 1*("0") [ chunk-ext ] CRLF
1896 
1897  chunk-ext      = *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
1898  chunk-ext-name = token
1899  chunk-ext-val  = token / quoted-str-nf
1900  chunk-data     = 1*OCTET ; a sequence of chunk-size octets
1901  trailer-part   = *( header-field CRLF )
1902 
1903  quoted-str-nf  = DQUOTE *( qdtext-nf / quoted-pair ) DQUOTE
1904                 ; like quoted-string, but disallowing line folding
1905  qdtext-nf      = HTAB / SP / %x21 / %x23-5B / %x5D-7E / obs-text
1906]]></artwork></figure>
1907<t>
1908   Chunk extensions within the chunked transfer coding are deprecated.
1909   Senders SHOULD NOT send chunk-ext.
1910   Definition of new chunk extensions is discouraged.
1911</t>
1912<t>
1913   The chunk-size field is a string of hex digits indicating the size of
1914   the chunk-data in octets. The chunked transfer coding is complete when a
1915   chunk with a chunk-size of zero is received, possibly followed by a
1916   trailer, and finally terminated by an empty line.
1917</t>
1918
1919<section title="Trailer" anchor="header.trailer">
1920  <iref primary="true" item="Trailer header field"/>
1921 
1922<t>
1923   A trailer allows the sender to include additional fields at the end of a
1924   chunked message in order to supply metadata that might be dynamically
1925   generated while the message body is sent, such as a message integrity
1926   check, digital signature, or post-processing status.
1927   The trailer MUST NOT contain fields that need to be known before a
1928   recipient processes the body, such as <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>,
1929   <xref target="header.content-length" format="none">Content-Length</xref>, and <xref target="header.trailer" format="none">Trailer</xref>.
1930</t>
1931<t>
1932   When a message includes a message body encoded with the chunked
1933   transfer coding and the sender desires to send metadata in the form of
1934   trailer fields at the end of the message, the sender SHOULD send a
1935   <xref target="header.trailer" format="none">Trailer</xref> header field before the message body to indicate
1936   which fields will be present in the trailers. This allows the recipient
1937   to prepare for receipt of that metadata before it starts processing the body,
1938   which is useful if the message is being streamed and the recipient wishes
1939   to confirm an integrity check on the fly.
1940</t>
1941<figure><iref primary="true" item="Grammar" subitem="Trailer"/><artwork type="abnf2616"><![CDATA[
1942  Trailer = 1#field-name
1943]]></artwork></figure>
1944<t>
1945   If no <xref target="header.trailer" format="none">Trailer</xref> header field is present, the sender of a
1946   chunked message body SHOULD send an empty trailer.
1947</t>
1948<t>
1949   A server MUST send an empty trailer with the chunked transfer coding
1950   unless at least one of the following is true:
1951  <list style="numbers">
1952    <t>the request included a <xref target="header.te" format="none">TE</xref> header field that indicates
1953    "trailers" is acceptable in the transfer coding of the response, as
1954    described in <xref target="header.te"/>; or,</t>
1955     
1956    <t>the trailer fields consist entirely of optional metadata and the
1957    recipient could use the message (in a manner acceptable to the server where
1958    the field originated) without receiving that metadata. In other words,
1959    the server that generated the header field is willing to accept the
1960    possibility that the trailer fields might be silently discarded along
1961    the path to the client.</t>
1962  </list>
1963</t>
1964<t>
1965   The above requirement prevents the need for an infinite buffer when a
1966   message is being received by an HTTP/1.1 (or later) proxy and forwarded to
1967   an HTTP/1.0 recipient.
1968</t>
1969</section>
1970
1971<section title="Decoding chunked" anchor="decoding.chunked">
1972<t>
1973   A process for decoding the chunked transfer coding
1974   can be represented in pseudo-code as:
1975</t>
1976<figure><artwork type="code"><![CDATA[
1977  length := 0
1978  read chunk-size, chunk-ext (if any), and CRLF
1979  while (chunk-size > 0) {
1980     read chunk-data and CRLF
1981     append chunk-data to decoded-body
1982     length := length + chunk-size
1983     read chunk-size, chunk-ext (if any), and CRLF
1984  }
1985  read header-field
1986  while (header-field not empty) {
1987     append header-field to existing header fields
1988     read header-field
1989  }
1990  Content-Length := length
1991  Remove "chunked" from Transfer-Encoding
1992  Remove Trailer from existing header fields
1993]]></artwork></figure>
1994<t>
1995   All recipients MUST be able to receive and decode the
1996   chunked transfer coding and MUST ignore chunk-ext extensions
1997   they do not understand.
1998</t>
1999</section>
2000</section>
2001
2002<section title="Compression Codings" anchor="compression.codings">
2003<t>
2004   The codings defined below can be used to compress the payload of a
2005   message.
2006</t>
2007
2008<section title="Compress Coding" anchor="compress.coding">
2009<iref item="compress (Coding Format)"/>
2010<t>
2011   The "compress" coding is an adaptive Lempel-Ziv-Welch (LZW) coding
2012   <xref target="Welch"/> that is commonly produced by the UNIX file
2013   compression program "compress".
2014   Recipients SHOULD consider "x-compress" to be equivalent to "compress".
2015</t>
2016</section>
2017
2018<section title="Deflate Coding" anchor="deflate.coding">
2019<iref item="deflate (Coding Format)"/>
2020<t>
2021   The "deflate" coding is a "zlib" data format <xref target="RFC1950"/>
2022   containing a "deflate" compressed data stream <xref target="RFC1951"/>
2023   that uses a combination of the Lempel-Ziv (LZ77) compression algorithm and
2024   Huffman coding.
2025</t>
2026<t><list>
2027  <t>
2028    Note: Some incorrect implementations send the "deflate"
2029    compressed data without the zlib wrapper.
2030   </t>
2031</list></t>
2032</section>
2033
2034<section title="Gzip Coding" anchor="gzip.coding">
2035<iref item="gzip (Coding Format)"/>
2036<t>
2037   The "gzip" coding is an LZ77 coding with a 32 bit CRC that is commonly
2038   produced by the gzip file compression program <xref target="RFC1952"/>.
2039   Recipients SHOULD consider "x-gzip" to be equivalent to "gzip".
2040</t>
2041</section>
2042
2043</section>
2044
2045<section title="TE" anchor="header.te">
2046  <iref primary="true" item="TE header field"/>
2047 
2048 
2049 
2050 
2051<t>
2052   The "TE" header field in a request indicates what transfer codings,
2053   besides chunked, the client is willing to accept in response, and
2054   whether or not the client is willing to accept trailer fields in a
2055   chunked transfer coding.
2056</t>
2057<t>
2058   The TE field-value consists of a comma-separated list of transfer coding
2059   names, each allowing for optional parameters (as described in
2060   <xref target="transfer.codings"/>), and/or the keyword "trailers".
2061   Clients MUST NOT send the chunked transfer coding name in TE;
2062   chunked is always acceptable for HTTP/1.1 recipients.
2063</t>
2064<figure><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="t-codings"/><iref primary="true" item="Grammar" subitem="t-ranking"/><iref primary="true" item="Grammar" subitem="rank"/><artwork type="abnf2616"><![CDATA[
2065  TE        = #t-codings
2066  t-codings = "trailers" / ( transfer-coding [ t-ranking ] )
2067  t-ranking = OWS ";" OWS "q=" rank
2068  rank      = ( "0" [ "." 0*3DIGIT ] )
2069             / ( "1" [ "." 0*3("0") ] )
2070]]></artwork></figure>
2071<t>
2072   Three examples of TE use are below.
2073</t>
2074<figure><artwork type="example"><![CDATA[
2075  TE: deflate
2076  TE:
2077  TE: trailers, deflate;q=0.5
2078]]></artwork></figure>
2079<t>
2080   The presence of the keyword "trailers" indicates that the client is willing
2081   to accept trailer fields in a chunked transfer coding, as defined in
2082   <xref target="chunked.encoding"/>, on behalf of itself and any downstream
2083   clients. For requests from an intermediary, this implies that either:
2084   (a) all downstream clients are willing to accept trailer fields in the
2085   forwarded response; or,
2086   (b) the intermediary will attempt to buffer the response on behalf of
2087   downstream recipients.
2088   Note that HTTP/1.1 does not define any means to limit the size of a
2089   chunked response such that an intermediary can be assured of buffering the
2090   entire response.
2091</t>
2092<t>
2093   When multiple transfer codings are acceptable, the client MAY rank the
2094   codings by preference using a case-insensitive "q" parameter (similar to
2095   the qvalues used in content negotiation fields, Section 5.3.1 of <xref target="Part2"/>). The rank value
2096   is a real number in the range 0 through 1, where 0.001 is the least
2097   preferred and 1 is the most preferred; a value of 0 means "not acceptable".
2098</t>
2099<t>
2100   If the TE field-value is empty or if no TE field is present, the only
2101   acceptable transfer coding is chunked. A message with no transfer coding
2102   is always acceptable.
2103</t>
2104<t>
2105   Since the TE header field only applies to the immediate connection,
2106   a sender of TE MUST also send a "TE" connection option within the
2107   <xref target="header.connection" format="none">Connection</xref> header field (<xref target="header.connection"/>)
2108   in order to prevent the TE field from being forwarded by intermediaries
2109   that do not support its semantics.
2110</t>
2111</section>
2112</section>
2113
2114<section title="Message Routing" anchor="message.routing">
2115<t>
2116   HTTP request message routing is determined by each client based on the
2117   target resource, the client's proxy configuration, and
2118   establishment or reuse of an inbound connection.  The corresponding
2119   response routing follows the same connection chain back to the client.
2120</t>
2121
2122<section title="Identifying a Target Resource" anchor="target-resource">
2123  <iref primary="true" item="target resource"/>
2124  <iref primary="true" item="target URI"/>
2125 
2126 
2127<t>
2128   HTTP is used in a wide variety of applications, ranging from
2129   general-purpose computers to home appliances.  In some cases,
2130   communication options are hard-coded in a client's configuration.
2131   However, most HTTP clients rely on the same resource identification
2132   mechanism and configuration techniques as general-purpose Web browsers.
2133</t>
2134<t>
2135   HTTP communication is initiated by a user agent for some purpose.
2136   The purpose is a combination of request semantics, which are defined in
2137   <xref target="Part2"/>, and a target resource upon which to apply those
2138   semantics.  A URI reference (<xref target="uri"/>) is typically used as
2139   an identifier for the "target resource", which a user agent
2140   would resolve to its absolute form in order to obtain the
2141   "target URI".  The target URI
2142   excludes the reference's fragment component, if any,
2143   since fragment identifiers are reserved for client-side processing
2144   (<xref target="RFC3986"/>, Section 3.5).
2145</t>
2146</section>
2147
2148<section title="Connecting Inbound" anchor="connecting.inbound">
2149<t>
2150   Once the target URI is determined, a client needs to decide whether
2151   a network request is necessary to accomplish the desired semantics and,
2152   if so, where that request is to be directed.
2153</t>
2154<t>
2155   If the client has a cache <xref target="Part6"/> and the request can be
2156   satisfied by it, then the request is
2157   usually directed there first.
2158</t>
2159<t>
2160   If the request is not satisfied by a cache, then a typical client will
2161   check its configuration to determine whether a proxy is to be used to
2162   satisfy the request.  Proxy configuration is implementation-dependent,
2163   but is often based on URI prefix matching, selective authority matching,
2164   or both, and the proxy itself is usually identified by an "http" or
2165   "https" URI.  If a proxy is applicable, the client connects inbound by
2166   establishing (or reusing) a connection to that proxy.
2167</t>
2168<t>
2169   If no proxy is applicable, a typical client will invoke a handler routine,
2170   usually specific to the target URI's scheme, to connect directly
2171   to an authority for the target resource.  How that is accomplished is
2172   dependent on the target URI scheme and defined by its associated
2173   specification, similar to how this specification defines origin server
2174   access for resolution of the "http" (<xref target="http.uri"/>) and
2175   "https" (<xref target="https.uri"/>) schemes.
2176</t>
2177<t>
2178   HTTP requirements regarding connection management are defined in
2179   <xref target="connection.management"/>.
2180</t>
2181</section>
2182
2183<section title="Request Target" anchor="request-target">
2184<t>
2185   Once an inbound connection is obtained,
2186   the client sends an HTTP request message (<xref target="http.message"/>)
2187   with a request-target derived from the target URI.
2188   There are four distinct formats for the request-target, depending on both
2189   the method being requested and whether the request is to a proxy.
2190</t>   
2191<figure><iref primary="true" item="Grammar" subitem="request-target"/><iref primary="true" item="Grammar" subitem="origin-form"/><iref primary="true" item="Grammar" subitem="absolute-form"/><iref primary="true" item="Grammar" subitem="authority-form"/><iref primary="true" item="Grammar" subitem="asterisk-form"/><artwork type="abnf2616"><![CDATA[
2192  request-target = origin-form
2193                 / absolute-form
2194                 / authority-form
2195                 / asterisk-form
2196
2197  origin-form    = absolute-path [ "?" query ]
2198  absolute-form  = absolute-URI
2199  authority-form = authority
2200  asterisk-form  = "*"
2201]]></artwork></figure>
2202<t anchor="origin-form"><iref item="origin-form (of request-target)"/>
2203  origin-form
2204</t>
2205<t>
2206   The most common form of request-target is the origin-form.
2207   When making a request directly to an origin server, other than a CONNECT
2208   or server-wide OPTIONS request (as detailed below),
2209   a client MUST send only the absolute path and query components of
2210   the target URI as the request-target.
2211   If the target URI's path component is empty, then the client MUST send
2212   "/" as the path within the origin-form of request-target.
2213   A <xref target="header.host" format="none">Host</xref> header field is also sent, as defined in
2214   <xref target="header.host"/>, containing the target URI's
2215   authority component (excluding any userinfo).
2216</t>
2217<t>
2218   For example, a client wishing to retrieve a representation of the resource
2219   identified as
2220</t>
2221<figure><artwork type="example"><![CDATA[
2222  http://www.example.org/where?q=now
2223  ]]></artwork></figure>
2224<t>
2225   directly from the origin server would open (or reuse) a TCP connection
2226   to port 80 of the host "www.example.org" and send the lines:
2227</t>
2228<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2229  GET /where?q=now HTTP/1.1
2230  Host: www.example.org
2231  ]]></artwork></figure>
2232<t>
2233   followed by the remainder of the request message.
2234</t>
2235<t anchor="absolute-form"><iref item="absolute-form (of request-target)"/>
2236  absolute-form
2237</t>
2238<t>
2239   When making a request to a proxy, other than a CONNECT or server-wide
2240   OPTIONS request (as detailed below), a client MUST send the target URI
2241   in absolute-form as the request-target.
2242   The proxy is requested to either service that request from a valid cache,
2243   if possible, or make the same request on the client's behalf to either
2244   the next inbound proxy server or directly to the origin server indicated
2245   by the request-target.  Requirements on such "forwarding" of messages are
2246   defined in <xref target="message.forwarding"/>.
2247</t>
2248<t>
2249   An example absolute-form of request-line would be:
2250</t>
2251<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2252  GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1
2253  ]]></artwork></figure>
2254<t>
2255   To allow for transition to the absolute-form for all requests in some
2256   future version of HTTP, HTTP/1.1 servers MUST accept the absolute-form
2257   in requests, even though HTTP/1.1 clients will only send them in requests
2258   to proxies.
2259</t>
2260<t anchor="authority-form"><iref item="authority-form (of request-target)"/>
2261  authority-form
2262</t>
2263<t>
2264   The authority-form of request-target is only used for CONNECT requests
2265   (Section 4.3.6 of <xref target="Part2"/>).  When making a CONNECT request to establish a tunnel through
2266   one or more proxies, a client MUST send only the target URI's
2267   authority component (excluding any userinfo) as the request-target.
2268   For example,
2269</t>
2270<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2271  CONNECT www.example.com:80 HTTP/1.1
2272  ]]></artwork></figure>
2273<t anchor="asterisk-form"><iref item="asterisk-form (of request-target)"/>
2274  asterisk-form
2275</t>
2276<t>
2277   The asterisk-form of request-target is only used for a server-wide
2278   OPTIONS request (Section 4.3.7 of <xref target="Part2"/>).  When a client wishes to request OPTIONS
2279   for the server as a whole, as opposed to a specific named resource of
2280   that server, the client MUST send only "*" (%x2A) as the request-target.
2281   For example,
2282</t>
2283<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2284  OPTIONS * HTTP/1.1
2285  ]]></artwork></figure>
2286<t>
2287   If a proxy receives an OPTIONS request with an absolute-form of
2288   request-target in which the URI has an empty path and no query component,
2289   then the last proxy on the request chain MUST send a request-target
2290   of "*" when it forwards the request to the indicated origin server.
2291</t>
2292<figure><preamble>  
2293   For example, the request
2294</preamble><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2295  OPTIONS http://www.example.org:8001 HTTP/1.1
2296  ]]></artwork></figure>
2297<figure><preamble>  
2298  would be forwarded by the final proxy as
2299</preamble><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2300  OPTIONS * HTTP/1.1
2301  Host: www.example.org:8001
2302  ]]></artwork>
2303<postamble>
2304   after connecting to port 8001 of host "www.example.org".
2305</postamble>
2306</figure>
2307</section>
2308
2309<section title="Host" anchor="header.host">
2310  <iref primary="true" item="Host header field"/>
2311 
2312<t>
2313   The "Host" header field in a request provides the host and port
2314   information from the target URI, enabling the origin
2315   server to distinguish among resources while servicing requests
2316   for multiple host names on a single IP address.  Since the Host
2317   field-value is critical information for handling a request, it
2318   SHOULD be sent as the first header field following the request-line.
2319</t>
2320<figure><iref primary="true" item="Grammar" subitem="Host"/><artwork type="abnf2616"><![CDATA[
2321  Host = uri-host [ ":" port ] ; Section 2.7.1
2322]]></artwork></figure>
2323<t>
2324   A client MUST send a Host header field in all HTTP/1.1 request
2325   messages.  If the target URI includes an authority component, then
2326   the Host field-value MUST be identical to that authority component
2327   after excluding any userinfo (<xref target="http.uri"/>).
2328   If the authority component is missing or undefined for the target URI,
2329   then the Host header field MUST be sent with an empty field-value.
2330</t>
2331<t>
2332   For example, a GET request to the origin server for
2333   &lt;http://www.example.org/pub/WWW/&gt; would begin with:
2334</t>
2335<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2336  GET /pub/WWW/ HTTP/1.1
2337  Host: www.example.org
2338  ]]></artwork></figure>
2339<t>
2340   The Host header field MUST be sent in an HTTP/1.1 request even
2341   if the request-target is in the absolute-form, since this
2342   allows the Host information to be forwarded through ancient HTTP/1.0
2343   proxies that might not have implemented Host.
2344</t>
2345<t>
2346   When a proxy receives a request with an absolute-form of
2347   request-target, the proxy MUST ignore the received
2348   Host header field (if any) and instead replace it with the host
2349   information of the request-target.  If the proxy forwards the request,
2350   it MUST generate a new Host field-value based on the received
2351   request-target rather than forward the received Host field-value.
2352</t>
2353<t>
2354   Since the Host header field acts as an application-level routing
2355   mechanism, it is a frequent target for malware seeking to poison
2356   a shared cache or redirect a request to an unintended server.
2357   An interception proxy is particularly vulnerable if it relies on
2358   the Host field-value for redirecting requests to internal
2359   servers, or for use as a cache key in a shared cache, without
2360   first verifying that the intercepted connection is targeting a
2361   valid IP address for that host.
2362</t>
2363<t>
2364   A server MUST respond with a 400 (Bad Request) status code
2365   to any HTTP/1.1 request message that lacks a Host header field and
2366   to any request message that contains more than one Host header field
2367   or a Host header field with an invalid field-value.
2368</t>
2369</section>
2370
2371<section title="Effective Request URI" anchor="effective.request.uri">
2372  <iref primary="true" item="effective request URI"/>
2373 
2374<t>
2375   A server that receives an HTTP request message MUST reconstruct
2376   the user agent's original target URI, based on the pieces of information
2377   learned from the request-target, <xref target="header.host" format="none">Host</xref> header field, and
2378   connection context, in order to identify the intended target resource and
2379   properly service the request. The URI derived from this reconstruction
2380   process is referred to as the "effective request URI".
2381</t>
2382<t>
2383   For a user agent, the effective request URI is the target URI.
2384</t>
2385<t>
2386   If the request-target is in absolute-form, then the effective request URI
2387   is the same as the request-target.  Otherwise, the effective request URI
2388   is constructed as follows.
2389</t>
2390<t>
2391   If the request is received over a TLS-secured TCP connection,
2392   then the effective request URI's scheme is "https"; otherwise, the
2393   scheme is "http".
2394</t>
2395<t>
2396   If the request-target is in authority-form, then the effective
2397   request URI's authority component is the same as the request-target.
2398   Otherwise, if a <xref target="header.host" format="none">Host</xref> header field is supplied with a
2399   non-empty field-value, then the authority component is the same as the
2400   Host field-value. Otherwise, the authority component is the concatenation of
2401   the default host name configured for the server, a colon (":"), and the
2402   connection's incoming TCP port number in decimal form.
2403</t>
2404<t>
2405   If the request-target is in authority-form or asterisk-form, then the
2406   effective request URI's combined path and query component is empty.
2407   Otherwise, the combined path and query component is the same as the
2408   request-target.
2409</t>
2410<t>
2411   The components of the effective request URI, once determined as above,
2412   can be combined into absolute-URI form by concatenating the scheme,
2413   "://", authority, and combined path and query component.
2414</t>
2415<figure>
2416<preamble>
2417   Example 1: the following message received over an insecure TCP connection
2418</preamble>
2419<artwork type="example"><![CDATA[
2420  GET /pub/WWW/TheProject.html HTTP/1.1
2421  Host: www.example.org:8080
2422  ]]></artwork>
2423</figure>
2424<figure>
2425<preamble>
2426  has an effective request URI of
2427</preamble>
2428<artwork type="example"><![CDATA[
2429  http://www.example.org:8080/pub/WWW/TheProject.html
2430  ]]></artwork>
2431</figure>
2432<figure>
2433<preamble>
2434   Example 2: the following message received over a TLS-secured TCP connection
2435</preamble>
2436<artwork type="example"><![CDATA[
2437  OPTIONS * HTTP/1.1
2438  Host: www.example.org
2439  ]]></artwork>
2440</figure>
2441<figure>
2442<preamble>
2443  has an effective request URI of
2444</preamble>
2445<artwork type="example"><![CDATA[
2446  https://www.example.org
2447  ]]></artwork>
2448</figure>
2449<t>
2450   An origin server that does not allow resources to differ by requested
2451   host MAY ignore the <xref target="header.host" format="none">Host</xref> field-value and instead replace it
2452   with a configured server name when constructing the effective request URI.
2453</t>
2454<t>
2455   Recipients of an HTTP/1.0 request that lacks a <xref target="header.host" format="none">Host</xref> header
2456   field MAY attempt to use heuristics (e.g., examination of the URI path for
2457   something unique to a particular host) in order to guess the
2458   effective request URI's authority component.
2459</t>
2460</section>
2461
2462<section title="Associating a Response to a Request" anchor="associating.response.to.request">
2463<t>
2464   HTTP does not include a request identifier for associating a given
2465   request message with its corresponding one or more response messages.
2466   Hence, it relies on the order of response arrival to correspond exactly
2467   to the order in which requests are made on the same connection.
2468   More than one response message per request only occurs when one or more
2469   informational responses (1xx, see Section 6.2 of <xref target="Part2"/>) precede a
2470   final response to the same request.
2471</t>
2472<t>
2473   A client that has more than one outstanding request on a connection MUST
2474   maintain a list of outstanding requests in the order sent and MUST
2475   associate each received response message on that connection to the highest
2476   ordered request that has not yet received a final (non-1xx)
2477   response.
2478</t>
2479</section>
2480
2481<section title="Message Forwarding" anchor="message.forwarding">
2482<t>
2483   As described in <xref target="intermediaries"/>, intermediaries can serve
2484   a variety of roles in the processing of HTTP requests and responses.
2485   Some intermediaries are used to improve performance or availability.
2486   Others are used for access control or to filter content.
2487   Since an HTTP stream has characteristics similar to a pipe-and-filter
2488   architecture, there are no inherent limits to the extent an intermediary
2489   can enhance (or interfere) with either direction of the stream.
2490</t>
2491<t>
2492   Intermediaries that forward a message MUST implement the
2493   <xref target="header.connection" format="none">Connection</xref> header field, as specified in
2494   <xref target="header.connection"/>, to exclude fields that are only
2495   intended for the incoming connection.
2496</t>
2497<t>
2498   In order to avoid request loops, a proxy that forwards requests to other
2499   proxies MUST be able to recognize and exclude all of its own server
2500   names, including any aliases, local variations, or literal IP addresses.
2501</t>
2502
2503<section title="Via" anchor="header.via">
2504  <iref primary="true" item="Via header field"/>
2505 
2506 
2507 
2508 
2509<t>
2510   The "Via" header field indicates the presence of intermediate protocols and
2511   recipients between the user agent and the server (on requests) or between
2512   the origin server and the client (on responses), similar to the
2513   "Received" header field in email
2514   (Section 3.6.7 of <xref target="RFC5322"/>).
2515   Via can be used for tracking message forwards,
2516   avoiding request loops, and identifying the protocol capabilities of
2517   senders along the request/response chain.
2518</t>
2519<figure><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"/><artwork type="abnf2616"><![CDATA[
2520  Via = 1#( received-protocol RWS received-by [ RWS comment ] )
2521
2522  received-protocol = [ protocol-name "/" ] protocol-version
2523                      ; see Section 6.7
2524  received-by       = ( uri-host [ ":" port ] ) / pseudonym
2525  pseudonym         = token
2526]]></artwork></figure>
2527<t>
2528   Multiple Via field values represent each proxy or gateway that has
2529   forwarded the message. Each intermediary appends its own information
2530   about how the message was received, such that the end result is ordered
2531   according to the sequence of forwarding recipients.
2532</t>
2533<t>
2534   A proxy MUST send an appropriate Via header field, as described below, in
2535   each message that it forwards.
2536   An HTTP-to-HTTP gateway MUST send an appropriate Via header field in
2537   each inbound request message and MAY send a Via header field in
2538   forwarded response messages.
2539</t>
2540<t>
2541   For each intermediary, the received-protocol indicates the protocol and
2542   protocol version used by the upstream sender of the message. Hence, the
2543   Via field value records the advertised protocol capabilities of the
2544   request/response chain such that they remain visible to downstream
2545   recipients; this can be useful for determining what backwards-incompatible
2546   features might be safe to use in response, or within a later request, as
2547   described in <xref target="http.version"/>. For brevity, the protocol-name
2548   is omitted when the received protocol is HTTP.
2549</t>
2550<t>
2551   The received-by field is normally the host and optional port number of a
2552   recipient server or client that subsequently forwarded the message.
2553   However, if the real host is considered to be sensitive information, it
2554   MAY be replaced by a pseudonym. If the port is not given, it MAY be
2555   assumed to be the default port of the received-protocol.
2556</t>
2557<t>
2558   Comments MAY be used in the Via header field to identify the software
2559   of each recipient, analogous to the User-Agent and
2560   Server header fields. However, all comments in the Via field
2561   are optional and MAY be removed by any recipient prior to forwarding the
2562   message.
2563</t>
2564<t>
2565   For example, a request message could be sent from an HTTP/1.0 user
2566   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
2567   forward the request to a public proxy at p.example.net, which completes
2568   the request by forwarding it to the origin server at www.example.com.
2569   The request received by www.example.com would then have the following
2570   Via header field:
2571</t>
2572<figure><artwork type="example"><![CDATA[
2573  Via: 1.0 fred, 1.1 p.example.net
2574]]></artwork></figure>
2575<t>
2576   A proxy or gateway used as a portal through a network firewall
2577   SHOULD NOT forward the names and ports of hosts within the firewall
2578   region unless it is explicitly enabled to do so. If not enabled, the
2579   received-by host of any host behind the firewall SHOULD be replaced
2580   by an appropriate pseudonym for that host.
2581</t>
2582<t>
2583   A proxy or gateway MAY combine an ordered subsequence of Via header
2584   field entries into a single such entry if the entries have identical
2585   received-protocol values. For example,
2586</t>
2587<figure><artwork type="example"><![CDATA[
2588  Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
2589]]></artwork></figure>
2590<t>
2591  could be collapsed to
2592</t>
2593<figure><artwork type="example"><![CDATA[
2594  Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
2595]]></artwork></figure>
2596<t>
2597   Senders SHOULD NOT combine multiple entries unless they are all
2598   under the same organizational control and the hosts have already been
2599   replaced by pseudonyms. Senders MUST NOT combine entries that
2600   have different received-protocol values.
2601</t>
2602</section>
2603
2604<section title="Transformations" anchor="message.transformations">
2605<t>
2606   Some intermediaries include features for transforming messages and their
2607   payloads.  A transforming proxy might, for example, convert between image
2608   formats in order to save cache space or to reduce the amount of traffic on
2609   a slow link. However, operational problems might occur when these
2610   transformations are applied to payloads intended for critical applications,
2611   such as medical imaging or scientific data analysis, particularly when
2612   integrity checks or digital signatures are used to ensure that the payload
2613   received is identical to the original.
2614</t>
2615<t>
2616   If a proxy receives a request-target with a host name that is not a
2617   fully qualified domain name, it MAY add its own domain to the host name
2618   it received when forwarding the request.  A proxy MUST NOT change the
2619   host name if it is a fully qualified domain name.
2620</t>
2621<t>
2622   A proxy MUST NOT modify the "absolute-path" and "query" parts of the
2623   received request-target when forwarding it to the next inbound server,
2624   except as noted above to replace an empty path with "/" or "*".
2625</t>
2626<t>
2627   A proxy MUST NOT modify header fields that provide information about the
2628   end points of the communication chain, the resource state, or the selected
2629   representation. A proxy MAY change the message body through application
2630   or removal of a transfer coding (<xref target="transfer.codings"/>).
2631</t>
2632<t>
2633   A non-transforming proxy MUST NOT modify the message payload (Section 3.3 of <xref target="Part2"/>).
2634   A transforming proxy MUST NOT modify the payload of a message that
2635   contains the no-transform cache-control directive.
2636</t>
2637<t>
2638   A transforming proxy MAY transform the payload of a message
2639   that does not contain the no-transform cache-control directive;
2640   if the payload is transformed, the transforming proxy MUST add a
2641   Warning header field with the warn-code of 214 ("Transformation Applied")
2642   if one does not already appear in the message (see Section 7.5 of <xref target="Part6"/>).
2643   If the payload of a 200 (OK) response is transformed, the
2644   transforming proxy can also inform downstream recipients that a
2645   transformation has been applied by changing the response status code to
2646   203 (Non-Authoritative Information) (Section 6.3.4 of <xref target="Part2"/>).
2647</t>
2648</section>
2649</section>
2650</section>
2651
2652<section title="Connection Management" anchor="connection.management">
2653<t>
2654   HTTP messaging is independent of the underlying transport or
2655   session-layer connection protocol(s).  HTTP only presumes a reliable
2656   transport with in-order delivery of requests and the corresponding
2657   in-order delivery of responses.  The mapping of HTTP request and
2658   response structures onto the data units of an underlying transport
2659   protocol is outside the scope of this specification.
2660</t>
2661<t>
2662   As described in <xref target="connecting.inbound"/>, the specific
2663   connection protocols to be used for an HTTP interaction are determined by
2664   client configuration and the <xref target="target-resource" format="none">target URI</xref>.
2665   For example, the "http" URI scheme
2666   (<xref target="http.uri"/>) indicates a default connection of TCP
2667   over IP, with a default TCP port of 80, but the client might be
2668   configured to use a proxy via some other connection, port, or protocol.
2669</t>
2670<t>
2671   HTTP implementations are expected to engage in connection management,
2672   which includes maintaining the state of current connections,
2673   establishing a new connection or reusing an existing connection,
2674   processing messages received on a connection, detecting connection
2675   failures, and closing each connection.
2676   Most clients maintain multiple connections in parallel, including
2677   more than one connection per server endpoint.
2678   Most servers are designed to maintain thousands of concurrent connections,
2679   while controlling request queues to enable fair use and detect
2680   denial of service attacks.
2681</t>
2682
2683<section title="Connection" anchor="header.connection">
2684  <iref primary="true" item="Connection header field"/>
2685  <iref primary="true" item="close"/>
2686 
2687 
2688 
2689<t>
2690   The "Connection" header field allows the sender to indicate desired
2691   control options for the current connection.  In order to avoid confusing
2692   downstream recipients, a proxy or gateway MUST remove or replace any
2693   received connection options before forwarding the message.
2694</t>
2695<t>
2696   When a header field aside from Connection is used to supply control
2697   information for or about the current connection, the sender MUST list
2698   the corresponding field-name within the "Connection" header field.
2699   A proxy or gateway MUST parse a received Connection
2700   header field before a message is forwarded and, for each
2701   connection-option in this field, remove any header field(s) from
2702   the message with the same name as the connection-option, and then
2703   remove the Connection header field itself (or replace it with the
2704   intermediary's own connection options for the forwarded message).
2705</t>
2706<t>
2707   Hence, the Connection header field provides a declarative way of
2708   distinguishing header fields that are only intended for the
2709   immediate recipient ("hop-by-hop") from those fields that are
2710   intended for all recipients on the chain ("end-to-end"), enabling the
2711   message to be self-descriptive and allowing future connection-specific
2712   extensions to be deployed without fear that they will be blindly
2713   forwarded by older intermediaries.
2714</t>
2715<t>
2716   The Connection header field's value has the following grammar:
2717</t>
2718<figure><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="connection-option"/><artwork type="abnf2616"><![CDATA[
2719  Connection        = 1#connection-option
2720  connection-option = token
2721]]></artwork></figure>
2722<t>
2723   Connection options are case-insensitive.
2724</t>
2725<t>
2726   A sender MUST NOT send a connection option corresponding to a header
2727   field that is intended for all recipients of the payload.
2728   For example, Cache-Control is never appropriate as a
2729   connection option (Section 7.2 of <xref target="Part6"/>).
2730</t>
2731<t>
2732   The connection options do not have to correspond to a header field
2733   present in the message, since a connection-specific header field
2734   might not be needed if there are no parameters associated with that
2735   connection option.  Recipients that trigger certain connection
2736   behavior based on the presence of connection options MUST do so
2737   based on the presence of the connection-option rather than only the
2738   presence of the optional header field.  In other words, if the
2739   connection option is received as a header field but not indicated
2740   within the Connection field-value, then the recipient MUST ignore
2741   the connection-specific header field because it has likely been
2742   forwarded by an intermediary that is only partially conformant.
2743</t>
2744<t>
2745   When defining new connection options, specifications ought to
2746   carefully consider existing deployed header fields and ensure
2747   that the new connection option does not share the same name as
2748   an unrelated header field that might already be deployed.
2749   Defining a new connection option essentially reserves that potential
2750   field-name for carrying additional information related to the
2751   connection option, since it would be unwise for senders to use
2752   that field-name for anything else.
2753</t>
2754<t>
2755   The "close" connection option is defined for a
2756   sender to signal that this connection will be closed after completion of
2757   the response. For example,
2758</t>
2759<figure><artwork type="example"><![CDATA[
2760  Connection: close
2761]]></artwork></figure>
2762<t>
2763   in either the request or the response header fields indicates that
2764   the connection MUST be closed after the current request/response
2765   is complete (<xref target="persistent.tear-down"/>).
2766</t>
2767<t>
2768   A client that does not support <xref target="persistent.connections" format="none">persistent connections</xref> MUST
2769   send the "close" connection option in every request message.
2770</t>
2771<t>
2772   A server that does not support <xref target="persistent.connections" format="none">persistent connections</xref> MUST
2773   send the "close" connection option in every response message that
2774   does not have a 1xx (Informational) status code.
2775</t>
2776</section>
2777
2778<section title="Establishment" anchor="persistent.establishment">
2779<t>
2780   It is beyond the scope of this specification to describe how connections
2781   are established via various transport or session-layer protocols.
2782   Each connection applies to only one transport link.
2783</t>
2784</section>
2785
2786<section title="Persistence" anchor="persistent.connections">
2787   
2788<t>
2789   HTTP/1.1 defaults to the use of "persistent connections",
2790   allowing multiple requests and responses to be carried over a single
2791   connection. The "<xref target="header.connection" format="none">close</xref>" connection-option is used to signal
2792   that a connection will not persist after the current request/response.
2793   HTTP implementations SHOULD support persistent connections.
2794</t>
2795<t>
2796   A recipient determines whether a connection is persistent or not based on
2797   the most recently received message's protocol version and
2798   <xref target="header.connection" format="none">Connection</xref> header field (if any):
2799   <list style="symbols">
2800     <t>If the <xref target="header.connection" format="none">close</xref> connection option is present, the
2801        connection will not persist after the current response; else,</t>
2802     <t>If the received protocol is HTTP/1.1 (or later), the connection will
2803        persist after the current response; else,</t>
2804     <t>If the received protocol is HTTP/1.0, the "keep-alive"
2805        connection option is present, the recipient is not a proxy, and
2806        the recipient wishes to honor the HTTP/1.0 "keep-alive" mechanism,
2807        the connection will persist after the current response; otherwise,</t>
2808     <t>The connection will close after the current response.</t>
2809   </list>
2810</t>
2811<t>
2812   A server MAY assume that an HTTP/1.1 client intends to maintain a
2813   persistent connection until a <xref target="header.connection" format="none">close</xref> connection option
2814   is received in a request.
2815</t>
2816<t>
2817   A client MAY reuse a persistent connection until it sends or receives
2818   a <xref target="header.connection" format="none">close</xref> connection option or receives an HTTP/1.0 response
2819   without a "keep-alive" connection option.
2820</t>
2821<t>
2822   In order to remain persistent, all messages on a connection MUST
2823   have a self-defined message length (i.e., one not defined by closure
2824   of the connection), as described in <xref target="message.body"/>.
2825   A server MUST read the entire request message body or close
2826   the connection after sending its response, since otherwise the
2827   remaining data on a persistent connection would be misinterpreted
2828   as the next request.  Likewise,
2829   a client MUST read the entire response message body if it intends
2830   to reuse the same connection for a subsequent request.
2831</t>
2832<t>
2833   A proxy server MUST NOT maintain a persistent connection with an
2834   HTTP/1.0 client (see Section 19.7.1 of <xref target="RFC2068"/> for
2835   information and discussion of the problems with the Keep-Alive header field
2836   implemented by many HTTP/1.0 clients).
2837</t>
2838<t>
2839   Clients and servers SHOULD NOT assume that a persistent connection is
2840   maintained for HTTP versions less than 1.1 unless it is explicitly
2841   signaled.
2842   See <xref target="compatibility.with.http.1.0.persistent.connections"/>
2843   for more information on backward compatibility with HTTP/1.0 clients.
2844</t>
2845
2846<section title="Retrying Requests" anchor="persistent.retrying.requests">
2847<t>
2848   Connections can be closed at any time, with or without intention.
2849   Implementations ought to anticipate the need to recover
2850   from asynchronous close events.
2851</t>
2852<t>
2853   When an inbound connection is closed prematurely, a client MAY open a new
2854   connection and automatically retransmit an aborted sequence of requests if
2855   all of those requests have idempotent methods (Section 4.2.2 of <xref target="Part2"/>).
2856   A proxy MUST NOT automatically retry non-idempotent requests.
2857</t>
2858<t>
2859   A user agent MUST NOT automatically retry a request with a non-idempotent
2860   method unless it has some means to know that the request semantics are
2861   actually idempotent, regardless of the method, or some means to detect that
2862   the original request was never applied. For example, a user agent that
2863   knows (through design or configuration) that a POST request to a given
2864   resource is safe can repeat that request automatically.
2865   Likewise, a user agent designed specifically to operate on a version
2866   control repository might be able to recover from partial failure conditions
2867   by checking the target resource revision(s) after a failed connection,
2868   reverting or fixing any changes that were partially applied, and then
2869   automatically retrying the requests that failed.
2870</t>
2871<t>
2872   An automatic retry SHOULD NOT be repeated if it fails.
2873</t>
2874</section>
2875
2876<section title="Pipelining" anchor="pipelining">
2877   
2878<t>
2879   A client that supports persistent connections MAY "pipeline"
2880   its requests (i.e., send multiple requests without waiting for each
2881   response). A server MAY process a sequence of pipelined requests in
2882   parallel if they all have safe methods (Section 4.2.1 of <xref target="Part2"/>), but MUST send
2883   the corresponding responses in the same order that the requests were
2884   received.
2885</t>
2886<t>
2887   A client that pipelines requests MUST be prepared to retry those
2888   requests if the connection closes before it receives all of the
2889   corresponding responses. A client that assumes a persistent connection and
2890   pipelines immediately after connection establishment MUST NOT pipeline
2891   on a retry connection until it knows the connection is persistent.
2892</t>
2893<t>
2894   Idempotent methods (Section 4.2.2 of <xref target="Part2"/>) are significant to pipelining
2895   because they can be automatically retried after a connection failure.
2896   A user agent SHOULD NOT pipeline requests after a non-idempotent method
2897   until the final response status code for that method has been received,
2898   unless the user agent has a means to detect and recover from partial
2899   failure conditions involving the pipelined sequence.
2900</t>
2901<t>
2902   An intermediary that receives pipelined requests MAY pipeline those
2903   requests when forwarding them inbound, since it can rely on the outbound
2904   user agent(s) to determine what requests can be safely pipelined. If the
2905   inbound connection fails before receiving a response, the pipelining
2906   intermediary MAY attempt to retry a sequence of requests that have yet
2907   to receive a response if the requests all have idempotent methods;
2908   otherwise, the pipelining intermediary SHOULD forward any received
2909   responses and then close the corresponding outbound connection(s) so that
2910   the outbound user agent(s) can recover accordingly.
2911</t>
2912</section>
2913</section>
2914   
2915<section title="Concurrency" anchor="persistent.concurrency">
2916<t>
2917   Clients SHOULD limit the number of simultaneous
2918   connections that they maintain to a given server.
2919</t>
2920<t>
2921   Previous revisions of HTTP gave a specific number of connections as a
2922   ceiling, but this was found to be impractical for many applications. As a
2923   result, this specification does not mandate a particular maximum number of
2924   connections, but instead encourages clients to be conservative when opening
2925   multiple connections.
2926</t>
2927<t>
2928   Multiple connections are typically used to avoid the "head-of-line
2929   blocking" problem, wherein a request that takes significant server-side
2930   processing and/or has a large payload blocks subsequent requests on the
2931   same connection. However, each connection consumes server resources.
2932   Furthermore, using multiple connections can cause undesirable side effects
2933   in congested networks.
2934</t>
2935<t>
2936   Note that servers might reject traffic that they deem abusive, including an
2937   excessive number of connections from a client.
2938</t>
2939</section>
2940
2941<section title="Failures and Time-outs" anchor="persistent.failures">
2942<t>
2943   Servers will usually have some time-out value beyond which they will
2944   no longer maintain an inactive connection. Proxy servers might make
2945   this a higher value since it is likely that the client will be making
2946   more connections through the same server. The use of persistent
2947   connections places no requirements on the length (or existence) of
2948   this time-out for either the client or the server.
2949</t>
2950<t>
2951   When a client or server wishes to time-out it SHOULD issue a graceful
2952   close on the transport connection. Clients and servers SHOULD both
2953   constantly watch for the other side of the transport close, and
2954   respond to it as appropriate. If a client or server does not detect
2955   the other side's close promptly it could cause unnecessary resource
2956   drain on the network.
2957</t>
2958<t>
2959   A client, server, or proxy MAY close the transport connection at any
2960   time. For example, a client might have started to send a new request
2961   at the same time that the server has decided to close the "idle"
2962   connection. From the server's point of view, the connection is being
2963   closed while it was idle, but from the client's point of view, a
2964   request is in progress.
2965</t>
2966<t>
2967   Servers SHOULD maintain persistent connections and allow the underlying
2968   transport's flow control mechanisms to resolve temporary overloads, rather
2969   than terminate connections with the expectation that clients will retry.
2970   The latter technique can exacerbate network congestion.
2971</t>
2972<t>
2973   A client sending a message body SHOULD monitor
2974   the network connection for an error response while it is transmitting
2975   the request. If the client sees an error response, it SHOULD
2976   immediately cease transmitting the body and close the connection.
2977</t>
2978</section>
2979   
2980<section title="Tear-down" anchor="persistent.tear-down">
2981  <iref primary="false" item="Connection header field"/>
2982  <iref primary="false" item="close"/>
2983<t>
2984   The <xref target="header.connection" format="none">Connection</xref> header field
2985   (<xref target="header.connection"/>) provides a "<xref target="header.connection" format="none">close</xref>"
2986   connection option that a sender SHOULD send when it wishes to close
2987   the connection after the current request/response pair.
2988</t>
2989<t>
2990   A client that sends a <xref target="header.connection" format="none">close</xref> connection option MUST NOT
2991   send further requests on that connection (after the one containing
2992   <xref target="header.connection" format="none">close</xref>) and MUST close the connection after reading the
2993   final response message corresponding to this request.
2994</t>
2995<t>
2996   A server that receives a <xref target="header.connection" format="none">close</xref> connection option MUST
2997   initiate a close of the connection (see below) after it sends the
2998   final response to the request that contained <xref target="header.connection" format="none">close</xref>.
2999   The server SHOULD send a <xref target="header.connection" format="none">close</xref> connection option
3000   in its final response on that connection. The server MUST NOT process
3001   any further requests received on that connection.
3002</t>
3003<t>
3004   A server that sends a <xref target="header.connection" format="none">close</xref> connection option MUST
3005   initiate a close of the connection (see below) after it sends the
3006   response containing <xref target="header.connection" format="none">close</xref>. The server MUST NOT process
3007   any further requests received on that connection.
3008</t>
3009<t>
3010   A client that receives a <xref target="header.connection" format="none">close</xref> connection option MUST
3011   cease sending requests on that connection and close the connection
3012   after reading the response message containing the close; if additional
3013   pipelined requests had been sent on the connection, the client SHOULD NOT
3014   assume that they will be processed by the server.
3015</t>
3016<t>
3017   If a server performs an immediate close of a TCP connection, there is a
3018   significant risk that the client will not be able to read the last HTTP
3019   response.  If the server receives additional data from the client on a
3020   fully-closed connection, such as another request that was sent by the
3021   client before receiving the server's response, the server's TCP stack will
3022   send a reset packet to the client; unfortunately, the reset packet might
3023   erase the client's unacknowledged input buffers before they can be read
3024   and interpreted by the client's HTTP parser.
3025</t>
3026<t>
3027   To avoid the TCP reset problem, servers typically close a connection in
3028   stages. First, the server performs a half-close by closing only the write
3029   side of the read/write connection. The server then continues to read from
3030   the connection until it receives a corresponding close by the client, or
3031   until the server is reasonably certain that its own TCP stack has received
3032   the client's acknowledgement of the packet(s) containing the server's last
3033   response. Finally, the server fully closes the connection.
3034</t>
3035<t>
3036   It is unknown whether the reset problem is exclusive to TCP or might also
3037   be found in other transport connection protocols.
3038</t>
3039</section>
3040
3041<section title="Upgrade" anchor="header.upgrade">
3042  <iref primary="true" item="Upgrade header field"/>
3043 
3044 
3045 
3046 
3047<t>
3048   The "Upgrade" header field is intended to provide a simple mechanism
3049   for transitioning from HTTP/1.1 to some other protocol on the same
3050   connection.  A client MAY send a list of protocols in the Upgrade
3051   header field of a request to invite the server to switch to one or
3052   more of those protocols, in order of descending preference, before sending
3053   the final response. A server MAY ignore a received Upgrade header field
3054   if it wishes to continue using the current protocol on that connection.
3055</t>
3056<figure><iref primary="true" item="Grammar" subitem="Upgrade"/><artwork type="abnf2616"><![CDATA[
3057  Upgrade          = 1#protocol
3058
3059  protocol         = protocol-name ["/" protocol-version]
3060  protocol-name    = token
3061  protocol-version = token
3062]]></artwork></figure>
3063<t>
3064   A server that sends a 101 (Switching Protocols) response
3065   MUST send an Upgrade header field to indicate the new protocol(s) to
3066   which the connection is being switched; if multiple protocol layers are
3067   being switched, the new protocols MUST be listed in layer-ascending
3068   order. A server MUST NOT switch to a protocol that was not indicated by
3069   the client in the corresponding request's Upgrade header field.
3070   A server MAY choose to ignore the order of preference indicated by the
3071   client and select the new protocol(s) based on other factors, such as the
3072   nature of the request or the current load on the server.
3073</t>
3074<t>
3075   A server that sends a 426 (Upgrade Required) response
3076   MUST send an Upgrade header field to indicate the acceptable protocols,
3077   in order of descending preference.
3078</t>
3079<t>
3080   A server MAY send an Upgrade header field in any other response to
3081   advertise that it implements support for upgrading to the listed protocols,
3082   in order of descending preference, when appropriate for a future request.
3083</t>
3084<figure><preamble>
3085   The following is a hypothetical example sent by a client:
3086</preamble><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
3087  GET /hello.txt HTTP/1.1
3088  Host: www.example.com
3089  Connection: upgrade
3090  Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
3091 
3092  ]]></artwork></figure>
3093<t>
3094   Upgrade cannot be used to insist on a protocol change; its acceptance and
3095   use by the server is optional. The capabilities and nature of the
3096   application-level communication after the protocol change is entirely
3097   dependent upon the new protocol(s) chosen, although the first action
3098   after changing the protocol MUST be a response to the initial HTTP
3099   request that contained the Upgrade header field.
3100</t>
3101<t>
3102   For example, if the Upgrade header field is received in a GET request
3103   and the server decides to switch protocols, it first responds
3104   with a 101 (Switching Protocols) message in HTTP/1.1 and
3105   then immediately follows that with the new protocol's equivalent of a
3106   response to a GET on the target resource.  This allows a connection to be
3107   upgraded to protocols with the same semantics as HTTP without the
3108   latency cost of an additional round-trip.  A server MUST NOT switch
3109   protocols unless the received message semantics can be honored by the new
3110   protocol; an OPTIONS request can be honored by any protocol.
3111</t>
3112<figure><preamble>
3113   The following is an example response to the above hypothetical request:
3114</preamble><artwork type="message/http; msgtype=&#34;response&#34;"><![CDATA[
3115  HTTP/1.1 101 Switching Protocols
3116  Connection: upgrade
3117  Upgrade: HTTP/2.0
3118 
3119  [... data stream switches to HTTP/2.0 with an appropriate response
3120  (as defined by new protocol) to the "GET /hello.txt" request ...]
3121  ]]></artwork></figure>
3122<t>
3123   When Upgrade is sent, the sender MUST also send a
3124   <xref target="header.connection" format="none">Connection</xref> header field (<xref target="header.connection"/>)
3125   that contains an "upgrade" connection option, in order to prevent Upgrade
3126   from being accidentally forwarded by intermediaries that might not implement
3127   the listed protocols.  A server MUST ignore an Upgrade header field that
3128   is received in an HTTP/1.0 request.
3129</t>
3130<t>
3131   The Upgrade header field only applies to switching protocols on top of the
3132   existing connection; it cannot be used to switch the underlying connection
3133   (transport) protocol, nor to switch the existing communication to a
3134   different connection. For those purposes, it is more appropriate to use a
3135   3xx (Redirection) response (Section 6.4 of <xref target="Part2"/>).
3136</t>
3137<t>
3138   This specification only defines the protocol name "HTTP" for use by
3139   the family of Hypertext Transfer Protocols, as defined by the HTTP
3140   version rules of <xref target="http.version"/> and future updates to this
3141   specification. Additional tokens ought to be registered with IANA using the
3142   registration procedure defined in <xref target="upgrade.token.registry"/>.
3143</t>
3144</section>
3145</section>
3146
3147<section title="IANA Considerations" anchor="IANA.considerations">
3148
3149<section title="Header Field Registration" anchor="header.field.registration">
3150<t>
3151   HTTP header fields are registered within the Message Header Field Registry
3152   maintained at
3153   <eref target="http://www.iana.org/assignments/message-headers/message-header-index.html"/>.
3154</t>
3155<t>
3156   This document defines the following HTTP header fields, so their
3157   associated registry entries shall be updated according to the permanent
3158   registrations below (see <xref target="BCP90"/>):
3159</t>
3160
3161<!--AUTOGENERATED FROM extract-header-defs.xslt, do not edit manually-->
3162<texttable align="left" suppress-title="true" anchor="iana.header.registration.table">
3163   <ttcol>Header Field Name</ttcol>
3164   <ttcol>Protocol</ttcol>
3165   <ttcol>Status</ttcol>
3166   <ttcol>Reference</ttcol>
3167
3168   <c>Connection</c>
3169   <c>http</c>
3170   <c>standard</c>
3171   <c>
3172      <xref target="header.connection"/>
3173   </c>
3174   <c>Content-Length</c>
3175   <c>http</c>
3176   <c>standard</c>
3177   <c>
3178      <xref target="header.content-length"/>
3179   </c>
3180   <c>Host</c>
3181   <c>http</c>
3182   <c>standard</c>
3183   <c>
3184      <xref target="header.host"/>
3185   </c>
3186   <c>TE</c>
3187   <c>http</c>
3188   <c>standard</c>
3189   <c>
3190      <xref target="header.te"/>
3191   </c>
3192   <c>Trailer</c>
3193   <c>http</c>
3194   <c>standard</c>
3195   <c>
3196      <xref target="header.trailer"/>
3197   </c>
3198   <c>Transfer-Encoding</c>
3199   <c>http</c>
3200   <c>standard</c>
3201   <c>
3202      <xref target="header.transfer-encoding"/>
3203   </c>
3204   <c>Upgrade</c>
3205   <c>http</c>
3206   <c>standard</c>
3207   <c>
3208      <xref target="header.upgrade"/>
3209   </c>
3210   <c>Via</c>
3211   <c>http</c>
3212   <c>standard</c>
3213   <c>
3214      <xref target="header.via"/>
3215   </c>
3216</texttable>
3217<!--(END)-->
3218
3219<t>
3220   Furthermore, the header field-name "Close" shall be registered as
3221   "reserved", since using that name as an HTTP header field might
3222   conflict with the "close" connection option of the "<xref target="header.connection" format="none">Connection</xref>"
3223   header field (<xref target="header.connection"/>).
3224</t>
3225<texttable align="left" suppress-title="true">
3226   <ttcol>Header Field Name</ttcol>
3227   <ttcol>Protocol</ttcol>
3228   <ttcol>Status</ttcol>
3229   <ttcol>Reference</ttcol>
3230
3231   <c>Close</c>
3232   <c>http</c>
3233   <c>reserved</c>
3234   <c>
3235      <xref target="header.field.registration"/>
3236   </c>
3237</texttable>
3238<t>
3239   The change controller is: "IETF (iesg@ietf.org) - Internet Engineering Task Force".
3240</t>
3241</section>
3242
3243<section title="URI Scheme Registration" anchor="uri.scheme.registration">
3244<t>
3245   IANA maintains the registry of URI Schemes <xref target="BCP115"/> at
3246   <eref target="http://www.iana.org/assignments/uri-schemes.html"/>.
3247</t>
3248<t>
3249   This document defines the following URI schemes, so their
3250   associated registry entries shall be updated according to the permanent
3251   registrations below:
3252</t>
3253<texttable align="left" suppress-title="true">
3254   <ttcol>URI Scheme</ttcol>
3255   <ttcol>Description</ttcol>
3256   <ttcol>Reference</ttcol>
3257
3258   <c>http</c>
3259   <c>Hypertext Transfer Protocol</c>
3260   <c><xref target="http.uri"/></c>
3261
3262   <c>https</c>
3263   <c>Hypertext Transfer Protocol Secure</c>
3264   <c><xref target="https.uri"/></c>
3265</texttable>
3266</section>
3267
3268<section title="Internet Media Type Registration" anchor="internet.media.type.http">
3269<t>
3270   This document serves as the specification for the Internet media types
3271   "message/http" and "application/http". The following is to be registered with
3272   IANA (see <xref target="BCP13"/>).
3273</t>
3274<section title="Internet Media Type message/http" anchor="internet.media.type.message.http">
3275<iref item="Media Type" subitem="message/http" primary="true"/>
3276<iref item="message/http Media Type" primary="true"/>
3277<t>
3278   The message/http type can be used to enclose a single HTTP request or
3279   response message, provided that it obeys the MIME restrictions for all
3280   "message" types regarding line length and encodings.
3281</t>
3282<t>
3283  <list style="hanging">
3284    <t hangText="Type name:">
3285      message
3286    </t>
3287    <t hangText="Subtype name:">
3288      http
3289    </t>
3290    <t hangText="Required parameters:">
3291      none
3292    </t>
3293    <t hangText="Optional parameters:">
3294      version, msgtype
3295      <list style="hanging">
3296        <t hangText="version:">
3297          The HTTP-version number of the enclosed message
3298          (e.g., "1.1"). If not present, the version can be
3299          determined from the first line of the body.
3300        </t>
3301        <t hangText="msgtype:">
3302          The message type — "request" or "response". If not
3303          present, the type can be determined from the first
3304          line of the body.
3305        </t>
3306      </list>
3307    </t>
3308    <t hangText="Encoding considerations:">
3309      only "7bit", "8bit", or "binary" are permitted
3310    </t>
3311    <t hangText="Security considerations:">
3312      none
3313    </t>
3314    <t hangText="Interoperability considerations:">
3315      none
3316    </t>
3317    <t hangText="Published specification:">
3318      This specification (see <xref target="internet.media.type.message.http"/>).
3319    </t>
3320    <t hangText="Applications that use this media type:">
3321    </t>
3322    <t hangText="Additional information:">
3323      <list style="hanging">
3324        <t hangText="Magic number(s):">none</t>
3325        <t hangText="File extension(s):">none</t>
3326        <t hangText="Macintosh file type code(s):">none</t>
3327      </list>
3328    </t>
3329    <t hangText="Person and email address to contact for further information:">
3330      See Authors Section.
3331    </t>
3332    <t hangText="Intended usage:">
3333      COMMON
3334    </t>
3335    <t hangText="Restrictions on usage:">
3336      none
3337    </t>
3338    <t hangText="Author:">
3339      See Authors Section.
3340    </t>
3341    <t hangText="Change controller:">
3342      IESG
3343    </t>
3344  </list>
3345</t>
3346</section>
3347<section title="Internet Media Type application/http" anchor="internet.media.type.application.http">
3348<iref item="Media Type" subitem="application/http" primary="true"/>
3349<iref item="application/http Media Type" primary="true"/>
3350<t>
3351   The application/http type can be used to enclose a pipeline of one or more
3352   HTTP request or response messages (not intermixed).
3353</t>
3354<t>
3355  <list style="hanging">
3356    <t hangText="Type name:">
3357      application
3358    </t>
3359    <t hangText="Subtype name:">
3360      http
3361    </t>
3362    <t hangText="Required parameters:">
3363      none
3364    </t>
3365    <t hangText="Optional parameters:">
3366      version, msgtype
3367      <list style="hanging">
3368        <t hangText="version:">
3369          The HTTP-version number of the enclosed messages
3370          (e.g., "1.1"). If not present, the version can be
3371          determined from the first line of the body.
3372        </t>
3373        <t hangText="msgtype:">
3374          The message type — "request" or "response". If not
3375          present, the type can be determined from the first
3376          line of the body.
3377        </t>
3378      </list>
3379    </t>
3380    <t hangText="Encoding considerations:">
3381      HTTP messages enclosed by this type
3382      are in "binary" format; use of an appropriate
3383      Content-Transfer-Encoding is required when
3384      transmitted via E-mail.
3385    </t>
3386    <t hangText="Security considerations:">
3387      none
3388    </t>
3389    <t hangText="Interoperability considerations:">
3390      none
3391    </t>
3392    <t hangText="Published specification:">
3393      This specification (see <xref target="internet.media.type.application.http"/>).
3394    </t>
3395    <t hangText="Applications that use this media type:">
3396    </t>
3397    <t hangText="Additional information:">
3398      <list style="hanging">
3399        <t hangText="Magic number(s):">none</t>
3400        <t hangText="File extension(s):">none</t>
3401        <t hangText="Macintosh file type code(s):">none</t>
3402      </list>
3403    </t>
3404    <t hangText="Person and email address to contact for further information:">
3405      See Authors Section.
3406    </t>
3407    <t hangText="Intended usage:">
3408      COMMON
3409    </t>
3410    <t hangText="Restrictions on usage:">
3411      none
3412    </t>
3413    <t hangText="Author:">
3414      See Authors Section.
3415    </t>
3416    <t hangText="Change controller:">
3417      IESG
3418    </t>
3419  </list>
3420</t>
3421</section>
3422</section>
3423
3424<section title="Transfer Coding Registry" anchor="transfer.coding.registry">
3425<t>
3426   The HTTP Transfer Coding Registry defines the name space for transfer
3427   coding names. It is maintained at <eref target="http://www.iana.org/assignments/http-parameters"/>.
3428</t>
3429
3430<section title="Procedure" anchor="transfer.coding.registry.procedure">
3431<t>
3432   Registrations MUST include the following fields:
3433   <list style="symbols">
3434     <t>Name</t>
3435     <t>Description</t>
3436     <t>Pointer to specification text</t>
3437   </list>
3438</t>
3439<t>
3440   Names of transfer codings MUST NOT overlap with names of content codings
3441   (Section 3.1.2.1 of <xref target="Part2"/>) unless the encoding transformation is identical, as
3442   is the case for the compression codings defined in
3443   <xref target="compression.codings"/>.
3444</t>
3445<t>
3446   Values to be added to this name space require IETF Review (see
3447   Section 4.1 of <xref target="RFC5226"/>), and MUST
3448   conform to the purpose of transfer coding defined in this specification.
3449</t>
3450<t>
3451   Use of program names for the identification of encoding formats
3452   is not desirable and is discouraged for future encodings.
3453</t>
3454</section>
3455
3456<section title="Registration" anchor="transfer.coding.registration">
3457<t>
3458   The HTTP Transfer Coding Registry shall be updated with the registrations
3459   below:
3460</t>
3461<texttable align="left" suppress-title="true" anchor="iana.transfer.coding.registration.table">
3462   <ttcol>Name</ttcol>
3463   <ttcol>Description</ttcol>
3464   <ttcol>Reference</ttcol>
3465   <c>chunked</c>
3466   <c>Transfer in a series of chunks</c>
3467   <c>
3468      <xref target="chunked.encoding"/>
3469   </c>
3470   <c>compress</c>
3471   <c>UNIX "compress" data format <xref target="Welch"/></c>
3472   <c>
3473      <xref target="compress.coding"/>
3474   </c>
3475   <c>deflate</c>
3476   <c>"deflate" compressed data (<xref target="RFC1951"/>) inside
3477   the "zlib" data format (<xref target="RFC1950"/>)
3478   </c>
3479   <c>
3480      <xref target="deflate.coding"/>
3481   </c>
3482   <c>gzip</c>
3483   <c>GZIP file format <xref target="RFC1952"/></c>
3484   <c>
3485      <xref target="gzip.coding"/>
3486   </c>
3487   <c>x-compress</c>
3488   <c>Deprecated (alias for compress)</c>
3489   <c>
3490      <xref target="compress.coding"/>
3491   </c>
3492   <c>x-gzip</c>
3493   <c>Deprecated (alias for gzip)</c>
3494   <c>
3495      <xref target="gzip.coding"/>
3496   </c>
3497</texttable>
3498</section>
3499</section>
3500
3501<section title="Upgrade Token Registry" anchor="upgrade.token.registry">
3502<t>
3503   The HTTP Upgrade Token Registry defines the name space for protocol-name
3504   tokens used to identify protocols in the <xref target="header.upgrade" format="none">Upgrade</xref> header
3505   field. The registry is maintained at <eref target="http://www.iana.org/assignments/http-upgrade-tokens"/>.
3506</t>
3507
3508<section title="Procedure" anchor="upgrade.token.registry.procedure">  
3509<t>
3510   Each registered protocol name is associated with contact information
3511   and an optional set of specifications that details how the connection
3512   will be processed after it has been upgraded.
3513</t>
3514<t>
3515   Registrations happen on a "First Come First Served" basis (see
3516   Section 4.1 of <xref target="RFC5226"/>) and are subject to the
3517   following rules:
3518  <list style="numbers">
3519    <t>A protocol-name token, once registered, stays registered forever.</t>
3520    <t>The registration MUST name a responsible party for the
3521       registration.</t>
3522    <t>The registration MUST name a point of contact.</t>
3523    <t>The registration MAY name a set of specifications associated with
3524       that token. Such specifications need not be publicly available.</t>
3525    <t>The registration SHOULD name a set of expected "protocol-version"
3526       tokens associated with that token at the time of registration.</t>
3527    <t>The responsible party MAY change the registration at any time.
3528       The IANA will keep a record of all such changes, and make them
3529       available upon request.</t>
3530    <t>The IESG MAY reassign responsibility for a protocol token.
3531       This will normally only be used in the case when a
3532       responsible party cannot be contacted.</t>
3533  </list>
3534</t>
3535<t>
3536   This registration procedure for HTTP Upgrade Tokens replaces that
3537   previously defined in Section 7.2 of <xref target="RFC2817"/>.
3538</t>
3539</section>
3540
3541<section title="Upgrade Token Registration" anchor="upgrade.token.registration">
3542<t>
3543   The HTTP Upgrade Token Registry shall be updated with the registration
3544   below:
3545</t>
3546<texttable align="left" suppress-title="true">
3547   <ttcol>Value</ttcol>
3548   <ttcol>Description</ttcol>
3549   <ttcol>Expected Version Tokens</ttcol>
3550   <ttcol>Reference</ttcol>
3551
3552   <c>HTTP</c>
3553   <c>Hypertext Transfer Protocol</c>
3554   <c>any DIGIT.DIGIT (e.g, "2.0")</c>
3555   <c><xref target="http.version"/></c>
3556</texttable>
3557<t>
3558   The responsible party is: "IETF (iesg@ietf.org) - Internet Engineering Task Force".
3559</t>
3560</section>
3561</section>
3562
3563</section>
3564
3565<section title="Security Considerations" anchor="security.considerations">
3566<t>
3567   This section is meant to inform developers, information providers, and
3568   users of known security concerns relevant to HTTP/1.1 message syntax,
3569   parsing, and routing.
3570</t>
3571
3572<section title="DNS-related Attacks" anchor="dns.related.attacks">
3573<t>
3574   HTTP clients rely heavily on the Domain Name Service (DNS), and are thus
3575   generally prone to security attacks based on the deliberate misassociation
3576   of IP addresses and DNS names not protected by DNSSEC. Clients need to be
3577   cautious in assuming the validity of an IP number/DNS name association unless
3578   the response is protected by DNSSEC (<xref target="RFC4033"/>).
3579</t>
3580</section>
3581
3582<section title="Intermediaries and Caching" anchor="attack.intermediaries">
3583<t>
3584   By their very nature, HTTP intermediaries are men-in-the-middle, and
3585   represent an opportunity for man-in-the-middle attacks. Compromise of
3586   the systems on which the intermediaries run can result in serious security
3587   and privacy problems. Intermediaries have access to security-related
3588   information, personal information about individual users and
3589   organizations, and proprietary information belonging to users and
3590   content providers. A compromised intermediary, or an intermediary
3591   implemented or configured without regard to security and privacy
3592   considerations, might be used in the commission of a wide range of
3593   potential attacks.
3594</t>
3595<t>
3596   Intermediaries that contain a shared cache are especially vulnerable
3597   to cache poisoning attacks.
3598</t>
3599<t>
3600   Implementers need to consider the privacy and security
3601   implications of their design and coding decisions, and of the
3602   configuration options they provide to operators (especially the
3603   default configuration).
3604</t>
3605<t>
3606   Users need to be aware that intermediaries are no more trustworthy than
3607   the people who run them; HTTP itself cannot solve this problem.
3608</t>
3609</section>
3610
3611<section title="Buffer Overflows" anchor="attack.protocol.element.size.overflows">
3612<t>
3613   Because HTTP uses mostly textual, character-delimited fields, attackers can
3614   overflow buffers in implementations, and/or perform a Denial of Service
3615   against implementations that accept fields with unlimited lengths.
3616</t>
3617<t>
3618   To promote interoperability, this specification makes specific
3619   recommendations for minimum size limits on request-line
3620   (<xref target="request.line"/>)
3621   and blocks of header fields (<xref target="header.fields"/>). These are
3622   minimum recommendations, chosen to be supportable even by implementations
3623   with limited resources; it is expected that most implementations will
3624   choose substantially higher limits.
3625</t>
3626<t>
3627   This specification also provides a way for servers to reject messages that
3628   have request-targets that are too long (Section 6.5.12 of <xref target="Part2"/>) or request entities
3629   that are too large (Section 6.5 of <xref target="Part2"/>). Additional status codes related to
3630   capacity limits have been defined by extensions to HTTP
3631   <xref target="RFC6585"/>.
3632</t>
3633<t>
3634   Recipients SHOULD carefully limit the extent to which they read other
3635   fields, including (but not limited to) request methods, response status
3636   phrases, header field-names, and body chunks, so as to avoid denial of
3637   service attacks without impeding interoperability.
3638</t>
3639</section>
3640
3641<section title="Message Integrity" anchor="message.integrity">
3642<t>
3643   HTTP does not define a specific mechanism for ensuring message integrity,
3644   instead relying on the error-detection ability of underlying transport
3645   protocols and the use of length or chunk-delimited framing to detect
3646   completeness. Additional integrity mechanisms, such as hash functions or
3647   digital signatures applied to the content, can be selectively added to
3648   messages via extensible metadata header fields. Historically, the lack of
3649   a single integrity mechanism has been justified by the informal nature of
3650   most HTTP communication.  However, the prevalence of HTTP as an information
3651   access mechanism has resulted in its increasing use within environments
3652   where verification of message integrity is crucial.
3653</t>
3654<t>
3655   User agents are encouraged to implement configurable means for detecting
3656   and reporting failures of message integrity such that those means can be
3657   enabled within environments for which integrity is necessary. For example,
3658   a browser being used to view medical history or drug interaction
3659   information needs to indicate to the user when such information is detected
3660   by the protocol to be incomplete, expired, or corrupted during transfer.
3661   Such mechanisms might be selectively enabled via user agent extensions or
3662   the presence of message integrity metadata in a response.
3663   At a minimum, user agents ought to provide some indication that allows a
3664   user to distinguish between a complete and incomplete response message
3665   (<xref target="incomplete.messages"/>) when such verification is desired.
3666</t>
3667</section>
3668
3669<section title="Server Log Information" anchor="abuse.of.server.log.information">
3670<t>
3671   A server is in the position to save personal data about a user's requests
3672   over time, which might identify their reading patterns or subjects of
3673   interest.  In particular, log information gathered at an intermediary
3674   often contains a history of user agent interaction, across a multitude
3675   of sites, that can be traced to individual users.
3676</t>
3677<t>
3678   HTTP log information is confidential in nature; its handling is often
3679   constrained by laws and regulations.  Log information needs to be securely
3680   stored and appropriate guidelines followed for its analysis.
3681   Anonymization of personal information within individual entries helps,
3682   but is generally not sufficient to prevent real log traces from being
3683   re-identified based on correlation with other access characteristics.
3684   As such, access traces that are keyed to a specific client should not
3685   be published even if the key is pseudonymous.
3686</t>
3687<t>
3688   To minimize the risk of theft or accidental publication, log information
3689   should be purged of personally identifiable information, including
3690   user identifiers, IP addresses, and user-provided query parameters,
3691   as soon as that information is no longer necessary to support operational
3692   needs for security, auditing, or fraud control.
3693</t>
3694</section>
3695</section>
3696
3697<section title="Acknowledgments" anchor="acks">
3698<t>
3699   This edition of HTTP/1.1 builds on the many contributions that went into
3700   <xref target="RFC1945" format="none">RFC 1945</xref>,
3701   <xref target="RFC2068" format="none">RFC 2068</xref>,
3702   <xref target="RFC2145" format="none">RFC 2145</xref>, and
3703   <xref target="RFC2616" format="none">RFC 2616</xref>, including
3704   substantial contributions made by the previous authors, editors, and
3705   working group chairs: Tim Berners-Lee, Ari Luotonen, Roy T. Fielding,
3706   Henrik Frystyk Nielsen, Jim Gettys, Jeffrey C. Mogul, Larry Masinter,
3707   and Paul J. Leach. Mark Nottingham oversaw this effort as working group chair.
3708</t>
3709<t>
3710   Since 1999, the following contributors have helped improve the HTTP
3711   specification by reporting bugs, asking smart questions, drafting or
3712   reviewing text, and evaluating open issues:
3713</t>
3714
3715<t>Adam Barth,
3716Adam Roach,
3717Addison Phillips,
3718Adrian Chadd,
3719Adrien W. de Croy,
3720Alan Ford,
3721Alan Ruttenberg,
3722Albert Lunde,
3723Alek Storm,
3724Alex Rousskov,
3725Alexandre Morgaut,
3726Alexey Melnikov,
3727Alisha Smith,
3728Amichai Rothman,
3729Amit Klein,
3730Amos Jeffries,
3731Andreas Maier,
3732Andreas Petersson,
3733Anil Sharma,
3734Anne van Kesteren,
3735Anthony Bryan,
3736Asbjorn Ulsberg,
3737Ashok Kumar,
3738Balachander Krishnamurthy,
3739Barry Leiba,
3740Ben Laurie,
3741Benjamin Carlyle,
3742Benjamin Niven-Jenkins,
3743Bil Corry,
3744Bill Burke,
3745Bjoern Hoehrmann,
3746Bob Scheifler,
3747Boris Zbarsky,
3748Brett Slatkin,
3749Brian Kell,
3750Brian McBarron,
3751Brian Pane,
3752Brian Raymor,
3753Brian Smith,
3754Bryce Nesbitt,
3755Cameron Heavon-Jones,
3756Carl Kugler,
3757Carsten Bormann,
3758Charles Fry,
3759Chris Newman,
3760Cyrus Daboo,
3761Dale Robert Anderson,
3762Dan Wing,
3763Dan Winship,
3764Daniel Stenberg,
3765Darrel Miller,
3766Dave Cridland,
3767Dave Crocker,
3768Dave Kristol,
3769Dave Thaler,
3770David Booth,
3771David Singer,
3772David W. Morris,
3773Diwakar Shetty,
3774Dmitry Kurochkin,
3775Drummond Reed,
3776Duane Wessels,
3777Edward Lee,
3778Eitan Adler,
3779Eliot Lear,
3780Eran Hammer-Lahav,
3781Eric D. Williams,
3782Eric J. Bowman,
3783Eric Lawrence,
3784Eric Rescorla,
3785Erik Aronesty,
3786Evan Prodromou,
3787Felix Geisendoerfer,
3788Florian Weimer,
3789Frank Ellermann,
3790Fred Akalin,
3791Fred Bohle,
3792Frederic Kayser,
3793Gabor Molnar,
3794Gabriel Montenegro,
3795Geoffrey Sneddon,
3796Gervase Markham,
3797Gili Tzabari,
3798Grahame Grieve,
3799Greg Wilkins,
3800Grzegorz Calkowski,
3801Harald Tveit Alvestrand,
3802Harry Halpin,
3803Helge Hess,
3804Henrik Nordstrom,
3805Henry S. Thompson,
3806Henry Story,
3807Herbert van de Sompel,
3808Herve Ruellan,
3809Howard Melman,
3810Hugo Haas,
3811Ian Fette,
3812Ian Hickson,
3813Ido Safruti,
3814Ilari Liusvaara,
3815Ilya Grigorik,
3816Ingo Struck,
3817J. Ross Nicoll,
3818James Cloos,
3819James H. Manger,
3820James Lacey,
3821James M. Snell,
3822Jamie Lokier,
3823Jan Algermissen,
3824Jeff Hodges (who came up with the term 'effective Request-URI'),
3825Jeff Pinner,
3826Jeff Walden,
3827Jim Luther,
3828Jitu Padhye,
3829Joe D. Williams,
3830Joe Gregorio,
3831Joe Orton,
3832John C. Klensin,
3833John C. Mallery,
3834John Cowan,
3835John Kemp,
3836John Panzer,
3837John Schneider,
3838John Stracke,
3839John Sullivan,
3840Jonas Sicking,
3841Jonathan A. Rees,
3842Jonathan Billington,
3843Jonathan Moore,
3844Jonathan Silvera,
3845Jordi Ros,
3846Joris Dobbelsteen,
3847Josh Cohen,
3848Julien Pierre,
3849Jungshik Shin,
3850Justin Chapweske,
3851Justin Erenkrantz,
3852Justin James,
3853Kalvinder Singh,
3854Karl Dubost,
3855Keith Hoffman,
3856Keith Moore,
3857Ken Murchison,
3858Koen Holtman,
3859Konstantin Voronkov,
3860Kris Zyp,
3861Lisa Dusseault,
3862Maciej Stachowiak,
3863Manu Sporny,
3864Marc Schneider,
3865Marc Slemko,
3866Mark Baker,
3867Mark Pauley,
3868Mark Watson,
3869Markus Isomaki,
3870Markus Lanthaler,
3871Martin J. Duerst,
3872Martin Musatov,
3873Martin Nilsson,
3874Martin Thomson,
3875Matt Lynch,
3876Matthew Cox,
3877Max Clark,
3878Michael Burrows,
3879Michael Hausenblas,
3880Michael Sweet,
3881Mike Amundsen,
3882Mike Belshe,
3883Mike Bishop,
3884Mike Kelly,
3885Mike Schinkel,
3886Miles Sabin,
3887Murray S. Kucherawy,
3888Mykyta Yevstifeyev,
3889Nathan Rixham,
3890Nicholas Shanks,
3891Nico Williams,
3892Nicolas Alvarez,
3893Nicolas Mailhot,
3894Noah Slater,
3895Osama Mazahir,
3896Pablo Castro,
3897Pat Hayes,
3898Patrick R. McManus,
3899Paul E. Jones,
3900Paul Hoffman,
3901Paul Marquess,
3902Peter Lepeska,
3903Peter Occil,
3904Peter Saint-Andre,
3905Peter Watkins,
3906Phil Archer,
3907Philippe Mougin,
3908Phillip Hallam-Baker,
3909Piotr Dobrogost,
3910Poul-Henning Kamp,
3911Preethi Natarajan,
3912Rajeev Bector,
3913Ray Polk,
3914Reto Bachmann-Gmuer,
3915Richard Cyganiak,
3916Robby Simpson,
3917Robert Brewer,
3918Robert Collins,
3919Robert Mattson,
3920Robert O'Callahan,
3921Robert Olofsson,
3922Robert Sayre,
3923Robert Siemer,
3924Robert de Wilde,
3925Roberto Javier Godoy,
3926Roberto Peon,
3927Roland Zink,
3928Ronny Widjaja,
3929S. Mike Dierken,
3930Salvatore Loreto,
3931Sam Johnston,
3932Sam Pullara,
3933Sam Ruby,
3934Scott Lawrence (who maintained the original issues list),
3935Sean B. Palmer,
3936Shane McCarron,
3937Shigeki Ohtsu,
3938Stefan Eissing,
3939Stefan Tilkov,
3940Stefanos Harhalakis,
3941Stephane Bortzmeyer,
3942Stephen Farrell,
3943Stephen Ludin,
3944Stuart Williams,
3945Subbu Allamaraju,
3946Sylvain Hellegouarch,
3947Tapan Divekar,
3948Tatsuya Hayashi,
3949Ted Hardie,
3950Thomas Broyer,
3951Thomas Fossati,
3952Thomas Maslen,
3953Thomas Nordin,
3954Thomas Roessler,
3955Tim Bray,
3956Tim Morgan,
3957Tim Olsen,
3958Tom Zhou,
3959Travis Snoozy,
3960Tyler Close,
3961Vincent Murphy,
3962Wenbo Zhu,
3963Werner Baumann,
3964Wilbur Streett,
3965Wilfredo Sanchez Vega,
3966William A. Rowe Jr.,
3967William Chan,
3968Willy Tarreau,
3969Xiaoshu Wang,
3970Yaron Goland,
3971Yngve Nysaeter Pettersen,
3972Yoav Nir,
3973Yogesh Bang,
3974Yutaka Oiwa,
3975Yves Lafon (long-time member of the editor team),
3976Zed A. Shaw, and
3977Zhong Yu.
3978</t>
3979
3980<t>
3981   See Section 16 of <xref target="RFC2616"/> for additional
3982   acknowledgements from prior revisions.
3983</t>
3984</section>
3985
3986</middle>
3987<back>
3988
3989<references title="Normative References">
3990
3991<reference anchor="Part2">
3992  <front>
3993    <title>Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content</title>
3994    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3995      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
3996      <address><email>fielding@gbiv.com</email></address>
3997    </author>
3998    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3999      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4000      <address><email>julian.reschke@greenbytes.de</email></address>
4001    </author>
4002    <date month="July" year="2013"/>
4003  </front>
4004  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-23"/>
4005 
4006</reference>
4007
4008<reference anchor="Part4">
4009  <front>
4010    <title>Hypertext Transfer Protocol (HTTP/1.1): Conditional Requests</title>
4011    <author fullname="Roy T. Fielding" initials="R." role="editor" surname="Fielding">
4012      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4013      <address><email>fielding@gbiv.com</email></address>
4014    </author>
4015    <author fullname="Julian F. Reschke" initials="J. F." role="editor" surname="Reschke">
4016      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4017      <address><email>julian.reschke@greenbytes.de</email></address>
4018    </author>
4019    <date month="July" year="2013"/>
4020  </front>
4021  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p4-conditional-23"/>
4022 
4023</reference>
4024
4025<reference anchor="Part5">
4026  <front>
4027    <title>Hypertext Transfer Protocol (HTTP/1.1): Range Requests</title>
4028    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4029      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4030      <address><email>fielding@gbiv.com</email></address>
4031    </author>
4032    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4033      <organization abbrev="W3C">World Wide Web Consortium</organization>
4034      <address><email>ylafon@w3.org</email></address>
4035    </author>
4036    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4037      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4038      <address><email>julian.reschke@greenbytes.de</email></address>
4039    </author>
4040    <date month="July" year="2013"/>
4041  </front>
4042  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-23"/>
4043 
4044</reference>
4045
4046<reference anchor="Part6">
4047  <front>
4048    <title>Hypertext Transfer Protocol (HTTP/1.1): Caching</title>
4049    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4050      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4051      <address><email>fielding@gbiv.com</email></address>
4052    </author>
4053    <author initials="M." surname="Nottingham" fullname="Mark Nottingham" role="editor">
4054      <organization>Akamai</organization>
4055      <address><email>mnot@mnot.net</email></address>
4056    </author>
4057    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4058      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4059      <address><email>julian.reschke@greenbytes.de</email></address>
4060    </author>
4061    <date month="July" year="2013"/>
4062  </front>
4063  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-23"/>
4064 
4065</reference>
4066
4067<reference anchor="Part7">
4068  <front>
4069    <title>Hypertext Transfer Protocol (HTTP/1.1): Authentication</title>
4070    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4071      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4072      <address><email>fielding@gbiv.com</email></address>
4073    </author>
4074    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4075      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4076      <address><email>julian.reschke@greenbytes.de</email></address>
4077    </author>
4078    <date month="July" year="2013"/>
4079  </front>
4080  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p7-auth-23"/>
4081 
4082</reference>
4083
4084<reference anchor="RFC5234">
4085  <front>
4086    <title abbrev="ABNF for Syntax Specifications">Augmented BNF for Syntax Specifications: ABNF</title>
4087    <author initials="D." surname="Crocker" fullname="Dave Crocker" role="editor">
4088      <organization>Brandenburg InternetWorking</organization>
4089      <address>
4090        <email>dcrocker@bbiw.net</email>
4091      </address> 
4092    </author>
4093    <author initials="P." surname="Overell" fullname="Paul Overell">
4094      <organization>THUS plc.</organization>
4095      <address>
4096        <email>paul.overell@thus.net</email>
4097      </address>
4098    </author>
4099    <date month="January" year="2008"/>
4100  </front>
4101  <seriesInfo name="STD" value="68"/>
4102  <seriesInfo name="RFC" value="5234"/>
4103</reference>
4104
4105<reference anchor="RFC2119">
4106  <front>
4107    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
4108    <author initials="S." surname="Bradner" fullname="Scott Bradner">
4109      <organization>Harvard University</organization>
4110      <address><email>sob@harvard.edu</email></address>
4111    </author>
4112    <date month="March" year="1997"/>
4113  </front>
4114  <seriesInfo name="BCP" value="14"/>
4115  <seriesInfo name="RFC" value="2119"/>
4116</reference>
4117
4118<reference anchor="RFC3986">
4119 <front>
4120  <title abbrev="URI Generic Syntax">Uniform Resource Identifier (URI): Generic Syntax</title>
4121  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4122    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
4123    <address>
4124       <email>timbl@w3.org</email>
4125       <uri>http://www.w3.org/People/Berners-Lee/</uri>
4126    </address>
4127  </author>
4128  <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
4129    <organization abbrev="Day Software">Day Software</organization>
4130    <address>
4131      <email>fielding@gbiv.com</email>
4132      <uri>http://roy.gbiv.com/</uri>
4133    </address>
4134  </author>
4135  <author initials="L." surname="Masinter" fullname="Larry Masinter">
4136    <organization abbrev="Adobe Systems">Adobe Systems Incorporated</organization>
4137    <address>
4138      <email>LMM@acm.org</email>
4139      <uri>http://larry.masinter.net/</uri>
4140    </address>
4141  </author>
4142  <date month="January" year="2005"/>
4143 </front>
4144 <seriesInfo name="STD" value="66"/>
4145 <seriesInfo name="RFC" value="3986"/>
4146</reference>
4147
4148<reference anchor="RFC0793">
4149  <front>
4150    <title>Transmission Control Protocol</title>
4151    <author initials="J." surname="Postel" fullname="Jon Postel">
4152      <organization>University of Southern California (USC)/Information Sciences Institute</organization>
4153    </author>
4154    <date year="1981" month="September"/>
4155  </front>
4156  <seriesInfo name="STD" value="7"/>
4157  <seriesInfo name="RFC" value="793"/>
4158</reference>
4159
4160<reference anchor="USASCII">
4161  <front>
4162    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
4163    <author>
4164      <organization>American National Standards Institute</organization>
4165    </author>
4166    <date year="1986"/>
4167  </front>
4168  <seriesInfo name="ANSI" value="X3.4"/>
4169</reference>
4170
4171<reference anchor="RFC1950">
4172  <front>
4173    <title>ZLIB Compressed Data Format Specification version 3.3</title>
4174    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4175      <organization>Aladdin Enterprises</organization>
4176      <address><email>ghost@aladdin.com</email></address>
4177    </author>
4178    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly"/>
4179    <date month="May" year="1996"/>
4180  </front>
4181  <seriesInfo name="RFC" value="1950"/>
4182  <!--<annotation>
4183    RFC 1950 is an Informational RFC, thus it might be less stable than
4184    this specification. On the other hand, this downward reference was
4185    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4186    therefore it is unlikely to cause problems in practice. See also
4187    <xref target="BCP97"/>.
4188  </annotation>-->
4189</reference>
4190
4191<reference anchor="RFC1951">
4192  <front>
4193    <title>DEFLATE Compressed Data Format Specification version 1.3</title>
4194    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4195      <organization>Aladdin Enterprises</organization>
4196      <address><email>ghost@aladdin.com</email></address>
4197    </author>
4198    <date month="May" year="1996"/>
4199  </front>
4200  <seriesInfo name="RFC" value="1951"/>
4201  <!--<annotation>
4202    RFC 1951 is an Informational RFC, thus it might be less stable than
4203    this specification. On the other hand, this downward reference was
4204    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4205    therefore it is unlikely to cause problems in practice. See also
4206    <xref target="BCP97"/>.
4207  </annotation>-->
4208</reference>
4209
4210<reference anchor="RFC1952">
4211  <front>
4212    <title>GZIP file format specification version 4.3</title>
4213    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4214      <organization>Aladdin Enterprises</organization>
4215      <address><email>ghost@aladdin.com</email></address>
4216    </author>
4217    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
4218      <address><email>gzip@prep.ai.mit.edu</email></address>
4219    </author>
4220    <author initials="M." surname="Adler" fullname="Mark Adler">
4221      <address><email>madler@alumni.caltech.edu</email></address>
4222    </author>
4223    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4224      <address><email>ghost@aladdin.com</email></address>
4225    </author>
4226    <author initials="G." surname="Randers-Pehrson" fullname="Glenn Randers-Pehrson">
4227      <address><email>randeg@alumni.rpi.edu</email></address>
4228    </author>
4229    <date month="May" year="1996"/>
4230  </front>
4231  <seriesInfo name="RFC" value="1952"/>
4232  <!--<annotation>
4233    RFC 1952 is an Informational RFC, thus it might be less stable than
4234    this specification. On the other hand, this downward reference was
4235    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4236    therefore it is unlikely to cause problems in practice. See also
4237    <xref target="BCP97"/>.
4238  </annotation>-->
4239</reference>
4240
4241<reference anchor="Welch">
4242  <front>
4243    <title>A Technique for High Performance Data Compression</title>
4244    <author initials="T.A." surname="Welch" fullname="Terry A. Welch"/>
4245    <date month="June" year="1984"/>
4246  </front>
4247  <seriesInfo name="IEEE Computer" value="17(6)"/>
4248</reference>
4249
4250</references>
4251
4252<references title="Informative References">
4253
4254<reference anchor="ISO-8859-1">
4255  <front>
4256    <title>
4257     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
4258    </title>
4259    <author>
4260      <organization>International Organization for Standardization</organization>
4261    </author>
4262    <date year="1998"/>
4263  </front>
4264  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
4265</reference>
4266
4267<reference anchor="RFC1919">
4268  <front>
4269    <title>Classical versus Transparent IP Proxies</title>
4270    <author initials="M." surname="Chatel" fullname="Marc Chatel">
4271      <address><email>mchatel@pax.eunet.ch</email></address>
4272    </author>
4273    <date year="1996" month="March"/>
4274  </front>
4275  <seriesInfo name="RFC" value="1919"/>
4276</reference>
4277
4278<reference anchor="RFC1945">
4279  <front>
4280    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
4281    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4282      <organization>MIT, Laboratory for Computer Science</organization>
4283      <address><email>timbl@w3.org</email></address>
4284    </author>
4285    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4286      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4287      <address><email>fielding@ics.uci.edu</email></address>
4288    </author>
4289    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4290      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
4291      <address><email>frystyk@w3.org</email></address>
4292    </author>
4293    <date month="May" year="1996"/>
4294  </front>
4295  <seriesInfo name="RFC" value="1945"/>
4296</reference>
4297
4298<reference anchor="RFC2045">
4299  <front>
4300    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
4301    <author initials="N." surname="Freed" fullname="Ned Freed">
4302      <organization>Innosoft International, Inc.</organization>
4303      <address><email>ned@innosoft.com</email></address>
4304    </author>
4305    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
4306      <organization>First Virtual Holdings</organization>
4307      <address><email>nsb@nsb.fv.com</email></address>
4308    </author>
4309    <date month="November" year="1996"/>
4310  </front>
4311  <seriesInfo name="RFC" value="2045"/>
4312</reference>
4313
4314<reference anchor="RFC2047">
4315  <front>
4316    <title abbrev="Message Header Extensions">MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text</title>
4317    <author initials="K." surname="Moore" fullname="Keith Moore">
4318      <organization>University of Tennessee</organization>
4319      <address><email>moore@cs.utk.edu</email></address>
4320    </author>
4321    <date month="November" year="1996"/>
4322  </front>
4323  <seriesInfo name="RFC" value="2047"/>
4324</reference>
4325
4326<reference anchor="RFC2068">
4327  <front>
4328    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4329    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
4330      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4331      <address><email>fielding@ics.uci.edu</email></address>
4332    </author>
4333    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4334      <organization>MIT Laboratory for Computer Science</organization>
4335      <address><email>jg@w3.org</email></address>
4336    </author>
4337    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
4338      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
4339      <address><email>mogul@wrl.dec.com</email></address>
4340    </author>
4341    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4342      <organization>MIT Laboratory for Computer Science</organization>
4343      <address><email>frystyk@w3.org</email></address>
4344    </author>
4345    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4346      <organization>MIT Laboratory for Computer Science</organization>
4347      <address><email>timbl@w3.org</email></address>
4348    </author>
4349    <date month="January" year="1997"/>
4350  </front>
4351  <seriesInfo name="RFC" value="2068"/>
4352</reference>
4353
4354<reference anchor="RFC2145">
4355  <front>
4356    <title abbrev="HTTP Version Numbers">Use and Interpretation of HTTP Version Numbers</title>
4357    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
4358      <organization>Western Research Laboratory</organization>
4359      <address><email>mogul@wrl.dec.com</email></address>
4360    </author>
4361    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4362      <organization>Department of Information and Computer Science</organization>
4363      <address><email>fielding@ics.uci.edu</email></address>
4364    </author>
4365    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4366      <organization>MIT Laboratory for Computer Science</organization>
4367      <address><email>jg@w3.org</email></address>
4368    </author>
4369    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4370      <organization>W3 Consortium</organization>
4371      <address><email>frystyk@w3.org</email></address>
4372    </author>
4373    <date month="May" year="1997"/>
4374  </front>
4375  <seriesInfo name="RFC" value="2145"/>
4376</reference>
4377
4378<reference anchor="RFC2616">
4379  <front>
4380    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4381    <author initials="R." surname="Fielding" fullname="R. Fielding">
4382      <organization>University of California, Irvine</organization>
4383      <address><email>fielding@ics.uci.edu</email></address>
4384    </author>
4385    <author initials="J." surname="Gettys" fullname="J. Gettys">
4386      <organization>W3C</organization>
4387      <address><email>jg@w3.org</email></address>
4388    </author>
4389    <author initials="J." surname="Mogul" fullname="J. Mogul">
4390      <organization>Compaq Computer Corporation</organization>
4391      <address><email>mogul@wrl.dec.com</email></address>
4392    </author>
4393    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
4394      <organization>MIT Laboratory for Computer Science</organization>
4395      <address><email>frystyk@w3.org</email></address>
4396    </author>
4397    <author initials="L." surname="Masinter" fullname="L. Masinter">
4398      <organization>Xerox Corporation</organization>
4399      <address><email>masinter@parc.xerox.com</email></address>
4400    </author>
4401    <author initials="P." surname="Leach" fullname="P. Leach">
4402      <organization>Microsoft Corporation</organization>
4403      <address><email>paulle@microsoft.com</email></address>
4404    </author>
4405    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
4406      <organization>W3C</organization>
4407      <address><email>timbl@w3.org</email></address>
4408    </author>
4409    <date month="June" year="1999"/>
4410  </front>
4411  <seriesInfo name="RFC" value="2616"/>
4412</reference>
4413
4414<reference anchor="RFC2817">
4415  <front>
4416    <title>Upgrading to TLS Within HTTP/1.1</title>
4417    <author initials="R." surname="Khare" fullname="R. Khare">
4418      <organization>4K Associates / UC Irvine</organization>
4419      <address><email>rohit@4K-associates.com</email></address>
4420    </author>
4421    <author initials="S." surname="Lawrence" fullname="S. Lawrence">
4422      <organization>Agranat Systems, Inc.</organization>
4423      <address><email>lawrence@agranat.com</email></address>
4424    </author>
4425    <date year="2000" month="May"/>
4426  </front>
4427  <seriesInfo name="RFC" value="2817"/>
4428</reference>
4429
4430<reference anchor="RFC2818">
4431  <front>
4432    <title>HTTP Over TLS</title>
4433    <author initials="E." surname="Rescorla" fullname="Eric Rescorla">
4434      <organization>RTFM, Inc.</organization>
4435      <address><email>ekr@rtfm.com</email></address>
4436    </author>
4437    <date year="2000" month="May"/>
4438  </front>
4439  <seriesInfo name="RFC" value="2818"/>
4440</reference>
4441
4442<reference anchor="RFC3040">
4443  <front>
4444    <title>Internet Web Replication and Caching Taxonomy</title>
4445    <author initials="I." surname="Cooper" fullname="I. Cooper">
4446      <organization>Equinix, Inc.</organization>
4447    </author>
4448    <author initials="I." surname="Melve" fullname="I. Melve">
4449      <organization>UNINETT</organization>
4450    </author>
4451    <author initials="G." surname="Tomlinson" fullname="G. Tomlinson">
4452      <organization>CacheFlow Inc.</organization>
4453    </author>
4454    <date year="2001" month="January"/>
4455  </front>
4456  <seriesInfo name="RFC" value="3040"/>
4457</reference>
4458
4459<reference anchor="BCP90">
4460  <front>
4461    <title>Registration Procedures for Message Header Fields</title>
4462    <author initials="G." surname="Klyne" fullname="G. Klyne">
4463      <organization>Nine by Nine</organization>
4464      <address><email>GK-IETF@ninebynine.org</email></address>
4465    </author>
4466    <author initials="M." surname="Nottingham" fullname="M. Nottingham">
4467      <organization>BEA Systems</organization>
4468      <address><email>mnot@pobox.com</email></address>
4469    </author>
4470    <author initials="J." surname="Mogul" fullname="J. Mogul">
4471      <organization>HP Labs</organization>
4472      <address><email>JeffMogul@acm.org</email></address>
4473    </author>
4474    <date year="2004" month="September"/>
4475  </front>
4476  <seriesInfo name="BCP" value="90"/>
4477  <seriesInfo name="RFC" value="3864"/>
4478</reference>
4479
4480<reference anchor="RFC4033">
4481  <front>
4482    <title>DNS Security Introduction and Requirements</title>
4483    <author initials="R." surname="Arends" fullname="R. Arends"/>
4484    <author initials="R." surname="Austein" fullname="R. Austein"/>
4485    <author initials="M." surname="Larson" fullname="M. Larson"/>
4486    <author initials="D." surname="Massey" fullname="D. Massey"/>
4487    <author initials="S." surname="Rose" fullname="S. Rose"/>
4488    <date year="2005" month="March"/>
4489  </front>
4490  <seriesInfo name="RFC" value="4033"/>
4491</reference>
4492
4493<reference anchor="BCP13">
4494  <front>
4495    <title>Media Type Specifications and Registration Procedures</title>
4496    <author initials="N." surname="Freed" fullname="Ned Freed">
4497      <organization>Oracle</organization>
4498      <address>
4499        <email>ned+ietf@mrochek.com</email>
4500      </address>
4501    </author>
4502    <author initials="J." surname="Klensin" fullname="John C. Klensin">
4503      <address>
4504        <email>john+ietf@jck.com</email>
4505      </address>
4506    </author>
4507    <author initials="T." surname="Hansen" fullname="Tony Hansen">
4508      <organization>AT&amp;T Laboratories</organization>
4509      <address>
4510        <email>tony+mtsuffix@maillennium.att.com</email>
4511      </address>
4512    </author>
4513    <date year="2013" month="January"/>
4514  </front>
4515  <seriesInfo name="BCP" value="13"/>
4516  <seriesInfo name="RFC" value="6838"/>
4517</reference>
4518
4519<reference anchor="BCP115">
4520  <front>
4521    <title>Guidelines and Registration Procedures for New URI Schemes</title>
4522    <author initials="T." surname="Hansen" fullname="T. Hansen">
4523      <organization>AT&amp;T Laboratories</organization>
4524      <address>
4525        <email>tony+urireg@maillennium.att.com</email>
4526      </address>
4527    </author>
4528    <author initials="T." surname="Hardie" fullname="T. Hardie">
4529      <organization>Qualcomm, Inc.</organization>
4530      <address>
4531        <email>hardie@qualcomm.com</email>
4532      </address>
4533    </author>
4534    <author initials="L." surname="Masinter" fullname="L. Masinter">
4535      <organization>Adobe Systems</organization>
4536      <address>
4537        <email>LMM@acm.org</email>
4538      </address>
4539    </author>
4540    <date year="2006" month="February"/>
4541  </front>
4542  <seriesInfo name="BCP" value="115"/>
4543  <seriesInfo name="RFC" value="4395"/>
4544</reference>
4545
4546<reference anchor="RFC4559">
4547  <front>
4548    <title>SPNEGO-based Kerberos and NTLM HTTP Authentication in Microsoft Windows</title>
4549    <author initials="K." surname="Jaganathan" fullname="K. Jaganathan"/>
4550    <author initials="L." surname="Zhu" fullname="L. Zhu"/>
4551    <author initials="J." surname="Brezak" fullname="J. Brezak"/>
4552    <date year="2006" month="June"/>
4553  </front>
4554  <seriesInfo name="RFC" value="4559"/>
4555</reference>
4556
4557<reference anchor="RFC5226">
4558  <front>
4559    <title>Guidelines for Writing an IANA Considerations Section in RFCs</title>
4560    <author initials="T." surname="Narten" fullname="T. Narten">
4561      <organization>IBM</organization>
4562      <address><email>narten@us.ibm.com</email></address>
4563    </author>
4564    <author initials="H." surname="Alvestrand" fullname="H. Alvestrand">
4565      <organization>Google</organization>
4566      <address><email>Harald@Alvestrand.no</email></address>
4567    </author>
4568    <date year="2008" month="May"/>
4569  </front>
4570  <seriesInfo name="BCP" value="26"/>
4571  <seriesInfo name="RFC" value="5226"/>
4572</reference>
4573
4574<reference anchor="RFC5246">
4575   <front>
4576      <title>The Transport Layer Security (TLS) Protocol Version 1.2</title>
4577      <author initials="T." surname="Dierks" fullname="T. Dierks">
4578         <organization/>
4579      </author>
4580      <author initials="E." surname="Rescorla" fullname="E. Rescorla">
4581         <organization>RTFM, Inc.</organization>
4582      </author>
4583      <date year="2008" month="August"/>
4584   </front>
4585   <seriesInfo name="RFC" value="5246"/>
4586</reference>
4587
4588<reference anchor="RFC5322">
4589  <front>
4590    <title>Internet Message Format</title>
4591    <author initials="P." surname="Resnick" fullname="P. Resnick">
4592      <organization>Qualcomm Incorporated</organization>
4593    </author>
4594    <date year="2008" month="October"/>
4595  </front>
4596  <seriesInfo name="RFC" value="5322"/>
4597</reference>
4598
4599<reference anchor="RFC6265">
4600  <front>
4601    <title>HTTP State Management Mechanism</title>
4602    <author initials="A." surname="Barth" fullname="Adam Barth">
4603      <organization abbrev="U.C. Berkeley">
4604        University of California, Berkeley
4605      </organization>
4606      <address><email>abarth@eecs.berkeley.edu</email></address>
4607    </author>
4608    <date year="2011" month="April"/>
4609  </front>
4610  <seriesInfo name="RFC" value="6265"/>
4611</reference>
4612
4613<reference anchor="RFC6585">
4614  <front>
4615    <title>Additional HTTP Status Codes</title>
4616    <author initials="M." surname="Nottingham" fullname="M. Nottingham">
4617      <organization>Rackspace</organization>
4618    </author>
4619    <author initials="R." surname="Fielding" fullname="R. Fielding">
4620      <organization>Adobe</organization>
4621    </author>
4622    <date year="2012" month="April"/>
4623   </front>
4624   <seriesInfo name="RFC" value="6585"/>
4625</reference>
4626
4627<!--<reference anchor='BCP97'>
4628  <front>
4629    <title>Handling Normative References to Standards-Track Documents</title>
4630    <author initials='J.' surname='Klensin' fullname='J. Klensin'>
4631      <address>
4632        <email>klensin+ietf@jck.com</email>
4633      </address>
4634    </author>
4635    <author initials='S.' surname='Hartman' fullname='S. Hartman'>
4636      <organization>MIT</organization>
4637      <address>
4638        <email>hartmans-ietf@mit.edu</email>
4639      </address>
4640    </author>
4641    <date year='2007' month='June' />
4642  </front>
4643  <seriesInfo name='BCP' value='97' />
4644  <seriesInfo name='RFC' value='4897' />
4645</reference>-->
4646
4647<reference anchor="Kri2001" target="http://arxiv.org/abs/cs.SE/0105018">
4648  <front>
4649    <title>HTTP Cookies: Standards, Privacy, and Politics</title>
4650    <author initials="D." surname="Kristol" fullname="David M. Kristol"/>
4651    <date year="2001" month="November"/>
4652  </front>
4653  <seriesInfo name="ACM Transactions on Internet Technology" value="1(2)"/>
4654</reference>
4655
4656</references>
4657
4658
4659<section title="HTTP Version History" anchor="compatibility">
4660<t>
4661   HTTP has been in use by the World-Wide Web global information initiative
4662   since 1990. The first version of HTTP, later referred to as HTTP/0.9,
4663   was a simple protocol for hypertext data transfer across the Internet
4664   with only a single request method (GET) and no metadata.
4665   HTTP/1.0, as defined by <xref target="RFC1945"/>, added a range of request
4666   methods and MIME-like messaging that could include metadata about the data
4667   transferred and modifiers on the request/response semantics. However,
4668   HTTP/1.0 did not sufficiently take into consideration the effects of
4669   hierarchical proxies, caching, the need for persistent connections, or
4670   name-based virtual hosts. The proliferation of incompletely-implemented
4671   applications calling themselves "HTTP/1.0" further necessitated a
4672   protocol version change in order for two communicating applications
4673   to determine each other's true capabilities.
4674</t>
4675<t>
4676   HTTP/1.1 remains compatible with HTTP/1.0 by including more stringent
4677   requirements that enable reliable implementations, adding only
4678   those new features that will either be safely ignored by an HTTP/1.0
4679   recipient or only sent when communicating with a party advertising
4680   conformance with HTTP/1.1.
4681</t>
4682<t>
4683   It is beyond the scope of a protocol specification to mandate
4684   conformance with previous versions. HTTP/1.1 was deliberately
4685   designed, however, to make supporting previous versions easy.
4686   We would expect a general-purpose HTTP/1.1 server to understand
4687   any valid request in the format of HTTP/1.0 and respond appropriately
4688   with an HTTP/1.1 message that only uses features understood (or
4689   safely ignored) by HTTP/1.0 clients.  Likewise, we would expect
4690   an HTTP/1.1 client to understand any valid HTTP/1.0 response.
4691</t>
4692<t>
4693   Since HTTP/0.9 did not support header fields in a request,
4694   there is no mechanism for it to support name-based virtual
4695   hosts (selection of resource by inspection of the <xref target="header.host" format="none">Host</xref> header
4696   field).  Any server that implements name-based virtual hosts
4697   ought to disable support for HTTP/0.9.  Most requests that
4698   appear to be HTTP/0.9 are, in fact, badly constructed HTTP/1.x
4699   requests wherein a buggy client failed to properly encode
4700   linear whitespace found in a URI reference and placed in
4701   the request-target.
4702</t>
4703
4704<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
4705<t>
4706   This section summarizes major differences between versions HTTP/1.0
4707   and HTTP/1.1.
4708</t>
4709
4710<section title="Multi-homed Web Servers" anchor="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses">
4711<t>
4712   The requirements that clients and servers support the <xref target="header.host" format="none">Host</xref>
4713   header field (<xref target="header.host"/>), report an error if it is
4714   missing from an HTTP/1.1 request, and accept absolute URIs (<xref target="request-target"/>)
4715   are among the most important changes defined by HTTP/1.1.
4716</t>
4717<t>
4718   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
4719   addresses and servers; there was no other established mechanism for
4720   distinguishing the intended server of a request than the IP address
4721   to which that request was directed. The <xref target="header.host" format="none">Host</xref> header field was
4722   introduced during the development of HTTP/1.1 and, though it was
4723   quickly implemented by most HTTP/1.0 browsers, additional requirements
4724   were placed on all HTTP/1.1 requests in order to ensure complete
4725   adoption.  At the time of this writing, most HTTP-based services
4726   are dependent upon the Host header field for targeting requests.
4727</t>
4728</section>
4729
4730<section title="Keep-Alive Connections" anchor="compatibility.with.http.1.0.persistent.connections">
4731<t>
4732   In HTTP/1.0, each connection is established by the client prior to the
4733   request and closed by the server after sending the response. However, some
4734   implementations implement the explicitly negotiated ("Keep-Alive") version
4735   of persistent connections described in Section 19.7.1 of <xref target="RFC2068"/>.
4736</t>
4737<t>
4738   Some clients and servers might wish to be compatible with these previous
4739   approaches to persistent connections, by explicitly negotiating for them
4740   with a "Connection: keep-alive" request header field. However, some
4741   experimental implementations of HTTP/1.0 persistent connections are faulty;
4742   for example, if an HTTP/1.0 proxy server doesn't understand
4743   <xref target="header.connection" format="none">Connection</xref>, it will erroneously forward that header field
4744   to the next inbound server, which would result in a hung connection.
4745</t>
4746<t>
4747   One attempted solution was the introduction of a Proxy-Connection header
4748   field, targeted specifically at proxies. In practice, this was also
4749   unworkable, because proxies are often deployed in multiple layers, bringing
4750   about the same problem discussed above.
4751</t>
4752<t>
4753   As a result, clients are encouraged not to send the Proxy-Connection header
4754   field in any requests.
4755</t>
4756<t>
4757   Clients are also encouraged to consider the use of Connection: keep-alive
4758   in requests carefully; while they can enable persistent connections with
4759   HTTP/1.0 servers, clients using them will need to monitor the
4760   connection for "hung" requests (which indicate that the client ought stop
4761   sending the header field), and this mechanism ought not be used by clients
4762   at all when a proxy is being used.
4763</t>
4764</section>
4765
4766<section title="Introduction of Transfer-Encoding" anchor="introduction.of.transfer-encoding">
4767<t>
4768   HTTP/1.1 introduces the <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field
4769   (<xref target="header.transfer-encoding"/>).
4770   Transfer codings need to be decoded prior to forwarding an HTTP message
4771   over a MIME-compliant protocol.
4772</t>
4773</section>
4774
4775</section>
4776
4777<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
4778<t>
4779  HTTP's approach to error handling has been explained.
4780  (<xref target="conformance"/>)
4781</t>
4782<t>
4783  The expectation to support HTTP/0.9 requests has been removed.
4784</t>
4785<t>
4786  The term "Effective Request URI" has been introduced.
4787  (<xref target="effective.request.uri"/>)
4788</t>
4789<t>
4790  HTTP messages can be (and often are) buffered by implementations; despite
4791  it sometimes being available as a stream, HTTP is fundamentally a
4792  message-oriented protocol.
4793  (<xref target="http.message"/>)
4794</t>
4795<t>
4796  Minimum supported sizes for various protocol elements have been
4797  suggested, to improve interoperability.
4798</t>
4799<t>
4800  Header fields that span multiple lines ("line folding") are deprecated.
4801  (<xref target="field.parsing"/>)
4802</t>
4803<t>
4804  The HTTP-version ABNF production has been clarified to be case-sensitive.
4805  Additionally, version numbers has been restricted to single digits, due
4806  to the fact that implementations are known to handle multi-digit version
4807  numbers incorrectly.
4808  (<xref target="http.version"/>)
4809</t>
4810<t>
4811  The HTTPS URI scheme is now defined by this specification; previously,
4812  it was done in  Section 2.4 of <xref target="RFC2818"/>.
4813  (<xref target="https.uri"/>)
4814</t>
4815<t>
4816  The HTTPS URI scheme implies end-to-end security.
4817  (<xref target="https.uri"/>)
4818</t>
4819<t>
4820  Userinfo (i.e., username and password) are now disallowed in HTTP and
4821  HTTPS URIs, because of security issues related to their transmission on the
4822  wire.
4823  (<xref target="http.uri"/>)
4824</t>
4825<t>
4826  Invalid whitespace around field-names is now required to be rejected,
4827  because accepting it represents a security vulnerability.
4828  (<xref target="header.fields"/>)
4829</t>
4830<t>
4831  The ABNF productions defining header fields now only list the field value.
4832  (<xref target="header.fields"/>)
4833</t>
4834<t>
4835  Rules about implicit linear whitespace between certain grammar productions
4836  have been removed; now whitespace is only allowed where specifically
4837  defined in the ABNF.
4838  (<xref target="whitespace"/>)
4839</t>
4840<t> 
4841  The NUL octet is no longer allowed in comment and quoted-string text, and
4842  handling of backslash-escaping in them has been clarified.
4843  (<xref target="field.components"/>)
4844</t> 
4845<t>
4846  The quoted-pair rule no longer allows escaping control characters other than
4847  HTAB.
4848  (<xref target="field.components"/>)
4849</t>
4850<t>
4851  Non-ASCII content in header fields and the reason phrase has been obsoleted
4852  and made opaque (the TEXT rule was removed).
4853  (<xref target="field.components"/>)
4854</t>
4855<t>
4856  Bogus "<xref target="header.content-length" format="none">Content-Length</xref>" header fields are now required to be
4857  handled as errors by recipients.
4858  (<xref target="header.content-length"/>)
4859</t>
4860<t>
4861  The "identity" transfer coding token has been removed.
4862  (Sections <xref format="counter" target="message.body"/> and
4863  <xref format="counter" target="transfer.codings"/>)
4864</t>
4865<t>
4866  The algorithm for determining the message body length has been clarified
4867  to indicate all of the special cases (e.g., driven by methods or status
4868  codes) that affect it, and that new protocol elements cannot define such
4869  special cases.
4870  (<xref target="message.body.length"/>)
4871</t>
4872<t>
4873  "multipart/byteranges" is no longer a way of determining message body length
4874  detection.
4875  (<xref target="message.body.length"/>)
4876</t>
4877<t>
4878  CONNECT is a new, special case in determining message body length.
4879  (<xref target="message.body.length"/>)
4880</t>
4881<t>
4882  Chunk length does not include the count of the octets in the
4883  chunk header and trailer.
4884  (<xref target="chunked.encoding"/>)
4885</t>
4886<t>
4887  Use of chunk extensions is deprecated, and line folding in them is
4888  disallowed.
4889  (<xref target="chunked.encoding"/>)
4890</t>
4891<t>
4892  The segment + query components of RFC3986 have been used to define the
4893  request-target, instead of abs_path from RFC 1808.
4894  (<xref target="request-target"/>)
4895</t>
4896<t>
4897  The asterisk form of the request-target is only allowed in the OPTIONS
4898  method.
4899  (<xref target="request-target"/>)
4900</t>
4901<t>
4902  Exactly when "close" connection options have to be sent has been clarified.
4903  (<xref target="header.connection"/>)
4904</t>
4905<t> 
4906  "hop-by-hop" header fields are required to appear in the Connection header
4907  field; just because they're defined as hop-by-hop in this specification
4908  doesn't exempt them.
4909  (<xref target="header.connection"/>)
4910</t>
4911<t>
4912  The limit of two connections per server has been removed.
4913  (<xref target="persistent.connections"/>)
4914</t>
4915<t>
4916  An idempotent sequence of requests is no longer required to be retried.
4917  (<xref target="persistent.connections"/>)
4918</t>
4919<t>
4920  The requirement to retry requests under certain circumstances when the
4921  server prematurely closes the connection has been removed.
4922  (<xref target="persistent.connections"/>)
4923</t>
4924<t>
4925  Some extraneous requirements about when servers are allowed to close
4926  connections prematurely have been removed.
4927  (<xref target="persistent.connections"/>)
4928</t>
4929<t>
4930  The semantics of the <xref target="header.upgrade" format="none">Upgrade</xref> header field is now defined in
4931  responses other than 101 (this was incorporated from <xref target="RFC2817"/>).
4932  (<xref target="header.upgrade"/>)
4933</t>
4934<t>
4935  Registration of Transfer Codings now requires IETF Review
4936  (<xref target="transfer.coding.registry"/>)
4937</t>
4938<t>
4939  The meaning of the "deflate" content coding has been clarified.
4940  (<xref target="deflate.coding"/>)
4941</t>
4942<t>
4943  This specification now defines the Upgrade Token Registry, previously
4944  defined in Section 7.2 of <xref target="RFC2817"/>.
4945  (<xref target="upgrade.token.registry"/>)
4946</t>
4947<t>
4948  Issues with the Keep-Alive and Proxy-Connection header fields in requests
4949  are pointed out, with use of the latter being discouraged altogether.
4950  (<xref target="compatibility.with.http.1.0.persistent.connections"/>)
4951</t>
4952<t>
4953  Empty list elements in list productions (e.g., a list header field containing
4954  ", ,") have been deprecated.
4955  (<xref target="abnf.extension"/>)
4956</t>
4957</section>
4958</section>
4959
4960<section title="ABNF list extension: #rule" anchor="abnf.extension">
4961<t>
4962  A #rule extension to the ABNF rules of <xref target="RFC5234"/> is used to
4963  improve readability in the definitions of some header field values.
4964</t>
4965<t>
4966  A construct "#" is defined, similar to "*", for defining comma-delimited
4967  lists of elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating
4968  at least &lt;n&gt; and at most &lt;m&gt; elements, each separated by a single
4969  comma (",") and optional whitespace (OWS).   
4970</t>
4971<figure><preamble>
4972  Thus,
4973</preamble><artwork type="example"><![CDATA[
4974  1#element => element *( OWS "," OWS element )
4975]]></artwork></figure>
4976<figure><preamble>
4977  and:
4978</preamble><artwork type="example"><![CDATA[
4979  #element => [ 1#element ]
4980]]></artwork></figure>
4981<figure><preamble>
4982  and for n &gt;= 1 and m &gt; 1:
4983</preamble><artwork type="example"><![CDATA[
4984  <n>#<m>element => element <n-1>*<m-1>( OWS "," OWS element )
4985]]></artwork></figure>
4986<t>
4987  For compatibility with legacy list rules, recipients SHOULD accept empty
4988  list elements. In other words, consumers would follow the list productions:
4989</t>
4990<figure><artwork type="example"><![CDATA[
4991  #element => [ ( "," / element ) *( OWS "," [ OWS element ] ) ]
4992 
4993  1#element => *( "," OWS ) element *( OWS "," [ OWS element ] )
4994]]></artwork></figure>
4995<t>
4996  Note that empty elements do not contribute to the count of elements present,
4997  though.
4998</t>
4999<t>
5000  For example, given these ABNF productions:
5001</t>
5002<figure><artwork type="example"><![CDATA[
5003  example-list      = 1#example-list-elmt
5004  example-list-elmt = token ; see Section 3.2.6
5005]]></artwork></figure>
5006<t>
5007  Then these are valid values for example-list (not including the double
5008  quotes, which are present for delimitation only):
5009</t>
5010<figure><artwork type="example"><![CDATA[
5011  "foo,bar"
5012  "foo ,bar,"
5013  "foo , ,bar,charlie   "
5014]]></artwork></figure>
5015<t>
5016  But these values would be invalid, as at least one non-empty element is
5017  required:
5018</t>
5019<figure><artwork type="example"><![CDATA[
5020  ""
5021  ","
5022  ",   ,"
5023]]></artwork></figure>
5024<t>
5025  <xref target="collected.abnf"/> shows the collected ABNF, with the list rules
5026  expanded as explained above.
5027</t>
5028</section>
5029
5030
5031<section title="Collected ABNF" anchor="collected.abnf">
5032<figure>
5033<artwork type="abnf" name="p1-messaging.parsed-abnf"><![CDATA[
5034BWS = OWS
5035
5036Connection = *( "," OWS ) connection-option *( OWS "," [ OWS
5037 connection-option ] )
5038Content-Length = 1*DIGIT
5039
5040HTTP-message = start-line *( header-field CRLF ) CRLF [ message-body
5041 ]
5042HTTP-name = %x48.54.54.50 ; HTTP
5043HTTP-version = HTTP-name "/" DIGIT "." DIGIT
5044Host = uri-host [ ":" port ]
5045
5046OWS = *( SP / HTAB )
5047
5048RWS = 1*( SP / HTAB )
5049
5050TE = [ ( "," / t-codings ) *( OWS "," [ OWS t-codings ] ) ]
5051Trailer = *( "," OWS ) field-name *( OWS "," [ OWS field-name ] )
5052Transfer-Encoding = *( "," OWS ) transfer-coding *( OWS "," [ OWS
5053 transfer-coding ] )
5054
5055URI-reference = <URI-reference, defined in [RFC3986], Section 4.1>
5056Upgrade = *( "," OWS ) protocol *( OWS "," [ OWS protocol ] )
5057
5058Via = *( "," OWS ) ( received-protocol RWS received-by [ RWS comment
5059 ] ) *( OWS "," [ OWS ( received-protocol RWS received-by [ RWS
5060 comment ] ) ] )
5061
5062absolute-URI = <absolute-URI, defined in [RFC3986], Section 4.3>
5063absolute-form = absolute-URI
5064absolute-path = 1*( "/" segment )
5065asterisk-form = "*"
5066attribute = token
5067authority = <authority, defined in [RFC3986], Section 3.2>
5068authority-form = authority
5069
5070chunk = chunk-size [ chunk-ext ] CRLF chunk-data CRLF
5071chunk-data = 1*OCTET
5072chunk-ext = *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
5073chunk-ext-name = token
5074chunk-ext-val = token / quoted-str-nf
5075chunk-size = 1*HEXDIG
5076chunked-body = *chunk last-chunk trailer-part CRLF
5077comment = "(" *( ctext / quoted-cpair / comment ) ")"
5078connection-option = token
5079ctext = HTAB / SP / %x21-27 ; '!'-'''
5080 / %x2A-5B ; '*'-'['
5081 / %x5D-7E ; ']'-'~'
5082 / obs-text
5083
5084field-content = *( HTAB / SP / VCHAR / obs-text )
5085field-name = token
5086field-value = *( field-content / obs-fold )
5087fragment = <fragment, defined in [RFC3986], Section 3.5>
5088
5089header-field = field-name ":" OWS field-value OWS
5090http-URI = "http://" authority path-abempty [ "?" query ] [ "#"
5091 fragment ]
5092https-URI = "https://" authority path-abempty [ "?" query ] [ "#"
5093 fragment ]
5094
5095last-chunk = 1*"0" [ chunk-ext ] CRLF
5096
5097message-body = *OCTET
5098method = token
5099
5100obs-fold = CRLF ( SP / HTAB )
5101obs-text = %x80-FF
5102origin-form = absolute-path [ "?" query ]
5103
5104partial-URI = relative-part [ "?" query ]
5105path-abempty = <path-abempty, defined in [RFC3986], Section 3.3>
5106port = <port, defined in [RFC3986], Section 3.2.3>
5107protocol = protocol-name [ "/" protocol-version ]
5108protocol-name = token
5109protocol-version = token
5110pseudonym = token
5111
5112qdtext = HTAB / SP / "!" / %x23-5B ; '#'-'['
5113 / %x5D-7E ; ']'-'~'
5114 / obs-text
5115qdtext-nf = HTAB / SP / "!" / %x23-5B ; '#'-'['
5116 / %x5D-7E ; ']'-'~'
5117 / obs-text
5118query = <query, defined in [RFC3986], Section 3.4>
5119quoted-cpair = "\" ( HTAB / SP / VCHAR / obs-text )
5120quoted-pair = "\" ( HTAB / SP / VCHAR / obs-text )
5121quoted-str-nf = DQUOTE *( qdtext-nf / quoted-pair ) DQUOTE
5122quoted-string = DQUOTE *( qdtext / quoted-pair ) DQUOTE
5123
5124rank = ( "0" [ "." *3DIGIT ] ) / ( "1" [ "." *3"0" ] )
5125reason-phrase = *( HTAB / SP / VCHAR / obs-text )
5126received-by = ( uri-host [ ":" port ] ) / pseudonym
5127received-protocol = [ protocol-name "/" ] protocol-version
5128relative-part = <relative-part, defined in [RFC3986], Section 4.2>
5129request-line = method SP request-target SP HTTP-version CRLF
5130request-target = origin-form / absolute-form / authority-form /
5131 asterisk-form
5132
5133segment = <segment, defined in [RFC3986], Section 3.3>
5134special = "(" / ")" / "<" / ">" / "@" / "," / ";" / ":" / "\" /
5135 DQUOTE / "/" / "[" / "]" / "?" / "=" / "{" / "}"
5136start-line = request-line / status-line
5137status-code = 3DIGIT
5138status-line = HTTP-version SP status-code SP reason-phrase CRLF
5139
5140t-codings = "trailers" / ( transfer-coding [ t-ranking ] )
5141t-ranking = OWS ";" OWS "q=" rank
5142tchar = "!" / "#" / "$" / "%" / "&" / "'" / "*" / "+" / "-" / "." /
5143 "^" / "_" / "`" / "|" / "~" / DIGIT / ALPHA
5144token = 1*tchar
5145trailer-part = *( header-field CRLF )
5146transfer-coding = "chunked" / "compress" / "deflate" / "gzip" /
5147 transfer-extension
5148transfer-extension = token *( OWS ";" OWS transfer-parameter )
5149transfer-parameter = attribute BWS "=" BWS value
5150
5151uri-host = <host, defined in [RFC3986], Section 3.2.2>
5152
5153value = word
5154
5155word = token / quoted-string
5156]]></artwork>
5157</figure>
5158</section>
5159
5160
5161<section title="Change Log (to be removed by RFC Editor before publication)" anchor="change.log">
5162
5163<section title="Since RFC 2616">
5164<t>
5165  Changes up to the first Working Group Last Call draft are summarized
5166  in <eref target="http://trac.tools.ietf.org/html/draft-ietf-httpbis-p1-messaging-21#appendix-D"/>.
5167</t>
5168</section>
5169
5170<section title="Since draft-ietf-httpbis-p1-messaging-21" anchor="changes.since.21">
5171<t>
5172  Closed issues:
5173  <list style="symbols">
5174    <t>
5175      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/128"/>:
5176      "Cite HTTPS URI scheme definition" (the spec now includes the HTTPs
5177      scheme definition and thus updates RFC 2818)
5178    </t>
5179    <t>
5180      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/389"/>:
5181      "mention of 'proxies' in section about caches"
5182    </t>
5183    <t>
5184      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/390"/>:
5185      "use of ABNF terms from RFC 3986"
5186    </t>
5187    <t>
5188      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/391"/>:
5189      "transferring URIs with userinfo in payload"
5190    </t>
5191    <t>
5192      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/392"/>:
5193      "editorial improvements to message length definition"
5194    </t>
5195    <t>
5196      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/395"/>:
5197      "Connection header field MUST vs SHOULD"
5198    </t>
5199    <t>
5200      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/396"/>:
5201      "editorial improvements to persistent connections section"
5202    </t>
5203    <t>
5204      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/397"/>:
5205      "URI normalization vs empty path"
5206    </t>
5207    <t>
5208      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/408"/>:
5209      "p1 feedback"
5210    </t>
5211    <t>
5212      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/409"/>:
5213      "is parsing OBS-FOLD mandatory?"
5214    </t>
5215    <t>
5216      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/410"/>:
5217      "HTTPS and Shared Caching"
5218    </t>
5219    <t>
5220      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/411"/>:
5221      "Requirements for recipients of ws between start-line and first header field"
5222    </t>
5223    <t>
5224      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/412"/>:
5225      "SP and HT when being tolerant"
5226    </t>
5227    <t>
5228      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/414"/>:
5229      "Message Parsing Strictness"
5230    </t>
5231    <t>
5232      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/415"/>:
5233      "'Render'"
5234    </t>
5235    <t>
5236      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/418"/>:
5237      "No-Transform"
5238    </t>
5239    <t>
5240      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/419"/>:
5241      "p2 editorial feedback"
5242    </t>
5243    <t>
5244      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/420"/>:
5245      "Content-Length SHOULD be sent"
5246    </t>
5247    <t>
5248      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/431"/>:
5249      "origin-form does not allow path starting with "//""
5250    </t>
5251    <t>
5252      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/433"/>:
5253      "ambiguity in part 1 example"
5254    </t>
5255  </list>
5256</t>
5257</section>
5258
5259<section title="Since draft-ietf-httpbis-p1-messaging-22" anchor="changes.since.22">
5260<t>
5261  Closed issues:
5262  <list style="symbols">
5263    <t>
5264      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/435"/>:
5265      "Part1 should have a reference to TCP (RFC 793)"
5266    </t>
5267    <t>
5268      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/438"/>:
5269      "media type registration template issues"
5270    </t>
5271    <t>
5272      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/442"/>:
5273      "BWS" (vs conformance)
5274    </t>
5275    <t>
5276      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/444"/>:
5277      "obs-fold language"
5278    </t>
5279    <t>
5280      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/445"/>:
5281      "Ordering in Upgrade"
5282    </t>
5283    <t>
5284      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/446"/>:
5285      "p1 editorial feedback"
5286    </t>
5287    <t>
5288      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/447"/>:
5289      "HTTP and TCP name delegation"
5290    </t>
5291    <t>
5292      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/449"/>:
5293      "Receiving a higher minor HTTP version number"
5294    </t>
5295    <t>
5296      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/451"/>:
5297      "HTTP(S) URIs and fragids"
5298    </t>
5299    <t>
5300      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/457"/>:
5301      "Registering x-gzip and x-deflate"
5302    </t>
5303    <t>
5304      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/460"/>:
5305      "Via and gateways"
5306    </t>
5307    <t>
5308      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/465"/>:
5309      "Mention 203 Non-Authoritative Information in p1"
5310    </t>
5311    <t>
5312      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/476"/>:
5313      "SHOULD and conformance"
5314    </t>
5315    <t>
5316      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/477"/>:
5317      "Pipelining language"
5318    </t>
5319    <t>
5320      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/482"/>:
5321      "proxy handling of a really bad Content-Length"
5322    </t>
5323  </list>
5324</t>
5325</section>
5326</section>
5327
5328</back>
5329</rfc>
Note: See TracBrowser for help on using the repository browser.