source: draft-ietf-httpbis/20/draft-ietf-httpbis-p1-messaging-20.xml @ 1807

Last change on this file since 1807 was 1807, checked in by julian.reschke@…, 8 years ago

Prepare release of -20 drafts

  • Property svn:eol-style set to native
  • Property svn:mime-type set to text/xml
File size: 235.2 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" category="std" ipr="pre5378Trust200902" docName="draft-ietf-httpbis-p1-messaging-20">
20
21
22<front>
23
24  <title abbrev="HTTP/1.1, Part 1">HTTP/1.1, part 1: Message Routing and Syntax"</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="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
42    <organization abbrev="W3C">World Wide Web Consortium</organization>
43    <address>
44      <postal>
45        <street>W3C / ERCIM</street>
46        <street>2004, rte des Lucioles</street>
47        <city>Sophia-Antipolis</city>
48        <region>AM</region>
49        <code>06902</code>
50        <country>France</country>
51      </postal>
52      <email>ylafon@w3.org</email>
53      <uri>http://www.raubacapeu.net/people/yves/</uri>
54    </address>
55  </author>
56
57  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
58    <organization abbrev="greenbytes">greenbytes GmbH</organization>
59    <address>
60      <postal>
61        <street>Hafenweg 16</street>
62        <city>Muenster</city><region>NW</region><code>48155</code>
63        <country>Germany</country>
64      </postal>
65      <email>julian.reschke@greenbytes.de</email>
66      <uri>http://greenbytes.de/tech/webdav/</uri>
67    </address>
68  </author>
69
70  <date month="July" year="2012" day="16"/>
71  <workgroup>HTTPbis Working Group</workgroup>
72
73<abstract>
74<t>
75   The Hypertext Transfer Protocol (HTTP) is an application-level protocol for
76   distributed, collaborative, hypertext information systems. HTTP has been in
77   use by the World Wide Web global information initiative since 1990.
78   This document provides an overview of HTTP architecture and its associated
79   terminology, defines the "http" and "https" Uniform Resource Identifier
80   (URI) schemes, defines the HTTP/1.1 message syntax and parsing requirements,
81   and describes general security concerns for implementations.
82</t>   
83</abstract>
84
85<note title="Editorial Note (To be removed by RFC Editor)">
86  <t>
87    Discussion of this draft takes place on the HTTPBIS working group
88    mailing list (ietf-http-wg@w3.org), which is archived at
89    <eref target="http://lists.w3.org/Archives/Public/ietf-http-wg/"/>.
90  </t>
91  <t>
92    The current issues list is at
93    <eref target="http://tools.ietf.org/wg/httpbis/trac/report/3"/> and related
94    documents (including fancy diffs) can be found at
95    <eref target="http://tools.ietf.org/wg/httpbis/"/>.
96  </t>
97  <t>
98    The changes in this draft are summarized in <xref target="changes.since.19"/>.
99  </t>
100</note>
101</front>
102<middle>
103<section title="Introduction" anchor="introduction">
104<t>
105   The Hypertext Transfer Protocol (HTTP) is an application-level
106   request/response protocol that uses extensible semantics and MIME-like
107   message payloads for flexible interaction with network-based hypertext
108   information systems. This document is the first in a series of documents
109   that collectively form the HTTP/1.1 specification:
110   <list style="empty">
111    <t>RFC xxx1: Message Routing and Syntax</t>
112    <t>RFC xxx2: Semantics and Payloads</t>
113    <t>RFC xxx3: Conditional Requests</t>
114    <t>RFC xxx4: Range Requests</t>
115    <t>RFC xxx5: Caching</t>
116    <t>RFC xxx6: Authentication</t>
117   </list>
118</t>
119<t>
120   This HTTP/1.1 specification obsoletes and moves to historic status
121   RFC 2616, its predecessor
122   RFC 2068,
123   RFC 2145 (on HTTP versioning),
124   and RFC 2817 (on using CONNECT
125   for TLS upgrades).
126</t>
127<t>
128   HTTP is a generic interface protocol for information systems. It is
129   designed to hide the details of how a service is implemented by presenting
130   a uniform interface to clients that is independent of the types of
131   resources provided. Likewise, servers do not need to be aware of each
132   client's purpose: an HTTP request can be considered in isolation rather
133   than being associated with a specific type of client or a predetermined
134   sequence of application steps. The result is a protocol that can be used
135   effectively in many different contexts and for which implementations can
136   evolve independently over time.
137</t>
138<t>
139   HTTP is also designed for use as an intermediation protocol for translating
140   communication to and from non-HTTP information systems.
141   HTTP proxies and gateways can provide access to alternative information
142   services by translating their diverse protocols into a hypertext
143   format that can be viewed and manipulated by clients in the same way
144   as HTTP services.
145</t>
146<t>
147   One consequence of HTTP flexibility is that the protocol cannot be
148   defined in terms of what occurs behind the interface. Instead, we
149   are limited to defining the syntax of communication, the intent
150   of received communication, and the expected behavior of recipients.
151   If the communication is considered in isolation, then successful
152   actions ought to be reflected in corresponding changes to the
153   observable interface provided by servers. However, since multiple
154   clients might act in parallel and perhaps at cross-purposes, we
155   cannot require that such changes be observable beyond the scope
156   of a single response.
157</t>
158<t>
159   This document describes the architectural elements that are used or
160   referred to in HTTP, defines the "http" and "https" URI schemes,
161   describes overall network operation and connection management,
162   and defines HTTP message framing and forwarding requirements.
163   Our goal is to define all of the mechanisms necessary for HTTP message
164   handling that are independent of message semantics, thereby defining the
165   complete set of requirements for message parsers and
166   message-forwarding intermediaries.
167</t>
168
169
170<section title="Requirement Notation" anchor="intro.requirements">
171<t>
172   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
173   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
174   document are to be interpreted as described in <xref target="RFC2119"/>.
175</t>
176</section>
177
178<section title="Syntax Notation" anchor="notation">
179<iref primary="true" item="Grammar" subitem="ALPHA"/>
180<iref primary="true" item="Grammar" subitem="CR"/>
181<iref primary="true" item="Grammar" subitem="CRLF"/>
182<iref primary="true" item="Grammar" subitem="CTL"/>
183<iref primary="true" item="Grammar" subitem="DIGIT"/>
184<iref primary="true" item="Grammar" subitem="DQUOTE"/>
185<iref primary="true" item="Grammar" subitem="HEXDIG"/>
186<iref primary="true" item="Grammar" subitem="HTAB"/>
187<iref primary="true" item="Grammar" subitem="LF"/>
188<iref primary="true" item="Grammar" subitem="OCTET"/>
189<iref primary="true" item="Grammar" subitem="SP"/>
190<iref primary="true" item="Grammar" subitem="VCHAR"/>
191<t>
192   This specification uses the Augmented Backus-Naur Form (ABNF) notation
193   of <xref target="RFC5234"/> with the list rule extension defined in
194   <xref target="abnf.extension"/>.  <xref target="collected.abnf"/> shows
195   the collected ABNF with the list rule expanded.
196</t>
197<t anchor="core.rules">
198 
199 
200 
201 
202 
203 
204 
205 
206 
207 
208 
209 
210   The following core rules are included by
211   reference, as defined in <xref target="RFC5234"/>, Appendix B.1:
212   ALPHA (letters), CR (carriage return), CRLF (CR LF), CTL (controls),
213   DIGIT (decimal 0-9), DQUOTE (double quote),
214   HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line feed),
215   OCTET (any 8-bit sequence of data), SP (space), and
216   VCHAR (any visible <xref target="USASCII"/> character).
217</t>
218<t>
219   As a convention, ABNF rule names prefixed with "obs-" denote
220   "obsolete" grammar rules that appear for historical reasons.
221</t>
222</section>
223</section>
224
225<section title="Architecture" anchor="architecture">
226<t>
227   HTTP was created for the World Wide Web architecture
228   and has evolved over time to support the scalability needs of a worldwide
229   hypertext system. Much of that architecture is reflected in the terminology
230   and syntax productions used to define HTTP.
231</t>
232
233<section title="Client/Server Messaging" anchor="operation">
234<iref primary="true" item="client"/>
235<iref primary="true" item="server"/>
236<iref primary="true" item="connection"/>
237<t>
238   HTTP is a stateless request/response protocol that operates by exchanging
239   messages (<xref target="http.message"/>) across a reliable
240   transport or session-layer
241   "connection". An HTTP "client" is a
242   program that establishes a connection to a server for the purpose of
243   sending one or more HTTP requests.  An HTTP "server" is a
244   program that accepts connections in order to service HTTP requests by
245   sending HTTP responses.
246</t>
247<iref primary="true" item="user agent"/>
248<iref primary="true" item="origin server"/>
249<iref primary="true" item="browser"/>
250<iref primary="true" item="spider"/>
251<iref primary="true" item="sender"/>
252<iref primary="true" item="recipient"/>
253<t>
254   The terms client and server refer only to the roles that
255   these programs perform for a particular connection.  The same program
256   might act as a client on some connections and a server on others.  We use
257   the term "user agent" to refer to the program that initiates a request,
258   such as a WWW browser, editor, or spider (web-traversing robot), and
259   the term "origin server" to refer to the program that can originate
260   authoritative responses to a request.  For general requirements, we use
261   the term "sender" to refer to whichever component sent a given message
262   and the term "recipient" to refer to any component that receives the
263   message.
264</t>
265<t>
266   HTTP relies upon the Uniform Resource Identifier (URI)
267   standard <xref target="RFC3986"/> to indicate the target resource
268   (<xref target="target-resource"/>) and relationships between resources.
269   Messages are passed in a format similar to that used by Internet mail
270   <xref target="RFC5322"/> and the Multipurpose Internet Mail Extensions
271   (MIME) <xref target="RFC2045"/> (see Appendix A of <xref target="Part2"/> for the differences
272   between HTTP and MIME messages).
273</t>
274<t>
275   Most HTTP communication consists of a retrieval request (GET) for
276   a representation of some resource identified by a URI.  In the
277   simplest case, this might be accomplished via a single bidirectional
278   connection (===) between the user agent (UA) and the origin server (O).
279</t>
280<figure><artwork type="drawing"><![CDATA[
281         request   >
282    UA ======================================= O
283                                <   response
284]]></artwork></figure>
285<iref primary="true" item="message"/>
286<iref primary="true" item="request"/>
287<iref primary="true" item="response"/>
288<t>
289   A client sends an HTTP request to a server in the form of a request
290   message, beginning with a request-line that includes a method, URI, and
291   protocol version (<xref target="request.line"/>),
292   followed by header fields containing
293   request modifiers, client information, and representation metadata
294   (<xref target="header.fields"/>),
295   an empty line to indicate the end of the header section, and finally
296   a message body containing the payload body (if any,
297   <xref target="message.body"/>).
298</t>
299<t>
300   A server responds to a client's request by sending one or more HTTP
301   response
302   messages, each beginning with a status line that
303   includes the protocol version, a success or error code, and textual
304   reason phrase (<xref target="status.line"/>),
305   possibly followed by header fields containing server
306   information, resource metadata, and representation metadata
307   (<xref target="header.fields"/>),
308   an empty line to indicate the end of the header section, and finally
309   a message body containing the payload body (if any,
310   <xref target="message.body"/>).
311</t>
312<t>
313   The following example illustrates a typical message exchange for a
314   GET request on the URI "http://www.example.com/hello.txt":
315</t>
316<figure><preamble>
317client request:
318</preamble><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
319  GET /hello.txt HTTP/1.1
320  User-Agent: curl/7.16.3 libcurl/7.16.3 OpenSSL/0.9.7l zlib/1.2.3
321  Host: www.example.com
322  Accept-Language: en, mi
323 
324  ]]></artwork></figure>
325<figure><preamble>
326server response:
327</preamble><artwork type="message/http; msgtype=&#34;response&#34;"><![CDATA[
328  HTTP/1.1 200 OK
329  Date: Mon, 27 Jul 2009 12:28:53 GMT
330  Server: Apache
331  Last-Modified: Wed, 22 Jul 2009 19:15:56 GMT
332  ETag: "34aa387-d-1568eb00"
333  Accept-Ranges: bytes
334  Content-Length: 14
335  Vary: Accept-Encoding
336  Content-Type: text/plain
337 
338  Hello World!
339  ]]></artwork></figure>
340</section>
341
342<section title="Implementation Diversity" anchor="implementation-diversity">
343<t>
344   When considering the design of HTTP, it is easy to fall into a trap of
345   thinking that all user agents are general-purpose browsers and all origin
346   servers are large public websites. That is not the case in practice.
347   Common HTTP user agents include household appliances, stereos, scales,
348   software/firmware updaters, command-line programs, mobile apps,
349   and communication devices in a multitude of shapes and sizes.  Likewise,
350   common HTTP origin servers include home automation units, configurable
351   networking components, office machines, autonomous robots, news feeds,
352   traffic cameras, ad selectors, and video delivery platforms.
353</t>
354<t>
355   The term "user agent" does not imply that there is a human user directly
356   interacting with the software agent at the time of a request. In many
357   cases, a user agent is installed or configured to run in the background
358   and save its results for later inspection (or save only a subset of those
359   results that might be interesting or erroneous). Spiders, for example, are
360   typically given a start URI and configured to follow certain behavior while
361   crawling the Web as a hypertext graph.
362</t>
363<t>
364   The implementation diversity of HTTP means that we cannot assume the
365   user agent can make interactive suggestions to a user or provide adequate
366   warning for security or privacy options.  In the few cases where this
367   specification requires reporting of errors to the user, it is acceptable
368   for such reporting to only be visible in an error console or log file.
369   Likewise, requirements that an automated action be confirmed by the user
370   before proceeding can me met via advance configuration choices,
371   run-time options, or simply not proceeding with the unsafe action.
372</t>
373</section>
374
375<section title="Connections and Transport Independence" anchor="transport-independence">
376<t>
377   HTTP messaging is independent of the underlying transport or
378   session-layer connection protocol(s).  HTTP only presumes a reliable
379   transport with in-order delivery of requests and the corresponding
380   in-order delivery of responses.  The mapping of HTTP request and
381   response structures onto the data units of the underlying transport
382   protocol is outside the scope of this specification.
383</t>
384<t>
385   The specific connection protocols to be used for an interaction
386   are determined by client configuration and the target URI
387   (<xref target="target-resource"/>).
388   For example, the "http" URI scheme
389   (<xref target="http.uri"/>) indicates a default connection of TCP
390   over IP, with a default TCP port of 80, but the client might be
391   configured to use a proxy via some other connection port or protocol
392   instead of using the defaults.
393</t>
394<t>
395   A connection might be used for multiple HTTP request/response exchanges,
396   as defined in <xref target="persistent.connections"/>.
397</t>
398</section>
399
400<section title="Intermediaries" anchor="intermediaries">
401<iref primary="true" item="intermediary"/>
402<t>
403   HTTP enables the use of intermediaries to satisfy requests through
404   a chain of connections.  There are three common forms of HTTP
405   intermediary: proxy, gateway, and tunnel.  In some cases,
406   a single intermediary might act as an origin server, proxy, gateway,
407   or tunnel, switching behavior based on the nature of each request.
408</t>
409<figure><artwork type="drawing"><![CDATA[
410         >             >             >             >
411    UA =========== A =========== B =========== C =========== O
412               <             <             <             <
413]]></artwork></figure>
414<t>
415   The figure above shows three intermediaries (A, B, and C) between the
416   user agent and origin server. A request or response message that
417   travels the whole chain will pass through four separate connections.
418   Some HTTP communication options
419   might apply only to the connection with the nearest, non-tunnel
420   neighbor, only to the end-points of the chain, or to all connections
421   along the chain. Although the diagram is linear, each participant might
422   be engaged in multiple, simultaneous communications. For example, B
423   might be receiving requests from many clients other than A, and/or
424   forwarding requests to servers other than C, at the same time that it
425   is handling A's request.
426</t>
427<t>
428<iref primary="true" item="upstream"/><iref primary="true" item="downstream"/>
429<iref primary="true" item="inbound"/><iref primary="true" item="outbound"/>
430   We use the terms "upstream" and "downstream"
431   to describe various requirements in relation to the directional flow of a
432   message: all messages flow from upstream to downstream.
433   Likewise, we use the terms inbound and outbound to refer to
434   directions in relation to the request path:
435   "inbound" means toward the origin server and
436   "outbound" means toward the user agent.
437</t>
438<t><iref primary="true" item="proxy"/>
439   A "proxy" is a message forwarding agent that is selected by the
440   client, usually via local configuration rules, to receive requests
441   for some type(s) of absolute URI and attempt to satisfy those
442   requests via translation through the HTTP interface.  Some translations
443   are minimal, such as for proxy requests for "http" URIs, whereas
444   other requests might require translation to and from entirely different
445   application-layer protocols. Proxies are often used to group an
446   organization's HTTP requests through a common intermediary for the
447   sake of security, annotation services, or shared caching.
448</t>
449<t>
450<iref primary="true" item="transforming proxy"/>
451<iref primary="true" item="non-transforming proxy"/>
452   An HTTP-to-HTTP proxy is called a "transforming proxy" if it is designed
453   or configured to modify request or response messages in a semantically
454   meaningful way (i.e., modifications, beyond those required by normal
455   HTTP processing, that change the message in a way that would be
456   significant to the original sender or potentially significant to
457   downstream recipients).  For example, a transforming proxy might be
458   acting as a shared annotation server (modifying responses to include
459   references to a local annotation database), a malware filter, a
460   format transcoder, or an intranet-to-Internet privacy filter.  Such
461   transformations are presumed to be desired by the client (or client
462   organization) that selected the proxy and are beyond the scope of
463   this specification.  However, when a proxy is not intended to transform
464   a given message, we use the term "non-transforming proxy" to target
465   requirements that preserve HTTP message semantics. See Section 4.4.4 of <xref target="Part2"/> and
466   Section 7.6 of <xref target="Part6"/> for status and warning codes related to transformations.
467</t>
468<t><iref primary="true" item="gateway"/><iref primary="true" item="reverse proxy"/>
469<iref primary="true" item="accelerator"/>
470   A "gateway" (a.k.a., "reverse proxy")
471   is a receiving agent that acts
472   as a layer above some other server(s) and translates the received
473   requests to the underlying server's protocol.  Gateways are often
474   used to encapsulate legacy or untrusted information services, to
475   improve server performance through "accelerator" caching, and to
476   enable partitioning or load-balancing of HTTP services across
477   multiple machines.
478</t>
479<t>
480   A gateway behaves as an origin server on its outbound connection and
481   as a user agent on its inbound connection.
482   All HTTP requirements applicable to an origin server
483   also apply to the outbound communication of a gateway.
484   A gateway communicates with inbound servers using any protocol that
485   it desires, including private extensions to HTTP that are outside
486   the scope of this specification.  However, an HTTP-to-HTTP gateway
487   that wishes to interoperate with third-party HTTP servers MUST
488   conform to HTTP user agent requirements on the gateway's inbound
489   connection and MUST implement the <xref target="header.connection" format="none">Connection</xref>
490   (<xref target="header.connection"/>) and <xref target="header.via" format="none">Via</xref>
491   (<xref target="header.via"/>) header fields for both connections.
492</t>
493<t><iref primary="true" item="tunnel"/>
494   A "tunnel" acts as a blind relay between two connections
495   without changing the messages. Once active, a tunnel is not
496   considered a party to the HTTP communication, though the tunnel might
497   have been initiated by an HTTP request. A tunnel ceases to exist when
498   both ends of the relayed connection are closed. Tunnels are used to
499   extend a virtual connection through an intermediary, such as when
500   transport-layer security is used to establish private communication
501   through a shared firewall proxy.
502</t>
503<t><iref primary="true" item="interception proxy"/>
504<iref primary="true" item="transparent proxy"/>
505<iref primary="true" item="captive portal"/>
506   The above categories for intermediary only consider those acting as
507   participants in the HTTP communication.  There are also intermediaries
508   that can act on lower layers of the network protocol stack, filtering or
509   redirecting HTTP traffic without the knowledge or permission of message
510   senders. Network intermediaries often introduce security flaws or
511   interoperability problems by violating HTTP semantics.  For example, an
512   "interception proxy" <xref target="RFC3040"/> (also commonly
513   known as a "transparent proxy" <xref target="RFC1919"/> or
514   "captive portal")
515   differs from an HTTP proxy because it is not selected by the client.
516   Instead, an interception proxy filters or redirects outgoing TCP port 80
517   packets (and occasionally other common port traffic).
518   Interception proxies are commonly found on public network access points,
519   as a means of enforcing account subscription prior to allowing use of
520   non-local Internet services, and within corporate firewalls to enforce
521   network usage policies.
522   They are indistinguishable from a man-in-the-middle attack.
523</t>
524<t>
525   HTTP is defined as a stateless protocol, meaning that each request message
526   can be understood in isolation.  Many implementations depend on HTTP's
527   stateless design in order to reuse proxied connections or dynamically
528   load balance requests across multiple servers.  Hence, servers MUST NOT
529   assume that two requests on the same connection are from the same user
530   agent unless the connection is secured and specific to that agent.
531   Some non-standard HTTP extensions (e.g., <xref target="RFC4559"/>) have
532   been known to violate this requirement, resulting in security and
533   interoperability problems.
534</t>
535</section>
536
537<section title="Caches" anchor="caches">
538<iref primary="true" item="cache"/>
539<t>
540   A "cache" is a local store of previous response messages and the
541   subsystem that controls its message storage, retrieval, and deletion.
542   A cache stores cacheable responses in order to reduce the response
543   time and network bandwidth consumption on future, equivalent
544   requests. Any client or server MAY employ a cache, though a cache
545   cannot be used by a server while it is acting as a tunnel.
546</t>
547<t>
548   The effect of a cache is that the request/response chain is shortened
549   if one of the participants along the chain has a cached response
550   applicable to that request. The following illustrates the resulting
551   chain if B has a cached copy of an earlier response from O (via C)
552   for a request which has not been cached by UA or A.
553</t>
554<figure><artwork type="drawing"><![CDATA[
555            >             >
556       UA =========== A =========== B - - - - - - C - - - - - - O
557                  <             <
558]]></artwork></figure>
559<t><iref primary="true" item="cacheable"/>
560   A response is "cacheable" if a cache is allowed to store a copy of
561   the response message for use in answering subsequent requests.
562   Even when a response is cacheable, there might be additional
563   constraints placed by the client or by the origin server on when
564   that cached response can be used for a particular request. HTTP
565   requirements for cache behavior and cacheable responses are
566   defined in Section 2 of <xref target="Part6"/>. 
567</t>
568<t>
569   There are a wide variety of architectures and configurations
570   of caches and proxies deployed across the World Wide Web and
571   inside large organizations. These systems include national hierarchies
572   of proxy caches to save transoceanic bandwidth, systems that
573   broadcast or multicast cache entries, organizations that distribute
574   subsets of cached data via optical media, and so on.
575</t>
576</section>
577
578<section title="Conformance and Error Handling" anchor="intro.conformance.and.error.handling">
579<t>
580   This specification targets conformance criteria according to the role of
581   a participant in HTTP communication.  Hence, HTTP requirements are placed
582   on senders, recipients, clients, servers, user agents, intermediaries,
583   origin servers, proxies, gateways, or caches, depending on what behavior
584   is being constrained by the requirement.
585</t>
586<t>
587   The verb "generate" is used instead of "send" where a requirement
588   differentiates between creating a protocol element and merely forwarding a
589   received element downstream.
590</t>
591<t>
592   An implementation is considered conformant if it complies with all of the
593   requirements associated with the roles it partakes in HTTP. Note that
594   SHOULD-level requirements are relevant here, unless one of the documented
595   exceptions is applicable.
596</t>
597<t>
598   In addition to the prose requirements placed upon them, senders MUST NOT
599   generate protocol elements that do not match the grammar defined by the
600   ABNF rules for those protocol elements that are applicable to the sender's
601   role. If a received protocol element is processed, the recipient MUST be
602   able to parse any value that would match the ABNF rules for that protocol
603   element, excluding only those rules not applicable to the recipient's role.
604</t>
605<t>
606   Unless noted otherwise, a recipient MAY attempt to recover a usable
607   protocol element from an invalid construct.  HTTP does not define
608   specific error handling mechanisms except when they have a direct impact
609   on security, since different applications of the protocol require
610   different error handling strategies.  For example, a Web browser might
611   wish to transparently recover from a response where the
612   Location header field doesn't parse according to the ABNF,
613   whereas a systems control client might consider any form of error recovery
614   to be dangerous.
615</t>
616</section>
617
618<section title="Protocol Versioning" anchor="http.version">
619 
620 
621<t>
622   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate
623   versions of the protocol. This specification defines version "1.1".
624   The protocol version as a whole indicates the sender's conformance
625   with the set of requirements laid out in that version's corresponding
626   specification of HTTP.
627</t>
628<t>
629   The version of an HTTP message is indicated by an HTTP-version field
630   in the first line of the message. HTTP-version is case-sensitive.
631</t>
632<figure><iref primary="true" item="Grammar" subitem="HTTP-version"/><iref primary="true" item="Grammar" subitem="HTTP-name"/><artwork type="abnf2616"><![CDATA[
633  HTTP-version  = HTTP-name "/" DIGIT "." DIGIT
634  HTTP-name     = %x48.54.54.50 ; "HTTP", case-sensitive
635]]></artwork></figure>
636<t>
637   The HTTP version number consists of two decimal digits separated by a "."
638   (period or decimal point).  The first digit ("major version") indicates the
639   HTTP messaging syntax, whereas the second digit ("minor version") indicates
640   the highest minor version to which the sender is
641   conformant and able to understand for future communication.  The minor
642   version advertises the sender's communication capabilities even when the
643   sender is only using a backwards-compatible subset of the protocol,
644   thereby letting the recipient know that more advanced features can
645   be used in response (by servers) or in future requests (by clients).
646</t>
647<t>
648   When an HTTP/1.1 message is sent to an HTTP/1.0 recipient
649   <xref target="RFC1945"/> or a recipient whose version is unknown,
650   the HTTP/1.1 message is constructed such that it can be interpreted
651   as a valid HTTP/1.0 message if all of the newer features are ignored.
652   This specification places recipient-version requirements on some
653   new features so that a conformant sender will only use compatible
654   features until it has determined, through configuration or the
655   receipt of a message, that the recipient supports HTTP/1.1.
656</t>
657<t>
658   The interpretation of a header field does not change between minor
659   versions of the same major HTTP version, though the default
660   behavior of a recipient in the absence of such a field can change.
661   Unless specified otherwise, header fields defined in HTTP/1.1 are
662   defined for all versions of HTTP/1.x.  In particular, the <xref target="header.host" format="none">Host</xref>
663   and <xref target="header.connection" format="none">Connection</xref> header fields ought to be implemented by all
664   HTTP/1.x implementations whether or not they advertise conformance with
665   HTTP/1.1.
666</t>
667<t>
668   New header fields can be defined such that, when they are
669   understood by a recipient, they might override or enhance the
670   interpretation of previously defined header fields.  When an
671   implementation receives an unrecognized header field, the recipient
672   MUST ignore that header field for local processing regardless of
673   the message's HTTP version.  An unrecognized header field received
674   by a proxy MUST be forwarded downstream unless the header field's
675   field-name is listed in the message's <xref target="header.connection" format="none">Connection</xref> header field
676   (see <xref target="header.connection"/>).
677   These requirements allow HTTP's functionality to be enhanced without
678   requiring prior update of deployed intermediaries.
679</t>
680<t>
681   Intermediaries that process HTTP messages (i.e., all intermediaries
682   other than those acting as tunnels) MUST send their own HTTP-version
683   in forwarded messages.  In other words, they MUST NOT blindly
684   forward the first line of an HTTP message without ensuring that the
685   protocol version in that message matches a version to which that
686   intermediary is conformant for both the receiving and
687   sending of messages.  Forwarding an HTTP message without rewriting
688   the HTTP-version might result in communication errors when downstream
689   recipients use the message sender's version to determine what features
690   are safe to use for later communication with that sender.
691</t>
692<t>
693   An HTTP client SHOULD send a request version equal to the highest
694   version to which the client is conformant and
695   whose major version is no higher than the highest version supported
696   by the server, if this is known.  An HTTP client MUST NOT send a
697   version to which it is not conformant.
698</t>
699<t>
700   An HTTP client MAY send a lower request version if it is known that
701   the server incorrectly implements the HTTP specification, but only
702   after the client has attempted at least one normal request and determined
703   from the response status or header fields (e.g., Server) that
704   the server improperly handles higher request versions.
705</t>
706<t>
707   An HTTP server SHOULD send a response version equal to the highest
708   version to which the server is conformant and
709   whose major version is less than or equal to the one received in the
710   request.  An HTTP server MUST NOT send a version to which it is not
711   conformant.  A server MAY send a 505 (HTTP Version Not
712   Supported) response if it cannot send a response using the
713   major version used in the client's request.
714</t>
715<t>
716   An HTTP server MAY send an HTTP/1.0 response to an HTTP/1.0 request
717   if it is known or suspected that the client incorrectly implements the
718   HTTP specification and is incapable of correctly processing later
719   version responses, such as when a client fails to parse the version
720   number correctly or when an intermediary is known to blindly forward
721   the HTTP-version even when it doesn't conform to the given minor
722   version of the protocol. Such protocol downgrades SHOULD NOT be
723   performed unless triggered by specific client attributes, such as when
724   one or more of the request header fields (e.g., User-Agent)
725   uniquely match the values sent by a client known to be in error.
726</t>
727<t>
728   The intention of HTTP's versioning design is that the major number
729   will only be incremented if an incompatible message syntax is
730   introduced, and that the minor number will only be incremented when
731   changes made to the protocol have the effect of adding to the message
732   semantics or implying additional capabilities of the sender.  However,
733   the minor version was not incremented for the changes introduced between
734   <xref target="RFC2068"/> and <xref target="RFC2616"/>, and this revision
735   is specifically avoiding any such changes to the protocol.
736</t>
737</section>
738
739<section title="Uniform Resource Identifiers" anchor="uri">
740<iref primary="true" item="resource"/>
741<t>
742   Uniform Resource Identifiers (URIs) <xref target="RFC3986"/> are used
743   throughout HTTP as the means for identifying resources. URI references
744   are used to target requests, indicate redirects, and define relationships.
745   HTTP does not limit what a resource might be; it merely defines an interface
746   that can be used to interact with a resource via HTTP. More information on
747   the scope of URIs and resources can be found in <xref target="RFC3986"/>.
748</t>
749 
750 
751 
752 
753 
754 
755 
756 
757 
758 
759<t>
760   This specification adopts the definitions of "URI-reference",
761   "absolute-URI", "relative-part", "port", "host",
762   "path-abempty", "path-absolute", "query", and "authority" from the
763   URI generic syntax <xref target="RFC3986"/>.
764   In addition, we define a partial-URI rule for protocol elements
765   that allow a relative URI but not a fragment.
766</t>
767<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="path-absolute"/><iref primary="true" item="Grammar" subitem="port"/><iref primary="true" item="Grammar" subitem="query"/><iref primary="true" item="Grammar" subitem="uri-host"/><iref primary="true" item="Grammar" subitem="partial-URI"><!--exported production--></iref><artwork type="abnf2616"><![CDATA[
768  URI-reference = <URI-reference, defined in [RFC3986], Section 4.1>
769  absolute-URI  = <absolute-URI, defined in [RFC3986], Section 4.3>
770  relative-part = <relative-part, defined in [RFC3986], Section 4.2>
771  authority     = <authority, defined in [RFC3986], Section 3.2>
772  path-abempty  = <path-abempty, defined in [RFC3986], Section 3.3>
773  path-absolute = <path-absolute, defined in [RFC3986], Section 3.3>
774  port          = <port, defined in [RFC3986], Section 3.2.3>
775  query         = <query, defined in [RFC3986], Section 3.4>
776  uri-host      = <host, defined in [RFC3986], Section 3.2.2>
777 
778  partial-URI   = relative-part [ "?" query ]
779]]></artwork></figure>
780<t>
781   Each protocol element in HTTP that allows a URI reference will indicate
782   in its ABNF production whether the element allows any form of reference
783   (URI-reference), only a URI in absolute form (absolute-URI), only the
784   path and optional query components, or some combination of the above.
785   Unless otherwise indicated, URI references are parsed
786   relative to the effective request URI
787   (<xref target="effective.request.uri"/>).
788</t>
789
790<section title="http URI scheme" anchor="http.uri">
791 
792  <iref item="http URI scheme" primary="true"/>
793  <iref item="URI scheme" subitem="http" primary="true"/>
794<t>
795   The "http" URI scheme is hereby defined for the purpose of minting
796   identifiers according to their association with the hierarchical
797   namespace governed by a potential HTTP origin server listening for
798   TCP connections on a given port.
799</t>
800<figure><iref primary="true" item="Grammar" subitem="http-URI"><!--terminal production--></iref><artwork type="abnf2616"><![CDATA[
801  http-URI = "http:" "//" authority path-abempty [ "?" query ]
802]]></artwork></figure>
803<t>
804   The HTTP origin server is identified by the generic syntax's
805   <xref target="uri" format="none">authority</xref> component, which includes a host identifier
806   and optional TCP port (<xref target="RFC3986"/>, Section 3.2.2).
807   The remainder of the URI, consisting of both the hierarchical path
808   component and optional query component, serves as an identifier for
809   a potential resource within that origin server's name space.
810</t>
811<t>
812   If the host identifier is provided as an IP literal or IPv4 address,
813   then the origin server is any listener on the indicated TCP port at
814   that IP address. If host is a registered name, then that name is
815   considered an indirect identifier and the recipient might use a name
816   resolution service, such as DNS, to find the address of a listener
817   for that host.
818   The host MUST NOT be empty; if an "http" URI is received with an
819   empty host, then it MUST be rejected as invalid.
820   If the port subcomponent is empty or not given, then TCP port 80 is
821   assumed (the default reserved port for WWW services).
822</t>
823<t>
824   Regardless of the form of host identifier, access to that host is not
825   implied by the mere presence of its name or address. The host might or might
826   not exist and, even when it does exist, might or might not be running an
827   HTTP server or listening to the indicated port. The "http" URI scheme
828   makes use of the delegated nature of Internet names and addresses to
829   establish a naming authority (whatever entity has the ability to place
830   an HTTP server at that Internet name or address) and allows that
831   authority to determine which names are valid and how they might be used.
832</t>
833<t>
834   When an "http" URI is used within a context that calls for access to the
835   indicated resource, a client MAY attempt access by resolving
836   the host to an IP address, establishing a TCP connection to that address
837   on the indicated port, and sending an HTTP request message
838   (<xref target="http.message"/>) containing the URI's identifying data
839   (<xref target="message.routing"/>) to the server.
840   If the server responds to that request with a non-interim HTTP response
841   message, as described in Section 4 of <xref target="Part2"/>, then that response
842   is considered an authoritative answer to the client's request.
843</t>
844<t>
845   Although HTTP is independent of the transport protocol, the "http"
846   scheme is specific to TCP-based services because the name delegation
847   process depends on TCP for establishing authority.
848   An HTTP service based on some other underlying connection protocol
849   would presumably be identified using a different URI scheme, just as
850   the "https" scheme (below) is used for servers that require an SSL/TLS
851   transport layer on a connection. Other protocols might also be used to
852   provide access to "http" identified resources — it is only the
853   authoritative interface used for mapping the namespace that is
854   specific to TCP.
855</t>
856<t>
857   The URI generic syntax for authority also includes a deprecated
858   userinfo subcomponent (<xref target="RFC3986"/>, Section 3.2.1)
859   for including user authentication information in the URI.  Some
860   implementations make use of the userinfo component for internal
861   configuration of authentication information, such as within command
862   invocation options, configuration files, or bookmark lists, even
863   though such usage might expose a user identifier or password.
864   Senders MUST NOT include a userinfo subcomponent (and its "@"
865   delimiter) when transmitting an "http" URI in a message.  Recipients
866   of HTTP messages that contain a URI reference SHOULD parse for the
867   existence of userinfo and treat its presence as an error, likely
868   indicating that the deprecated subcomponent is being used to obscure
869   the authority for the sake of phishing attacks.
870</t>
871</section>
872
873<section title="https URI scheme" anchor="https.uri">
874   
875   <iref item="https URI scheme"/>
876   <iref item="URI scheme" subitem="https"/>
877<t>
878   The "https" URI scheme is hereby defined for the purpose of minting
879   identifiers according to their association with the hierarchical
880   namespace governed by a potential HTTP origin server listening for
881   SSL/TLS-secured connections on a given TCP port.
882</t>
883<t>
884   All of the requirements listed above for the "http" scheme are also
885   requirements for the "https" scheme, except that a default TCP port
886   of 443 is assumed if the port subcomponent is empty or not given,
887   and the TCP connection MUST be secured for privacy through the
888   use of strong encryption prior to sending the first HTTP request.
889</t>
890<figure><iref primary="true" item="Grammar" subitem="https-URI"><!--terminal production--></iref><artwork type="abnf2616"><![CDATA[
891  https-URI = "https:" "//" authority path-abempty [ "?" query ]
892]]></artwork></figure>
893<t>
894   Unlike the "http" scheme, responses to "https" identified requests
895   are never "public" and thus MUST NOT be reused for shared caching.
896   They can, however, be reused in a private cache if the message is
897   cacheable by default in HTTP or specifically indicated as such by
898   the Cache-Control header field (Section 7.2 of <xref target="Part6"/>).
899</t>
900<t>
901   Resources made available via the "https" scheme have no shared
902   identity with the "http" scheme even if their resource identifiers
903   indicate the same authority (the same host listening to the same
904   TCP port).  They are distinct name spaces and are considered to be
905   distinct origin servers.  However, an extension to HTTP that is
906   defined to apply to entire host domains, such as the Cookie protocol
907   <xref target="RFC6265"/>, can allow information
908   set by one service to impact communication with other services
909   within a matching group of host domains.
910</t>
911<t>
912   The process for authoritative access to an "https" identified
913   resource is defined in <xref target="RFC2818"/>.
914</t>
915</section>
916
917<section title="http and https URI Normalization and Comparison" anchor="uri.comparison">
918<t>
919   Since the "http" and "https" schemes conform to the URI generic syntax,
920   such URIs are normalized and compared according to the algorithm defined
921   in <xref target="RFC3986"/>, Section 6, using the defaults
922   described above for each scheme.
923</t>
924<t>
925   If the port is equal to the default port for a scheme, the normal
926   form is to elide the port subcomponent. Likewise, an empty path
927   component is equivalent to an absolute path of "/", so the normal
928   form is to provide a path of "/" instead. The scheme and host
929   are case-insensitive and normally provided in lowercase; all
930   other components are compared in a case-sensitive manner.
931   Characters other than those in the "reserved" set are equivalent
932   to their percent-encoded octets (see <xref target="RFC3986"/>, Section 2.1): the normal form is to not encode them.
933</t>
934<t>
935   For example, the following three URIs are equivalent:
936</t>
937<figure><artwork type="example"><![CDATA[
938   http://example.com:80/~smith/home.html
939   http://EXAMPLE.com/%7Esmith/home.html
940   http://EXAMPLE.com:/%7esmith/home.html
941]]></artwork></figure>
942</section>
943</section>
944</section>
945
946<section title="Message Format" anchor="http.message">
947
948
949
950
951<iref item="header section"/>
952<iref item="headers"/>
953<iref item="header field"/>
954<t>
955   All HTTP/1.1 messages consist of a start-line followed by a sequence of
956   octets in a format similar to the Internet Message Format
957   <xref target="RFC5322"/>: zero or more header fields (collectively
958   referred to as the "headers" or the "header section"), an empty line
959   indicating the end of the header section, and an optional message body.
960</t>
961<figure><iref primary="true" item="Grammar" subitem="HTTP-message"><!--terminal production--></iref><artwork type="abnf2616"><![CDATA[
962  HTTP-message   = start-line
963                   *( header-field CRLF )
964                   CRLF
965                   [ message-body ]
966]]></artwork></figure>
967<t>
968   The normal procedure for parsing an HTTP message is to read the
969   start-line into a structure, read each header field into a hash
970   table by field name until the empty line, and then use the parsed
971   data to determine if a message body is expected.  If a message body
972   has been indicated, then it is read as a stream until an amount
973   of octets equal to the message body length is read or the connection
974   is closed.
975</t>
976<t>
977   Recipients MUST parse an HTTP message as a sequence of octets in an
978   encoding that is a superset of US-ASCII <xref target="USASCII"/>.
979   Parsing an HTTP message as a stream of Unicode characters, without regard
980   for the specific encoding, creates security vulnerabilities due to the
981   varying ways that string processing libraries handle invalid multibyte
982   character sequences that contain the octet LF (%x0A).  String-based
983   parsers can only be safely used within protocol elements after the element
984   has been extracted from the message, such as within a header field-value
985   after message parsing has delineated the individual fields.
986</t>
987<t>
988   An HTTP message can be parsed as a stream for incremental processing or
989   forwarding downstream.  However, recipients cannot rely on incremental
990   delivery of partial messages, since some implementations will buffer or
991   delay message forwarding for the sake of network efficiency, security
992   checks, or payload transformations.
993</t>
994
995<section title="Start Line" anchor="start.line">
996 
997<t>
998   An HTTP message can either be a request from client to server or a
999   response from server to client.  Syntactically, the two types of message
1000   differ only in the start-line, which is either a request-line (for requests)
1001   or a status-line (for responses), and in the algorithm for determining
1002   the length of the message body (<xref target="message.body"/>).
1003   In theory, a client could receive requests and a server could receive
1004   responses, distinguishing them by their different start-line formats,
1005   but in practice servers are implemented to only expect a request
1006   (a response is interpreted as an unknown or invalid request method)
1007   and clients are implemented to only expect a response.
1008</t>
1009<figure><iref primary="true" item="Grammar" subitem="start-line"/><artwork type="abnf2616"><![CDATA[
1010  start-line     = request-line / status-line
1011]]></artwork></figure>
1012<t>
1013   Implementations MUST NOT send whitespace between the start-line and
1014   the first header field. The presence of such whitespace in a request
1015   might be an attempt to trick a server into ignoring that field or
1016   processing the line after it as a new request, either of which might
1017   result in a security vulnerability if other implementations within
1018   the request chain interpret the same message differently.
1019   Likewise, the presence of such whitespace in a response might be
1020   ignored by some clients or cause others to cease parsing.
1021</t>
1022
1023<section title="Request Line" anchor="request.line">
1024 
1025 
1026<t>
1027   A request-line begins with a method token, followed by a single
1028   space (SP), the request-target, another single space (SP), the
1029   protocol version, and ending with CRLF.
1030</t>
1031<figure><iref primary="true" item="Grammar" subitem="request-line"/><artwork type="abnf2616"><![CDATA[
1032  request-line   = method SP request-target SP HTTP-version CRLF
1033]]></artwork></figure>
1034<t>
1035   A server MUST be able to parse any received message that begins
1036   with a request-line and matches the ABNF rule for HTTP-message.
1037</t>
1038<iref primary="true" item="method"/>
1039<t anchor="method">
1040   The method token indicates the request method to be performed on the
1041   target resource. The request method is case-sensitive.
1042</t>
1043<figure><iref primary="true" item="Grammar" subitem="method"/><artwork type="abnf2616"><![CDATA[
1044  method         = token
1045]]></artwork></figure>
1046<t>
1047   The methods defined by this specification can be found in
1048   Section 2 of <xref target="Part2"/>, along with information regarding the HTTP method registry
1049   and considerations for defining new methods.
1050</t>
1051<iref item="request-target"/>
1052<t>
1053   The request-target identifies the target resource upon which to apply
1054   the request, as defined in <xref target="request-target"/>.
1055</t>
1056<t>
1057   No whitespace is allowed inside the method, request-target, and
1058   protocol version.  Hence, recipients typically parse the request-line
1059   into its component parts by splitting on the SP characters.
1060</t>
1061<t>
1062   Unfortunately, some user agents fail to properly encode hypertext
1063   references that have embedded whitespace, sending the characters
1064   directly instead of properly percent-encoding the disallowed characters.
1065   Recipients of an invalid request-line SHOULD respond with either a
1066   400 (Bad Request) error or a 301 (Moved Permanently)
1067   redirect with the request-target properly encoded.  Recipients SHOULD NOT
1068   attempt to autocorrect and then process the request without a redirect,
1069   since the invalid request-line might be deliberately crafted to bypass
1070   security filters along the request chain.
1071</t>
1072<t>
1073   HTTP does not place a pre-defined limit on the length of a request-line.
1074   A server that receives a method longer than any that it implements
1075   SHOULD respond with either a 405 (Method Not Allowed), if it is an origin
1076   server, or a 501 (Not Implemented) status code.
1077   A server MUST be prepared to receive URIs of unbounded length and
1078   respond with the 414 (URI Too Long) status code if the received
1079   request-target would be longer than the server wishes to handle
1080   (see Section 4.6.12 of <xref target="Part2"/>).
1081</t>
1082<t>
1083   Various ad-hoc limitations on request-line length are found in practice.
1084   It is RECOMMENDED that all HTTP senders and recipients support, at a
1085   minimum, request-line lengths of up to 8000 octets.
1086</t>
1087</section>
1088
1089<section title="Status Line" anchor="status.line">
1090 
1091 
1092 
1093 
1094<t>
1095   The first line of a response message is the status-line, consisting
1096   of the protocol version, a space (SP), the status code, another space,
1097   a possibly-empty textual phrase describing the status code, and
1098   ending with CRLF.
1099</t>
1100<figure><iref primary="true" item="Grammar" subitem="status-line"/><artwork type="abnf2616"><![CDATA[
1101  status-line = HTTP-version SP status-code SP reason-phrase CRLF
1102]]></artwork></figure>
1103<t>
1104   A client MUST be able to parse any received message that begins
1105   with a status-line and matches the ABNF rule for HTTP-message.
1106</t>
1107<t>
1108   The status-code element is a 3-digit integer code describing the
1109   result of the server's attempt to understand and satisfy the client's
1110   corresponding request. The rest of the response message is to be
1111   interpreted in light of the semantics defined for that status code.
1112   See Section 4 of <xref target="Part2"/> for information about the semantics of status codes,
1113   including the classes of status code (indicated by the first digit),
1114   the status codes defined by this specification, considerations for the
1115   definition of new status codes, and the IANA registry.
1116</t>
1117<figure><iref primary="true" item="Grammar" subitem="status-code"/><artwork type="abnf2616"><![CDATA[
1118  status-code    = 3DIGIT
1119]]></artwork></figure>
1120<t>  
1121   The reason-phrase element exists for the sole purpose of providing a
1122   textual description associated with the numeric status code, mostly
1123   out of deference to earlier Internet application protocols that were more
1124   frequently used with interactive text clients. A client SHOULD ignore
1125   the reason-phrase content.
1126</t>
1127<figure><iref primary="true" item="Grammar" subitem="reason-phrase"/><artwork type="abnf2616"><![CDATA[
1128  reason-phrase  = *( HTAB / SP / VCHAR / obs-text )
1129]]></artwork></figure>
1130</section>
1131</section>
1132
1133<section title="Header Fields" anchor="header.fields">
1134 
1135 
1136 
1137 
1138 
1139<t>
1140   Each HTTP header field consists of a case-insensitive field name
1141   followed by a colon (":"), optional whitespace, and the field value.
1142</t>
1143<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[
1144  header-field   = field-name ":" OWS field-value BWS
1145  field-name     = token
1146  field-value    = *( field-content / obs-fold )
1147  field-content  = *( HTAB / SP / VCHAR / obs-text )
1148  obs-fold       = CRLF ( SP / HTAB )
1149                 ; obsolete line folding
1150                 ; see Section 3.2.2
1151]]></artwork></figure>
1152<t>
1153   The field-name token labels the corresponding field-value as having the
1154   semantics defined by that header field.  For example, the Date
1155   header field is defined in Section 9.10 of <xref target="Part2"/> as containing the origination
1156   timestamp for the message in which it appears.
1157</t>
1158<t>
1159   HTTP header fields are fully extensible: there is no limit on the
1160   introduction of new field names, each presumably defining new semantics,
1161   or on the number of header fields used in a given message.  Existing
1162   fields are defined in each part of this specification and in many other
1163   specifications outside the standards process.
1164   New header fields can be introduced without changing the protocol version
1165   if their defined semantics allow them to be safely ignored by recipients
1166   that do not recognize them.
1167</t>
1168<t>
1169   New HTTP header fields SHOULD be registered with IANA according
1170   to the procedures in Section 3.1 of <xref target="Part2"/>.
1171   Unrecognized header fields MUST be forwarded by a proxy unless the
1172   field-name is listed in the <xref target="header.connection" format="none">Connection</xref> header field
1173   (<xref target="header.connection"/>) or the proxy is specifically
1174   configured to block or otherwise transform such fields.
1175   Unrecognized header fields SHOULD be ignored by other recipients.
1176</t>
1177<t>
1178   The order in which header fields with differing field names are
1179   received is not significant. However, it is "good practice" to send
1180   header fields that contain control data first, such as <xref target="header.host" format="none">Host</xref>
1181   on requests and Date on responses, so that implementations
1182   can decide when not to handle a message as early as possible.  A server
1183   MUST wait until the entire header section is received before interpreting
1184   a request message, since later header fields might include conditionals,
1185   authentication credentials, or deliberately misleading duplicate
1186   header fields that would impact request processing.
1187</t>
1188<t>
1189   Multiple header fields with the same field name MUST NOT be
1190   sent in a message unless the entire field value for that
1191   header field is defined as a comma-separated list [i.e., #(values)].
1192   Multiple header fields with the same field name can be combined into
1193   one "field-name: field-value" pair, without changing the semantics of the
1194   message, by appending each subsequent field value to the combined
1195   field value in order, separated by a comma. The order in which
1196   header fields with the same field name are received is therefore
1197   significant to the interpretation of the combined field value;
1198   a proxy MUST NOT change the order of these field values when
1199   forwarding a message.
1200</t>
1201<t><list>
1202  <t>
1203   Note: The "Set-Cookie" header field as implemented in
1204   practice can occur multiple times, but does not use the list syntax, and
1205   thus cannot be combined into a single line (<xref target="RFC6265"/>). (See Appendix A.2.3 of <xref target="Kri2001"/>
1206   for details.) Also note that the Set-Cookie2 header field specified in
1207   <xref target="RFC2965"/> does not share this problem.
1208  </t>
1209</list></t>
1210
1211<section title="Whitespace" anchor="whitespace">
1212<t anchor="rule.LWS">
1213   This specification uses three rules to denote the use of linear
1214   whitespace: OWS (optional whitespace), RWS (required whitespace), and
1215   BWS ("bad" whitespace).
1216</t>
1217<t anchor="rule.OWS">
1218   The OWS rule is used where zero or more linear whitespace octets might
1219   appear. OWS SHOULD either not be produced or be produced as a single
1220   SP. Multiple OWS octets that occur within field-content SHOULD either
1221   be replaced with a single SP or transformed to all SP octets (each
1222   octet other than SP replaced with SP) before interpreting the field value
1223   or forwarding the message downstream.
1224</t>
1225<t anchor="rule.RWS">
1226   RWS is used when at least one linear whitespace octet is required to
1227   separate field tokens. RWS SHOULD be produced as a single SP.
1228   Multiple RWS octets that occur within field-content SHOULD either
1229   be replaced with a single SP or transformed to all SP octets before
1230   interpreting the field value or forwarding the message downstream.
1231</t>
1232<t anchor="rule.BWS">
1233   BWS is used where the grammar allows optional whitespace for historical
1234   reasons but senders SHOULD NOT produce it in messages. HTTP/1.1
1235   recipients MUST accept such bad optional whitespace and remove it before
1236   interpreting the field value or forwarding the message downstream.
1237</t>
1238<t anchor="rule.whitespace">
1239 
1240 
1241 
1242</t>
1243<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[
1244  OWS            = *( SP / HTAB )
1245                 ; "optional" whitespace
1246  RWS            = 1*( SP / HTAB )
1247                 ; "required" whitespace
1248  BWS            = OWS
1249                 ; "bad" whitespace
1250]]></artwork></figure>
1251</section>
1252
1253<section title="Field Parsing" anchor="field.parsing">
1254<t>
1255   No whitespace is allowed between the header field-name and colon.
1256   In the past, differences in the handling of such whitespace have led to
1257   security vulnerabilities in request routing and response handling.
1258   Any received request message that contains whitespace between a header
1259   field-name and colon MUST be rejected with a response code of 400
1260   (Bad Request).  A proxy MUST remove any such whitespace from a response
1261   message before forwarding the message downstream.
1262</t>
1263<t>
1264   A field value MAY be preceded by optional whitespace (OWS); a single SP is
1265   preferred. The field value does not include any leading or trailing white
1266   space: OWS occurring before the first non-whitespace octet of the
1267   field value or after the last non-whitespace octet of the field value
1268   is ignored and SHOULD be removed before further processing (as this does
1269   not change the meaning of the header field).
1270</t>
1271<t>
1272   Historically, HTTP header field values could be extended over multiple
1273   lines by preceding each extra line with at least one space or horizontal
1274   tab (obs-fold). This specification deprecates such line
1275   folding except within the message/http media type
1276   (<xref target="internet.media.type.message.http"/>).
1277   HTTP senders MUST NOT produce messages that include line folding
1278   (i.e., that contain any field-value that matches the obs-fold rule) unless
1279   the message is intended for packaging within the message/http media type.
1280   HTTP recipients SHOULD accept line folding and replace any embedded
1281   obs-fold whitespace with either a single SP or a matching number of SP
1282   octets (to avoid buffer copying) prior to interpreting the field value or
1283   forwarding the message downstream.
1284</t>
1285<t>
1286   Historically, HTTP has allowed field content with text in the ISO-8859-1
1287   <xref target="ISO-8859-1"/> character encoding and supported other
1288   character sets only through use of <xref target="RFC2047"/> encoding.
1289   In practice, most HTTP header field values use only a subset of the
1290   US-ASCII character encoding <xref target="USASCII"/>. Newly defined
1291   header fields SHOULD limit their field values to US-ASCII octets.
1292   Recipients SHOULD treat other (obs-text) octets in field content as
1293   opaque data.
1294</t>
1295</section>
1296
1297<section title="Field Length" anchor="field.length">
1298<t>
1299   HTTP does not place a pre-defined limit on the length of header fields,
1300   either in isolation or as a set. A server MUST be prepared to receive
1301   request header fields of unbounded length and respond with a 4xx
1302   (Client Error) status code if the received header field(s) would be
1303   longer than the server wishes to handle.
1304</t>
1305<t>
1306   A client that receives response header fields that are longer than it wishes
1307   to handle can only treat it as a server error.
1308</t>
1309<t>
1310   Various ad-hoc limitations on header field length are found in practice. It
1311   is RECOMMENDED that all HTTP senders and recipients support messages whose
1312   combined header fields have 4000 or more octets.
1313</t>
1314</section>
1315
1316<section title="Field value components" anchor="field.components">
1317<t anchor="rule.token.separators">
1318 
1319 
1320 
1321 
1322   Many HTTP/1.1 header field values consist of words (token or quoted-string)
1323   separated by whitespace or special characters. These special characters
1324   MUST be in a quoted string to be used within a parameter value (as defined
1325   in <xref target="transfer.codings"/>).
1326</t>
1327<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[
1328  word           = token / quoted-string
1329
1330  token          = 1*tchar
1331
1332  tchar          = "!" / "#" / "$" / "%" / "&" / "'" / "*"
1333                 / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
1334                 / DIGIT / ALPHA
1335                 ; any VCHAR, except special
1336
1337  special        = "(" / ")" / "<" / ">" / "@" / ","
1338                 / ";" / ":" / "\" / DQUOTE / "/" / "["
1339                 / "]" / "?" / "=" / "{" / "}"
1340]]></artwork></figure>
1341<t anchor="rule.quoted-string">
1342 
1343 
1344 
1345   A string of text is parsed as a single word if it is quoted using
1346   double-quote marks.
1347</t>
1348<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[
1349  quoted-string  = DQUOTE *( qdtext / quoted-pair ) DQUOTE
1350  qdtext         = OWS / %x21 / %x23-5B / %x5D-7E / obs-text
1351  obs-text       = %x80-FF
1352]]></artwork></figure>
1353<t anchor="rule.quoted-pair">
1354 
1355   The backslash octet ("\") can be used as a single-octet
1356   quoting mechanism within quoted-string constructs:
1357</t>
1358<figure><iref primary="true" item="Grammar" subitem="quoted-pair"/><artwork type="abnf2616"><![CDATA[
1359  quoted-pair    = "\" ( HTAB / SP / VCHAR / obs-text )
1360]]></artwork></figure>
1361<t>
1362   Recipients that process the value of the quoted-string MUST handle a
1363   quoted-pair as if it were replaced by the octet following the backslash.
1364</t>
1365<t>
1366   Senders SHOULD NOT escape octets in quoted-strings that do not require
1367   escaping (i.e., other than DQUOTE and the backslash octet).
1368</t>
1369<t anchor="rule.comment">
1370 
1371 
1372   Comments can be included in some HTTP header fields by surrounding
1373   the comment text with parentheses. Comments are only allowed in
1374   fields containing "comment" as part of their field value definition.
1375</t>
1376<figure><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/><artwork type="abnf2616"><![CDATA[
1377  comment        = "(" *( ctext / quoted-cpair / comment ) ")"
1378  ctext          = OWS / %x21-27 / %x2A-5B / %x5D-7E / obs-text
1379]]></artwork></figure>
1380<t anchor="rule.quoted-cpair">
1381 
1382   The backslash octet ("\") can be used as a single-octet
1383   quoting mechanism within comment constructs:
1384</t>
1385<figure><iref primary="true" item="Grammar" subitem="quoted-cpair"/><artwork type="abnf2616"><![CDATA[
1386  quoted-cpair   = "\" ( HTAB / SP / VCHAR / obs-text )
1387]]></artwork></figure>
1388<t>
1389   Senders SHOULD NOT escape octets in comments that do not require escaping
1390   (i.e., other than the backslash octet "\" and the parentheses "(" and ")").
1391</t>
1392</section>
1393
1394</section>
1395
1396<section title="Message Body" anchor="message.body">
1397 
1398<t>
1399   The message body (if any) of an HTTP message is used to carry the
1400   payload body of that request or response.  The message body is
1401   identical to the payload body unless a transfer coding has been
1402   applied, as described in <xref target="header.transfer-encoding"/>.
1403</t>
1404<figure><iref primary="true" item="Grammar" subitem="message-body"/><artwork type="abnf2616"><![CDATA[
1405  message-body = *OCTET
1406]]></artwork></figure>
1407<t>
1408   The rules for when a message body is allowed in a message differ for
1409   requests and responses.
1410</t>
1411<t>
1412   The presence of a message body in a request is signaled by a
1413   a <xref target="header.content-length" format="none">Content-Length</xref> or <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header
1414   field. Request message framing is independent of method semantics,
1415   even if the method does not define any use for a message body.
1416</t>
1417<t>
1418   The presence of a message body in a response depends on both
1419   the request method to which it is responding and the response
1420   status code (<xref target="status.line"/>).
1421   Responses to the HEAD request method never include a message body
1422   because the associated response header fields (e.g.,
1423   <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>, <xref target="header.content-length" format="none">Content-Length</xref>, etc.) only
1424   indicate what their values would have been if the request method had been
1425   GET. 2xx (Successful) responses to CONNECT switch to tunnel
1426   mode instead of having a message body.
1427   All 1xx (Informational), 204 (No Content), and
1428   304 (Not Modified) responses MUST NOT include a message body.
1429   All other responses do include a message body, although the body
1430   MAY be of zero length.
1431</t>
1432
1433<section title="Transfer-Encoding" anchor="header.transfer-encoding">
1434  <iref primary="true" item="Transfer-Encoding header field"/>
1435  <iref primary="true" item="Header Fields" subitem="Transfer-Encoding"/>
1436 
1437<t>
1438   When one or more transfer codings are applied to a payload body in order
1439   to form the message body, a Transfer-Encoding header field MUST be sent
1440   in the message and MUST contain the list of corresponding
1441   transfer-coding names in the same order that they were applied.
1442   Transfer codings are defined in <xref target="transfer.codings"/>.
1443</t>
1444<figure><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/><artwork type="abnf2616"><![CDATA[
1445  Transfer-Encoding = 1#transfer-coding
1446]]></artwork></figure>
1447<t>
1448   Transfer-Encoding is analogous to the Content-Transfer-Encoding field of
1449   MIME, which was designed to enable safe transport of binary data over a
1450   7-bit transport service (<xref target="RFC2045"/>, Section 6).
1451   However, safe transport has a different focus for an 8bit-clean transfer
1452   protocol. In HTTP's case, Transfer-Encoding is primarily intended to
1453   accurately delimit a dynamically generated payload and to distinguish
1454   payload encodings that are only applied for transport efficiency or
1455   security from those that are characteristics of the target resource.
1456</t>
1457<t>
1458   The "chunked" transfer-coding (<xref target="chunked.encoding"/>)
1459   MUST be implemented by all HTTP/1.1 recipients because it plays a
1460   crucial role in delimiting messages when the payload body size is not
1461   known in advance.
1462   When the "chunked" transfer-coding is used, it MUST be the last
1463   transfer-coding applied to form the message body and MUST NOT
1464   be applied more than once in a message body.
1465   If any transfer-coding is applied to a request payload body,
1466   the final transfer-coding applied MUST be "chunked".
1467   If any transfer-coding is applied to a response payload body, then either
1468   the final transfer-coding applied MUST be "chunked" or
1469   the message MUST be terminated by closing the connection.
1470</t>
1471<figure><preamble>
1472   For example,
1473</preamble><artwork type="example"><![CDATA[
1474  Transfer-Encoding: gzip, chunked
1475]]></artwork><postamble>
1476   indicates that the payload body has been compressed using the gzip
1477   coding and then chunked using the chunked coding while forming the
1478   message body.
1479</postamble></figure>
1480<t>
1481   If more than one Transfer-Encoding header field is present in a message,
1482   the multiple field-values MUST be combined into one field-value,
1483   according to the algorithm defined in <xref target="header.fields"/>,
1484   before determining the message body length.
1485</t>
1486<t>
1487   Unlike Content-Encoding (Section 5.4 of <xref target="Part2"/>),
1488   Transfer-Encoding is a property of the message, not of the payload, and thus
1489   MAY be added or removed by any implementation along the request/response
1490   chain. Additional information about the encoding parameters MAY be
1491   provided by other header fields not defined by this specification.
1492</t>
1493<t>
1494   Transfer-Encoding MAY be sent in a response to a HEAD request or in a
1495   304 (Not Modified) response (Section 4.1 of <xref target="Part4"/>) to a GET request,
1496   neither of which includes a message body,
1497   to indicate that the origin server would have applied a transfer coding
1498   to the message body if the request had been an unconditional GET.
1499   This indication is not required, however, because any recipient on
1500   the response chain (including the origin server) can remove transfer
1501   codings when they are not needed.
1502</t>
1503<t>
1504   Transfer-Encoding was added in HTTP/1.1.  It is generally assumed that
1505   implementations advertising only HTTP/1.0 support will not understand
1506   how to process a transfer-encoded payload.
1507   A client MUST NOT send a request containing Transfer-Encoding unless it
1508   knows the server will handle HTTP/1.1 (or later) requests; such knowledge
1509   might be in the form of specific user configuration or by remembering the
1510   version of a prior received response.
1511   A server MUST NOT send a response containing Transfer-Encoding unless
1512   the corresponding request indicates HTTP/1.1 (or later).
1513</t>
1514<t>
1515   A server that receives a request message with a transfer-coding it does
1516   not understand SHOULD respond with 501 (Not Implemented) and then
1517   close the connection.
1518</t>
1519</section>
1520
1521<section title="Content-Length" anchor="header.content-length">
1522  <iref primary="true" item="Content-Length header field"/>
1523  <iref primary="true" item="Header Fields" subitem="Content-Length"/>
1524 
1525<t>
1526   When a message does not have a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field
1527   and the payload body length can be determined prior to being transferred, a
1528   Content-Length header field SHOULD be sent to indicate the length of the
1529   payload body that is either present as the message body, for requests
1530   and non-HEAD responses other than 304 (Not Modified), or
1531   would have been present had the request been an unconditional GET.  The
1532   length is expressed as a decimal number of octets.
1533</t>
1534<figure><iref primary="true" item="Grammar" subitem="Content-Length"/><artwork type="abnf2616"><![CDATA[
1535  Content-Length = 1*DIGIT
1536]]></artwork></figure>
1537<t>
1538   An example is
1539</t>
1540<figure><artwork type="example"><![CDATA[
1541  Content-Length: 3495
1542]]></artwork></figure>
1543<t>
1544   In the case of a response to a HEAD request, Content-Length indicates
1545   the size of the payload body (without any potential transfer-coding)
1546   that would have been sent had the request been a GET.
1547   In the case of a 304 (Not Modified) response (Section 4.1 of <xref target="Part4"/>)
1548   to a GET request, Content-Length indicates the size of the payload body (without
1549   any potential transfer-coding) that would have been sent in a 200 (OK)
1550   response.
1551</t>
1552<t>
1553   Any Content-Length field value greater than or equal to zero is valid.
1554   Since there is no predefined limit to the length of an HTTP payload,
1555   recipients SHOULD anticipate potentially large decimal numerals and
1556   prevent parsing errors due to integer conversion overflows
1557   (<xref target="attack.protocol.element.size.overflows"/>).
1558</t>
1559<t>
1560   If a message is received that has multiple Content-Length header fields
1561   with field-values consisting of the same decimal value, or a single
1562   Content-Length header field with a field value containing a list of
1563   identical decimal values (e.g., "Content-Length: 42, 42"), indicating that
1564   duplicate Content-Length header fields have been generated or combined by an
1565   upstream message processor, then the recipient MUST either reject the
1566   message as invalid or replace the duplicated field-values with a single
1567   valid Content-Length field containing that decimal value prior to
1568   determining the message body length.
1569</t>
1570<t><list>
1571  <t>
1572   Note: HTTP's use of Content-Length for message framing differs
1573   significantly from the same field's use in MIME, where it is an optional
1574   field used only within the "message/external-body" media-type.
1575  </t>
1576</list></t>
1577</section>
1578
1579<section title="Message Body Length" anchor="message.body.length">
1580<t>
1581   The length of a message body is determined by one of the following
1582   (in order of precedence):
1583</t>
1584<t>
1585  <list style="numbers">
1586    <t>
1587     Any response to a HEAD request and any response with a
1588     1xx (Informational), 204 (No Content), or
1589     304 (Not Modified) status code is always
1590     terminated by the first empty line after the header fields, regardless of
1591     the header fields present in the message, and thus cannot contain a
1592     message body.
1593    </t>
1594    <t>
1595     Any 2xx (Successful) response to a CONNECT request implies that the
1596     connection will become a tunnel immediately after the empty line that
1597     concludes the header fields.  A client MUST ignore any
1598     <xref target="header.content-length" format="none">Content-Length</xref> or <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header
1599     fields received in such a message.
1600    </t>
1601    <t>
1602     If a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field is present
1603     and the "chunked" transfer-coding (<xref target="chunked.encoding"/>)
1604     is the final encoding, the message body length is determined by reading
1605     and decoding the chunked data until the transfer-coding indicates the
1606     data is complete.
1607    <vspace blankLines="1"/>
1608     If a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field is present in a
1609     response and the "chunked" transfer-coding is not the final encoding, the
1610     message body length is determined by reading the connection until it is
1611     closed by the server.
1612     If a Transfer-Encoding header field is present in a request and the
1613     "chunked" transfer-coding is not the final encoding, the message body
1614     length cannot be determined reliably; the server MUST respond with
1615     the 400 (Bad Request) status code and then close the connection.
1616    <vspace blankLines="1"/>
1617     If a message is received with both a <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>
1618     and a <xref target="header.content-length" format="none">Content-Length</xref> header field, the
1619     Transfer-Encoding overrides the Content-Length.
1620     Such a message might indicate an attempt to perform request or response
1621     smuggling (bypass of security-related checks on message routing or content)
1622     and thus ought to be handled as an error.  The provided Content-Length MUST
1623     be removed, prior to forwarding the message downstream, or replaced with
1624     the real message body length after the transfer-coding is decoded.
1625    </t>
1626    <t>
1627     If a message is received without <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> and with
1628     either multiple <xref target="header.content-length" format="none">Content-Length</xref> header fields having
1629     differing field-values or a single Content-Length header field having an
1630     invalid value, then the message framing is invalid and MUST be treated
1631     as an error to prevent request or response smuggling.
1632     If this is a request message, the server MUST respond with
1633     a 400 (Bad Request) status code and then close the connection.
1634     If this is a response message received by a proxy, the proxy
1635     MUST discard the received response, send a 502 (Bad Gateway)
1636     status code as its downstream response, and then close the connection.
1637     If this is a response message received by a user-agent, it MUST be
1638     treated as an error by discarding the message and closing the connection.
1639    </t>
1640    <t>
1641     If a valid <xref target="header.content-length" format="none">Content-Length</xref> header field is present without
1642     <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>, its decimal value defines the
1643     message body length in octets.  If the actual number of octets sent in
1644     the message is less than the indicated Content-Length, the recipient
1645     MUST consider the message to be incomplete and treat the connection
1646     as no longer usable.
1647     If the actual number of octets sent in the message is more than the indicated
1648     Content-Length, the recipient MUST only process the message body up to the
1649     field value's number of octets; the remainder of the message MUST either
1650     be discarded or treated as the next message in a pipeline.  For the sake of
1651     robustness, a user-agent MAY attempt to detect and correct such an error
1652     in message framing if it is parsing the response to the last request on
1653     a connection and the connection has been closed by the server.
1654    </t>
1655    <t>
1656     If this is a request message and none of the above are true, then the
1657     message body length is zero (no message body is present).
1658    </t>
1659    <t>
1660     Otherwise, this is a response message without a declared message body
1661     length, so the message body length is determined by the number of octets
1662     received prior to the server closing the connection.
1663    </t>
1664  </list>
1665</t>
1666<t>
1667   Since there is no way to distinguish a successfully completed,
1668   close-delimited message from a partially-received message interrupted
1669   by network failure, implementations SHOULD use encoding or
1670   length-delimited messages whenever possible.  The close-delimiting
1671   feature exists primarily for backwards compatibility with HTTP/1.0.
1672</t>
1673<t>
1674   A server MAY reject a request that contains a message body but
1675   not a <xref target="header.content-length" format="none">Content-Length</xref> by responding with
1676   411 (Length Required).
1677</t>
1678<t>
1679   Unless a transfer-coding other than "chunked" has been applied,
1680   a client that sends a request containing a message body SHOULD
1681   use a valid <xref target="header.content-length" format="none">Content-Length</xref> header field if the message body
1682   length is known in advance, rather than the "chunked" encoding, since some
1683   existing services respond to "chunked" with a 411 (Length Required)
1684   status code even though they understand the chunked encoding.  This
1685   is typically because such services are implemented via a gateway that
1686   requires a content-length in advance of being called and the server
1687   is unable or unwilling to buffer the entire request before processing.
1688</t>
1689<t>
1690   A client that sends a request containing a message body MUST include a
1691   valid <xref target="header.content-length" format="none">Content-Length</xref> header field if it does not know the
1692   server will handle HTTP/1.1 (or later) requests; such knowledge can be in
1693   the form of specific user configuration or by remembering the version of a
1694   prior received response.
1695</t>
1696</section>
1697</section>
1698
1699<section anchor="incomplete.messages" title="Handling Incomplete Messages">
1700<t>
1701   Request messages that are prematurely terminated, possibly due to a
1702   canceled connection or a server-imposed time-out exception, MUST
1703   result in closure of the connection; sending an HTTP/1.1 error response
1704   prior to closing the connection is OPTIONAL.
1705</t>
1706<t>
1707   Response messages that are prematurely terminated, usually by closure
1708   of the connection prior to receiving the expected number of octets or by
1709   failure to decode a transfer-encoded message body, MUST be recorded
1710   as incomplete.  A response that terminates in the middle of the header
1711   block (before the empty line is received) cannot be assumed to convey the
1712   full semantics of the response and MUST be treated as an error.
1713</t>
1714<t>
1715   A message body that uses the chunked transfer encoding is
1716   incomplete if the zero-sized chunk that terminates the encoding has not
1717   been received.  A message that uses a valid <xref target="header.content-length" format="none">Content-Length</xref> is
1718   incomplete if the size of the message body received (in octets) is less than
1719   the value given by Content-Length.  A response that has neither chunked
1720   transfer encoding nor Content-Length is terminated by closure of the
1721   connection, and thus is considered complete regardless of the number of
1722   message body octets received, provided that the header block was received
1723   intact.
1724</t>
1725<t>
1726   A user agent MUST NOT render an incomplete response message body as if
1727   it were complete (i.e., some indication needs to be given to the user that an
1728   error occurred).  Cache requirements for incomplete responses are defined
1729   in Section 3 of <xref target="Part6"/>.
1730</t>
1731<t>
1732   A server MUST read the entire request message body or close
1733   the connection after sending its response, since otherwise the
1734   remaining data on a persistent connection would be misinterpreted
1735   as the next request.  Likewise,
1736   a client MUST read the entire response message body if it intends
1737   to reuse the same connection for a subsequent request.  Pipelining
1738   multiple requests on a connection is described in <xref target="pipelining"/>.
1739</t>
1740</section>
1741
1742<section title="Message Parsing Robustness" anchor="message.robustness">
1743<t>
1744   Older HTTP/1.0 client implementations might send an extra CRLF
1745   after a POST request as a lame workaround for some early server
1746   applications that failed to read message body content that was
1747   not terminated by a line-ending. An HTTP/1.1 client MUST NOT
1748   preface or follow a request with an extra CRLF.  If terminating
1749   the request message body with a line-ending is desired, then the
1750   client MUST include the terminating CRLF octets as part of the
1751   message body length.
1752</t>
1753<t>
1754   In the interest of robustness, servers SHOULD ignore at least one
1755   empty line received where a request-line is expected. In other words, if
1756   the server is reading the protocol stream at the beginning of a
1757   message and receives a CRLF first, it SHOULD ignore the CRLF.
1758   Likewise, although the line terminator for the start-line and header
1759   fields is the sequence CRLF, we recommend that recipients recognize a
1760   single LF as a line terminator and ignore any CR.
1761</t>
1762<t>
1763   When a server listening only for HTTP request messages, or processing
1764   what appears from the start-line to be an HTTP request message,
1765   receives a sequence of octets that does not match the HTTP-message
1766   grammar aside from the robustness exceptions listed above, the
1767   server MUST respond with an HTTP/1.1 400 (Bad Request) response. 
1768</t>
1769</section>
1770</section>
1771
1772<section title="Transfer Codings" anchor="transfer.codings">
1773 
1774 
1775<t>
1776   Transfer-coding values are used to indicate an encoding
1777   transformation that has been, can be, or might need to be applied to a
1778   payload body in order to ensure "safe transport" through the network.
1779   This differs from a content coding in that the transfer-coding is a
1780   property of the message rather than a property of the representation
1781   that is being transferred.
1782</t>
1783<figure><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/><artwork type="abnf2616"><![CDATA[
1784  transfer-coding    = "chunked" ; Section 4.1
1785                     / "compress" ; Section 4.2.1
1786                     / "deflate" ; Section 4.2.2
1787                     / "gzip" ; Section 4.2.3
1788                     / transfer-extension
1789  transfer-extension = token *( OWS ";" OWS transfer-parameter )
1790]]></artwork></figure>
1791<t anchor="rule.parameter">
1792 
1793 
1794 
1795   Parameters are in the form of attribute/value pairs.
1796</t>
1797<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[
1798  transfer-parameter = attribute BWS "=" BWS value
1799  attribute          = token
1800  value              = word
1801]]></artwork></figure>
1802<t>
1803   All transfer-coding values are case-insensitive.
1804   The HTTP Transfer Coding registry is defined in
1805   <xref target="transfer.coding.registry"/>.
1806   HTTP/1.1 uses transfer-coding values in the <xref target="header.te" format="none">TE</xref> header field
1807   (<xref target="header.te"/>) and in the <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref>
1808   header field (<xref target="header.transfer-encoding"/>).
1809</t>
1810
1811<section title="Chunked Transfer Coding" anchor="chunked.encoding">
1812  <iref item="chunked (Coding Format)"/>
1813  <iref item="Coding Format" subitem="chunked"/>
1814 
1815 
1816 
1817 
1818 
1819 
1820 
1821 
1822 
1823 
1824 
1825<t>
1826   The chunked encoding modifies the body of a message in order to
1827   transfer it as a series of chunks, each with its own size indicator,
1828   followed by an OPTIONAL trailer containing header fields. This
1829   allows dynamically produced content to be transferred along with the
1830   information necessary for the recipient to verify that it has
1831   received the full message.
1832</t>
1833<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[
1834  chunked-body   = *chunk
1835                   last-chunk
1836                   trailer-part
1837                   CRLF
1838 
1839  chunk          = chunk-size [ chunk-ext ] CRLF
1840                   chunk-data CRLF
1841  chunk-size     = 1*HEXDIG
1842  last-chunk     = 1*("0") [ chunk-ext ] CRLF
1843 
1844  chunk-ext      = *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
1845  chunk-ext-name = token
1846  chunk-ext-val  = token / quoted-str-nf
1847  chunk-data     = 1*OCTET ; a sequence of chunk-size octets
1848  trailer-part   = *( header-field CRLF )
1849 
1850  quoted-str-nf  = DQUOTE *( qdtext-nf / quoted-pair ) DQUOTE
1851                 ; like quoted-string, but disallowing line folding
1852  qdtext-nf      = HTAB / SP / %x21 / %x23-5B / %x5D-7E / obs-text
1853]]></artwork></figure>
1854<t>
1855   The chunk-size field is a string of hex digits indicating the size of
1856   the chunk-data in octets. The chunked encoding is ended by any chunk whose size is
1857   zero, followed by the trailer, which is terminated by an empty line.
1858</t>
1859<t>
1860   The trailer allows the sender to include additional HTTP header
1861   fields at the end of the message. The <xref target="header.trailer" format="none">Trailer</xref> header field
1862   can be used to indicate which header fields are included in a trailer (see
1863   <xref target="header.trailer"/>).
1864</t>
1865<t>
1866   A server using chunked transfer-coding in a response MUST NOT use the
1867   trailer for any header fields unless at least one of the following is
1868   true:
1869  <list style="numbers">
1870    <t>the request included a <xref target="header.te" format="none">TE</xref> header field that indicates
1871    "trailers" is acceptable in the transfer-coding of the response, as
1872    described in <xref target="header.te"/>; or,</t>
1873     
1874    <t>the trailer fields consist entirely of optional metadata, and the
1875    recipient could use the message (in a manner acceptable to the server where
1876    the field originated) without receiving it. In other words, the server that
1877    generated the header field (often but not always the origin server) is
1878    willing to accept the possibility that the trailer fields might be silently
1879    discarded along the path to the client.</t>
1880  </list>
1881</t>
1882<t>
1883   This requirement prevents an interoperability failure when the
1884   message is being received by an HTTP/1.1 (or later) proxy and
1885   forwarded to an HTTP/1.0 recipient. It avoids a situation where
1886   conformance with the protocol would have necessitated a possibly
1887   infinite buffer on the proxy.
1888</t>
1889<t>
1890   A process for decoding the "chunked" transfer-coding
1891   can be represented in pseudo-code as:
1892</t>
1893<figure><artwork type="code"><![CDATA[
1894  length := 0
1895  read chunk-size, chunk-ext (if any) and CRLF
1896  while (chunk-size > 0) {
1897     read chunk-data and CRLF
1898     append chunk-data to decoded-body
1899     length := length + chunk-size
1900     read chunk-size and CRLF
1901  }
1902  read header-field
1903  while (header-field not empty) {
1904     append header-field to existing header fields
1905     read header-field
1906  }
1907  Content-Length := length
1908  Remove "chunked" from Transfer-Encoding
1909]]></artwork></figure>
1910<t>
1911   All HTTP/1.1 applications MUST be able to receive and decode the
1912   "chunked" transfer-coding and MUST ignore chunk-ext extensions
1913   they do not understand.
1914</t>
1915<t>
1916   Use of chunk-ext extensions by senders is deprecated; they SHOULD NOT be
1917   sent and definition of new chunk-extensions is discouraged.
1918</t>
1919</section>
1920
1921<section title="Compression Codings" anchor="compression.codings">
1922<t>
1923   The codings defined below can be used to compress the payload of a
1924   message.
1925</t>
1926<t><list><t>
1927   Note: Use of program names for the identification of encoding formats
1928   is not desirable and is discouraged for future encodings. Their
1929   use here is representative of historical practice, not good
1930   design.
1931</t></list></t>
1932<t><list><t>
1933   Note: For compatibility with previous implementations of HTTP,
1934   applications SHOULD consider "x-gzip" and "x-compress" to be
1935   equivalent to "gzip" and "compress" respectively.
1936</t></list></t>
1937
1938<section title="Compress Coding" anchor="compress.coding">
1939<iref item="compress (Coding Format)"/>
1940<iref item="Coding Format" subitem="compress"/>
1941<t>
1942   The "compress" format is produced by the common UNIX file compression
1943   program "compress". This format is an adaptive Lempel-Ziv-Welch
1944   coding (LZW).
1945</t>
1946</section>
1947
1948<section title="Deflate Coding" anchor="deflate.coding">
1949<iref item="deflate (Coding Format)"/>
1950<iref item="Coding Format" subitem="deflate"/>
1951<t>
1952   The "deflate" format is defined as the "deflate" compression mechanism
1953   (described in <xref target="RFC1951"/>) used inside the "zlib"
1954   data format (<xref target="RFC1950"/>).
1955</t>
1956<t><list>
1957  <t>
1958    Note: Some incorrect implementations send the "deflate"
1959    compressed data without the zlib wrapper.
1960   </t>
1961</list></t>
1962</section>
1963
1964<section title="Gzip Coding" anchor="gzip.coding">
1965<iref item="gzip (Coding Format)"/>
1966<iref item="Coding Format" subitem="gzip"/>
1967<t>
1968   The "gzip" format is produced by the file compression program
1969   "gzip" (GNU zip), as described in <xref target="RFC1952"/>. This format is a
1970   Lempel-Ziv coding (LZ77) with a 32 bit CRC.
1971</t>
1972</section>
1973
1974</section>
1975
1976<section title="TE" anchor="header.te">
1977  <iref primary="true" item="TE header field"/>
1978  <iref primary="true" item="Header Fields" subitem="TE"/>
1979 
1980 
1981 
1982 
1983<t>
1984   The "TE" header field indicates what extension transfer-codings
1985   the client is willing to accept in the response, and whether or not it is
1986   willing to accept trailer fields in a chunked transfer-coding.
1987</t>
1988<t>
1989   Its value consists of the keyword "trailers" and/or a comma-separated
1990   list of extension transfer-coding names with optional accept
1991   parameters (as described in <xref target="transfer.codings"/>).
1992</t>
1993<figure><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="t-codings"/><iref primary="true" item="Grammar" subitem="te-params"/><iref primary="true" item="Grammar" subitem="te-ext"/><artwork type="abnf2616"><![CDATA[
1994  TE        = #t-codings
1995  t-codings = "trailers" / ( transfer-extension [ te-params ] )
1996  te-params = OWS ";" OWS "q=" qvalue *( te-ext )
1997  te-ext    = OWS ";" OWS token [ "=" word ]
1998]]></artwork></figure>
1999<t>
2000   The presence of the keyword "trailers" indicates that the client is
2001   willing to accept trailer fields in a chunked transfer-coding, as
2002   defined in <xref target="chunked.encoding"/>. This keyword is reserved for use with
2003   transfer-coding values even though it does not itself represent a
2004   transfer-coding.
2005</t>
2006<t>
2007   Examples of its use are:
2008</t>
2009<figure><artwork type="example"><![CDATA[
2010  TE: deflate
2011  TE:
2012  TE: trailers, deflate;q=0.5
2013]]></artwork></figure>
2014<t>
2015   The TE header field only applies to the immediate connection.
2016   Therefore, the keyword MUST be supplied within a <xref target="header.connection" format="none">Connection</xref>
2017   header field (<xref target="header.connection"/>) whenever TE is present in
2018   an HTTP/1.1 message.
2019</t>
2020<t>
2021   A server tests whether a transfer-coding is acceptable, according to
2022   a TE field, using these rules:
2023  <list style="numbers">
2024    <t>The "chunked" transfer-coding is always acceptable. If the
2025         keyword "trailers" is listed, the client indicates that it is
2026         willing to accept trailer fields in the chunked response on
2027         behalf of itself and any downstream clients. The implication is
2028         that, if given, the client is stating that either all
2029         downstream clients are willing to accept trailer fields in the
2030         forwarded response, or that it will attempt to buffer the
2031         response on behalf of downstream recipients.
2032      <vspace blankLines="1"/>
2033         Note: HTTP/1.1 does not define any means to limit the size of a
2034         chunked response such that a client can be assured of buffering
2035         the entire response.</t>
2036    <t>If the transfer-coding being tested is one of the transfer-codings
2037         listed in the TE field, then it is acceptable unless it
2038         is accompanied by a qvalue of 0. (As defined in <xref target="quality.values"/>, a
2039         qvalue of 0 means "not acceptable".)</t>
2040    <t>If multiple transfer-codings are acceptable, then the
2041         acceptable transfer-coding with the highest non-zero qvalue is
2042         preferred.  The "chunked" transfer-coding always has a qvalue
2043         of 1.</t>
2044  </list>
2045</t>
2046<t>
2047   If the TE field-value is empty or if no TE field is present, the only
2048   acceptable transfer-coding is "chunked". A message with no transfer-coding is
2049   always acceptable.
2050</t>
2051
2052<section title="Quality Values" anchor="quality.values">
2053 
2054<t>
2055   Both transfer codings (<xref target="header.te" format="none">TE</xref> request header field,
2056   <xref target="header.te"/>) and content negotiation (Section 8 of <xref target="Part2"/>)
2057   use short "floating point" numbers to indicate the relative importance
2058   ("weight") of various negotiable parameters.  A weight is normalized to a
2059   real number in the range 0 through 1, where 0 is the minimum and 1 the
2060   maximum value. If a parameter has a quality value of 0, then content with
2061   this parameter is "not acceptable" for the client. HTTP/1.1
2062   applications MUST NOT generate more than three digits after the
2063   decimal point. User configuration of these values SHOULD also be
2064   limited in this fashion.
2065</t>
2066<figure><iref primary="true" item="Grammar" subitem="qvalue"/><artwork type="abnf2616"><![CDATA[
2067  qvalue         = ( "0" [ "." 0*3DIGIT ] )
2068                 / ( "1" [ "." 0*3("0") ] )
2069]]></artwork></figure>
2070<t><list>
2071  <t>
2072     Note: "Quality values" is a misnomer, since these values merely represent
2073     relative degradation in desired quality.
2074  </t>
2075</list></t>
2076</section>
2077</section>
2078
2079<section title="Trailer" anchor="header.trailer">
2080  <iref primary="true" item="Trailer header field"/>
2081  <iref primary="true" item="Header Fields" subitem="Trailer"/>
2082 
2083<t>
2084   The "Trailer" header field indicates that the given set of
2085   header fields is present in the trailer of a message encoded with
2086   chunked transfer-coding.
2087</t>
2088<figure><iref primary="true" item="Grammar" subitem="Trailer"/><artwork type="abnf2616"><![CDATA[
2089  Trailer = 1#field-name
2090]]></artwork></figure>
2091<t>
2092   An HTTP/1.1 message SHOULD include a Trailer header field in a
2093   message using chunked transfer-coding with a non-empty trailer. Doing
2094   so allows the recipient to know which header fields to expect in the
2095   trailer.
2096</t>
2097<t>
2098   If no Trailer header field is present, the trailer SHOULD NOT  include
2099   any header fields. See <xref target="chunked.encoding"/> for restrictions on the use of
2100   trailer fields in a "chunked" transfer-coding.
2101</t>
2102<t>
2103   Message header fields listed in the Trailer header field MUST NOT
2104   include the following header fields:
2105  <list style="symbols">
2106    <t><xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref></t>
2107    <t><xref target="header.content-length" format="none">Content-Length</xref></t>
2108    <t><xref target="header.trailer" format="none">Trailer</xref></t>
2109  </list>
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<t>
2126   HTTP is used in a wide variety of applications, ranging from
2127   general-purpose computers to home appliances.  In some cases,
2128   communication options are hard-coded in a client's configuration.
2129   However, most HTTP clients rely on the same resource identification
2130   mechanism and configuration techniques as general-purpose Web browsers.
2131</t>
2132<t>
2133   HTTP communication is initiated by a user agent for some purpose.
2134   The purpose is a combination of request semantics, which are defined in
2135   <xref target="Part2"/>, and a target resource upon which to apply those
2136   semantics.  A URI reference (<xref target="uri"/>) is typically used as
2137   an identifier for the "target resource", which a user agent would resolve
2138   to its absolute form in order to obtain the "target URI".  The target URI
2139   excludes the reference's fragment identifier component, if any,
2140   since fragment identifiers are reserved for client-side processing
2141   (<xref target="RFC3986"/>, Section 3.5).
2142</t>
2143<t>
2144   HTTP intermediaries obtain the request semantics and target URI
2145   from the request-line of an incoming request message.
2146</t>
2147</section>
2148
2149<section title="Connecting Inbound" anchor="connecting.inbound">
2150<t>
2151   Once the target URI is determined, a client needs to decide whether
2152   a network request is necessary to accomplish the desired semantics and,
2153   if so, where that request is to be directed.
2154</t>
2155<t>
2156   If the client has a response cache and the request semantics can be
2157   satisfied by a cache (<xref target="Part6"/>), then the request is
2158   usually directed to the cache first.
2159</t>
2160<t>
2161   If the request is not satisfied by a cache, then a typical client will
2162   check its configuration to determine whether a proxy is to be used to
2163   satisfy the request.  Proxy configuration is implementation-dependent,
2164   but is often based on URI prefix matching, selective authority matching,
2165   or both, and the proxy itself is usually identified by an "http" or
2166   "https" URI.  If a proxy is applicable, the client connects inbound by
2167   establishing (or reusing) a connection to that proxy.
2168</t>
2169<t>
2170   If no proxy is applicable, a typical client will invoke a handler routine,
2171   usually specific to the target URI's scheme, to connect directly
2172   to an authority for the target resource.  How that is accomplished is
2173   dependent on the target URI scheme and defined by its associated
2174   specification, similar to how this specification defines origin server
2175   access for resolution of the "http" (<xref target="http.uri"/>) and
2176   "https" (<xref target="https.uri"/>) schemes.
2177</t>
2178</section>
2179
2180<section title="Request Target" anchor="request-target">
2181<t>
2182   Once an inbound connection is obtained
2183   (<xref target="connection.management"/>),
2184   the client sends an HTTP request message (<xref target="http.message"/>)
2185   with a request-target derived from the target URI.
2186   There are four distinct formats for the request-target, depending on both
2187   the method being requested and whether the request is to a proxy.
2188</t>   
2189<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[
2190  request-target = origin-form
2191                 / absolute-form
2192                 / authority-form
2193                 / asterisk-form
2194
2195  origin-form    = path-absolute [ "?" query ]
2196  absolute-form  = absolute-URI
2197  authority-form = authority
2198  asterisk-form  = "*"
2199]]></artwork></figure>
2200<t anchor="origin-form"><iref item="origin-form (of request-target)"/>
2201   The most common form of request-target is the origin-form.
2202   When making a request directly to an origin server, other than a CONNECT
2203   or server-wide OPTIONS request (as detailed below),
2204   a client MUST send only the absolute path and query components of
2205   the target URI as the request-target.
2206   If the target URI's path component is empty, then the client MUST send
2207   "/" as the path within the origin-form of request-target.
2208   A <xref target="header.host" format="none">Host</xref> header field is also sent, as defined in
2209   <xref target="header.host"/>, containing the target URI's
2210   authority component (excluding any userinfo).
2211</t>
2212<t>
2213   For example, a client wishing to retrieve a representation of the resource
2214   identified as
2215</t>
2216<figure><artwork type="example"><![CDATA[
2217  http://www.example.org/where?q=now
2218  ]]></artwork></figure>
2219<t>
2220   directly from the origin server would open (or reuse) a TCP connection
2221   to port 80 of the host "www.example.org" and send the lines:
2222</t>
2223<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2224  GET /where?q=now HTTP/1.1
2225  Host: www.example.org
2226  ]]></artwork></figure>
2227<t>
2228   followed by the remainder of the request message.
2229</t>
2230<t anchor="absolute-form"><iref item="absolute-form (of request-target)"/>
2231   When making a request to a proxy, other than a CONNECT or server-wide
2232   OPTIONS request (as detailed below), a client MUST send the target URI
2233   in absolute-form as the request-target.
2234   The proxy is requested to either service that request from a valid cache,
2235   if possible, or make the same request on the client's behalf to either
2236   the next inbound proxy server or directly to the origin server indicated
2237   by the request-target.  Requirements on such "forwarding" of messages are
2238   defined in <xref target="intermediary.forwarding"/>.
2239</t>
2240<t>
2241   An example absolute-form of request-line would be:
2242</t>
2243<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2244  GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1
2245  ]]></artwork></figure>
2246<t>
2247   To allow for transition to the absolute-form for all requests in some
2248   future version of HTTP, HTTP/1.1 servers MUST accept the absolute-form
2249   in requests, even though HTTP/1.1 clients will only send them in requests
2250   to proxies.
2251</t>
2252<t anchor="authority-form"><iref item="authority-form (of request-target)"/>
2253   The authority-form of request-target is only used for CONNECT requests
2254   (Section 2.3.8 of <xref target="Part2"/>).  When making a CONNECT request to establish a tunnel through
2255   one or more proxies, a client MUST send only the target URI's
2256   authority component (excluding any userinfo) as the request-target.
2257   For example,
2258</t>
2259<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2260  CONNECT www.example.com:80 HTTP/1.1
2261  ]]></artwork></figure>
2262<t anchor="asterisk-form"><iref item="asterisk-form (of request-target)"/>
2263   The asterisk-form of request-target is only used for a server-wide
2264   OPTIONS request (Section 2.3.1 of <xref target="Part2"/>).  When a client wishes to request OPTIONS
2265   for the server as a whole, as opposed to a specific named resource of
2266   that server, the client MUST send only "*" (%x2A) as the request-target.
2267   For example,
2268</t>
2269<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2270  OPTIONS * HTTP/1.1
2271  ]]></artwork></figure>
2272<t>
2273   If a proxy receives an OPTIONS request with an absolute-form of
2274   request-target in which the URI has an empty path and no query component,
2275   then the last proxy on the request chain MUST send a request-target
2276   of "*" when it forwards the request to the indicated origin server.
2277</t>
2278<figure><preamble>  
2279   For example, the request
2280</preamble><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2281  OPTIONS http://www.example.org:8001 HTTP/1.1
2282  ]]></artwork></figure>
2283<figure><preamble>  
2284  would be forwarded by the final proxy as
2285</preamble><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2286  OPTIONS * HTTP/1.1
2287  Host: www.example.org:8001
2288  ]]></artwork>
2289<postamble>
2290   after connecting to port 8001 of host "www.example.org".
2291</postamble>
2292</figure>
2293</section>
2294
2295<section title="Host" anchor="header.host">
2296  <iref primary="true" item="Host header field"/>
2297  <iref primary="true" item="Header Fields" subitem="Host"/>
2298 
2299<t>
2300   The "Host" header field in a request provides the host and port
2301   information from the target URI, enabling the origin
2302   server to distinguish among resources while servicing requests
2303   for multiple host names on a single IP address.  Since the Host
2304   field-value is critical information for handling a request, it
2305   SHOULD be sent as the first header field following the request-line.
2306</t>
2307<figure><iref primary="true" item="Grammar" subitem="Host"/><artwork type="abnf2616"><![CDATA[
2308  Host = uri-host [ ":" port ] ; Section 2.8.1
2309]]></artwork></figure>
2310<t>
2311   A client MUST send a Host header field in all HTTP/1.1 request
2312   messages.  If the target URI includes an authority component, then
2313   the Host field-value MUST be identical to that authority component
2314   after excluding any userinfo (<xref target="http.uri"/>).
2315   If the authority component is missing or undefined for the target URI,
2316   then the Host header field MUST be sent with an empty field-value.
2317</t>
2318<t>
2319   For example, a GET request to the origin server for
2320   &lt;http://www.example.org/pub/WWW/&gt; would begin with:
2321</t>
2322<figure><artwork type="message/http; msgtype=&#34;request&#34;"><![CDATA[
2323  GET /pub/WWW/ HTTP/1.1
2324  Host: www.example.org
2325  ]]></artwork></figure>
2326<t>
2327   The Host header field MUST be sent in an HTTP/1.1 request even
2328   if the request-target is in the absolute-form, since this
2329   allows the Host information to be forwarded through ancient HTTP/1.0
2330   proxies that might not have implemented Host.
2331</t>
2332<t>
2333   When an HTTP/1.1 proxy receives a request with an absolute-form of
2334   request-target, the proxy MUST ignore the received
2335   Host header field (if any) and instead replace it with the host
2336   information of the request-target.  If the proxy forwards the request,
2337   it MUST generate a new Host field-value based on the received
2338   request-target rather than forward the received Host field-value.
2339</t>
2340<t>
2341   Since the Host header field acts as an application-level routing
2342   mechanism, it is a frequent target for malware seeking to poison
2343   a shared cache or redirect a request to an unintended server.
2344   An interception proxy is particularly vulnerable if it relies on
2345   the Host field-value for redirecting requests to internal
2346   servers, or for use as a cache key in a shared cache, without
2347   first verifying that the intercepted connection is targeting a
2348   valid IP address for that host.
2349</t>
2350<t>
2351   A server MUST respond with a 400 (Bad Request) status code
2352   to any HTTP/1.1 request message that lacks a Host header field and
2353   to any request message that contains more than one Host header field
2354   or a Host header field with an invalid field-value.
2355</t>
2356</section>
2357
2358<section title="Effective Request URI" anchor="effective.request.uri">
2359  <iref primary="true" item="effective request URI"/>
2360<t>
2361   A server that receives an HTTP request message MUST reconstruct
2362   the user agent's original target URI, based on the pieces of information
2363   learned from the request-target, <xref target="header.host" format="none">Host</xref> header field, and
2364   connection context, in order to identify the intended target resource and
2365   properly service the request. The URI derived from this reconstruction
2366   process is referred to as the "effective request URI".
2367</t>
2368<t>
2369   For a user agent, the effective request URI is the target URI.
2370</t>
2371<t>
2372   If the request-target is in absolute-form, then the effective request URI
2373   is the same as the request-target.  Otherwise, the effective request URI
2374   is constructed as follows.
2375</t>
2376<t>
2377   If the request is received over an SSL/TLS-secured TCP connection,
2378   then the effective request URI's scheme is "https"; otherwise, the
2379   scheme is "http".
2380</t>
2381<t>
2382   If the request-target is in authority-form, then the effective
2383   request URI's authority component is the same as the request-target.
2384   Otherwise, if a <xref target="header.host" format="none">Host</xref> header field is supplied with a
2385   non-empty field-value, then the authority component is the same as the
2386   Host field-value. Otherwise, the authority component is the concatenation of
2387   the default host name configured for the server, a colon (":"), and the
2388   connection's incoming TCP port number in decimal form.
2389</t>
2390<t>
2391   If the request-target is in authority-form or asterisk-form, then the
2392   effective request URI's combined path and query component is empty.
2393   Otherwise, the combined path and query component is the same as the
2394   request-target.
2395</t>
2396<t>
2397   The components of the effective request URI, once determined as above,
2398   can be combined into absolute-URI form by concatenating the scheme,
2399   "://", authority, and combined path and query component.
2400</t>
2401<figure>
2402<preamble>
2403   Example 1: the following message received over an insecure TCP connection
2404</preamble>
2405<artwork type="example"><![CDATA[
2406  GET /pub/WWW/TheProject.html HTTP/1.1
2407  Host: www.example.org:8080
2408  ]]></artwork>
2409</figure>
2410<figure>
2411<preamble>
2412  has an effective request URI of
2413</preamble>
2414<artwork type="example"><![CDATA[
2415  http://www.example.org:8080/pub/WWW/TheProject.html
2416  ]]></artwork>
2417</figure>
2418<figure>
2419<preamble>
2420   Example 2: the following message received over an SSL/TLS-secured TCP
2421   connection
2422</preamble>
2423<artwork type="example"><![CDATA[
2424  OPTIONS * HTTP/1.1
2425  Host: www.example.org
2426  ]]></artwork>
2427</figure>
2428<figure>
2429<preamble>
2430  has an effective request URI of
2431</preamble>
2432<artwork type="example"><![CDATA[
2433  https://www.example.org
2434  ]]></artwork>
2435</figure>
2436<t>
2437   An origin server that does not allow resources to differ by requested
2438   host MAY ignore the <xref target="header.host" format="none">Host</xref> field-value and instead replace it
2439   with a configured server name when constructing the effective request URI.
2440</t>
2441<t>
2442   Recipients of an HTTP/1.0 request that lacks a <xref target="header.host" format="none">Host</xref> header
2443   field MAY attempt to use heuristics (e.g., examination of the URI path for
2444   something unique to a particular host) in order to guess the
2445   effective request URI's authority component.
2446</t>
2447</section>
2448
2449<section title="Intermediary Forwarding" anchor="intermediary.forwarding">
2450<t>
2451   As described in <xref target="intermediaries"/>, intermediaries can serve
2452   a variety of roles in the processing of HTTP requests and responses.
2453   Some intermediaries are used to improve performance or availability.
2454   Others are used for access control or to filter content.
2455   Since an HTTP stream has characteristics similar to a pipe-and-filter
2456   architecture, there are no inherent limits to the extent an intermediary
2457   can enhance (or interfere) with either direction of the stream.
2458</t>
2459<t>
2460   In order to avoid request loops, a proxy that forwards requests to other
2461   proxies MUST be able to recognize and exclude all of its own server
2462   names, including any aliases, local variations, or literal IP addresses.
2463</t>
2464<t>
2465   If a proxy receives a request-target with a host name that is not a
2466   fully qualified domain name, it MAY add its domain to the host name
2467   it received when forwarding the request.  A proxy MUST NOT change the
2468   host name if it is a fully qualified domain name.
2469</t>
2470<t>
2471   A non-transforming proxy MUST NOT rewrite the "path-absolute" and "query"
2472   parts of the received request-target when forwarding it to the next inbound
2473   server, except as noted above to replace an empty path with "/" or "*".
2474</t>
2475<t>
2476   Intermediaries that forward a message MUST implement the
2477   <xref target="header.connection" format="none">Connection</xref> header field as specified in
2478   <xref target="header.connection"/>.
2479</t>
2480
2481<section title="End-to-end and Hop-by-hop Header Fields" anchor="end-to-end.and.hop-by-hop.header-fields">
2482<!--<t>
2483  <cref anchor="TODO-end-to-end" source="jre">
2484    Restored from <eref target="http://tools.ietf.org/html/draft-ietf-httpbis-p6-cache-05#section-7.1"/>.
2485    See also <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/60"/>.
2486  </cref>
2487</t>-->
2488<t>
2489   For the purpose of defining the behavior of caches and non-caching
2490   proxies, we divide HTTP header fields into two categories:
2491  <list style="symbols">
2492      <t>End-to-end header fields, which are  transmitted to the ultimate
2493        recipient of a request or response. End-to-end header fields in
2494        responses MUST be stored as part of a cache entry and MUST be
2495        transmitted in any response formed from a cache entry.</t>
2496
2497      <t>Hop-by-hop header fields, which are meaningful only for a single
2498        transport-level connection, and are not stored by caches or
2499        forwarded by proxies.</t>
2500  </list>
2501</t>
2502<t>
2503   The following HTTP/1.1 header fields are hop-by-hop header fields:
2504  <list style="symbols">
2505      <t><xref target="header.connection" format="none">Connection</xref></t>
2506      <t>Keep-Alive (Section 19.7.1.1 of <xref target="RFC2068"/>)</t>
2507      <t>Proxy-Authenticate (Section 4.2 of <xref target="Part7"/>)</t>
2508      <t>Proxy-Authorization (Section 4.3 of <xref target="Part7"/>)</t>
2509      <t><xref target="header.te" format="none">TE</xref></t>
2510      <t><xref target="header.trailer" format="none">Trailer</xref></t>
2511      <t><xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref></t>
2512      <t><xref target="header.upgrade" format="none">Upgrade</xref></t>
2513  </list>
2514</t>
2515<t>
2516   All other header fields defined by HTTP/1.1 are end-to-end header fields.
2517</t>
2518<t>
2519   Other hop-by-hop header fields MUST be listed in a
2520   <xref target="header.connection" format="none">Connection</xref> header field (<xref target="header.connection"/>).
2521</t>
2522</section>
2523
2524<section title="Non-modifiable Header Fields" anchor="non-modifiable.header-fields">
2525<!--<t>
2526  <cref anchor="TODO-non-mod-headers" source="jre">
2527    Restored from <eref target="http://tools.ietf.org/html/draft-ietf-httpbis-p6-cache-05#section-7.2"/>.
2528    See also <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/60"/>.
2529  </cref>
2530</t>-->
2531<t>
2532   Some features of HTTP/1.1, such as Digest Authentication, depend on the
2533   value of certain end-to-end header fields. A non-transforming proxy SHOULD NOT
2534   modify an end-to-end header field unless the definition of that header field requires
2535   or specifically allows that.
2536</t>
2537<t>
2538   A non-transforming proxy MUST NOT modify any of the following fields in a
2539   request or response, and it MUST NOT add any of these fields if not
2540   already present:
2541  <list style="symbols">
2542    <t>Allow (Section 9.5 of <xref target="Part2"/>)</t>
2543    <t>Content-Location (Section 9.8 of <xref target="Part2"/>)</t>
2544    <t>Content-MD5 (Section 14.15 of <xref target="RFC2616"/>)</t>
2545    <t>ETag (Section 2.3 of <xref target="Part4"/>)</t>
2546    <t>Last-Modified (Section 2.2 of <xref target="Part4"/>)</t>
2547    <t>Server (Section 9.17 of <xref target="Part2"/>)</t>
2548  </list>
2549</t>
2550<t>
2551   A non-transforming proxy MUST NOT modify any of the following fields in a
2552   response:
2553  <list style="symbols">
2554    <t>Expires (Section 7.3 of <xref target="Part6"/>)</t>
2555  </list>
2556</t>
2557<t>
2558   but it MAY add any of these fields if not already present. If an
2559   Expires header field is added, it MUST be given a
2560   field value identical to that of the Date header field in
2561   that response.
2562</t>
2563<t>
2564   A proxy MUST NOT modify or add any of the following fields in a
2565   message that contains the no-transform cache-control directive, or in
2566   any request:
2567  <list style="symbols">
2568    <t>Content-Encoding (Section 9.6 of <xref target="Part2"/>)</t>
2569    <t>Content-Range (Section 5.2 of <xref target="Part5"/>)</t>
2570    <t>Content-Type (Section 9.9 of <xref target="Part2"/>)</t>
2571  </list>
2572</t>
2573<t>
2574   A transforming proxy MAY modify or add these fields to a message
2575   that does not include no-transform, but if it does so, it MUST add a
2576   Warning 214 (Transformation applied) if one does not already appear
2577   in the message (see Section 7.6 of <xref target="Part6"/>).
2578</t>
2579<t><list>
2580  <t>
2581    Warning: Unnecessary modification of end-to-end header fields might
2582    cause authentication failures if stronger authentication
2583    mechanisms are introduced in later versions of HTTP. Such
2584    authentication mechanisms MAY rely on the values of header fields
2585    not listed here.
2586  </t>
2587</list></t>
2588<t>
2589   A non-transforming proxy MUST preserve the message payload (<xref target="Part2"/>),
2590   though it MAY change the message body through application or removal
2591   of a transfer-coding (<xref target="transfer.codings"/>).
2592</t>
2593</section>
2594
2595</section>
2596
2597<section title="Associating a Response to a Request" anchor="associating.response.to.request">
2598<t>
2599   HTTP does not include a request identifier for associating a given
2600   request message with its corresponding one or more response messages.
2601   Hence, it relies on the order of response arrival to correspond exactly
2602   to the order in which requests are made on the same connection.
2603   More than one response message per request only occurs when one or more
2604   informational responses (1xx, see Section 4.3 of <xref target="Part2"/>) precede a final response
2605   to the same request.
2606</t>
2607<t>
2608   A client that uses persistent connections and sends more than one request
2609   per connection MUST maintain a list of outstanding requests in the
2610   order sent on that connection and MUST associate each received response
2611   message to the highest ordered request that has not yet received a final
2612   (non-1xx) response.
2613</t>
2614</section>
2615</section>
2616
2617<section title="Connection Management" anchor="connection.management">
2618
2619<section title="Connection" anchor="header.connection">
2620  <iref primary="true" item="Connection header field"/>
2621  <iref primary="true" item="Header Fields" subitem="Connection"/>
2622 
2623 
2624<t>
2625   The "Connection" header field allows the sender to specify
2626   options that are desired only for that particular connection.
2627   Such connection options MUST be removed or replaced before the
2628   message can be forwarded downstream by a proxy or gateway.
2629   This mechanism also allows the sender to indicate which HTTP
2630   header fields used in the message are only intended for the
2631   immediate recipient ("hop-by-hop"), as opposed to all recipients
2632   on the chain ("end-to-end"), enabling the message to be
2633   self-descriptive and allowing future connection-specific extensions
2634   to be deployed in HTTP without fear that they will be blindly
2635   forwarded by previously deployed intermediaries.
2636</t>
2637<t>
2638   The Connection header field's value has the following grammar:
2639</t>
2640<figure><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="connection-option"/><artwork type="abnf2616"><![CDATA[
2641  Connection        = 1#connection-option
2642  connection-option = token
2643]]></artwork></figure>
2644<t>
2645   Connection options are compared case-insensitively.
2646</t>
2647<t>
2648   A proxy or gateway MUST parse a received Connection
2649   header field before a message is forwarded and, for each
2650   connection-option in this field, remove any header field(s) from
2651   the message with the same name as the connection-option, and then
2652   remove the Connection header field itself or replace it with the
2653   sender's own connection options for the forwarded message.
2654</t>
2655<t>
2656   A sender MUST NOT include field-names in the Connection header
2657   field-value for fields that are defined as expressing constraints
2658   for all recipients in the request or response chain, such as the
2659   Cache-Control header field (Section 7.2 of <xref target="Part6"/>).
2660</t>
2661<t>
2662   The connection options do not have to correspond to a header field
2663   present in the message, since a connection-specific header field
2664   might not be needed if there are no parameters associated with that
2665   connection option.  Recipients that trigger certain connection
2666   behavior based on the presence of connection options MUST do so
2667   based on the presence of the connection-option rather than only the
2668   presence of the optional header field.  In other words, if the
2669   connection option is received as a header field but not indicated
2670   within the Connection field-value, then the recipient MUST ignore
2671   the connection-specific header field because it has likely been
2672   forwarded by an intermediary that is only partially conformant.
2673</t>
2674<t>
2675   When defining new connection options, specifications ought to
2676   carefully consider existing deployed header fields and ensure
2677   that the new connection option does not share the same name as
2678   an unrelated header field that might already be deployed.
2679   Defining a new connection option essentially reserves that potential
2680   field-name for carrying additional information related to the
2681   connection option, since it would be unwise for senders to use
2682   that field-name for anything else.
2683</t>
2684<t>
2685   HTTP/1.1 defines the "close" connection option for the sender to
2686   signal that the connection will be closed after completion of the
2687   response. For example,
2688</t>
2689<figure><artwork type="example"><![CDATA[
2690  Connection: close
2691]]></artwork></figure>
2692<t>
2693   in either the request or the response header fields indicates that
2694   the connection SHOULD NOT  be considered "persistent" (<xref target="persistent.connections"/>)
2695   after the current request/response is complete.
2696</t>
2697<t>
2698   An HTTP/1.1 client that does not support persistent connections MUST
2699   include the "close" connection option in every request message.
2700</t>
2701<t>
2702   An HTTP/1.1 server that does not support persistent connections MUST
2703   include the "close" connection option in every response message that
2704   does not have a 1xx (Informational) status code.
2705</t>
2706</section>
2707
2708<section title="Via" anchor="header.via">
2709  <iref primary="true" item="Via header field"/>
2710  <iref primary="true" item="Header Fields" subitem="Via"/>
2711 
2712 
2713 
2714 
2715<t>
2716   The "Via" header field MUST be sent by a proxy or gateway to
2717   indicate the intermediate protocols and recipients between the user
2718   agent and the server on requests, and between the origin server and
2719   the client on responses. It is analogous to the "Received" field
2720   used by email systems (Section 3.6.7 of <xref target="RFC5322"/>)
2721   and is intended to be used for tracking message forwards,
2722   avoiding request loops, and identifying the protocol capabilities of
2723   all senders along the request/response chain.
2724</t>
2725<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[
2726  Via               = 1#( received-protocol RWS received-by
2727                          [ RWS comment ] )
2728  received-protocol = [ protocol-name "/" ] protocol-version
2729  received-by       = ( uri-host [ ":" port ] ) / pseudonym
2730  pseudonym         = token
2731]]></artwork></figure>
2732<t>
2733   The received-protocol indicates the protocol version of the message
2734   received by the server or client along each segment of the
2735   request/response chain. The received-protocol version is appended to
2736   the Via field value when the message is forwarded so that information
2737   about the protocol capabilities of upstream applications remains
2738   visible to all recipients.
2739</t>
2740<t>
2741   The protocol-name is excluded if and only if it would be "HTTP". The
2742   received-by field is normally the host and optional port number of a
2743   recipient server or client that subsequently forwarded the message.
2744   However, if the real host is considered to be sensitive information,
2745   it MAY be replaced by a pseudonym. If the port is not given, it MAY
2746   be assumed to be the default port of the received-protocol.
2747</t>
2748<t>
2749   Multiple Via field values represent each proxy or gateway that has
2750   forwarded the message. Each recipient MUST append its information
2751   such that the end result is ordered according to the sequence of
2752   forwarding applications.
2753</t>
2754<t>
2755   Comments MAY be used in the Via header field to identify the software
2756   of each recipient, analogous to the User-Agent and
2757   Server header fields. However, all comments in the Via field
2758   are optional and MAY be removed by any recipient prior to forwarding the
2759   message.
2760</t>
2761<t>
2762   For example, a request message could be sent from an HTTP/1.0 user
2763   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
2764   forward the request to a public proxy at p.example.net, which completes
2765   the request by forwarding it to the origin server at www.example.com.
2766   The request received by www.example.com would then have the following
2767   Via header field:
2768</t>
2769<figure><artwork type="example"><![CDATA[
2770  Via: 1.0 fred, 1.1 p.example.net (Apache/1.1)
2771]]></artwork></figure>
2772<t>
2773   A proxy or gateway used as a portal through a network firewall
2774   SHOULD NOT forward the names and ports of hosts within the firewall
2775   region unless it is explicitly enabled to do so. If not enabled, the
2776   received-by host of any host behind the firewall SHOULD be replaced
2777   by an appropriate pseudonym for that host.
2778</t>
2779<t>
2780   For organizations that have strong privacy requirements for hiding
2781   internal structures, a proxy or gateway MAY combine an ordered
2782   subsequence of Via header field entries with identical received-protocol
2783   values into a single such entry. For example,
2784</t>
2785<figure><artwork type="example"><![CDATA[
2786  Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
2787]]></artwork></figure>
2788<t>
2789  could be collapsed to
2790</t>
2791<figure><artwork type="example"><![CDATA[
2792  Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
2793]]></artwork></figure>
2794<t>
2795   Senders SHOULD NOT combine multiple entries unless they are all
2796   under the same organizational control and the hosts have already been
2797   replaced by pseudonyms. Senders MUST NOT combine entries which
2798   have different received-protocol values.
2799</t>
2800</section>
2801
2802<section title="Persistent Connections" anchor="persistent.connections">
2803
2804<section title="Purpose" anchor="persistent.purpose">
2805<t>
2806   Prior to persistent connections, a separate TCP connection was
2807   established for each request, increasing the load on HTTP servers
2808   and causing congestion on the Internet. The use of inline images and
2809   other associated data often requires a client to make multiple
2810   requests of the same server in a short amount of time. Analysis of
2811   these performance problems and results from a prototype
2812   implementation are available <xref target="Pad1995"/> <xref target="Spe"/>. Implementation experience and
2813   measurements of actual HTTP/1.1 implementations show good
2814   results <xref target="Nie1997"/>. Alternatives have also been explored, for example,
2815   T/TCP <xref target="Tou1998"/>.
2816</t>
2817<t>
2818   Persistent HTTP connections have a number of advantages:
2819  <list style="symbols">
2820      <t>
2821        By opening and closing fewer TCP connections, CPU time is saved
2822        in routers and hosts (clients, servers, proxies, gateways,
2823        tunnels, or caches), and memory used for TCP protocol control
2824        blocks can be saved in hosts.
2825      </t>
2826      <t>
2827        HTTP requests and responses can be pipelined on a connection.
2828        Pipelining allows a client to make multiple requests without
2829        waiting for each response, allowing a single TCP connection to
2830        be used much more efficiently, with much lower elapsed time.
2831      </t>
2832      <t>
2833        Network congestion is reduced by reducing the number of packets
2834        caused by TCP opens, and by allowing TCP sufficient time to
2835        determine the congestion state of the network.
2836      </t>
2837      <t>
2838        Latency on subsequent requests is reduced since there is no time
2839        spent in TCP's connection opening handshake.
2840      </t>
2841      <t>
2842        HTTP can evolve more gracefully, since errors can be reported
2843        without the penalty of closing the TCP connection. Clients using
2844        future versions of HTTP might optimistically try a new feature,
2845        but if communicating with an older server, retry with old
2846        semantics after an error is reported.
2847      </t>
2848    </list>
2849</t>
2850<t>
2851   HTTP implementations SHOULD implement persistent connections.
2852</t>
2853</section>
2854
2855<section title="Overall Operation" anchor="persistent.overall">
2856<t>
2857   A significant difference between HTTP/1.1 and earlier versions of
2858   HTTP is that persistent connections are the default behavior of any
2859   HTTP connection. That is, unless otherwise indicated, the client
2860   SHOULD assume that the server will maintain a persistent connection,
2861   even after error responses from the server.
2862</t>
2863<t>
2864   Persistent connections provide a mechanism by which a client and a
2865   server can signal the close of a TCP connection. This signaling takes
2866   place using the <xref target="header.connection" format="none">Connection</xref> header field
2867   (<xref target="header.connection"/>). Once a close has been signaled, the
2868   client MUST NOT send any more requests on that
2869   connection.
2870</t>
2871
2872<section title="Negotiation" anchor="persistent.negotiation">
2873<t>
2874   An HTTP/1.1 server MAY assume that a HTTP/1.1 client intends to
2875   maintain a persistent connection unless a <xref target="header.connection" format="none">Connection</xref> header
2876   field including the connection option "close" was sent in the request. If
2877   the server chooses to close the connection immediately after sending the
2878   response, it SHOULD send a Connection header field including the
2879   connection option "close".
2880</t>
2881<t>
2882   An HTTP/1.1 client MAY expect a connection to remain open, but would
2883   decide to keep it open based on whether the response from a server
2884   contains a <xref target="header.connection" format="none">Connection</xref> header field with the connection option
2885   "close". In case the client does not want to maintain a connection for more
2886   than that request, it SHOULD send a Connection header field including the
2887   connection option "close".
2888</t>
2889<t>
2890   If either the client or the server sends the "close" option in the
2891   <xref target="header.connection" format="none">Connection</xref> header field, that request becomes the last one
2892   for the connection.
2893</t>
2894<t>
2895   Clients and servers SHOULD NOT  assume that a persistent connection is
2896   maintained for HTTP versions less than 1.1 unless it is explicitly
2897   signaled. See <xref target="compatibility.with.http.1.0.persistent.connections"/> for more information on backward
2898   compatibility with HTTP/1.0 clients.
2899</t>
2900<t>
2901   Each persistent connection applies to only one transport link.
2902</t>
2903<t>
2904   A proxy server MUST NOT establish a HTTP/1.1 persistent connection
2905   with an HTTP/1.0 client (but see Section 19.7.1 of <xref target="RFC2068"/>
2906   for information and discussion of the problems with the Keep-Alive header field
2907   implemented by many HTTP/1.0 clients).
2908</t>
2909<t>
2910   In order to remain persistent, all messages on the connection MUST
2911   have a self-defined message length (i.e., one not defined by closure
2912   of the connection), as described in <xref target="message.body"/>.
2913</t>
2914</section>
2915
2916<section title="Pipelining" anchor="pipelining">
2917<t>
2918   A client that supports persistent connections MAY "pipeline" its
2919   requests (i.e., send multiple requests without waiting for each
2920   response). A server MUST send its responses to those requests in the
2921   same order that the requests were received.
2922</t>
2923<t>
2924   Clients which assume persistent connections and pipeline immediately
2925   after connection establishment SHOULD be prepared to retry their
2926   connection if the first pipelined attempt fails. If a client does
2927   such a retry, it MUST NOT pipeline before it knows the connection is
2928   persistent. Clients MUST also be prepared to resend their requests if
2929   the server closes the connection before sending all of the
2930   corresponding responses.
2931</t>
2932<t>
2933   Clients SHOULD NOT pipeline requests using non-idempotent request methods or
2934   non-idempotent sequences of request methods (see Section 2.1.2 of <xref target="Part2"/>). Otherwise, a
2935   premature termination of the transport connection could lead to
2936   indeterminate results. A client wishing to send a non-idempotent
2937   request SHOULD wait to send that request until it has received the
2938   response status line for the previous request.
2939</t>
2940</section>
2941</section>
2942
2943<section title="Practical Considerations" anchor="persistent.practical">
2944<t>
2945   Servers will usually have some time-out value beyond which they will
2946   no longer maintain an inactive connection. Proxy servers might make
2947   this a higher value since it is likely that the client will be making
2948   more connections through the same server. The use of persistent
2949   connections places no requirements on the length (or existence) of
2950   this time-out for either the client or the server.
2951</t>
2952<t>
2953   When a client or server wishes to time-out it SHOULD issue a graceful
2954   close on the transport connection. Clients and servers SHOULD both
2955   constantly watch for the other side of the transport close, and
2956   respond to it as appropriate. If a client or server does not detect
2957   the other side's close promptly it could cause unnecessary resource
2958   drain on the network.
2959</t>
2960<t>
2961   A client, server, or proxy MAY close the transport connection at any
2962   time. For example, a client might have started to send a new request
2963   at the same time that the server has decided to close the "idle"
2964   connection. From the server's point of view, the connection is being
2965   closed while it was idle, but from the client's point of view, a
2966   request is in progress.
2967</t>
2968<t>
2969   Clients (including proxies) SHOULD limit the number of simultaneous
2970   connections that they maintain to a given server (including proxies).
2971</t>
2972<t>
2973   Previous revisions of HTTP gave a specific number of connections as a
2974   ceiling, but this was found to be impractical for many applications. As a
2975   result, this specification does not mandate a particular maximum number of
2976   connections, but instead encourages clients to be conservative when opening
2977   multiple connections.
2978</t>
2979<t>
2980   In particular, while using multiple connections avoids the "head-of-line
2981   blocking" problem (whereby a request that takes significant server-side
2982   processing and/or has a large payload can block subsequent requests on the
2983   same connection), each connection used consumes server resources (sometimes
2984   significantly), and furthermore using multiple connections can cause
2985   undesirable side effects in congested networks.
2986</t>
2987<t>
2988   Note that servers might reject traffic that they deem abusive, including an
2989   excessive number of connections from a client.
2990</t>
2991</section>
2992
2993<section title="Retrying Requests" anchor="persistent.retrying.requests">
2994<t>
2995   Senders can close the transport connection at any time. Therefore,
2996   clients, servers, and proxies MUST be able to recover
2997   from asynchronous close events. Client software MAY reopen the
2998   transport connection and retransmit the aborted sequence of requests
2999   without user interaction so long as the request sequence is
3000   idempotent (see Section 2.1.2 of <xref target="Part2"/>). Non-idempotent request methods or sequences
3001   MUST NOT be automatically retried, although user agents MAY offer a
3002   human operator the choice of retrying the request(s). Confirmation by
3003   user-agent software with semantic understanding of the application
3004   MAY substitute for user confirmation. The automatic retry SHOULD NOT
3005   be repeated if the second sequence of requests fails.
3006</t>
3007</section>
3008</section>
3009
3010<section title="Message Transmission Requirements" anchor="message.transmission.requirements">
3011
3012<section title="Persistent Connections and Flow Control" anchor="persistent.flow">
3013<t>
3014   HTTP/1.1 servers SHOULD maintain persistent connections and use TCP's
3015   flow control mechanisms to resolve temporary overloads, rather than
3016   terminating connections with the expectation that clients will retry.
3017   The latter technique can exacerbate network congestion.
3018</t>
3019</section>
3020
3021<section title="Monitoring Connections for Error Status Messages" anchor="persistent.monitor">
3022<t>
3023   An HTTP/1.1 (or later) client sending a message body SHOULD monitor
3024   the network connection for an error status code while it is transmitting
3025   the request. If the client sees an error status code, it SHOULD
3026   immediately cease transmitting the body. If the body is being sent
3027   using a "chunked" encoding (<xref target="transfer.codings"/>), a zero length chunk and
3028   empty trailer MAY be used to prematurely mark the end of the message.
3029   If the body was preceded by a Content-Length header field, the client MUST
3030   close the connection.
3031</t>
3032</section>
3033
3034<section title="Use of the 100 (Continue) Status" anchor="use.of.the.100.status">
3035<t>
3036   The purpose of the 100 (Continue) status code (see Section 4.3.1 of <xref target="Part2"/>)
3037   is to allow a client that is sending a request message with a request body
3038   to determine if the origin server is willing to accept the request
3039   (based on the request header fields) before the client sends the request
3040   body. In some cases, it might either be inappropriate or highly
3041   inefficient for the client to send the body if the server will reject
3042   the message without looking at the body.
3043</t>
3044<t>
3045   Requirements for HTTP/1.1 clients:
3046  <list style="symbols">
3047    <t>
3048        If a client will wait for a 100 (Continue) response before
3049        sending the request body, it MUST send an Expect header
3050        field (Section 9.11 of <xref target="Part2"/>) with the "100-continue" expectation.
3051    </t>
3052    <t>
3053        A client MUST NOT send an Expect header field with
3054        the "100-continue" expectation if it does not intend to send a request
3055        body.
3056    </t>
3057  </list>
3058</t>
3059<t>
3060   Because of the presence of older implementations, the protocol allows
3061   ambiguous situations in which a client might send "Expect: 100-continue"
3062   without receiving either a 417 (Expectation Failed)
3063   or a 100 (Continue) status code. Therefore, when a client sends this
3064   header field to an origin server (possibly via a proxy) from which it
3065   has never seen a 100 (Continue) status code, the client SHOULD NOT 
3066   wait for an indefinite period before sending the request body.
3067</t>
3068<t>
3069   Requirements for HTTP/1.1 origin servers:
3070  <list style="symbols">
3071    <t> Upon receiving a request which includes an Expect header
3072        field with the "100-continue" expectation, an origin server MUST
3073        either respond with 100 (Continue) status code and continue to read
3074        from the input stream, or respond with a final status code. The
3075        origin server MUST NOT wait for the request body before sending
3076        the 100 (Continue) response. If it responds with a final status
3077        code, it MAY close the transport connection or it MAY continue
3078        to read and discard the rest of the request.  It MUST NOT
3079        perform the request method if it returns a final status code.
3080    </t>
3081    <t> An origin server SHOULD NOT  send a 100 (Continue) response if
3082        the request message does not include an Expect header
3083        field with the "100-continue" expectation, and MUST NOT send a
3084        100 (Continue) response if such a request comes from an HTTP/1.0
3085        (or earlier) client. There is an exception to this rule: for
3086        compatibility with <xref target="RFC2068"/>, a server MAY send a 100 (Continue)
3087        status code in response to an HTTP/1.1 PUT or POST request that does
3088        not include an Expect header field with the "100-continue"
3089        expectation. This exception, the purpose of which is
3090        to minimize any client processing delays associated with an
3091        undeclared wait for 100 (Continue) status code, applies only to
3092        HTTP/1.1 requests, and not to requests with any other HTTP-version
3093        value.
3094    </t>
3095    <t> An origin server MAY omit a 100 (Continue) response if it has
3096        already received some or all of the request body for the
3097        corresponding request.
3098    </t>
3099    <t> An origin server that sends a 100 (Continue) response MUST
3100        ultimately send a final status code, once the request body is
3101        received and processed, unless it terminates the transport
3102        connection prematurely.
3103    </t>
3104    <t> If an origin server receives a request that does not include an
3105        Expect header field with the "100-continue" expectation,
3106        the request includes a request body, and the server responds
3107        with a final status code before reading the entire request body
3108        from the transport connection, then the server SHOULD NOT  close
3109        the transport connection until it has read the entire request,
3110        or until the client closes the connection. Otherwise, the client
3111        might not reliably receive the response message. However, this
3112        requirement ought not be construed as preventing a server from
3113        defending itself against denial-of-service attacks, or from
3114        badly broken client implementations.
3115      </t>
3116    </list>
3117</t>
3118<t>
3119   Requirements for HTTP/1.1 proxies:
3120  <list style="symbols">
3121    <t> If a proxy receives a request that includes an Expect
3122        header field with the "100-continue" expectation, and the proxy
3123        either knows that the next-hop server complies with HTTP/1.1 or
3124        higher, or does not know the HTTP version of the next-hop
3125        server, it MUST forward the request, including the Expect header
3126        field.
3127    </t>
3128    <t> If the proxy knows that the version of the next-hop server is
3129        HTTP/1.0 or lower, it MUST NOT forward the request, and it MUST
3130        respond with a 417 (Expectation Failed) status code.
3131    </t>
3132    <t> Proxies SHOULD maintain a record of the HTTP version
3133        numbers received from recently-referenced next-hop servers.
3134    </t>
3135    <t> A proxy MUST NOT forward a 100 (Continue) response if the
3136        request message was received from an HTTP/1.0 (or earlier)
3137        client and did not include an Expect header field with
3138        the "100-continue" expectation. This requirement overrides the
3139        general rule for forwarding of 1xx responses (see Section 4.3 of <xref target="Part2"/>).
3140    </t>
3141  </list>
3142</t>
3143</section>
3144
3145<section title="Closing Connections on Error" anchor="closing.connections.on.error">
3146<t>
3147   If the client is sending data, a server implementation using TCP
3148   SHOULD be careful to ensure that the client acknowledges receipt of
3149   the packet(s) containing the response, before the server closes the
3150   input connection. If the client continues sending data to the server
3151   after the close, the server's TCP stack will send a reset packet to
3152   the client, which might erase the client's unacknowledged input buffers
3153   before they can be read and interpreted by the HTTP application.
3154</t>
3155</section>
3156
3157</section>
3158
3159<section title="Upgrade" anchor="header.upgrade">
3160  <iref primary="true" item="Upgrade header field"/>
3161  <iref primary="true" item="Header Fields" subitem="Upgrade"/>
3162 
3163 
3164 
3165 
3166<t>
3167   The "Upgrade" header field allows the client to specify what
3168   additional communication protocols it would like to use, if the server
3169   chooses to switch protocols. Servers can use it to indicate what protocols
3170   they are willing to switch to.
3171</t>
3172<figure><iref primary="true" item="Grammar" subitem="Upgrade"/><artwork type="abnf2616"><![CDATA[
3173  Upgrade          = 1#protocol
3174
3175  protocol         = protocol-name ["/" protocol-version]
3176  protocol-name    = token
3177  protocol-version = token
3178]]></artwork></figure>
3179<t>
3180   For example,
3181</t>
3182<figure><artwork type="example"><![CDATA[
3183  Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
3184]]></artwork></figure>
3185<t>
3186   The Upgrade header field is intended to provide a simple mechanism
3187   for transitioning from HTTP/1.1 to some other, incompatible protocol. It
3188   does so by allowing the client to advertise its desire to use another
3189   protocol, such as a later version of HTTP with a higher major version
3190   number, even though the current request has been made using HTTP/1.1.
3191   This eases the difficult transition between incompatible protocols by
3192   allowing the client to initiate a request in the more commonly
3193   supported protocol while indicating to the server that it would like
3194   to use a "better" protocol if available (where "better" is determined
3195   by the server, possibly according to the nature of the request method
3196   or target resource).
3197</t>
3198<t>
3199   The Upgrade header field only applies to switching application-layer
3200   protocols upon the existing transport-layer connection. Upgrade
3201   cannot be used to insist on a protocol change; its acceptance and use
3202   by the server is optional. The capabilities and nature of the
3203   application-layer communication after the protocol change is entirely
3204   dependent upon the new protocol chosen, although the first action
3205   after changing the protocol MUST be a response to the initial HTTP
3206   request containing the Upgrade header field.
3207</t>
3208<t>
3209   The Upgrade header field only applies to the immediate connection.
3210   Therefore, the upgrade keyword MUST be supplied within a
3211   <xref target="header.connection" format="none">Connection</xref> header field (<xref target="header.connection"/>)
3212   whenever Upgrade is present in an HTTP/1.1 message.
3213</t>
3214<t>
3215   The Upgrade header field cannot be used to indicate a switch to a
3216   protocol on a different connection. For that purpose, it is more
3217   appropriate to use a 3xx (Redirection) response (Section 4.5 of <xref target="Part2"/>).
3218</t>
3219<t>
3220   Servers MUST include the "Upgrade" header field in 101 (Switching
3221   Protocols) responses to indicate which protocol(s) are being switched to,
3222   and MUST include it in 426 (Upgrade Required) responses to indicate
3223   acceptable protocols to upgrade to. Servers MAY include it in any other
3224   response to indicate that they are willing to upgrade to one of the
3225   specified protocols.
3226</t>
3227<t>
3228   This specification only defines the protocol name "HTTP" for use by
3229   the family of Hypertext Transfer Protocols, as defined by the HTTP
3230   version rules of <xref target="http.version"/> and future updates to this
3231   specification. Additional tokens can be registered with IANA using the
3232   registration procedure defined in <xref target="upgrade.token.registry"/>.
3233</t>
3234</section>
3235
3236</section>
3237
3238<section title="IANA Considerations" anchor="IANA.considerations">
3239
3240<section title="Header Field Registration" anchor="header.field.registration">
3241<t>
3242   HTTP header fields are registered within the Message Header Field Registry
3243   <xref target="RFC3864"/> maintained by IANA at
3244   <eref target="http://www.iana.org/assignments/message-headers/message-header-index.html"/>.
3245</t>
3246<t>
3247   This document defines the following HTTP header fields, so their
3248   associated registry entries shall be updated according to the permanent
3249   registrations below:
3250</t>
3251
3252<!--AUTOGENERATED FROM extract-header-defs.xslt, do not edit manually-->
3253<texttable align="left" suppress-title="true" anchor="iana.header.registration.table">
3254   <ttcol>Header Field Name</ttcol>
3255   <ttcol>Protocol</ttcol>
3256   <ttcol>Status</ttcol>
3257   <ttcol>Reference</ttcol>
3258
3259   <c>Connection</c>
3260   <c>http</c>
3261   <c>standard</c>
3262   <c>
3263      <xref target="header.connection"/>
3264   </c>
3265   <c>Content-Length</c>
3266   <c>http</c>
3267   <c>standard</c>
3268   <c>
3269      <xref target="header.content-length"/>
3270   </c>
3271   <c>Host</c>
3272   <c>http</c>
3273   <c>standard</c>
3274   <c>
3275      <xref target="header.host"/>
3276   </c>
3277   <c>TE</c>
3278   <c>http</c>
3279   <c>standard</c>
3280   <c>
3281      <xref target="header.te"/>
3282   </c>
3283   <c>Trailer</c>
3284   <c>http</c>
3285   <c>standard</c>
3286   <c>
3287      <xref target="header.trailer"/>
3288   </c>
3289   <c>Transfer-Encoding</c>
3290   <c>http</c>
3291   <c>standard</c>
3292   <c>
3293      <xref target="header.transfer-encoding"/>
3294   </c>
3295   <c>Upgrade</c>
3296   <c>http</c>
3297   <c>standard</c>
3298   <c>
3299      <xref target="header.upgrade"/>
3300   </c>
3301   <c>Via</c>
3302   <c>http</c>
3303   <c>standard</c>
3304   <c>
3305      <xref target="header.via"/>
3306   </c>
3307</texttable>
3308<!--(END)-->
3309
3310<t>
3311   Furthermore, the header field-name "Close" shall be registered as
3312   "reserved", since using that name as an HTTP header field might
3313   conflict with the "close" connection option of the "<xref target="header.connection" format="none">Connection</xref>"
3314   header field (<xref target="header.connection"/>).
3315</t>
3316<texttable align="left" suppress-title="true">
3317   <ttcol>Header Field Name</ttcol>
3318   <ttcol>Protocol</ttcol>
3319   <ttcol>Status</ttcol>
3320   <ttcol>Reference</ttcol>
3321
3322   <c>Close</c>
3323   <c>http</c>
3324   <c>reserved</c>
3325   <c>
3326      <xref target="header.field.registration"/>
3327   </c>
3328</texttable>
3329<t>
3330   The change controller is: "IETF (iesg@ietf.org) - Internet Engineering Task Force".
3331</t>
3332</section>
3333
3334<section title="URI Scheme Registration" anchor="uri.scheme.registration">
3335<t>
3336   IANA maintains the registry of URI Schemes <xref target="RFC4395"/> at
3337   <eref target="http://www.iana.org/assignments/uri-schemes.html"/>.
3338</t>
3339<t>
3340   This document defines the following URI schemes, so their
3341   associated registry entries shall be updated according to the permanent
3342   registrations below:
3343</t>
3344<texttable align="left" suppress-title="true">
3345   <ttcol>URI Scheme</ttcol>
3346   <ttcol>Description</ttcol>
3347   <ttcol>Reference</ttcol>
3348
3349   <c>http</c>
3350   <c>Hypertext Transfer Protocol</c>
3351   <c><xref target="http.uri"/></c>
3352
3353   <c>https</c>
3354   <c>Hypertext Transfer Protocol Secure</c>
3355   <c><xref target="https.uri"/></c>
3356</texttable>
3357</section>
3358
3359<section title="Internet Media Type Registrations" anchor="internet.media.type.http">
3360<t>
3361   This document serves as the specification for the Internet media types
3362   "message/http" and "application/http". The following is to be registered with
3363   IANA (see <xref target="RFC4288"/>).
3364</t>
3365<section title="Internet Media Type message/http" anchor="internet.media.type.message.http">
3366<iref item="Media Type" subitem="message/http" primary="true"/>
3367<iref item="message/http Media Type" primary="true"/>
3368<t>
3369   The message/http type can be used to enclose a single HTTP request or
3370   response message, provided that it obeys the MIME restrictions for all
3371   "message" types regarding line length and encodings.
3372</t>
3373<t>
3374  <list style="hanging">
3375    <t hangText="Type name:">
3376      message
3377    </t>
3378    <t hangText="Subtype name:">
3379      http
3380    </t>
3381    <t hangText="Required parameters:">
3382      none
3383    </t>
3384    <t hangText="Optional parameters:">
3385      version, msgtype
3386      <list style="hanging">
3387        <t hangText="version:">
3388          The HTTP-version number of the enclosed message
3389          (e.g., "1.1"). If not present, the version can be
3390          determined from the first line of the body.
3391        </t>
3392        <t hangText="msgtype:">
3393          The message type — "request" or "response". If not
3394          present, the type can be determined from the first
3395          line of the body.
3396        </t>
3397      </list>
3398    </t>
3399    <t hangText="Encoding considerations:">
3400      only "7bit", "8bit", or "binary" are permitted
3401    </t>
3402    <t hangText="Security considerations:">
3403      none
3404    </t>
3405    <t hangText="Interoperability considerations:">
3406      none
3407    </t>
3408    <t hangText="Published specification:">
3409      This specification (see <xref target="internet.media.type.message.http"/>).
3410    </t>
3411    <t hangText="Applications that use this media type:">
3412    </t>
3413    <t hangText="Additional information:">
3414      <list style="hanging">
3415        <t hangText="Magic number(s):">none</t>
3416        <t hangText="File extension(s):">none</t>
3417        <t hangText="Macintosh file type code(s):">none</t>
3418      </list>
3419    </t>
3420    <t hangText="Person and email address to contact for further information:">
3421      See Authors Section.
3422    </t>
3423    <t hangText="Intended usage:">
3424      COMMON
3425    </t>
3426    <t hangText="Restrictions on usage:">
3427      none
3428    </t>
3429    <t hangText="Author/Change controller:">
3430      IESG
3431    </t>
3432  </list>
3433</t>
3434</section>
3435<section title="Internet Media Type application/http" anchor="internet.media.type.application.http">
3436<iref item="Media Type" subitem="application/http" primary="true"/>
3437<iref item="application/http Media Type" primary="true"/>
3438<t>
3439   The application/http type can be used to enclose a pipeline of one or more
3440   HTTP request or response messages (not intermixed).
3441</t>
3442<t>
3443  <list style="hanging">
3444    <t hangText="Type name:">
3445      application
3446    </t>
3447    <t hangText="Subtype name:">
3448      http
3449    </t>
3450    <t hangText="Required parameters:">
3451      none
3452    </t>
3453    <t hangText="Optional parameters:">
3454      version, msgtype
3455      <list style="hanging">
3456        <t hangText="version:">
3457          The HTTP-version number of the enclosed messages
3458          (e.g., "1.1"). If not present, the version can be
3459          determined from the first line of the body.
3460        </t>
3461        <t hangText="msgtype:">
3462          The message type — "request" or "response". If not
3463          present, the type can be determined from the first
3464          line of the body.
3465        </t>
3466      </list>
3467    </t>
3468    <t hangText="Encoding considerations:">
3469      HTTP messages enclosed by this type
3470      are in "binary" format; use of an appropriate
3471      Content-Transfer-Encoding is required when
3472      transmitted via E-mail.
3473    </t>
3474    <t hangText="Security considerations:">
3475      none
3476    </t>
3477    <t hangText="Interoperability considerations:">
3478      none
3479    </t>
3480    <t hangText="Published specification:">
3481      This specification (see <xref target="internet.media.type.application.http"/>).
3482    </t>
3483    <t hangText="Applications that use this media type:">
3484    </t>
3485    <t hangText="Additional information:">
3486      <list style="hanging">
3487        <t hangText="Magic number(s):">none</t>
3488        <t hangText="File extension(s):">none</t>
3489        <t hangText="Macintosh file type code(s):">none</t>
3490      </list>
3491    </t>
3492    <t hangText="Person and email address to contact for further information:">
3493      See Authors Section.
3494    </t>
3495    <t hangText="Intended usage:">
3496      COMMON
3497    </t>
3498    <t hangText="Restrictions on usage:">
3499      none
3500    </t>
3501    <t hangText="Author/Change controller:">
3502      IESG
3503    </t>
3504  </list>
3505</t>
3506</section>
3507</section>
3508
3509<section title="Transfer Coding Registry" anchor="transfer.coding.registry">
3510<t>
3511   The HTTP Transfer Coding Registry defines the name space for transfer
3512   coding names.
3513</t>
3514<t>
3515   Registrations MUST include the following fields:
3516   <list style="symbols">
3517     <t>Name</t>
3518     <t>Description</t>
3519     <t>Pointer to specification text</t>
3520   </list>
3521</t>
3522<t>
3523   Names of transfer codings MUST NOT overlap with names of content codings
3524   (Section 5.4 of <xref target="Part2"/>) unless the encoding transformation is identical, as it
3525   is the case for the compression codings defined in
3526   <xref target="compression.codings"/>.
3527</t>
3528<t>
3529   Values to be added to this name space require IETF Review (see
3530   Section 4.1 of <xref target="RFC5226"/>), and MUST
3531   conform to the purpose of transfer coding defined in this section.
3532</t>
3533<t>
3534   The registry itself is maintained at
3535   <eref target="http://www.iana.org/assignments/http-parameters"/>.
3536</t>
3537</section>
3538
3539<section title="Transfer Coding Registrations" anchor="transfer.coding.registration">
3540<t>
3541   The HTTP Transfer Coding Registry shall be updated with the registrations
3542   below:
3543</t>
3544<texttable align="left" suppress-title="true" anchor="iana.transfer.coding.registration.table">
3545   <ttcol>Name</ttcol>
3546   <ttcol>Description</ttcol>
3547   <ttcol>Reference</ttcol>
3548   <c>chunked</c>
3549   <c>Transfer in a series of chunks</c>
3550   <c>
3551      <xref target="chunked.encoding"/>
3552   </c>
3553   <c>compress</c>
3554   <c>UNIX "compress" program method</c>
3555   <c>
3556      <xref target="compress.coding"/>
3557   </c>
3558   <c>deflate</c>
3559   <c>"deflate" compression mechanism (<xref target="RFC1951"/>) used inside
3560   the "zlib" data format (<xref target="RFC1950"/>)
3561   </c>
3562   <c>
3563      <xref target="deflate.coding"/>
3564   </c>
3565   <c>gzip</c>
3566   <c>Same as GNU zip <xref target="RFC1952"/></c>
3567   <c>
3568      <xref target="gzip.coding"/>
3569   </c>
3570</texttable>
3571</section>
3572
3573<section title="Upgrade Token Registry" anchor="upgrade.token.registry">
3574<t>
3575   The HTTP Upgrade Token Registry defines the name space for protocol-name
3576   tokens used to identify protocols in the <xref target="header.upgrade" format="none">Upgrade</xref> header
3577   field. Each registered protocol name is associated with contact information
3578   and an optional set of specifications that details how the connection
3579   will be processed after it has been upgraded.
3580</t>
3581<t>
3582   Registrations happen on a "First Come First Served" basis (see
3583   Section 4.1 of <xref target="RFC5226"/>) and are subject to the
3584   following rules:
3585  <list style="numbers">
3586    <t>A protocol-name token, once registered, stays registered forever.</t>
3587    <t>The registration MUST name a responsible party for the
3588       registration.</t>
3589    <t>The registration MUST name a point of contact.</t>
3590    <t>The registration MAY name a set of specifications associated with
3591       that token. Such specifications need not be publicly available.</t>
3592    <t>The registration SHOULD name a set of expected "protocol-version"
3593       tokens associated with that token at the time of registration.</t>
3594    <t>The responsible party MAY change the registration at any time.
3595       The IANA will keep a record of all such changes, and make them
3596       available upon request.</t>
3597    <t>The IESG MAY reassign responsibility for a protocol token.
3598       This will normally only be used in the case when a
3599       responsible party cannot be contacted.</t>
3600  </list>
3601</t>
3602<t>
3603   This registration procedure for HTTP Upgrade Tokens replaces that
3604   previously defined in Section 7.2 of <xref target="RFC2817"/>.
3605</t>
3606</section>
3607
3608<section title="Upgrade Token Registration" anchor="upgrade.token.registration">
3609<t>
3610   The HTTP Upgrade Token Registry shall be updated with the registration
3611   below:
3612</t>
3613<texttable align="left" suppress-title="true">
3614   <ttcol>Value</ttcol>
3615   <ttcol>Description</ttcol>
3616   <ttcol>Expected Version Tokens</ttcol>
3617   <ttcol>Reference</ttcol>
3618
3619   <c>HTTP</c>
3620   <c>Hypertext Transfer Protocol</c>
3621   <c>any DIGIT.DIGIT (e.g, "2.0")</c>
3622   <c><xref target="http.version"/></c>
3623</texttable>
3624<t>
3625   The responsible party is: "IETF (iesg@ietf.org) - Internet Engineering Task Force".
3626</t>
3627</section>
3628
3629</section>
3630
3631<section title="Security Considerations" anchor="security.considerations">
3632<t>
3633   This section is meant to inform application developers, information
3634   providers, and users of the security limitations in HTTP/1.1 as
3635   described by this document. The discussion does not include
3636   definitive solutions to the problems revealed, though it does make
3637   some suggestions for reducing security risks.
3638</t>
3639
3640<section title="Personal Information" anchor="personal.information">
3641<t>
3642   HTTP clients are often privy to large amounts of personal information
3643   (e.g., the user's name, location, mail address, passwords, encryption
3644   keys, etc.), and SHOULD be very careful to prevent unintentional
3645   leakage of this information.
3646   We very strongly recommend that a convenient interface be provided
3647   for the user to control dissemination of such information, and that
3648   designers and implementers be particularly careful in this area.
3649   History shows that errors in this area often create serious security
3650   and/or privacy problems and generate highly adverse publicity for the
3651   implementer's company.
3652</t>
3653</section>
3654
3655<section title="Abuse of Server Log Information" anchor="abuse.of.server.log.information">
3656<t>
3657   A server is in the position to save personal data about a user's
3658   requests which might identify their reading patterns or subjects of
3659   interest.  In particular, log information gathered at an intermediary
3660   often contains a history of user agent interaction, across a multitude
3661   of sites, that can be traced to individual users.
3662</t>
3663<t>
3664   HTTP log information is confidential in nature; its handling is often
3665   constrained by laws and regulations.  Log information needs to be securely
3666   stored and appropriate guidelines followed for its analysis.
3667   Anonymization of personal information within individual entries helps,
3668   but is generally not sufficient to prevent real log traces from being
3669   re-identified based on correlation with other access characteristics.
3670   As such, access traces that are keyed to a specific client should not
3671   be published even if the key is pseudonymous.
3672</t>
3673<t>
3674   To minimize the risk of theft or accidental publication, log information
3675   should be purged of personally identifiable information, including
3676   user identifiers, IP addresses, and user-provided query parameters,
3677   as soon as that information is no longer necessary to support operational
3678   needs for security, auditing, or fraud control.
3679</t>
3680</section>
3681
3682<section title="Attacks Based On File and Path Names" anchor="attack.pathname">
3683<t>
3684   Implementations of HTTP origin servers SHOULD be careful to restrict
3685   the documents returned by HTTP requests to be only those that were
3686   intended by the server administrators. If an HTTP server translates
3687   HTTP URIs directly into file system calls, the server MUST take
3688   special care not to serve files that were not intended to be
3689   delivered to HTTP clients. For example, UNIX, Microsoft Windows, and
3690   other operating systems use ".." as a path component to indicate a
3691   directory level above the current one. On such a system, an HTTP
3692   server MUST disallow any such construct in the request-target if it
3693   would otherwise allow access to a resource outside those intended to
3694   be accessible via the HTTP server. Similarly, files intended for
3695   reference only internally to the server (such as access control
3696   files, configuration files, and script code) MUST be protected from
3697   inappropriate retrieval, since they might contain sensitive
3698   information. Experience has shown that minor bugs in such HTTP server
3699   implementations have turned into security risks.
3700</t>
3701</section>
3702
3703<section title="DNS-related Attacks" anchor="dns.related.attacks">
3704<t>
3705   HTTP clients rely heavily on the Domain Name Service (DNS), and are thus
3706   generally prone to security attacks based on the deliberate misassociation
3707   of IP addresses and DNS names not protected by DNSSec. Clients need to be
3708   cautious in assuming the validity of an IP number/DNS name association unless
3709   the response is protected by DNSSec (<xref target="RFC4033"/>).
3710</t>
3711</section>
3712
3713<section title="Intermediaries and Caching" anchor="attack.intermediaries">
3714<t>
3715   By their very nature, HTTP intermediaries are men-in-the-middle, and
3716   represent an opportunity for man-in-the-middle attacks. Compromise of
3717   the systems on which the intermediaries run can result in serious security
3718   and privacy problems. Intermediaries have access to security-related
3719   information, personal information about individual users and
3720   organizations, and proprietary information belonging to users and
3721   content providers. A compromised intermediary, or an intermediary
3722   implemented or configured without regard to security and privacy
3723   considerations, might be used in the commission of a wide range of
3724   potential attacks.
3725</t>
3726<t>
3727   Intermediaries that contain a shared cache are especially vulnerable
3728   to cache poisoning attacks.
3729</t>
3730<t>
3731   Implementers need to consider the privacy and security
3732   implications of their design and coding decisions, and of the
3733   configuration options they provide to operators (especially the
3734   default configuration).
3735</t>
3736<t>
3737   Users need to be aware that intermediaries are no more trustworthy than
3738   the people who run them; HTTP itself cannot solve this problem.
3739</t>
3740<t>
3741   The judicious use of cryptography, when appropriate, might suffice to
3742   protect against a broad range of security and privacy attacks. Such
3743   cryptography is beyond the scope of the HTTP/1.1 specification.
3744</t>
3745</section>
3746
3747<section title="Protocol Element Size Overflows" anchor="attack.protocol.element.size.overflows">
3748<t>
3749   Because HTTP uses mostly textual, character-delimited fields, attackers can
3750   overflow buffers in implementations, and/or perform a Denial of Service
3751   against implementations that accept fields with unlimited lengths.
3752</t>
3753<t>
3754   To promote interoperability, this specification makes specific
3755   recommendations for minimum size limits on request-line
3756   (<xref target="request.line"/>)
3757   and blocks of header fields (<xref target="header.fields"/>). These are
3758   minimum recommendations, chosen to be supportable even by implementations
3759   with limited resources; it is expected that most implementations will
3760   choose substantially higher limits.
3761</t>
3762<t>
3763   This specification also provides a way for servers to reject messages that
3764   have request-targets that are too long (Section 4.6.12 of <xref target="Part2"/>) or request entities
3765   that are too large (Section 4.6 of <xref target="Part2"/>).
3766</t>
3767<t>
3768   Other fields (including but not limited to request methods, response status
3769   phrases, header field-names, and body chunks) SHOULD be limited by
3770   implementations carefully, so as to not impede interoperability.
3771</t>
3772</section>
3773</section>
3774
3775<section title="Acknowledgments" anchor="acks">
3776<t>
3777   This edition of HTTP builds on the many contributions that went into
3778   <xref target="RFC1945" format="none">RFC 1945</xref>,
3779   <xref target="RFC2068" format="none">RFC 2068</xref>,
3780   <xref target="RFC2145" format="none">RFC 2145</xref>, and
3781   <xref target="RFC2616" format="none">RFC 2616</xref>, including
3782   substantial contributions made by the previous authors, editors, and
3783   working group chairs: Tim Berners-Lee, Ari Luotonen, Roy T. Fielding,
3784   Henrik Frystyk Nielsen, Jim Gettys, Jeffrey C. Mogul, Larry Masinter,
3785   Paul J. Leach, and Mark Nottingham.
3786   See Section 16 of <xref target="RFC2616"/> for additional
3787   acknowledgements from prior revisions.
3788</t>
3789<t>
3790   Since 1999, the following contributors have helped improve the HTTP
3791   specification by reporting bugs, asking smart questions, drafting or
3792   reviewing text, and evaluating open issues:
3793</t>
3794
3795<t>Adam Barth,
3796Adam Roach,
3797Addison Phillips,
3798Adrian Chadd,
3799Adrien W. de Croy,
3800Alan Ford,
3801Alan Ruttenberg,
3802Albert Lunde,
3803Alek Storm,
3804Alex Rousskov,
3805Alexandre Morgaut,
3806Alexey Melnikov,
3807Alisha Smith,
3808Amichai Rothman,
3809Amit Klein,
3810Amos Jeffries,
3811Andreas Maier,
3812Andreas Petersson,
3813Anne van Kesteren,
3814Anthony Bryan,
3815Asbjorn Ulsberg,
3816Balachander Krishnamurthy,
3817Barry Leiba,
3818Ben Laurie,
3819Benjamin Niven-Jenkins,
3820Bil Corry,
3821Bill Burke,
3822Bjoern Hoehrmann,
3823Bob Scheifler,
3824Boris Zbarsky,
3825Brett Slatkin,
3826Brian Kell,
3827Brian McBarron,
3828Brian Pane,
3829Brian Smith,
3830Bryce Nesbitt,
3831Cameron Heavon-Jones,
3832Carl Kugler,
3833Carsten Bormann,
3834Charles Fry,
3835Chris Newman,
3836Cyrus Daboo,
3837Dale Robert Anderson,
3838Dan Winship,
3839Daniel Stenberg,
3840Dave Cridland,
3841Dave Crocker,
3842Dave Kristol,
3843David Booth,
3844David Singer,
3845David W. Morris,
3846Diwakar Shetty,
3847Dmitry Kurochkin,
3848Drummond Reed,
3849Duane Wessels,
3850Edward Lee,
3851Eliot Lear,
3852Eran Hammer-Lahav,
3853Eric D. Williams,
3854Eric J. Bowman,
3855Eric Lawrence,
3856Eric Rescorla,
3857Erik Aronesty,
3858Florian Weimer,
3859Frank Ellermann,
3860Fred Bohle,
3861Geoffrey Sneddon,
3862Gervase Markham,
3863Greg Wilkins,
3864Harald Tveit Alvestrand,
3865Harry Halpin,
3866Helge Hess,
3867Henrik Nordstrom,
3868Henry S. Thompson,
3869Henry Story,
3870Herbert van de Sompel,
3871Howard Melman,
3872Hugo Haas,
3873Ian Hickson,
3874Ingo Struck,
3875J. Ross Nicoll,
3876James H. Manger,
3877James Lacey,
3878James M. Snell,
3879Jamie Lokier,
3880Jan Algermissen,
3881Jeff Hodges (who came up with the term 'effective Request-URI'),
3882Jeff Walden,
3883Jim Luther,
3884Joe D. Williams,
3885Joe Gregorio,
3886Joe Orton,
3887John C. Klensin,
3888John C. Mallery,
3889John Cowan,
3890John Kemp,
3891John Panzer,
3892John Schneider,
3893John Stracke,
3894John Sullivan,
3895Jonas Sicking,
3896Jonathan Billington,
3897Jonathan Moore,
3898Jonathan Rees,
3899Jonathan Silvera,
3900Jordi Ros,
3901Joris Dobbelsteen,
3902Josh Cohen,
3903Julien Pierre,
3904Jungshik Shin,
3905Justin Chapweske,
3906Justin Erenkrantz,
3907Justin James,
3908Kalvinder Singh,
3909Karl Dubost,
3910Keith Hoffman,
3911Keith Moore,
3912Koen Holtman,
3913Konstantin Voronkov,
3914Kris Zyp,
3915Lisa Dusseault,
3916Maciej Stachowiak,
3917Marc Schneider,
3918Marc Slemko,
3919Mark Baker,
3920Mark Pauley,
3921Mark Watson,
3922Markus Isomaki,
3923Markus Lanthaler,
3924Martin J. Duerst,
3925Martin Musatov,
3926Martin Nilsson,
3927Martin Thomson,
3928Matt Lynch,
3929Matthew Cox,
3930Max Clark,
3931Michael Burrows,
3932Michael Hausenblas,
3933Mike Amundsen,
3934Mike Belshe,
3935Mike Kelly,
3936Mike Schinkel,
3937Miles Sabin,
3938Murray S. Kucherawy,
3939Mykyta Yevstifeyev,
3940Nathan Rixham,
3941Nicholas Shanks,
3942Nico Williams,
3943Nicolas Alvarez,
3944Nicolas Mailhot,
3945Noah Slater,
3946Pablo Castro,
3947Pat Hayes,
3948Patrick R. McManus,
3949Paul E. Jones,
3950Paul Hoffman,
3951Paul Marquess,
3952Peter Lepeska,
3953Peter Saint-Andre,
3954Peter Watkins,
3955Phil Archer,
3956Phillip Hallam-Baker,
3957Poul-Henning Kamp,
3958Preethi Natarajan,
3959Ray Polk,
3960Reto Bachmann-Gmuer,
3961Richard Cyganiak,
3962Robert Brewer,
3963Robert Collins,
3964Robert O'Callahan,
3965Robert Olofsson,
3966Robert Sayre,
3967Robert Siemer,
3968Robert de Wilde,
3969Roberto Javier Godoy,
3970Roberto Peon,
3971Ronny Widjaja,
3972S. Mike Dierken,
3973Salvatore Loreto,
3974Sam Johnston,
3975Sam Ruby,
3976Scott Lawrence (who maintained the original issues list),
3977Sean B. Palmer,
3978Shane McCarron,
3979Stefan Eissing,
3980Stefan Tilkov,
3981Stefanos Harhalakis,
3982Stephane Bortzmeyer,
3983Stephen Farrell,
3984Stephen Ludin,
3985Stuart Williams,
3986Subbu Allamaraju,
3987Sylvain Hellegouarch,
3988Tapan Divekar,
3989Tatsuya Hayashi,
3990Ted Hardie,
3991Thomas Broyer,
3992Thomas Nordin,
3993Thomas Roessler,
3994Tim Bray,
3995Tim Morgan,
3996Tim Olsen,
3997Tom Zhou,
3998Travis Snoozy,
3999Tyler Close,
4000Vincent Murphy,
4001Wenbo Zhu,
4002Werner Baumann,
4003Wilbur Streett,
4004Wilfredo Sanchez Vega,
4005William A. Rowe Jr.,
4006William Chan,
4007Willy Tarreau,
4008Xiaoshu Wang,
4009Yaron Goland,
4010Yngve Nysaeter Pettersen,
4011Yoav Nir,
4012Yogesh Bang,
4013Yutaka Oiwa,
4014Zed A. Shaw, and
4015Zhong Yu.
4016</t>
4017
4018</section>
4019
4020</middle>
4021<back>
4022
4023<references title="Normative References">
4024
4025<reference anchor="Part2">
4026  <front>
4027    <title>HTTP/1.1, part 2: Semantics and Payloads</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="2012"/>
4041  </front>
4042  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-20"/>
4043 
4044</reference>
4045
4046<reference anchor="Part4">
4047  <front>
4048    <title>HTTP/1.1, part 4: Conditional Requests</title>
4049    <author fullname="Roy T. Fielding" initials="R." role="editor" surname="Fielding">
4050      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4051      <address><email>fielding@gbiv.com</email></address>
4052    </author>
4053    <author fullname="Yves Lafon" initials="Y." role="editor" surname="Lafon">
4054      <organization abbrev="W3C">World Wide Web Consortium</organization>
4055      <address><email>ylafon@w3.org</email></address>
4056    </author>
4057    <author fullname="Julian F. Reschke" initials="J. F." role="editor" surname="Reschke">
4058      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4059      <address><email>julian.reschke@greenbytes.de</email></address>
4060    </author>
4061    <date month="July" year="2012"/>
4062  </front>
4063  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p4-conditional-20"/>
4064 
4065</reference>
4066
4067<reference anchor="Part5">
4068  <front>
4069    <title>HTTP/1.1, part 5: Range Requests</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="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4075      <organization abbrev="W3C">World Wide Web Consortium</organization>
4076      <address><email>ylafon@w3.org</email></address>
4077    </author>
4078    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4079      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4080      <address><email>julian.reschke@greenbytes.de</email></address>
4081    </author>
4082    <date month="July" year="2012"/>
4083  </front>
4084  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-20"/>
4085 
4086</reference>
4087
4088<reference anchor="Part6">
4089  <front>
4090    <title>HTTP/1.1, part 6: Caching</title>
4091    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4092      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4093      <address><email>fielding@gbiv.com</email></address>
4094    </author>
4095    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4096      <organization abbrev="W3C">World Wide Web Consortium</organization>
4097      <address><email>ylafon@w3.org</email></address>
4098    </author>
4099    <author initials="M." surname="Nottingham" fullname="Mark Nottingham" role="editor">
4100      <organization>Rackspace</organization>
4101      <address><email>mnot@mnot.net</email></address>
4102    </author>
4103    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4104      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4105      <address><email>julian.reschke@greenbytes.de</email></address>
4106    </author>
4107    <date month="July" year="2012"/>
4108  </front>
4109  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-20"/>
4110 
4111</reference>
4112
4113<reference anchor="Part7">
4114  <front>
4115    <title>HTTP/1.1, part 7: Authentication</title>
4116    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4117      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4118      <address><email>fielding@gbiv.com</email></address>
4119    </author>
4120    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4121      <organization abbrev="W3C">World Wide Web Consortium</organization>
4122      <address><email>ylafon@w3.org</email></address>
4123    </author>
4124    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4125      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4126      <address><email>julian.reschke@greenbytes.de</email></address>
4127    </author>
4128    <date month="July" year="2012"/>
4129  </front>
4130  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p7-auth-20"/>
4131 
4132</reference>
4133
4134<reference anchor="RFC5234">
4135  <front>
4136    <title abbrev="ABNF for Syntax Specifications">Augmented BNF for Syntax Specifications: ABNF</title>
4137    <author initials="D." surname="Crocker" fullname="Dave Crocker" role="editor">
4138      <organization>Brandenburg InternetWorking</organization>
4139      <address>
4140        <email>dcrocker@bbiw.net</email>
4141      </address> 
4142    </author>
4143    <author initials="P." surname="Overell" fullname="Paul Overell">
4144      <organization>THUS plc.</organization>
4145      <address>
4146        <email>paul.overell@thus.net</email>
4147      </address>
4148    </author>
4149    <date month="January" year="2008"/>
4150  </front>
4151  <seriesInfo name="STD" value="68"/>
4152  <seriesInfo name="RFC" value="5234"/>
4153</reference>
4154
4155<reference anchor="RFC2119">
4156  <front>
4157    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
4158    <author initials="S." surname="Bradner" fullname="Scott Bradner">
4159      <organization>Harvard University</organization>
4160      <address><email>sob@harvard.edu</email></address>
4161    </author>
4162    <date month="March" year="1997"/>
4163  </front>
4164  <seriesInfo name="BCP" value="14"/>
4165  <seriesInfo name="RFC" value="2119"/>
4166</reference>
4167
4168<reference anchor="RFC3986">
4169 <front>
4170  <title abbrev="URI Generic Syntax">Uniform Resource Identifier (URI): Generic Syntax</title>
4171  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4172    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
4173    <address>
4174       <email>timbl@w3.org</email>
4175       <uri>http://www.w3.org/People/Berners-Lee/</uri>
4176    </address>
4177  </author>
4178  <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
4179    <organization abbrev="Day Software">Day Software</organization>
4180    <address>
4181      <email>fielding@gbiv.com</email>
4182      <uri>http://roy.gbiv.com/</uri>
4183    </address>
4184  </author>
4185  <author initials="L." surname="Masinter" fullname="Larry Masinter">
4186    <organization abbrev="Adobe Systems">Adobe Systems Incorporated</organization>
4187    <address>
4188      <email>LMM@acm.org</email>
4189      <uri>http://larry.masinter.net/</uri>
4190    </address>
4191  </author>
4192  <date month="January" year="2005"/>
4193 </front>
4194 <seriesInfo name="STD" value="66"/>
4195 <seriesInfo name="RFC" value="3986"/>
4196</reference>
4197
4198<reference anchor="USASCII">
4199  <front>
4200    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
4201    <author>
4202      <organization>American National Standards Institute</organization>
4203    </author>
4204    <date year="1986"/>
4205  </front>
4206  <seriesInfo name="ANSI" value="X3.4"/>
4207</reference>
4208
4209<reference anchor="RFC1950">
4210  <front>
4211    <title>ZLIB Compressed Data Format Specification version 3.3</title>
4212    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4213      <organization>Aladdin Enterprises</organization>
4214      <address><email>ghost@aladdin.com</email></address>
4215    </author>
4216    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly"/>
4217    <date month="May" year="1996"/>
4218  </front>
4219  <seriesInfo name="RFC" value="1950"/>
4220  <!--<annotation>
4221    RFC 1950 is an Informational RFC, thus it might be less stable than
4222    this specification. On the other hand, this downward reference was
4223    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4224    therefore it is unlikely to cause problems in practice. See also
4225    <xref target="BCP97"/>.
4226  </annotation>-->
4227</reference>
4228
4229<reference anchor="RFC1951">
4230  <front>
4231    <title>DEFLATE Compressed Data Format Specification version 1.3</title>
4232    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4233      <organization>Aladdin Enterprises</organization>
4234      <address><email>ghost@aladdin.com</email></address>
4235    </author>
4236    <date month="May" year="1996"/>
4237  </front>
4238  <seriesInfo name="RFC" value="1951"/>
4239  <!--<annotation>
4240    RFC 1951 is an Informational RFC, thus it might be less stable than
4241    this specification. On the other hand, this downward reference was
4242    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4243    therefore it is unlikely to cause problems in practice. See also
4244    <xref target="BCP97"/>.
4245  </annotation>-->
4246</reference>
4247
4248<reference anchor="RFC1952">
4249  <front>
4250    <title>GZIP file format specification version 4.3</title>
4251    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4252      <organization>Aladdin Enterprises</organization>
4253      <address><email>ghost@aladdin.com</email></address>
4254    </author>
4255    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
4256      <address><email>gzip@prep.ai.mit.edu</email></address>
4257    </author>
4258    <author initials="M." surname="Adler" fullname="Mark Adler">
4259      <address><email>madler@alumni.caltech.edu</email></address>
4260    </author>
4261    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4262      <address><email>ghost@aladdin.com</email></address>
4263    </author>
4264    <author initials="G." surname="Randers-Pehrson" fullname="Glenn Randers-Pehrson">
4265      <address><email>randeg@alumni.rpi.edu</email></address>
4266    </author>
4267    <date month="May" year="1996"/>
4268  </front>
4269  <seriesInfo name="RFC" value="1952"/>
4270  <!--<annotation>
4271    RFC 1952 is an Informational RFC, thus it might be less stable than
4272    this specification. On the other hand, this downward reference was
4273    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4274    therefore it is unlikely to cause problems in practice. See also
4275    <xref target="BCP97"/>.
4276  </annotation>-->
4277</reference>
4278
4279</references>
4280
4281<references title="Informative References">
4282
4283<reference anchor="ISO-8859-1">
4284  <front>
4285    <title>
4286     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
4287    </title>
4288    <author>
4289      <organization>International Organization for Standardization</organization>
4290    </author>
4291    <date year="1998"/>
4292  </front>
4293  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
4294</reference>
4295
4296<reference anchor="Nie1997" target="http://doi.acm.org/10.1145/263105.263157">
4297  <front>
4298    <title>Network Performance Effects of HTTP/1.1, CSS1, and PNG</title>
4299    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen"/>
4300    <author initials="J." surname="Gettys" fullname="J. Gettys"/>
4301    <author initials="E." surname="Prud'hommeaux" fullname="E. Prud'hommeaux"/>
4302    <author initials="H." surname="Lie" fullname="H. Lie"/>
4303    <author initials="C." surname="Lilley" fullname="C. Lilley"/>
4304    <date year="1997" month="September"/>
4305  </front>
4306  <seriesInfo name="ACM" value="Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication SIGCOMM '97"/>
4307</reference>
4308
4309<reference anchor="Pad1995" target="http://portal.acm.org/citation.cfm?id=219094">
4310  <front>
4311    <title>Improving HTTP Latency</title>
4312    <author initials="V.N." surname="Padmanabhan" fullname="Venkata N. Padmanabhan"/>
4313    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul"/>
4314    <date year="1995" month="December"/>
4315  </front>
4316  <seriesInfo name="Computer Networks and ISDN Systems" value="v. 28, pp. 25-35"/>
4317</reference>
4318
4319<reference anchor="RFC1919">
4320  <front>
4321    <title>Classical versus Transparent IP Proxies</title>
4322    <author initials="M." surname="Chatel" fullname="Marc Chatel">
4323      <address><email>mchatel@pax.eunet.ch</email></address>
4324    </author>
4325    <date year="1996" month="March"/>
4326  </front>
4327  <seriesInfo name="RFC" value="1919"/>
4328</reference>
4329
4330<reference anchor="RFC1945">
4331  <front>
4332    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
4333    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4334      <organization>MIT, Laboratory for Computer Science</organization>
4335      <address><email>timbl@w3.org</email></address>
4336    </author>
4337    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4338      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4339      <address><email>fielding@ics.uci.edu</email></address>
4340    </author>
4341    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4342      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
4343      <address><email>frystyk@w3.org</email></address>
4344    </author>
4345    <date month="May" year="1996"/>
4346  </front>
4347  <seriesInfo name="RFC" value="1945"/>
4348</reference>
4349
4350<reference anchor="RFC2045">
4351  <front>
4352    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
4353    <author initials="N." surname="Freed" fullname="Ned Freed">
4354      <organization>Innosoft International, Inc.</organization>
4355      <address><email>ned@innosoft.com</email></address>
4356    </author>
4357    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
4358      <organization>First Virtual Holdings</organization>
4359      <address><email>nsb@nsb.fv.com</email></address>
4360    </author>
4361    <date month="November" year="1996"/>
4362  </front>
4363  <seriesInfo name="RFC" value="2045"/>
4364</reference>
4365
4366<reference anchor="RFC2047">
4367  <front>
4368    <title abbrev="Message Header Extensions">MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text</title>
4369    <author initials="K." surname="Moore" fullname="Keith Moore">
4370      <organization>University of Tennessee</organization>
4371      <address><email>moore@cs.utk.edu</email></address>
4372    </author>
4373    <date month="November" year="1996"/>
4374  </front>
4375  <seriesInfo name="RFC" value="2047"/>
4376</reference>
4377
4378<reference anchor="RFC2068">
4379  <front>
4380    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4381    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
4382      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4383      <address><email>fielding@ics.uci.edu</email></address>
4384    </author>
4385    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4386      <organization>MIT Laboratory for Computer Science</organization>
4387      <address><email>jg@w3.org</email></address>
4388    </author>
4389    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
4390      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
4391      <address><email>mogul@wrl.dec.com</email></address>
4392    </author>
4393    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4394      <organization>MIT Laboratory for Computer Science</organization>
4395      <address><email>frystyk@w3.org</email></address>
4396    </author>
4397    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4398      <organization>MIT Laboratory for Computer Science</organization>
4399      <address><email>timbl@w3.org</email></address>
4400    </author>
4401    <date month="January" year="1997"/>
4402  </front>
4403  <seriesInfo name="RFC" value="2068"/>
4404</reference>
4405
4406<reference anchor="RFC2145">
4407  <front>
4408    <title abbrev="HTTP Version Numbers">Use and Interpretation of HTTP Version Numbers</title>
4409    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
4410      <organization>Western Research Laboratory</organization>
4411      <address><email>mogul@wrl.dec.com</email></address>
4412    </author>
4413    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4414      <organization>Department of Information and Computer Science</organization>
4415      <address><email>fielding@ics.uci.edu</email></address>
4416    </author>
4417    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4418      <organization>MIT Laboratory for Computer Science</organization>
4419      <address><email>jg@w3.org</email></address>
4420    </author>
4421    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4422      <organization>W3 Consortium</organization>
4423      <address><email>frystyk@w3.org</email></address>
4424    </author>
4425    <date month="May" year="1997"/>
4426  </front>
4427  <seriesInfo name="RFC" value="2145"/>
4428</reference>
4429
4430<reference anchor="RFC2616">
4431  <front>
4432    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4433    <author initials="R." surname="Fielding" fullname="R. Fielding">
4434      <organization>University of California, Irvine</organization>
4435      <address><email>fielding@ics.uci.edu</email></address>
4436    </author>
4437    <author initials="J." surname="Gettys" fullname="J. Gettys">
4438      <organization>W3C</organization>
4439      <address><email>jg@w3.org</email></address>
4440    </author>
4441    <author initials="J." surname="Mogul" fullname="J. Mogul">
4442      <organization>Compaq Computer Corporation</organization>
4443      <address><email>mogul@wrl.dec.com</email></address>
4444    </author>
4445    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
4446      <organization>MIT Laboratory for Computer Science</organization>
4447      <address><email>frystyk@w3.org</email></address>
4448    </author>
4449    <author initials="L." surname="Masinter" fullname="L. Masinter">
4450      <organization>Xerox Corporation</organization>
4451      <address><email>masinter@parc.xerox.com</email></address>
4452    </author>
4453    <author initials="P." surname="Leach" fullname="P. Leach">
4454      <organization>Microsoft Corporation</organization>
4455      <address><email>paulle@microsoft.com</email></address>
4456    </author>
4457    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
4458      <organization>W3C</organization>
4459      <address><email>timbl@w3.org</email></address>
4460    </author>
4461    <date month="June" year="1999"/>
4462  </front>
4463  <seriesInfo name="RFC" value="2616"/>
4464</reference>
4465
4466<reference anchor="RFC2817">
4467  <front>
4468    <title>Upgrading to TLS Within HTTP/1.1</title>
4469    <author initials="R." surname="Khare" fullname="R. Khare">
4470      <organization>4K Associates / UC Irvine</organization>
4471      <address><email>rohit@4K-associates.com</email></address>
4472    </author>
4473    <author initials="S." surname="Lawrence" fullname="S. Lawrence">
4474      <organization>Agranat Systems, Inc.</organization>
4475      <address><email>lawrence@agranat.com</email></address>
4476    </author>
4477    <date year="2000" month="May"/>
4478  </front>
4479  <seriesInfo name="RFC" value="2817"/>
4480</reference>
4481
4482<reference anchor="RFC2818">
4483  <front>
4484    <title>HTTP Over TLS</title>
4485    <author initials="E." surname="Rescorla" fullname="Eric Rescorla">
4486      <organization>RTFM, Inc.</organization>
4487      <address><email>ekr@rtfm.com</email></address>
4488    </author>
4489    <date year="2000" month="May"/>
4490  </front>
4491  <seriesInfo name="RFC" value="2818"/>
4492</reference>
4493
4494<reference anchor="RFC2965">
4495  <front>
4496    <title>HTTP State Management Mechanism</title>
4497    <author initials="D. M." surname="Kristol" fullname="David M. Kristol">
4498      <organization>Bell Laboratories, Lucent Technologies</organization>
4499      <address><email>dmk@bell-labs.com</email></address>
4500    </author>
4501    <author initials="L." surname="Montulli" fullname="Lou Montulli">
4502      <organization>Epinions.com, Inc.</organization>
4503      <address><email>lou@montulli.org</email></address>
4504    </author>
4505    <date year="2000" month="October"/>
4506  </front>
4507  <seriesInfo name="RFC" value="2965"/>
4508</reference>
4509
4510<reference anchor="RFC3040">
4511  <front>
4512    <title>Internet Web Replication and Caching Taxonomy</title>
4513    <author initials="I." surname="Cooper" fullname="I. Cooper">
4514      <organization>Equinix, Inc.</organization>
4515    </author>
4516    <author initials="I." surname="Melve" fullname="I. Melve">
4517      <organization>UNINETT</organization>
4518    </author>
4519    <author initials="G." surname="Tomlinson" fullname="G. Tomlinson">
4520      <organization>CacheFlow Inc.</organization>
4521    </author>
4522    <date year="2001" month="January"/>
4523  </front>
4524  <seriesInfo name="RFC" value="3040"/>
4525</reference>
4526
4527<reference anchor="RFC3864">
4528  <front>
4529    <title>Registration Procedures for Message Header Fields</title>
4530    <author initials="G." surname="Klyne" fullname="G. Klyne">
4531      <organization>Nine by Nine</organization>
4532      <address><email>GK-IETF@ninebynine.org</email></address>
4533    </author>
4534    <author initials="M." surname="Nottingham" fullname="M. Nottingham">
4535      <organization>BEA Systems</organization>
4536      <address><email>mnot@pobox.com</email></address>
4537    </author>
4538    <author initials="J." surname="Mogul" fullname="J. Mogul">
4539      <organization>HP Labs</organization>
4540      <address><email>JeffMogul@acm.org</email></address>
4541    </author>
4542    <date year="2004" month="September"/>
4543  </front>
4544  <seriesInfo name="BCP" value="90"/>
4545  <seriesInfo name="RFC" value="3864"/>
4546</reference>
4547
4548<reference anchor="RFC4033">
4549  <front>
4550    <title>DNS Security Introduction and Requirements</title>
4551    <author initials="R." surname="Arends" fullname="R. Arends"/>
4552    <author initials="R." surname="Austein" fullname="R. Austein"/>
4553    <author initials="M." surname="Larson" fullname="M. Larson"/>
4554    <author initials="D." surname="Massey" fullname="D. Massey"/>
4555    <author initials="S." surname="Rose" fullname="S. Rose"/>
4556    <date year="2005" month="March"/>
4557  </front>
4558  <seriesInfo name="RFC" value="4033"/>
4559</reference>
4560
4561<reference anchor="RFC4288">
4562  <front>
4563    <title>Media Type Specifications and Registration Procedures</title>
4564    <author initials="N." surname="Freed" fullname="N. Freed">
4565      <organization>Sun Microsystems</organization>
4566      <address>
4567        <email>ned.freed@mrochek.com</email>
4568      </address>
4569    </author>
4570    <author initials="J." surname="Klensin" fullname="J. Klensin">
4571      <address>
4572        <email>klensin+ietf@jck.com</email>
4573      </address>
4574    </author>
4575    <date year="2005" month="December"/>
4576  </front>
4577  <seriesInfo name="BCP" value="13"/>
4578  <seriesInfo name="RFC" value="4288"/>
4579</reference>
4580
4581<reference anchor="RFC4395">
4582  <front>
4583    <title>Guidelines and Registration Procedures for New URI Schemes</title>
4584    <author initials="T." surname="Hansen" fullname="T. Hansen">
4585      <organization>AT&amp;T Laboratories</organization>
4586      <address>
4587        <email>tony+urireg@maillennium.att.com</email>
4588      </address>
4589    </author>
4590    <author initials="T." surname="Hardie" fullname="T. Hardie">
4591      <organization>Qualcomm, Inc.</organization>
4592      <address>
4593        <email>hardie@qualcomm.com</email>
4594      </address>
4595    </author>
4596    <author initials="L." surname="Masinter" fullname="L. Masinter">
4597      <organization>Adobe Systems</organization>
4598      <address>
4599        <email>LMM@acm.org</email>
4600      </address>
4601    </author>
4602    <date year="2006" month="February"/>
4603  </front>
4604  <seriesInfo name="BCP" value="115"/>
4605  <seriesInfo name="RFC" value="4395"/>
4606</reference>
4607
4608<reference anchor="RFC4559">
4609  <front>
4610    <title>SPNEGO-based Kerberos and NTLM HTTP Authentication in Microsoft Windows</title>
4611    <author initials="K." surname="Jaganathan" fullname="K. Jaganathan"/>
4612    <author initials="L." surname="Zhu" fullname="L. Zhu"/>
4613    <author initials="J." surname="Brezak" fullname="J. Brezak"/>
4614    <date year="2006" month="June"/>
4615  </front>
4616  <seriesInfo name="RFC" value="4559"/>
4617</reference>
4618
4619<reference anchor="RFC5226">
4620  <front>
4621    <title>Guidelines for Writing an IANA Considerations Section in RFCs</title>
4622    <author initials="T." surname="Narten" fullname="T. Narten">
4623      <organization>IBM</organization>
4624      <address><email>narten@us.ibm.com</email></address>
4625    </author>
4626    <author initials="H." surname="Alvestrand" fullname="H. Alvestrand">
4627      <organization>Google</organization>
4628      <address><email>Harald@Alvestrand.no</email></address>
4629    </author>
4630    <date year="2008" month="May"/>
4631  </front>
4632  <seriesInfo name="BCP" value="26"/>
4633  <seriesInfo name="RFC" value="5226"/>
4634</reference>
4635
4636<reference anchor="RFC5322">
4637  <front>
4638    <title>Internet Message Format</title>
4639    <author initials="P." surname="Resnick" fullname="P. Resnick">
4640      <organization>Qualcomm Incorporated</organization>
4641    </author>
4642    <date year="2008" month="October"/>
4643  </front>
4644  <seriesInfo name="RFC" value="5322"/>
4645</reference>
4646
4647<reference anchor="RFC6265">
4648  <front>
4649    <title>HTTP State Management Mechanism</title>
4650    <author initials="A." surname="Barth" fullname="Adam Barth">
4651      <organization abbrev="U.C. Berkeley">
4652        University of California, Berkeley
4653      </organization>
4654      <address><email>abarth@eecs.berkeley.edu</email></address>
4655    </author>
4656    <date year="2011" month="April"/>
4657  </front>
4658  <seriesInfo name="RFC" value="6265"/>
4659</reference>
4660
4661<!--<reference anchor='BCP97'>
4662  <front>
4663    <title>Handling Normative References to Standards-Track Documents</title>
4664    <author initials='J.' surname='Klensin' fullname='J. Klensin'>
4665      <address>
4666        <email>klensin+ietf@jck.com</email>
4667      </address>
4668    </author>
4669    <author initials='S.' surname='Hartman' fullname='S. Hartman'>
4670      <organization>MIT</organization>
4671      <address>
4672        <email>hartmans-ietf@mit.edu</email>
4673      </address>
4674    </author>
4675    <date year='2007' month='June' />
4676  </front>
4677  <seriesInfo name='BCP' value='97' />
4678  <seriesInfo name='RFC' value='4897' />
4679</reference>-->
4680
4681<reference anchor="Kri2001" target="http://arxiv.org/abs/cs.SE/0105018">
4682  <front>
4683    <title>HTTP Cookies: Standards, Privacy, and Politics</title>
4684    <author initials="D." surname="Kristol" fullname="David M. Kristol"/>
4685    <date year="2001" month="November"/>
4686  </front>
4687  <seriesInfo name="ACM Transactions on Internet Technology" value="Vol. 1, #2"/>
4688</reference>
4689
4690<reference anchor="Spe" target="http://sunsite.unc.edu/mdma-release/http-prob.html">
4691  <front>
4692    <title>Analysis of HTTP Performance Problems</title>
4693    <author initials="S." surname="Spero" fullname="Simon E. Spero"/>
4694    <date/>
4695  </front>
4696</reference>
4697
4698<reference anchor="Tou1998" target="http://www.isi.edu/touch/pubs/http-perf96/">
4699  <front>
4700  <title>Analysis of HTTP Performance</title>
4701  <author initials="J." surname="Touch" fullname="Joe Touch">
4702    <organization>USC/Information Sciences Institute</organization>
4703    <address><email>touch@isi.edu</email></address>
4704  </author>
4705  <author initials="J." surname="Heidemann" fullname="John Heidemann">
4706    <organization>USC/Information Sciences Institute</organization>
4707    <address><email>johnh@isi.edu</email></address>
4708  </author>
4709  <author initials="K." surname="Obraczka" fullname="Katia Obraczka">
4710    <organization>USC/Information Sciences Institute</organization>
4711    <address><email>katia@isi.edu</email></address>
4712  </author>
4713  <date year="1998" month="Aug"/>
4714  </front>
4715  <seriesInfo name="ISI Research Report" value="ISI/RR-98-463"/>
4716  <annotation>(original report dated Aug. 1996)</annotation>
4717</reference>
4718
4719</references>
4720
4721
4722<section title="HTTP Version History" anchor="compatibility">
4723<t>
4724   HTTP has been in use by the World-Wide Web global information initiative
4725   since 1990. The first version of HTTP, later referred to as HTTP/0.9,
4726   was a simple protocol for hypertext data transfer across the Internet
4727   with only a single request method (GET) and no metadata.
4728   HTTP/1.0, as defined by <xref target="RFC1945"/>, added a range of request
4729   methods and MIME-like messaging that could include metadata about the data
4730   transferred and modifiers on the request/response semantics. However,
4731   HTTP/1.0 did not sufficiently take into consideration the effects of
4732   hierarchical proxies, caching, the need for persistent connections, or
4733   name-based virtual hosts. The proliferation of incompletely-implemented
4734   applications calling themselves "HTTP/1.0" further necessitated a
4735   protocol version change in order for two communicating applications
4736   to determine each other's true capabilities.
4737</t>
4738<t>
4739   HTTP/1.1 remains compatible with HTTP/1.0 by including more stringent
4740   requirements that enable reliable implementations, adding only
4741   those new features that will either be safely ignored by an HTTP/1.0
4742   recipient or only sent when communicating with a party advertising
4743   conformance with HTTP/1.1.
4744</t>
4745<t>
4746   It is beyond the scope of a protocol specification to mandate
4747   conformance with previous versions. HTTP/1.1 was deliberately
4748   designed, however, to make supporting previous versions easy.
4749   We would expect a general-purpose HTTP/1.1 server to understand
4750   any valid request in the format of HTTP/1.0 and respond appropriately
4751   with an HTTP/1.1 message that only uses features understood (or
4752   safely ignored) by HTTP/1.0 clients.  Likewise, we would expect
4753   an HTTP/1.1 client to understand any valid HTTP/1.0 response.
4754</t>
4755<t>
4756   Since HTTP/0.9 did not support header fields in a request,
4757   there is no mechanism for it to support name-based virtual
4758   hosts (selection of resource by inspection of the <xref target="header.host" format="none">Host</xref> header
4759   field).  Any server that implements name-based virtual hosts
4760   ought to disable support for HTTP/0.9.  Most requests that
4761   appear to be HTTP/0.9 are, in fact, badly constructed HTTP/1.x
4762   requests wherein a buggy client failed to properly encode
4763   linear whitespace found in a URI reference and placed in
4764   the request-target.
4765</t>
4766
4767<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
4768<t>
4769   This section summarizes major differences between versions HTTP/1.0
4770   and HTTP/1.1.
4771</t>
4772
4773<section title="Multi-homed Web Servers" anchor="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses">
4774<t>
4775   The requirements that clients and servers support the <xref target="header.host" format="none">Host</xref>
4776   header field (<xref target="header.host"/>), report an error if it is
4777   missing from an HTTP/1.1 request, and accept absolute URIs (<xref target="request-target"/>)
4778   are among the most important changes defined by HTTP/1.1.
4779</t>
4780<t>
4781   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
4782   addresses and servers; there was no other established mechanism for
4783   distinguishing the intended server of a request than the IP address
4784   to which that request was directed. The <xref target="header.host" format="none">Host</xref> header field was
4785   introduced during the development of HTTP/1.1 and, though it was
4786   quickly implemented by most HTTP/1.0 browsers, additional requirements
4787   were placed on all HTTP/1.1 requests in order to ensure complete
4788   adoption.  At the time of this writing, most HTTP-based services
4789   are dependent upon the Host header field for targeting requests.
4790</t>
4791</section>
4792
4793<section title="Keep-Alive Connections" anchor="compatibility.with.http.1.0.persistent.connections">
4794<t>
4795   In HTTP/1.0, each connection is established by the client prior to the
4796   request and closed by the server after sending the response. However, some
4797   implementations implement the explicitly negotiated ("Keep-Alive") version
4798   of persistent connections described in Section 19.7.1 of <xref target="RFC2068"/>.
4799</t>
4800<t>
4801   Some clients and servers might wish to be compatible with these previous
4802   approaches to persistent connections, by explicitly negotiating for them
4803   with a "Connection: keep-alive" request header field. However, some
4804   experimental implementations of HTTP/1.0 persistent connections are faulty;
4805   for example, if a HTTP/1.0 proxy server doesn't understand
4806   <xref target="header.connection" format="none">Connection</xref>, it will erroneously forward that header field
4807   to the next inbound server, which would result in a hung connection.
4808</t>
4809<t>
4810   One attempted solution was the introduction of a Proxy-Connection header
4811   field, targeted specifically at proxies. In practice, this was also
4812   unworkable, because proxies are often deployed in multiple layers, bringing
4813   about the same problem discussed above.
4814</t>
4815<t>
4816   As a result, clients are encouraged not to send the Proxy-Connection header
4817   field in any requests.
4818</t>
4819<t>
4820   Clients are also encouraged to consider the use of Connection: keep-alive
4821   in requests carefully; while they can enable persistent connections with
4822   HTTP/1.0 servers, clients using them need will need to monitor the
4823   connection for "hung" requests (which indicate that the client ought stop
4824   sending the header field), and this mechanism ought not be used by clients
4825   at all when a proxy is being used.
4826</t>
4827</section>
4828
4829<section title="Introduction of Transfer-Encoding" anchor="introduction.of.transfer-encoding">
4830<t>
4831   HTTP/1.1 introduces the <xref target="header.transfer-encoding" format="none">Transfer-Encoding</xref> header field
4832   (<xref target="header.transfer-encoding"/>). Proxies/gateways MUST remove
4833   any transfer-coding prior to forwarding a message via a MIME-compliant
4834   protocol.
4835</t>
4836</section>
4837
4838</section>
4839
4840<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
4841<t>
4842  Clarify that the string "HTTP" in the HTTP-version ABNF production is case
4843  sensitive. Restrict the version numbers to be single digits due to the fact
4844  that implementations are known to handle multi-digit version numbers
4845  incorrectly.
4846  (<xref target="http.version"/>)
4847</t>
4848<t>
4849  Update use of abs_path production from RFC 1808 to the path-absolute + query
4850  components of RFC 3986. State that the asterisk form is allowed for the OPTIONS
4851  request method only.
4852  (<xref target="request-target"/>)
4853</t>
4854<t>
4855  Require that invalid whitespace around field-names be rejected.
4856  (<xref target="header.fields"/>)
4857</t>
4858<t>
4859  Rules about implicit linear whitespace between certain grammar productions
4860  have been removed; now whitespace is only allowed where specifically
4861  defined in the ABNF.
4862  (<xref target="whitespace"/>)
4863</t>
4864<t> 
4865  The NUL octet is no longer allowed in comment and quoted-string
4866  text. The quoted-pair rule no longer allows escaping control characters other than HTAB.
4867  Non-ASCII content in header fields and reason phrase has been obsoleted and
4868  made opaque (the TEXT rule was removed).
4869  (<xref target="field.components"/>)
4870</t>
4871<t>
4872  Empty list elements in list productions have been deprecated.
4873  (<xref target="abnf.extension"/>)
4874</t>
4875<t>
4876  Require recipients to handle bogus <xref target="header.content-length" format="none">Content-Length</xref> header
4877  fields as errors.
4878  (<xref target="message.body"/>)
4879</t>
4880<t>
4881  Remove reference to non-existent identity transfer-coding value tokens.
4882  (Sections <xref format="counter" target="message.body"/> and
4883  <xref format="counter" target="transfer.codings"/>)
4884</t>
4885<t>
4886  Clarification that the chunk length does not include the count of the octets
4887  in the chunk header and trailer. Furthermore disallowed line folding
4888  in chunk extensions, and deprecate their use.
4889  (<xref target="chunked.encoding"/>)
4890</t>
4891<t>
4892  Registration of Transfer Codings now requires IETF Review
4893  (<xref target="transfer.coding.registry"/>)
4894</t>
4895<t>
4896  Remove hard limit of two connections per server.
4897  Remove requirement to retry a sequence of requests as long it was idempotent.
4898  Remove requirements about when servers are allowed to close connections
4899  prematurely.
4900  (<xref target="persistent.practical"/>)
4901</t>
4902<t>
4903  Remove requirement to retry requests under certain circumstances when the
4904  server prematurely closes the connection.
4905  (<xref target="message.transmission.requirements"/>)
4906</t>
4907<t>
4908  Change ABNF productions for header fields to only define the field value.
4909</t>
4910<t>
4911  Clarify exactly when "close" connection options have to be sent.
4912  (<xref target="header.connection"/>)
4913</t>
4914<t>
4915  Define the semantics of the <xref target="header.upgrade" format="none">Upgrade</xref> header field in responses
4916  other than 101 (this was incorporated from <xref target="RFC2817"/>).
4917  (<xref target="header.upgrade"/>)
4918</t>
4919<t>
4920  Take over the Upgrade Token Registry, previously defined in
4921  Section 7.2 of <xref target="RFC2817"/>.
4922  (<xref target="upgrade.token.registry"/>)
4923</t>
4924</section>
4925</section>
4926
4927<section title="ABNF list extension: #rule" anchor="abnf.extension">
4928<t>
4929  A #rule extension to the ABNF rules of <xref target="RFC5234"/> is used to
4930  improve readability in the definitions of some header field values.
4931</t>
4932<t>
4933  A construct "#" is defined, similar to "*", for defining comma-delimited
4934  lists of elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating
4935  at least &lt;n&gt; and at most &lt;m&gt; elements, each separated by a single
4936  comma (",") and optional whitespace (OWS).   
4937</t>
4938<figure><preamble>
4939  Thus,
4940</preamble><artwork type="example"><![CDATA[
4941  1#element => element *( OWS "," OWS element )
4942]]></artwork></figure>
4943<figure><preamble>
4944  and:
4945</preamble><artwork type="example"><![CDATA[
4946  #element => [ 1#element ]
4947]]></artwork></figure>
4948<figure><preamble>
4949  and for n &gt;= 1 and m &gt; 1:
4950</preamble><artwork type="example"><![CDATA[
4951  <n>#<m>element => element <n-1>*<m-1>( OWS "," OWS element )
4952]]></artwork></figure>
4953<t>
4954  For compatibility with legacy list rules, recipients SHOULD accept empty
4955  list elements. In other words, consumers would follow the list productions:
4956</t>
4957<figure><artwork type="example"><![CDATA[
4958  #element => [ ( "," / element ) *( OWS "," [ OWS element ] ) ]
4959 
4960  1#element => *( "," OWS ) element *( OWS "," [ OWS element ] )
4961]]></artwork></figure>
4962<t>
4963  Note that empty elements do not contribute to the count of elements present,
4964  though.
4965</t>
4966<t>
4967  For example, given these ABNF productions:
4968</t>
4969<figure><artwork type="example"><![CDATA[
4970  example-list      = 1#example-list-elmt
4971  example-list-elmt = token ; see Section 3.2.4
4972]]></artwork></figure>
4973<t>
4974  Then these are valid values for example-list (not including the double
4975  quotes, which are present for delimitation only):
4976</t>
4977<figure><artwork type="example"><![CDATA[
4978  "foo,bar"
4979  "foo ,bar,"
4980  "foo , ,bar,charlie   "
4981]]></artwork></figure>
4982<t>
4983  But these values would be invalid, as at least one non-empty element is
4984  required:
4985</t>
4986<figure><artwork type="example"><![CDATA[
4987  ""
4988  ","
4989  ",   ,"
4990]]></artwork></figure>
4991<t>
4992  <xref target="collected.abnf"/> shows the collected ABNF, with the list rules
4993  expanded as explained above.
4994</t>
4995</section>
4996
4997
4998<section title="Collected ABNF" anchor="collected.abnf">
4999<figure>
5000<artwork type="abnf" name="p1-messaging.parsed-abnf"><![CDATA[
5001BWS = OWS
5002
5003Connection = *( "," OWS ) connection-option *( OWS "," [ OWS
5004 connection-option ] )
5005Content-Length = 1*DIGIT
5006
5007HTTP-message = start-line *( header-field CRLF ) CRLF [ message-body
5008 ]
5009HTTP-name = %x48.54.54.50 ; HTTP
5010HTTP-version = HTTP-name "/" DIGIT "." DIGIT
5011Host = uri-host [ ":" port ]
5012
5013OWS = *( SP / HTAB )
5014
5015RWS = 1*( SP / HTAB )
5016
5017TE = [ ( "," / t-codings ) *( OWS "," [ OWS t-codings ] ) ]
5018Trailer = *( "," OWS ) field-name *( OWS "," [ OWS field-name ] )
5019Transfer-Encoding = *( "," OWS ) transfer-coding *( OWS "," [ OWS
5020 transfer-coding ] )
5021
5022URI-reference = <URI-reference, defined in [RFC3986], Section 4.1>
5023Upgrade = *( "," OWS ) protocol *( OWS "," [ OWS protocol ] )
5024
5025Via = *( "," OWS ) ( received-protocol RWS received-by [ RWS comment
5026 ] ) *( OWS "," [ OWS ( received-protocol RWS received-by [ RWS
5027 comment ] ) ] )
5028
5029absolute-URI = <absolute-URI, defined in [RFC3986], Section 4.3>
5030absolute-form = absolute-URI
5031asterisk-form = "*"
5032attribute = token
5033authority = <authority, defined in [RFC3986], Section 3.2>
5034authority-form = authority
5035
5036chunk = chunk-size [ chunk-ext ] CRLF chunk-data CRLF
5037chunk-data = 1*OCTET
5038chunk-ext = *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
5039chunk-ext-name = token
5040chunk-ext-val = token / quoted-str-nf
5041chunk-size = 1*HEXDIG
5042chunked-body = *chunk last-chunk trailer-part CRLF
5043comment = "(" *( ctext / quoted-cpair / comment ) ")"
5044connection-option = token
5045ctext = OWS / %x21-27 ; '!'-'''
5046 / %x2A-5B ; '*'-'['
5047 / %x5D-7E ; ']'-'~'
5048 / obs-text
5049
5050field-content = *( HTAB / SP / VCHAR / obs-text )
5051field-name = token
5052field-value = *( field-content / obs-fold )
5053
5054header-field = field-name ":" OWS field-value BWS
5055http-URI = "http://" authority path-abempty [ "?" query ]
5056https-URI = "https://" authority path-abempty [ "?" query ]
5057
5058last-chunk = 1*"0" [ chunk-ext ] CRLF
5059
5060message-body = *OCTET
5061method = token
5062
5063obs-fold = CRLF ( SP / HTAB )
5064obs-text = %x80-FF
5065origin-form = path-absolute [ "?" query ]
5066
5067partial-URI = relative-part [ "?" query ]
5068path-abempty = <path-abempty, defined in [RFC3986], Section 3.3>
5069path-absolute = <path-absolute, defined in [RFC3986], Section 3.3>
5070port = <port, defined in [RFC3986], Section 3.2.3>
5071protocol = protocol-name [ "/" protocol-version ]
5072protocol-name = token
5073protocol-version = token
5074pseudonym = token
5075
5076qdtext = OWS / "!" / %x23-5B ; '#'-'['
5077 / %x5D-7E ; ']'-'~'
5078 / obs-text
5079qdtext-nf = HTAB / SP / "!" / %x23-5B ; '#'-'['
5080 / %x5D-7E ; ']'-'~'
5081 / obs-text
5082query = <query, defined in [RFC3986], Section 3.4>
5083quoted-cpair = "\" ( HTAB / SP / VCHAR / obs-text )
5084quoted-pair = "\" ( HTAB / SP / VCHAR / obs-text )
5085quoted-str-nf = DQUOTE *( qdtext-nf / quoted-pair ) DQUOTE
5086quoted-string = DQUOTE *( qdtext / quoted-pair ) DQUOTE
5087qvalue = ( "0" [ "." *3DIGIT ] ) / ( "1" [ "." *3"0" ] )
5088
5089reason-phrase = *( HTAB / SP / VCHAR / obs-text )
5090received-by = ( uri-host [ ":" port ] ) / pseudonym
5091received-protocol = [ protocol-name "/" ] protocol-version
5092relative-part = <relative-part, defined in [RFC3986], Section 4.2>
5093request-line = method SP request-target SP HTTP-version CRLF
5094request-target = origin-form / absolute-form / authority-form /
5095 asterisk-form
5096
5097special = "(" / ")" / "<" / ">" / "@" / "," / ";" / ":" / "\" /
5098 DQUOTE / "/" / "[" / "]" / "?" / "=" / "{" / "}"
5099start-line = request-line / status-line
5100status-code = 3DIGIT
5101status-line = HTTP-version SP status-code SP reason-phrase CRLF
5102
5103t-codings = "trailers" / ( transfer-extension [ te-params ] )
5104tchar = "!" / "#" / "$" / "%" / "&" / "'" / "*" / "+" / "-" / "." /
5105 "^" / "_" / "`" / "|" / "~" / DIGIT / ALPHA
5106te-ext = OWS ";" OWS token [ "=" word ]
5107te-params = OWS ";" OWS "q=" qvalue *te-ext
5108token = 1*tchar
5109trailer-part = *( header-field CRLF )
5110transfer-coding = "chunked" / "compress" / "deflate" / "gzip" /
5111 transfer-extension
5112transfer-extension = token *( OWS ";" OWS transfer-parameter )
5113transfer-parameter = attribute BWS "=" BWS value
5114
5115uri-host = <host, defined in [RFC3986], Section 3.2.2>
5116
5117value = word
5118
5119word = token / quoted-string
5120]]></artwork>
5121</figure>
5122</section>
5123
5124
5125<section title="Change Log (to be removed by RFC Editor before publication)" anchor="change.log">
5126
5127<section title="Since RFC 2616">
5128<t>
5129  Extracted relevant partitions from <xref target="RFC2616"/>.
5130</t>
5131</section>
5132
5133<section title="Since draft-ietf-httpbis-p1-messaging-00">
5134<t>
5135  Closed issues:
5136  <list style="symbols">
5137    <t>
5138      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/1"/>:
5139      "HTTP Version should be case sensitive"
5140      (<eref target="http://purl.org/NET/http-errata#verscase"/>)
5141    </t>
5142    <t>
5143      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/2"/>:
5144      "'unsafe' characters"
5145      (<eref target="http://purl.org/NET/http-errata#unsafe-uri"/>)
5146    </t>
5147    <t>
5148      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/3"/>:
5149      "Chunk Size Definition"
5150      (<eref target="http://purl.org/NET/http-errata#chunk-size"/>)
5151    </t>
5152    <t>
5153      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/4"/>:
5154      "Message Length"
5155      (<eref target="http://purl.org/NET/http-errata#msg-len-chars"/>)
5156    </t>
5157    <t>
5158      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/8"/>:
5159      "Media Type Registrations"
5160      (<eref target="http://purl.org/NET/http-errata#media-reg"/>)
5161    </t>
5162    <t>
5163      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/11"/>:
5164      "URI includes query"
5165      (<eref target="http://purl.org/NET/http-errata#uriquery"/>)
5166    </t>
5167    <t>
5168      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/15"/>:
5169      "No close on 1xx responses"
5170      (<eref target="http://purl.org/NET/http-errata#noclose1xx"/>)
5171    </t>
5172    <t>
5173      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/16"/>:
5174      "Remove 'identity' token references"
5175      (<eref target="http://purl.org/NET/http-errata#identity"/>)
5176    </t>
5177    <t>
5178      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/26"/>:
5179      "Import query BNF"
5180    </t>
5181    <t>
5182      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/31"/>:
5183      "qdtext BNF"
5184    </t>
5185    <t>
5186      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/35"/>:
5187      "Normative and Informative references"
5188    </t>
5189    <t>
5190      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/42"/>:
5191      "RFC2606 Compliance"
5192    </t>
5193    <t>
5194      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/45"/>:
5195      "RFC977 reference"
5196    </t>
5197    <t>
5198      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/46"/>:
5199      "RFC1700 references"
5200    </t>
5201    <t>
5202      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/47"/>:
5203      "inconsistency in date format explanation"
5204    </t>
5205    <t>
5206      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/48"/>:
5207      "Date reference typo"
5208    </t>
5209    <t>
5210      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/65"/>:
5211      "Informative references"
5212    </t>
5213    <t>
5214      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/66"/>:
5215      "ISO-8859-1 Reference"
5216    </t>
5217    <t>
5218      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/86"/>:
5219      "Normative up-to-date references"
5220    </t>
5221  </list>
5222</t>
5223<t>
5224  Other changes:
5225  <list style="symbols">
5226    <t>
5227      Update media type registrations to use RFC4288 template.
5228    </t>
5229    <t>
5230      Use names of RFC4234 core rules DQUOTE and HTAB,
5231      fix broken ABNF for chunk-data
5232      (work in progress on <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/36"/>)
5233    </t>
5234  </list>
5235</t>
5236</section>
5237
5238<section title="Since draft-ietf-httpbis-p1-messaging-01">
5239<t>
5240  Closed issues:
5241  <list style="symbols">
5242    <t>
5243      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/19"/>:
5244      "Bodies on GET (and other) requests"
5245    </t>
5246    <t>
5247      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/55"/>:
5248      "Updating to RFC4288"
5249    </t>
5250    <t>
5251      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/57"/>:
5252      "Status Code and Reason Phrase"
5253    </t>
5254    <t>
5255      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/82"/>:
5256      "rel_path not used"
5257    </t>
5258  </list>
5259</t>
5260<t>
5261  Ongoing work on ABNF conversion (<eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/36"/>):
5262  <list style="symbols">
5263    <t>
5264      Get rid of duplicate BNF rule names ("host" -&gt; "uri-host", "trailer" -&gt;
5265      "trailer-part").
5266    </t>
5267    <t>
5268      Avoid underscore character in rule names ("http_URL" -&gt;
5269      "http-URL", "abs_path" -&gt; "path-absolute").
5270    </t>
5271    <t>
5272      Add rules for terms imported from URI spec ("absoluteURI", "authority",
5273      "path-absolute", "port", "query", "relativeURI", "host) — these will
5274      have to be updated when switching over to RFC3986.
5275    </t>
5276    <t>
5277      Synchronize core rules with RFC5234.
5278    </t>
5279    <t>
5280      Get rid of prose rules that span multiple lines.
5281    </t>
5282    <t>
5283      Get rid of unused rules LOALPHA and UPALPHA.
5284    </t>
5285    <t>
5286      Move "Product Tokens" section (back) into Part 1, as "token" is used
5287      in the definition of the Upgrade header field.
5288    </t>
5289    <t>
5290      Add explicit references to BNF syntax and rules imported from other parts of the specification.
5291    </t>
5292    <t>
5293      Rewrite prose rule "token" in terms of "tchar", rewrite prose rule "TEXT".
5294    </t>
5295  </list>
5296</t>
5297</section>
5298
5299<section title="Since draft-ietf-httpbis-p1-messaging-02" anchor="changes.since.02">
5300<t>
5301  Closed issues:
5302  <list style="symbols">
5303    <t>
5304      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/51"/>:
5305      "HTTP-date vs. rfc1123-date"
5306    </t>
5307    <t>
5308      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/64"/>:
5309      "WS in quoted-pair"
5310    </t>
5311  </list>
5312</t>
5313<t>
5314  Ongoing work on IANA Message Header Field Registration (<eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/40"/>):
5315  <list style="symbols">
5316    <t>
5317      Reference RFC 3984, and update header field registrations for header
5318      fields defined in this document.
5319    </t>
5320  </list>
5321</t>
5322<t>
5323  Ongoing work on ABNF conversion (<eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/36"/>):
5324  <list style="symbols">
5325    <t>
5326      Replace string literals when the string really is case-sensitive (HTTP-version).
5327    </t>
5328  </list>
5329</t>
5330</section>
5331
5332<section title="Since draft-ietf-httpbis-p1-messaging-03" anchor="changes.since.03">
5333<t>
5334  Closed issues:
5335  <list style="symbols">
5336    <t>
5337      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/28"/>:
5338      "Connection closing"
5339    </t>
5340    <t>
5341      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/97"/>:
5342      "Move registrations and registry information to IANA Considerations"
5343    </t>
5344    <t>
5345      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/120"/>:
5346      "need new URL for PAD1995 reference"
5347    </t>
5348    <t>
5349      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/127"/>:
5350      "IANA Considerations: update HTTP URI scheme registration"
5351    </t>
5352    <t>
5353      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/128"/>:
5354      "Cite HTTPS URI scheme definition"
5355    </t>
5356    <t>
5357      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/129"/>:
5358      "List-type header fields vs Set-Cookie"
5359    </t>
5360  </list>
5361</t>
5362<t>
5363  Ongoing work on ABNF conversion (<eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/36"/>):
5364  <list style="symbols">
5365    <t>
5366      Replace string literals when the string really is case-sensitive (HTTP-Date).
5367    </t>
5368    <t>
5369      Replace HEX by HEXDIG for future consistence with RFC 5234's core rules.
5370    </t>
5371  </list>
5372</t>
5373</section>
5374
5375<section title="Since draft-ietf-httpbis-p1-messaging-04" anchor="changes.since.04">
5376<t>
5377  Closed issues:
5378  <list style="symbols">
5379    <t>
5380      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/34"/>:
5381      "Out-of-date reference for URIs"
5382    </t>
5383    <t>
5384      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/132"/>:
5385      "RFC 2822 is updated by RFC 5322"
5386    </t>
5387  </list>
5388</t>
5389<t>
5390  Ongoing work on ABNF conversion (<eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/36"/>):
5391  <list style="symbols">
5392    <t>
5393      Use "/" instead of "|" for alternatives.
5394    </t>
5395    <t>
5396      Get rid of RFC822 dependency; use RFC5234 plus extensions instead.
5397    </t>
5398    <t>
5399      Only reference RFC 5234's core rules.
5400    </t>
5401    <t>
5402      Introduce new ABNF rules for "bad" whitespace ("BWS"), optional
5403      whitespace ("OWS") and required whitespace ("RWS").
5404    </t>
5405    <t>
5406      Rewrite ABNFs to spell out whitespace rules, factor out
5407      header field value format definitions.
5408    </t>
5409  </list>
5410</t>
5411</section>
5412
5413<section title="Since draft-ietf-httpbis-p1-messaging-05" anchor="changes.since.05">
5414<t>
5415  Closed issues:
5416  <list style="symbols">
5417    <t>
5418      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/30"/>:
5419      "Header LWS"
5420    </t>
5421    <t>
5422      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/52"/>:
5423      "Sort 1.3 Terminology"
5424    </t>
5425    <t>
5426      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/63"/>:
5427      "RFC2047 encoded words"
5428    </t>
5429    <t>
5430      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/74"/>:
5431      "Character Encodings in TEXT"
5432    </t>
5433    <t>
5434      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/77"/>:
5435      "Line Folding"
5436    </t>
5437    <t>
5438      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/83"/>:
5439      "OPTIONS * and proxies"
5440    </t>
5441    <t>
5442      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/94"/>:
5443      "reason-phrase BNF"
5444    </t>
5445    <t>
5446      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/111"/>:
5447      "Use of TEXT"
5448    </t>
5449    <t>
5450      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/118"/>:
5451      "Join "Differences Between HTTP Entities and RFC 2045 Entities"?"
5452    </t>
5453    <t>
5454      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/134"/>:
5455      "RFC822 reference left in discussion of date formats"
5456    </t>
5457  </list>
5458</t>
5459<t>
5460  Final work on ABNF conversion (<eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/36"/>):
5461  <list style="symbols">
5462    <t>
5463      Rewrite definition of list rules, deprecate empty list elements.
5464    </t>
5465    <t>
5466      Add appendix containing collected and expanded ABNF.
5467    </t>
5468  </list>
5469</t>
5470<t>
5471  Other changes:
5472  <list style="symbols">
5473    <t>
5474      Rewrite introduction; add mostly new Architecture Section.
5475    </t>
5476    <t>
5477      Move definition of quality values from Part 3 into Part 1;
5478      make TE request header field grammar independent of accept-params (defined in Part 3).
5479    </t>
5480  </list>
5481</t>
5482</section>
5483
5484<section title="Since draft-ietf-httpbis-p1-messaging-06" anchor="changes.since.06">
5485<t>
5486  Closed issues:
5487  <list style="symbols">
5488    <t>
5489      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/161"/>:
5490      "base for numeric protocol elements"
5491    </t>
5492    <t>
5493      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/162"/>:
5494      "comment ABNF"
5495    </t>
5496  </list>
5497</t>
5498<t>
5499  Partly resolved issues:
5500  <list style="symbols">
5501    <t>
5502      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/88"/>:
5503      "205 Bodies" (took out language that implied that there might be
5504      methods for which a request body MUST NOT be included)
5505    </t>
5506    <t>
5507      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/163"/>:
5508      "editorial improvements around HTTP-date"
5509    </t>
5510  </list>
5511</t>
5512</section>
5513
5514<section title="Since draft-ietf-httpbis-p1-messaging-07" anchor="changes.since.07">
5515<t>
5516  Closed issues:
5517  <list style="symbols">
5518    <t>
5519      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/93"/>:
5520      "Repeating single-value header fields"
5521    </t>
5522    <t>
5523      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/131"/>:
5524      "increase connection limit"
5525    </t>
5526    <t>
5527      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/157"/>:
5528      "IP addresses in URLs"
5529    </t>
5530    <t>
5531      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/172"/>:
5532      "take over HTTP Upgrade Token Registry"
5533    </t>
5534    <t>
5535      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/173"/>:
5536      "CR and LF in chunk extension values"
5537    </t>
5538    <t>
5539      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/184"/>:
5540      "HTTP/0.9 support"
5541    </t>
5542    <t>
5543      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/188"/>:
5544      "pick IANA policy (RFC5226) for Transfer Coding / Content Coding"
5545    </t>
5546    <t>
5547      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/189"/>:
5548      "move definitions of gzip/deflate/compress to part 1"
5549    </t>
5550    <t>
5551      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/194"/>:
5552      "disallow control characters in quoted-pair"
5553    </t>
5554  </list>
5555</t>
5556<t>
5557  Partly resolved issues:
5558  <list style="symbols">
5559    <t>
5560      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/148"/>:
5561      "update IANA requirements wrt Transfer-Coding values" (add the
5562      IANA Considerations subsection)
5563    </t>
5564  </list>
5565</t>
5566</section>
5567
5568<section title="Since draft-ietf-httpbis-p1-messaging-08" anchor="changes.since.08">
5569<t>
5570  Closed issues:
5571  <list style="symbols">
5572    <t>
5573      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/201"/>:
5574      "header parsing, treatment of leading and trailing OWS"
5575    </t>
5576  </list>
5577</t>
5578<t>
5579  Partly resolved issues:
5580  <list style="symbols">
5581    <t>
5582      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/60"/>:
5583      "Placement of 13.5.1 and 13.5.2"
5584    </t>
5585    <t>
5586      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/200"/>:
5587      "use of term "word" when talking about header field structure"
5588    </t>
5589  </list>
5590</t>
5591</section>
5592
5593<section title="Since draft-ietf-httpbis-p1-messaging-09" anchor="changes.since.09">
5594<t>
5595  Closed issues:
5596  <list style="symbols">
5597    <t>
5598      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/73"/>:
5599      "Clarification of the term 'deflate'"
5600    </t>
5601    <t>
5602      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/83"/>:
5603      "OPTIONS * and proxies"
5604    </t>
5605    <t>
5606      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/122"/>:
5607      "MIME-Version not listed in P1, general header fields"
5608    </t>
5609    <t>
5610      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/143"/>:
5611      "IANA registry for content/transfer encodings"
5612    </t>
5613    <t>
5614      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/165"/>:
5615      "Case-sensitivity of HTTP-date"
5616    </t>
5617    <t>
5618      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/200"/>:
5619      "use of term "word" when talking about header field structure"
5620    </t>
5621  </list>
5622</t>
5623<t>
5624  Partly resolved issues:
5625  <list style="symbols">
5626    <t>
5627      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/196"/>:
5628      "Term for the requested resource's URI"
5629    </t>
5630  </list>
5631</t>
5632</section>
5633
5634<section title="Since draft-ietf-httpbis-p1-messaging-10" anchor="changes.since.10">
5635<t>
5636  Closed issues:
5637  <list style="symbols">
5638    <t>
5639      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/28"/>:
5640      "Connection Closing"
5641    </t>
5642    <t>
5643      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/90"/>:
5644      "Delimiting messages with multipart/byteranges"
5645    </t>
5646    <t>
5647      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/95"/>:
5648      "Handling multiple Content-Length header fields"
5649    </t>
5650    <t>
5651      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/109"/>:
5652      "Clarify entity / representation / variant terminology"
5653    </t>
5654    <t>
5655      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/220"/>:
5656      "consider removing the 'changes from 2068' sections"
5657    </t>
5658  </list>
5659</t>
5660<t>
5661  Partly resolved issues:
5662  <list style="symbols">
5663    <t>
5664      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/159"/>:
5665      "HTTP(s) URI scheme definitions"
5666    </t>
5667  </list>
5668</t>
5669</section>
5670
5671<section title="Since draft-ietf-httpbis-p1-messaging-11" anchor="changes.since.11">
5672<t>
5673  Closed issues:
5674  <list style="symbols">
5675    <t>
5676      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/193"/>:
5677      "Trailer requirements"
5678    </t>
5679    <t>
5680      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/204"/>:
5681      "Text about clock requirement for caches belongs in p6"
5682    </t>
5683    <t>
5684      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/221"/>:
5685      "effective request URI: handling of missing host in HTTP/1.0"
5686    </t>
5687    <t>
5688      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/248"/>:
5689      "confusing Date requirements for clients"
5690    </t>
5691  </list>
5692</t>
5693<t>
5694  Partly resolved issues:
5695  <list style="symbols">
5696    <t>
5697      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/95"/>:
5698      "Handling multiple Content-Length header fields"
5699    </t>
5700  </list>
5701</t>
5702</section>
5703
5704<section title="Since draft-ietf-httpbis-p1-messaging-12" anchor="changes.since.12">
5705<t>
5706  Closed issues:
5707  <list style="symbols">
5708    <t>
5709      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/75"/>:
5710      "RFC2145 Normative"
5711    </t>
5712    <t>
5713      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/159"/>:
5714      "HTTP(s) URI scheme definitions" (tune the requirements on userinfo)
5715    </t>
5716    <t>
5717      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/210"/>:
5718      "define 'transparent' proxy"
5719    </t>
5720    <t>
5721      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/224"/>:
5722      "Header Field Classification"
5723    </t>
5724    <t>
5725      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/233"/>:
5726      "Is * usable as a request-uri for new methods?"
5727    </t>
5728    <t>
5729      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/240"/>:
5730      "Migrate Upgrade details from RFC2817"
5731    </t>
5732    <t>
5733      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/276"/>:
5734      "untangle ABNFs for header fields"
5735    </t>
5736    <t>
5737      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/279"/>:
5738      "update RFC 2109 reference"
5739    </t>
5740  </list>
5741</t>
5742</section>
5743
5744<section title="Since draft-ietf-httpbis-p1-messaging-13" anchor="changes.since.13">
5745<t>
5746  Closed issues:
5747  <list style="symbols">
5748    <t>
5749      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/53"/>:
5750      "Allow is not in 13.5.2"
5751    </t>
5752    <t>
5753      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/95"/>:
5754      "Handling multiple Content-Length header fields"
5755    </t>
5756    <t>
5757      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/276"/>:
5758      "untangle ABNFs for header fields"
5759    </t>
5760    <t>
5761      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/286"/>:
5762      "Content-Length ABNF broken"
5763    </t>
5764  </list>
5765</t>
5766</section>
5767
5768<section title="Since draft-ietf-httpbis-p1-messaging-14" anchor="changes.since.14">
5769<t>
5770  Closed issues:
5771  <list style="symbols">
5772    <t>
5773      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/273"/>:
5774      "HTTP-version should be redefined as fixed length pair of DIGIT . DIGIT"
5775    </t>
5776    <t>
5777      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/282"/>:
5778      "Recommend minimum sizes for protocol elements"
5779    </t>
5780    <t>
5781      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/283"/>:
5782      "Set expectations around buffering"
5783    </t>
5784    <t>
5785      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/288"/>:
5786      "Considering messages in isolation"
5787    </t>
5788  </list>
5789</t>
5790</section>
5791
5792<section title="Since draft-ietf-httpbis-p1-messaging-15" anchor="changes.since.15">
5793<t>
5794  Closed issues:
5795  <list style="symbols">
5796    <t>
5797      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/100"/>:
5798      "DNS Spoofing / DNS Binding advice"
5799    </t>
5800    <t>
5801      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/254"/>:
5802      "move RFCs 2145, 2616, 2817 to Historic status"
5803    </t>
5804    <t>
5805      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/270"/>:
5806      "\-escaping in quoted strings"
5807    </t>
5808    <t>
5809      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/305"/>:
5810      "'Close' should be reserved in the HTTP header field registry"
5811    </t>
5812  </list>
5813</t>
5814</section>
5815
5816<section title="Since draft-ietf-httpbis-p1-messaging-16" anchor="changes.since.16">
5817<t>
5818  Closed issues:
5819  <list style="symbols">
5820    <t>
5821      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/186"/>:
5822      "Document HTTP's error-handling philosophy"
5823    </t>
5824    <t>
5825      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/215"/>:
5826      "Explain header field registration"
5827    </t>
5828    <t>
5829      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/219"/>:
5830      "Revise Acknowledgements Sections"
5831    </t>
5832    <t>
5833      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/297"/>:
5834      "Retrying Requests"
5835    </t>
5836    <t>
5837      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/318"/>:
5838      "Closing the connection on server error"
5839    </t>
5840  </list>
5841</t>
5842</section>
5843
5844<section title="Since draft-ietf-httpbis-p1-messaging-17" anchor="changes.since.17">
5845<t>
5846  Closed issues:
5847  <list style="symbols">
5848    <t>
5849      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/158"/>:
5850      "Proxy-Connection and Keep-Alive"
5851    </t>
5852    <t>
5853      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/166"/>:
5854      "Clarify 'User Agent'"
5855    </t>
5856    <t>
5857      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/300"/>:
5858      "Define non-final responses"
5859    </t>
5860    <t>
5861      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/323"/>:
5862      "intended maturity level vs normative references"
5863    </t>
5864    <t>
5865      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/324"/>:
5866      "Intermediary rewriting of queries"
5867    </t>
5868  </list>
5869</t>
5870</section>
5871
5872<section title="Since draft-ietf-httpbis-p1-messaging-18" anchor="changes.since.18">
5873<t>
5874  Closed issues:
5875  <list style="symbols">
5876    <t>
5877      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/250"/>:
5878      "message-body in CONNECT response"
5879    </t>
5880    <t>
5881      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/302"/>:
5882      "Misplaced text on connection handling in p2"
5883    </t>
5884    <t>
5885      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/335"/>:
5886      "wording of line folding rule"
5887    </t>
5888    <t>
5889      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/343"/>:
5890      "chunk-extensions"
5891    </t>
5892    <t>
5893      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/346"/>:
5894      "make IANA policy definitions consistent"
5895    </t>
5896  </list>
5897</t>
5898</section>
5899
5900<section title="Since draft-ietf-httpbis-p1-messaging-19" anchor="changes.since.19">
5901<t>
5902  Closed issues:
5903  <list style="symbols">
5904    <t>
5905      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/346"/>:
5906      "make IANA policy definitions consistent"
5907    </t>
5908    <t>
5909      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/359"/>:
5910      "clarify connection header field values are case-insensitive"
5911    </t>
5912    <t>
5913      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/361"/>:
5914      "ABNF requirements for recipients"
5915    </t>
5916    <t>
5917      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/368"/>:
5918      "note introduction of new IANA registries as normative changes"
5919    </t>
5920    <t>
5921      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/374"/>:
5922      "Reference to ISO-8859-1 is informative"
5923    </t>
5924  </list>
5925</t>
5926</section>
5927
5928</section>
5929
5930</back>
5931</rfc>
Note: See TracBrowser for help on using the repository browser.