source: draft-ietf-httpbis/latest/p1-messaging.xml @ 2038

Last change on this file since 2038 was 2038, checked in by fielding@…, 7 years ago

Special case of OPTIONS for normalizing empty path; addresses #397

  • Property svn:eol-style set to native
  • Property svn:mime-type set to text/xml
File size: 223.1 KB
Line 
1<?xml version="1.0" encoding="utf-8"?>
2<?xml-stylesheet type='text/xsl' href='../myxml2rfc.xslt'?>
3<!DOCTYPE rfc [
4  <!ENTITY MAY "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MAY</bcp14>">
5  <!ENTITY MUST "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST</bcp14>">
6  <!ENTITY MUST-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST NOT</bcp14>">
7  <!ENTITY OPTIONAL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>OPTIONAL</bcp14>">
8  <!ENTITY RECOMMENDED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>RECOMMENDED</bcp14>">
9  <!ENTITY REQUIRED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>REQUIRED</bcp14>">
10  <!ENTITY SHALL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL</bcp14>">
11  <!ENTITY SHALL-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL NOT</bcp14>">
12  <!ENTITY SHOULD "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD</bcp14>">
13  <!ENTITY SHOULD-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD NOT</bcp14>">
14  <!ENTITY ID-VERSION "latest">
15  <!ENTITY ID-MONTH "December">
16  <!ENTITY ID-YEAR "2012">
17  <!ENTITY mdash "&#8212;">
18  <!ENTITY Note "<x:h xmlns:x='http://purl.org/net/xml2rfc/ext'>Note:</x:h>">
19  <!ENTITY caching-overview       "<xref target='Part6' x:rel='#caching.overview' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
20  <!ENTITY cache-incomplete       "<xref target='Part6' x:rel='#response.cacheability' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
21  <!ENTITY payload                "<xref target='Part2' x:rel='#payload' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
22  <!ENTITY media-type            "<xref target='Part2' x:rel='#media.type' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
23  <!ENTITY content-codings        "<xref target='Part2' x:rel='#content.codings' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
24  <!ENTITY CONNECT                "<xref target='Part2' x:rel='#CONNECT' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
25  <!ENTITY content.negotiation    "<xref target='Part2' x:rel='#content.negotiation' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
26  <!ENTITY diff-mime              "<xref target='Part2' x:rel='#differences.between.http.and.mime' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
27  <!ENTITY representation         "<xref target='Part2' x:rel='#representation' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
28  <!ENTITY HEAD                   "<xref target='Part2' x:rel='#HEAD' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
29  <!ENTITY header-allow           "<xref target='Part2' x:rel='#header.allow' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
30  <!ENTITY header-cache-control   "<xref target='Part6' x:rel='#header.cache-control' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
31  <!ENTITY header-content-encoding    "<xref target='Part2' x:rel='#header.content-encoding' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
32  <!ENTITY header-content-location    "<xref target='Part2' x:rel='#header.content-location' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
33  <!ENTITY header-content-range   "<xref target='Part5' x:rel='#header.content-range' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
34  <!ENTITY header-content-type    "<xref target='Part2' x:rel='#header.content-type' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
35  <!ENTITY header-date            "<xref target='Part2' x:rel='#header.date' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
36  <!ENTITY header-etag            "<xref target='Part4' x:rel='#header.etag' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
37  <!ENTITY header-expires         "<xref target='Part6' x:rel='#header.expires' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
38  <!ENTITY header-last-modified   "<xref target='Part4' x:rel='#header.last-modified' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
39  <!ENTITY header-mime-version    "<xref target='Part2' x:rel='#mime-version' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
40  <!ENTITY header-pragma          "<xref target='Part6' x:rel='#header.pragma' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
41  <!ENTITY header-proxy-authenticate  "<xref target='Part7' x:rel='#header.proxy-authenticate' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
42  <!ENTITY header-proxy-authorization "<xref target='Part7' x:rel='#header.proxy-authorization' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
43  <!ENTITY header-server          "<xref target='Part2' x:rel='#header.server' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
44  <!ENTITY header-warning         "<xref target='Part6' x:rel='#header.warning' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
45  <!ENTITY idempotent-methods     "<xref target='Part2' x:rel='#idempotent.methods' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
46  <!ENTITY methods                "<xref target='Part2' x:rel='#methods' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
47  <!ENTITY OPTIONS                "<xref target='Part2' x:rel='#OPTIONS' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
48  <!ENTITY qvalue                 "<xref target='Part2' x:rel='#quality.values' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
49  <!ENTITY resource               "<xref target='Part2' x:rel='#resource' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
50  <!ENTITY selected-representation    "<xref target='Part2' x:rel='#selected.representation' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
51  <!ENTITY status-codes           "<xref target='Part2' x:rel='#status.codes' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
52  <!ENTITY status-1xx             "<xref target='Part2' x:rel='#status.1xx' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
53  <!ENTITY status-203             "<xref target='Part2' x:rel='#status.203' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
54  <!ENTITY status-3xx             "<xref target='Part2' x:rel='#status.3xx' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
55  <!ENTITY status-304             "<xref target='Part4' x:rel='#status.304' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
56  <!ENTITY status-4xx             "<xref target='Part2' x:rel='#status.4xx' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
57  <!ENTITY status-414             "<xref target='Part2' x:rel='#status.414' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
58  <!ENTITY iana-header-registry   "<xref target='Part2' x:rel='#header.field.registry' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
59]>
60<?rfc toc="yes" ?>
61<?rfc symrefs="yes" ?>
62<?rfc sortrefs="yes" ?>
63<?rfc compact="yes"?>
64<?rfc subcompact="no" ?>
65<?rfc linkmailto="no" ?>
66<?rfc editing="no" ?>
67<?rfc comments="yes"?>
68<?rfc inline="yes"?>
69<?rfc rfcedstyle="yes"?>
70<?rfc-ext allow-markup-in-artwork="yes" ?>
71<?rfc-ext include-references-in-index="yes" ?>
72<rfc obsoletes="2145,2616" updates="2817,2818" category="std" x:maturity-level="proposed"
73     ipr="pre5378Trust200902" docName="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"
74     xmlns:x='http://purl.org/net/xml2rfc/ext'>
75<x:link rel="next" basename="p2-semantics"/>
76<x:feedback template="mailto:ietf-http-wg@w3.org?subject={docname},%20%22{section}%22&amp;body=&lt;{ref}&gt;:"/>
77<front>
78
79  <title abbrev="HTTP/1.1 Message Syntax and Routing">Hypertext Transfer Protocol (HTTP/1.1): Message Syntax and Routing</title>
80
81  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
82    <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
83    <address>
84      <postal>
85        <street>345 Park Ave</street>
86        <city>San Jose</city>
87        <region>CA</region>
88        <code>95110</code>
89        <country>USA</country>
90      </postal>
91      <email>fielding@gbiv.com</email>
92      <uri>http://roy.gbiv.com/</uri>
93    </address>
94  </author>
95
96  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
97    <organization abbrev="greenbytes">greenbytes GmbH</organization>
98    <address>
99      <postal>
100        <street>Hafenweg 16</street>
101        <city>Muenster</city><region>NW</region><code>48155</code>
102        <country>Germany</country>
103      </postal>
104      <email>julian.reschke@greenbytes.de</email>
105      <uri>http://greenbytes.de/tech/webdav/</uri>
106    </address>
107  </author>
108
109  <date month="&ID-MONTH;" year="&ID-YEAR;"/>
110  <workgroup>HTTPbis Working Group</workgroup>
111
112<abstract>
113<t>
114   The Hypertext Transfer Protocol (HTTP) is an application-level protocol for
115   distributed, collaborative, hypertext information systems. HTTP has been in
116   use by the World Wide Web global information initiative since 1990.
117   This document provides an overview of HTTP architecture and its associated
118   terminology, defines the "http" and "https" Uniform Resource Identifier
119   (URI) schemes, defines the HTTP/1.1 message syntax and parsing requirements,
120   and describes general security concerns for implementations.
121</t>   
122</abstract>
123
124<note title="Editorial Note (To be removed by RFC Editor)">
125  <t>
126    Discussion of this draft takes place on the HTTPBIS working group
127    mailing list (ietf-http-wg@w3.org), which is archived at
128    <eref target="http://lists.w3.org/Archives/Public/ietf-http-wg/"/>.
129  </t>
130  <t>
131    The current issues list is at
132    <eref target="http://tools.ietf.org/wg/httpbis/trac/report/3"/> and related
133    documents (including fancy diffs) can be found at
134    <eref target="http://tools.ietf.org/wg/httpbis/"/>.
135  </t>
136  <t>
137    The changes in this draft are summarized in <xref target="changes.since.21"/>.
138  </t>
139</note>
140</front>
141<middle>
142<section title="Introduction" anchor="introduction">
143<t>
144   The Hypertext Transfer Protocol (HTTP) is an application-level
145   request/response protocol that uses extensible semantics and MIME-like
146   message payloads for flexible interaction with network-based hypertext
147   information systems. This document is the first in a series of documents
148   that collectively form the HTTP/1.1 specification:
149   <list style="empty">
150    <t>RFC xxx1: Message Syntax and Routing</t>
151    <t><xref target="Part2" x:fmt="none">RFC xxx2</xref>: Semantics and Content</t>
152    <t><xref target="Part4" x:fmt="none">RFC xxx3</xref>: Conditional Requests</t>
153    <t><xref target="Part5" x:fmt="none">RFC xxx4</xref>: Range Requests</t>
154    <t><xref target="Part6" x:fmt="none">RFC xxx5</xref>: Caching</t>
155    <t><xref target="Part7" x:fmt="none">RFC xxx6</xref>: Authentication</t>
156   </list>
157</t>
158<t>
159   This HTTP/1.1 specification obsoletes and moves to historic status
160   <xref target="RFC2616" x:fmt="none">RFC 2616</xref>, its predecessor
161   <xref target="RFC2068" x:fmt="none">RFC 2068</xref>, and
162   <xref target="RFC2145" x:fmt="none">RFC 2145</xref> (on HTTP versioning).
163   This specification also updates the use of CONNECT to establish a tunnel,
164   previously defined in <xref target="RFC2817" x:fmt="none">RFC 2817</xref>,
165   and defines the "https" URI scheme that was described informally in
166   <xref target="RFC2818" x:fmt="none">RFC 2818</xref>.
167</t>
168<t>
169   HTTP is a generic interface protocol for information systems. It is
170   designed to hide the details of how a service is implemented by presenting
171   a uniform interface to clients that is independent of the types of
172   resources provided. Likewise, servers do not need to be aware of each
173   client's purpose: an HTTP request can be considered in isolation rather
174   than being associated with a specific type of client or a predetermined
175   sequence of application steps. The result is a protocol that can be used
176   effectively in many different contexts and for which implementations can
177   evolve independently over time.
178</t>
179<t>
180   HTTP is also designed for use as an intermediation protocol for translating
181   communication to and from non-HTTP information systems.
182   HTTP proxies and gateways can provide access to alternative information
183   services by translating their diverse protocols into a hypertext
184   format that can be viewed and manipulated by clients in the same way
185   as HTTP services.
186</t>
187<t>
188   One consequence of this flexibility is that the protocol cannot be
189   defined in terms of what occurs behind the interface. Instead, we
190   are limited to defining the syntax of communication, the intent
191   of received communication, and the expected behavior of recipients.
192   If the communication is considered in isolation, then successful
193   actions ought to be reflected in corresponding changes to the
194   observable interface provided by servers. However, since multiple
195   clients might act in parallel and perhaps at cross-purposes, we
196   cannot require that such changes be observable beyond the scope
197   of a single response.
198</t>
199<t>
200   This document describes the architectural elements that are used or
201   referred to in HTTP, defines the "http" and "https" URI schemes,
202   describes overall network operation and connection management,
203   and defines HTTP message framing and forwarding requirements.
204   Our goal is to define all of the mechanisms necessary for HTTP message
205   handling that are independent of message semantics, thereby defining the
206   complete set of requirements for message parsers and
207   message-forwarding intermediaries.
208</t>
209
210
211<section title="Requirement Notation" anchor="intro.requirements">
212<t>
213   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
214   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
215   document are to be interpreted as described in <xref target="RFC2119"/>.
216</t>
217<t>
218   Conformance criteria and considerations regarding error handling
219   are defined in <xref target="conformance"/>.
220</t>
221</section>
222
223<section title="Syntax Notation" anchor="notation">
224<iref primary="true" item="Grammar" subitem="ALPHA"/>
225<iref primary="true" item="Grammar" subitem="CR"/>
226<iref primary="true" item="Grammar" subitem="CRLF"/>
227<iref primary="true" item="Grammar" subitem="CTL"/>
228<iref primary="true" item="Grammar" subitem="DIGIT"/>
229<iref primary="true" item="Grammar" subitem="DQUOTE"/>
230<iref primary="true" item="Grammar" subitem="HEXDIG"/>
231<iref primary="true" item="Grammar" subitem="HTAB"/>
232<iref primary="true" item="Grammar" subitem="LF"/>
233<iref primary="true" item="Grammar" subitem="OCTET"/>
234<iref primary="true" item="Grammar" subitem="SP"/>
235<iref primary="true" item="Grammar" subitem="VCHAR"/>
236<t>
237   This specification uses the Augmented Backus-Naur Form (ABNF) notation
238   of <xref target="RFC5234"/> with the list rule extension defined in
239   <xref target="abnf.extension"/><xref target="collected.abnf"/> shows
240   the collected ABNF with the list rule expanded.
241</t>
242<t anchor="core.rules">
243  <x:anchor-alias value="ALPHA"/>
244  <x:anchor-alias value="CTL"/>
245  <x:anchor-alias value="CR"/>
246  <x:anchor-alias value="CRLF"/>
247  <x:anchor-alias value="DIGIT"/>
248  <x:anchor-alias value="DQUOTE"/>
249  <x:anchor-alias value="HEXDIG"/>
250  <x:anchor-alias value="HTAB"/>
251  <x:anchor-alias value="LF"/>
252  <x:anchor-alias value="OCTET"/>
253  <x:anchor-alias value="SP"/>
254  <x:anchor-alias value="VCHAR"/>
255   The following core rules are included by
256   reference, as defined in <xref target="RFC5234" x:fmt="," x:sec="B.1"/>:
257   ALPHA (letters), CR (carriage return), CRLF (CR LF), CTL (controls),
258   DIGIT (decimal 0-9), DQUOTE (double quote),
259   HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line feed),
260   OCTET (any 8-bit sequence of data), SP (space), and
261   VCHAR (any visible <xref target="USASCII"/> character).
262</t>
263<t>
264   As a convention, ABNF rule names prefixed with "obs-" denote
265   "obsolete" grammar rules that appear for historical reasons.
266</t>
267</section>
268</section>
269
270<section title="Architecture" anchor="architecture">
271<t>
272   HTTP was created for the World Wide Web architecture
273   and has evolved over time to support the scalability needs of a worldwide
274   hypertext system. Much of that architecture is reflected in the terminology
275   and syntax productions used to define HTTP.
276</t>
277
278<section title="Client/Server Messaging" anchor="operation">
279<iref primary="true" item="client"/>
280<iref primary="true" item="server"/>
281<iref primary="true" item="connection"/>
282<t>
283   HTTP is a stateless request/response protocol that operates by exchanging
284   <x:dfn>messages</x:dfn> (<xref target="http.message"/>) across a reliable
285   transport or session-layer
286   "<x:dfn>connection</x:dfn>" (<xref target="connection.management"/>).
287   An HTTP "<x:dfn>client</x:dfn>" is a program that establishes a connection
288   to a server for the purpose of sending one or more HTTP requests.
289   An HTTP "<x:dfn>server</x:dfn>" is a program that accepts connections
290   in order to service HTTP requests by sending HTTP responses.
291</t>
292<iref primary="true" item="user agent"/>
293<iref primary="true" item="origin server"/>
294<iref primary="true" item="browser"/>
295<iref primary="true" item="spider"/>
296<iref primary="true" item="sender"/>
297<iref primary="true" item="recipient"/>
298<t>
299   The terms client and server refer only to the roles that
300   these programs perform for a particular connection.  The same program
301   might act as a client on some connections and a server on others.
302   We use the term "<x:dfn>user agent</x:dfn>" to refer to any of the various
303   client programs that initiate a request, including (but not limited to)
304   browsers, spiders (web-based robots), command-line tools, native
305   applications, and mobile apps.  The term "<x:dfn>origin server</x:dfn>" is
306   used to refer to the program that can originate authoritative responses to
307   a request. For general requirements, we use the terms
308   "<x:dfn>sender</x:dfn>" and "<x:dfn>recipient</x:dfn>" to refer to any
309   component that sends or receives, respectively, a given message.
310</t>
311<t>
312   HTTP relies upon the Uniform Resource Identifier (URI)
313   standard <xref target="RFC3986"/> to indicate the target resource
314   (<xref target="target-resource"/>) and relationships between resources.
315   Messages are passed in a format similar to that used by Internet mail
316   <xref target="RFC5322"/> and the Multipurpose Internet Mail Extensions
317   (MIME) <xref target="RFC2045"/> (see &diff-mime; for the differences
318   between HTTP and MIME messages).
319</t>
320<t>
321   Most HTTP communication consists of a retrieval request (GET) for
322   a representation of some resource identified by a URI.  In the
323   simplest case, this might be accomplished via a single bidirectional
324   connection (===) between the user agent (UA) and the origin server (O).
325</t>
326<figure><artwork type="drawing">
327         request   &gt;
328    <x:highlight>UA</x:highlight> ======================================= <x:highlight>O</x:highlight>
329                                &lt;   response
330</artwork></figure>
331<iref primary="true" item="message"/>
332<iref primary="true" item="request"/>
333<iref primary="true" item="response"/>
334<t>
335   A client sends an HTTP request to a server in the form of a <x:dfn>request</x:dfn>
336   message, beginning with a request-line that includes a method, URI, and
337   protocol version (<xref target="request.line"/>),
338   followed by header fields containing
339   request modifiers, client information, and representation metadata
340   (<xref target="header.fields"/>),
341   an empty line to indicate the end of the header section, and finally
342   a message body containing the payload body (if any,
343   <xref target="message.body"/>).
344</t>
345<t>
346   A server responds to a client's request by sending one or more HTTP
347   <x:dfn>response</x:dfn>
348   messages, each beginning with a status line that
349   includes the protocol version, a success or error code, and textual
350   reason phrase (<xref target="status.line"/>),
351   possibly followed by header fields containing server
352   information, resource metadata, and representation metadata
353   (<xref target="header.fields"/>),
354   an empty line to indicate the end of the header section, and finally
355   a message body containing the payload body (if any,
356   <xref target="message.body"/>).
357</t>
358<t>
359   A connection might be used for multiple request/response exchanges,
360   as defined in <xref target="persistent.connections"/>.
361</t>
362<t>
363   The following example illustrates a typical message exchange for a
364   GET request on the URI "http://www.example.com/hello.txt":
365</t>
366<figure><preamble>
367client request:
368</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
369GET /hello.txt HTTP/1.1
370User-Agent: curl/7.16.3 libcurl/7.16.3 OpenSSL/0.9.7l zlib/1.2.3
371Host: www.example.com
372Accept-Language: en, mi
373
374</artwork></figure>
375<figure><preamble>
376server response:
377</preamble><artwork type="message/http; msgtype=&#34;response&#34;" x:indent-with="  ">
378HTTP/1.1 200 OK
379Date: Mon, 27 Jul 2009 12:28:53 GMT
380Server: Apache
381Last-Modified: Wed, 22 Jul 2009 19:15:56 GMT
382ETag: "34aa387-d-1568eb00"
383Accept-Ranges: bytes
384Content-Length: <x:length-of target="exbody"/>
385Vary: Accept-Encoding
386Content-Type: text/plain
387
388<x:span anchor="exbody">Hello World!
389</x:span></artwork></figure>
390</section>
391
392<section title="Implementation Diversity" anchor="implementation-diversity">
393<t>
394   When considering the design of HTTP, it is easy to fall into a trap of
395   thinking that all user agents are general-purpose browsers and all origin
396   servers are large public websites. That is not the case in practice.
397   Common HTTP user agents include household appliances, stereos, scales,
398   firmware update scripts, command-line programs, mobile apps,
399   and communication devices in a multitude of shapes and sizes.  Likewise,
400   common HTTP origin servers include home automation units, configurable
401   networking components, office machines, autonomous robots, news feeds,
402   traffic cameras, ad selectors, and video delivery platforms.
403</t>
404<t>
405   The term "user agent" does not imply that there is a human user directly
406   interacting with the software agent at the time of a request. In many
407   cases, a user agent is installed or configured to run in the background
408   and save its results for later inspection (or save only a subset of those
409   results that might be interesting or erroneous). Spiders, for example, are
410   typically given a start URI and configured to follow certain behavior while
411   crawling the Web as a hypertext graph.
412</t>
413<t>
414   The implementation diversity of HTTP means that we cannot assume the
415   user agent can make interactive suggestions to a user or provide adequate
416   warning for security or privacy options.  In the few cases where this
417   specification requires reporting of errors to the user, it is acceptable
418   for such reporting to only be observable in an error console or log file.
419   Likewise, requirements that an automated action be confirmed by the user
420   before proceeding can be met via advance configuration choices,
421   run-time options, or simply not proceeding with the unsafe action.
422</t>
423</section>
424
425<section title="Intermediaries" anchor="intermediaries">
426<iref primary="true" item="intermediary"/>
427<t>
428   HTTP enables the use of intermediaries to satisfy requests through
429   a chain of connections.  There are three common forms of HTTP
430   <x:dfn>intermediary</x:dfn>: proxy, gateway, and tunnel.  In some cases,
431   a single intermediary might act as an origin server, proxy, gateway,
432   or tunnel, switching behavior based on the nature of each request.
433</t>
434<figure><artwork type="drawing">
435         &gt;             &gt;             &gt;             &gt;
436    <x:highlight>UA</x:highlight> =========== <x:highlight>A</x:highlight> =========== <x:highlight>B</x:highlight> =========== <x:highlight>C</x:highlight> =========== <x:highlight>O</x:highlight>
437               &lt;             &lt;             &lt;             &lt;
438</artwork></figure>
439<t>
440   The figure above shows three intermediaries (A, B, and C) between the
441   user agent and origin server. A request or response message that
442   travels the whole chain will pass through four separate connections.
443   Some HTTP communication options
444   might apply only to the connection with the nearest, non-tunnel
445   neighbor, only to the end-points of the chain, or to all connections
446   along the chain. Although the diagram is linear, each participant might
447   be engaged in multiple, simultaneous communications. For example, B
448   might be receiving requests from many clients other than A, and/or
449   forwarding requests to servers other than C, at the same time that it
450   is handling A's request.
451</t>
452<t>
453<iref primary="true" item="upstream"/><iref primary="true" item="downstream"/>
454<iref primary="true" item="inbound"/><iref primary="true" item="outbound"/>
455   We use the terms "<x:dfn>upstream</x:dfn>" and "<x:dfn>downstream</x:dfn>"
456   to describe various requirements in relation to the directional flow of a
457   message: all messages flow from upstream to downstream.
458   Likewise, we use the terms inbound and outbound to refer to
459   directions in relation to the request path:
460   "<x:dfn>inbound</x:dfn>" means toward the origin server and
461   "<x:dfn>outbound</x:dfn>" means toward the user agent.
462</t>
463<t><iref primary="true" item="proxy"/>
464   A "<x:dfn>proxy</x:dfn>" is a message forwarding agent that is selected by the
465   client, usually via local configuration rules, to receive requests
466   for some type(s) of absolute URI and attempt to satisfy those
467   requests via translation through the HTTP interface.  Some translations
468   are minimal, such as for proxy requests for "http" URIs, whereas
469   other requests might require translation to and from entirely different
470   application-level protocols. Proxies are often used to group an
471   organization's HTTP requests through a common intermediary for the
472   sake of security, annotation services, or shared caching.
473</t>
474<t>
475<iref primary="true" item="transforming proxy"/>
476<iref primary="true" item="non-transforming proxy"/>
477   An HTTP-to-HTTP proxy is called a "<x:dfn>transforming proxy</x:dfn>" if it is designed
478   or configured to modify request or response messages in a semantically
479   meaningful way (i.e., modifications, beyond those required by normal
480   HTTP processing, that change the message in a way that would be
481   significant to the original sender or potentially significant to
482   downstream recipients).  For example, a transforming proxy might be
483   acting as a shared annotation server (modifying responses to include
484   references to a local annotation database), a malware filter, a
485   format transcoder, or an intranet-to-Internet privacy filter.  Such
486   transformations are presumed to be desired by the client (or client
487   organization) that selected the proxy and are beyond the scope of
488   this specification.  However, when a proxy is not intended to transform
489   a given message, we use the term "<x:dfn>non-transforming proxy</x:dfn>" to target
490   requirements that preserve HTTP message semantics. See &status-203; and
491   &header-warning; for status and warning codes related to transformations.
492</t>
493<t><iref primary="true" item="gateway"/><iref primary="true" item="reverse proxy"/>
494<iref primary="true" item="accelerator"/>
495   A "<x:dfn>gateway</x:dfn>" (a.k.a., "<x:dfn>reverse proxy</x:dfn>")
496   is a receiving agent that acts
497   as a layer above some other server(s) and translates the received
498   requests to the underlying server's protocol.  Gateways are often
499   used to encapsulate legacy or untrusted information services, to
500   improve server performance through "<x:dfn>accelerator</x:dfn>" caching, and to
501   enable partitioning or load-balancing of HTTP services across
502   multiple machines.
503</t>
504<t>
505   A gateway behaves as an origin server on its outbound connection and
506   as a user agent on its inbound connection.
507   All HTTP requirements applicable to an origin server
508   also apply to the outbound communication of a gateway.
509   A gateway communicates with inbound servers using any protocol that
510   it desires, including private extensions to HTTP that are outside
511   the scope of this specification.  However, an HTTP-to-HTTP gateway
512   that wishes to interoperate with third-party HTTP servers &MUST;
513   conform to HTTP user agent requirements on the gateway's inbound
514   connection and &MUST; implement the <x:ref>Connection</x:ref>
515   (<xref target="header.connection"/>) and <x:ref>Via</x:ref>
516   (<xref target="header.via"/>) header fields for both connections.
517</t>
518<t><iref primary="true" item="tunnel"/>
519   A "<x:dfn>tunnel</x:dfn>" acts as a blind relay between two connections
520   without changing the messages. Once active, a tunnel is not
521   considered a party to the HTTP communication, though the tunnel might
522   have been initiated by an HTTP request. A tunnel ceases to exist when
523   both ends of the relayed connection are closed. Tunnels are used to
524   extend a virtual connection through an intermediary, such as when
525   Transport Layer Security (TLS, <xref target="RFC5246"/>) is used to
526   establish confidential communication through a shared firewall proxy.
527</t>
528<t><iref primary="true" item="interception proxy"/>
529<iref primary="true" item="transparent proxy"/>
530<iref primary="true" item="captive portal"/>
531   The above categories for intermediary only consider those acting as
532   participants in the HTTP communication.  There are also intermediaries
533   that can act on lower layers of the network protocol stack, filtering or
534   redirecting HTTP traffic without the knowledge or permission of message
535   senders. Network intermediaries often introduce security flaws or
536   interoperability problems by violating HTTP semantics.  For example, an
537   "<x:dfn>interception proxy</x:dfn>" <xref target="RFC3040"/> (also commonly
538   known as a "<x:dfn>transparent proxy</x:dfn>" <xref target="RFC1919"/> or
539   "<x:dfn>captive portal</x:dfn>")
540   differs from an HTTP proxy because it is not selected by the client.
541   Instead, an interception proxy filters or redirects outgoing TCP port 80
542   packets (and occasionally other common port traffic).
543   Interception proxies are commonly found on public network access points,
544   as a means of enforcing account subscription prior to allowing use of
545   non-local Internet services, and within corporate firewalls to enforce
546   network usage policies.
547   They are indistinguishable from a man-in-the-middle attack.
548</t>
549<t>
550   HTTP is defined as a stateless protocol, meaning that each request message
551   can be understood in isolation.  Many implementations depend on HTTP's
552   stateless design in order to reuse proxied connections or dynamically
553   load-balance requests across multiple servers.  Hence, servers &MUST-NOT;
554   assume that two requests on the same connection are from the same user
555   agent unless the connection is secured and specific to that agent.
556   Some non-standard HTTP extensions (e.g., <xref target="RFC4559"/>) have
557   been known to violate this requirement, resulting in security and
558   interoperability problems.
559</t>
560</section>
561
562<section title="Caches" anchor="caches">
563<iref primary="true" item="cache"/>
564<t>
565   A "<x:dfn>cache</x:dfn>" is a local store of previous response messages and the
566   subsystem that controls its message storage, retrieval, and deletion.
567   A cache stores cacheable responses in order to reduce the response
568   time and network bandwidth consumption on future, equivalent
569   requests. Any client or server &MAY; employ a cache, though a cache
570   cannot be used by a server while it is acting as a tunnel.
571</t>
572<t>
573   The effect of a cache is that the request/response chain is shortened
574   if one of the participants along the chain has a cached response
575   applicable to that request. The following illustrates the resulting
576   chain if B has a cached copy of an earlier response from O (via C)
577   for a request which has not been cached by UA or A.
578</t>
579<figure><artwork type="drawing">
580            &gt;             &gt;
581       <x:highlight>UA</x:highlight> =========== <x:highlight>A</x:highlight> =========== <x:highlight>B</x:highlight> - - - - - - <x:highlight>C</x:highlight> - - - - - - <x:highlight>O</x:highlight>
582                  &lt;             &lt;
583</artwork></figure>
584<t><iref primary="true" item="cacheable"/>
585   A response is "<x:dfn>cacheable</x:dfn>" if a cache is allowed to store a copy of
586   the response message for use in answering subsequent requests.
587   Even when a response is cacheable, there might be additional
588   constraints placed by the client or by the origin server on when
589   that cached response can be used for a particular request. HTTP
590   requirements for cache behavior and cacheable responses are
591   defined in &caching-overview;
592</t>
593<t>
594   There are a wide variety of architectures and configurations
595   of caches deployed across the World Wide Web and
596   inside large organizations. These include national hierarchies
597   of proxy caches to save transoceanic bandwidth, collaborative systems that
598   broadcast or multicast cache entries, archives of pre-fetched cache
599   entries for use in off-line or high-latency environments, and so on.
600</t>
601</section>
602
603<section title="Conformance and Error Handling" anchor="conformance">
604<t>
605   This specification targets conformance criteria according to the role of
606   a participant in HTTP communication.  Hence, HTTP requirements are placed
607   on senders, recipients, clients, servers, user agents, intermediaries,
608   origin servers, proxies, gateways, or caches, depending on what behavior
609   is being constrained by the requirement. Additional (social) requirements
610   are placed on implementations, resource owners, and protocol element
611   registrations when they apply beyond the scope of a single communication.
612</t>
613<t>
614   The verb "generate" is used instead of "send" where a requirement
615   differentiates between creating a protocol element and merely forwarding a
616   received element downstream.
617</t>
618<t>
619   An implementation is considered conformant if it complies with all of the
620   requirements associated with the roles it partakes in HTTP. Note that
621   SHOULD-level requirements are relevant here, unless one of the documented
622   exceptions is applicable.
623</t>
624<t>
625   Conformance applies to both the syntax and semantics of HTTP protocol
626   elements. A sender &MUST-NOT; generate protocol elements that convey a
627   meaning that is known by that sender to be false. A sender &MUST-NOT;
628   generate protocol elements that do not match the grammar defined by the
629   ABNF rules for those protocol elements that are applicable to the sender's
630   role. If a received protocol element is processed, the recipient &MUST; be
631   able to parse any value that would match the ABNF rules for that protocol
632   element, excluding only those rules not applicable to the recipient's role.
633</t>
634<t>
635   Unless noted otherwise, a recipient &MAY; attempt to recover a usable
636   protocol element from an invalid construct.  HTTP does not define
637   specific error handling mechanisms except when they have a direct impact
638   on security, since different applications of the protocol require
639   different error handling strategies.  For example, a Web browser might
640   wish to transparently recover from a response where the
641   <x:ref>Location</x:ref> header field doesn't parse according to the ABNF,
642   whereas a systems control client might consider any form of error recovery
643   to be dangerous.
644</t>
645</section>
646
647<section title="Protocol Versioning" anchor="http.version">
648  <x:anchor-alias value="HTTP-version"/>
649  <x:anchor-alias value="HTTP-name"/>
650<t>
651   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate
652   versions of the protocol. This specification defines version "1.1".
653   The protocol version as a whole indicates the sender's conformance
654   with the set of requirements laid out in that version's corresponding
655   specification of HTTP.
656</t>
657<t>
658   The version of an HTTP message is indicated by an HTTP-version field
659   in the first line of the message. HTTP-version is case-sensitive.
660</t>
661<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-version"/><iref primary="true" item="Grammar" subitem="HTTP-name"/>
662  <x:ref>HTTP-version</x:ref>  = <x:ref>HTTP-name</x:ref> "/" <x:ref>DIGIT</x:ref> "." <x:ref>DIGIT</x:ref>
663  <x:ref>HTTP-name</x:ref>     = <x:abnf-char-sequence>"HTTP"</x:abnf-char-sequence> ; "HTTP", case-sensitive
664</artwork></figure>
665<t>
666   The HTTP version number consists of two decimal digits separated by a "."
667   (period or decimal point).  The first digit ("major version") indicates the
668   HTTP messaging syntax, whereas the second digit ("minor version") indicates
669   the highest minor version to which the sender is
670   conformant and able to understand for future communication.  The minor
671   version advertises the sender's communication capabilities even when the
672   sender is only using a backwards-compatible subset of the protocol,
673   thereby letting the recipient know that more advanced features can
674   be used in response (by servers) or in future requests (by clients).
675</t>
676<t>
677   When an HTTP/1.1 message is sent to an HTTP/1.0 recipient
678   <xref target="RFC1945"/> or a recipient whose version is unknown,
679   the HTTP/1.1 message is constructed such that it can be interpreted
680   as a valid HTTP/1.0 message if all of the newer features are ignored.
681   This specification places recipient-version requirements on some
682   new features so that a conformant sender will only use compatible
683   features until it has determined, through configuration or the
684   receipt of a message, that the recipient supports HTTP/1.1.
685</t>
686<t>
687   The interpretation of a header field does not change between minor
688   versions of the same major HTTP version, though the default
689   behavior of a recipient in the absence of such a field can change.
690   Unless specified otherwise, header fields defined in HTTP/1.1 are
691   defined for all versions of HTTP/1.x.  In particular, the <x:ref>Host</x:ref>
692   and <x:ref>Connection</x:ref> header fields ought to be implemented by all
693   HTTP/1.x implementations whether or not they advertise conformance with
694   HTTP/1.1.
695</t>
696<t>
697   New header fields can be defined such that, when they are
698   understood by a recipient, they might override or enhance the
699   interpretation of previously defined header fields.  When an
700   implementation receives an unrecognized header field, the recipient
701   &MUST; ignore that header field for local processing regardless of
702   the message's HTTP version.  An unrecognized header field received
703   by a proxy &MUST; be forwarded downstream unless the header field's
704   field-name is listed in the message's <x:ref>Connection</x:ref> header field
705   (see <xref target="header.connection"/>).
706   These requirements allow HTTP's functionality to be enhanced without
707   requiring prior update of deployed intermediaries.
708</t>
709<t>
710   Intermediaries that process HTTP messages (i.e., all intermediaries
711   other than those acting as tunnels) &MUST; send their own HTTP-version
712   in forwarded messages.  In other words, they &MUST-NOT; blindly
713   forward the first line of an HTTP message without ensuring that the
714   protocol version in that message matches a version to which that
715   intermediary is conformant for both the receiving and
716   sending of messages.  Forwarding an HTTP message without rewriting
717   the HTTP-version might result in communication errors when downstream
718   recipients use the message sender's version to determine what features
719   are safe to use for later communication with that sender.
720</t>
721<t>
722   An HTTP client &SHOULD; send a request version equal to the highest
723   version to which the client is conformant and
724   whose major version is no higher than the highest version supported
725   by the server, if this is known.  An HTTP client &MUST-NOT; send a
726   version to which it is not conformant.
727</t>
728<t>
729   An HTTP client &MAY; send a lower request version if it is known that
730   the server incorrectly implements the HTTP specification, but only
731   after the client has attempted at least one normal request and determined
732   from the response status or header fields (e.g., <x:ref>Server</x:ref>) that
733   the server improperly handles higher request versions.
734</t>
735<t>
736   An HTTP server &SHOULD; send a response version equal to the highest
737   version to which the server is conformant and
738   whose major version is less than or equal to the one received in the
739   request.  An HTTP server &MUST-NOT; send a version to which it is not
740   conformant.  A server &MAY; send a <x:ref>505 (HTTP Version Not
741   Supported)</x:ref> response if it cannot send a response using the
742   major version used in the client's request.
743</t>
744<t>
745   An HTTP server &MAY; send an HTTP/1.0 response to an HTTP/1.0 request
746   if it is known or suspected that the client incorrectly implements the
747   HTTP specification and is incapable of correctly processing later
748   version responses, such as when a client fails to parse the version
749   number correctly or when an intermediary is known to blindly forward
750   the HTTP-version even when it doesn't conform to the given minor
751   version of the protocol. Such protocol downgrades &SHOULD-NOT; be
752   performed unless triggered by specific client attributes, such as when
753   one or more of the request header fields (e.g., <x:ref>User-Agent</x:ref>)
754   uniquely match the values sent by a client known to be in error.
755</t>
756<t>
757   The intention of HTTP's versioning design is that the major number
758   will only be incremented if an incompatible message syntax is
759   introduced, and that the minor number will only be incremented when
760   changes made to the protocol have the effect of adding to the message
761   semantics or implying additional capabilities of the sender.  However,
762   the minor version was not incremented for the changes introduced between
763   <xref target="RFC2068"/> and <xref target="RFC2616"/>, and this revision
764   has specifically avoiding any such changes to the protocol.
765</t>
766</section>
767
768<section title="Uniform Resource Identifiers" anchor="uri">
769<iref primary="true" item="resource"/>
770<t>
771   Uniform Resource Identifiers (URIs) <xref target="RFC3986"/> are used
772   throughout HTTP as the means for identifying resources (&resource;).
773   URI references are used to target requests, indicate redirects, and define
774   relationships.
775</t>
776  <x:anchor-alias value="URI-reference"/>
777  <x:anchor-alias value="absolute-URI"/>
778  <x:anchor-alias value="relative-part"/>
779  <x:anchor-alias value="authority"/>
780  <x:anchor-alias value="path-abempty"/>
781  <x:anchor-alias value="path-absolute"/>
782  <x:anchor-alias value="port"/>
783  <x:anchor-alias value="query"/>
784  <x:anchor-alias value="uri-host"/>
785  <x:anchor-alias value="partial-URI"/>
786<t>
787   This specification adopts the definitions of "URI-reference",
788   "absolute-URI", "relative-part", "port", "host",
789   "path-abempty", "path-absolute", "query", and "authority" from the
790   URI generic syntax.
791   In addition, we define a partial-URI rule for protocol elements
792   that allow a relative URI but not a fragment.
793</t>
794<figure><artwork type="abnf2616"><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>
795  <x:ref>URI-reference</x:ref> = &lt;URI-reference, defined in <xref target="RFC3986" x:fmt="," x:sec="4.1"/>&gt;
796  <x:ref>absolute-URI</x:ref>  = &lt;absolute-URI, defined in <xref target="RFC3986" x:fmt="," x:sec="4.3"/>&gt;
797  <x:ref>relative-part</x:ref> = &lt;relative-part, defined in <xref target="RFC3986" x:fmt="," x:sec="4.2"/>&gt;
798  <x:ref>authority</x:ref>     = &lt;authority, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2"/>&gt;
799  <x:ref>path-abempty</x:ref>  = &lt;path-abempty, defined in <xref target="RFC3986" x:fmt="," x:sec="3.3"/>&gt;
800  <x:ref>path-absolute</x:ref> = &lt;path-absolute, defined in <xref target="RFC3986" x:fmt="," x:sec="3.3"/>&gt;
801  <x:ref>port</x:ref>          = &lt;port, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2.3"/>&gt;
802  <x:ref>query</x:ref>         = &lt;query, defined in <xref target="RFC3986" x:fmt="," x:sec="3.4"/>&gt;
803  <x:ref>uri-host</x:ref>      = &lt;host, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2.2"/>&gt;
804 
805  <x:ref>partial-URI</x:ref>   = relative-part [ "?" query ]
806</artwork></figure>
807<t>
808   Each protocol element in HTTP that allows a URI reference will indicate
809   in its ABNF production whether the element allows any form of reference
810   (URI-reference), only a URI in absolute form (absolute-URI), only the
811   path and optional query components, or some combination of the above.
812   Unless otherwise indicated, URI references are parsed
813   relative to the effective request URI
814   (<xref target="effective.request.uri"/>).
815</t>
816
817<section title="http URI scheme" anchor="http.uri">
818  <x:anchor-alias value="http-URI"/>
819  <iref item="http URI scheme" primary="true"/>
820  <iref item="URI scheme" subitem="http" primary="true"/>
821<t>
822   The "http" URI scheme is hereby defined for the purpose of minting
823   identifiers according to their association with the hierarchical
824   namespace governed by a potential HTTP origin server listening for
825   TCP connections on a given port.
826</t>
827<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="http-URI"><!--terminal production--></iref>
828  <x:ref>http-URI</x:ref> = "http:" "//" <x:ref>authority</x:ref> <x:ref>path-abempty</x:ref> [ "?" <x:ref>query</x:ref> ]
829</artwork></figure>
830<t>
831   The HTTP origin server is identified by the generic syntax's
832   <x:ref>authority</x:ref> component, which includes a host identifier
833   and optional TCP port (<xref target="RFC3986" x:fmt="," x:sec="3.2.2"/>).
834   The remainder of the URI, consisting of both the hierarchical path
835   component and optional query component, serves as an identifier for
836   a potential resource within that origin server's name space.
837</t>
838<t>
839   If the host identifier is provided as an IP address,
840   then the origin server is any listener on the indicated TCP port at
841   that IP address. If host is a registered name, then that name is
842   considered an indirect identifier and the recipient might use a name
843   resolution service, such as DNS, to find the address of a listener
844   for that host.
845   The host &MUST-NOT; be empty; if an "http" URI is received with an
846   empty host, then it &MUST; be rejected as invalid.
847   If the port subcomponent is empty or not given, then TCP port 80 is
848   assumed (the default reserved port for WWW services).
849</t>
850<t>
851   Regardless of the form of host identifier, access to that host is not
852   implied by the mere presence of its name or address. The host might or might
853   not exist and, even when it does exist, might or might not be running an
854   HTTP server or listening to the indicated port. The "http" URI scheme
855   makes use of the delegated nature of Internet names and addresses to
856   establish a naming authority (whatever entity has the ability to place
857   an HTTP server at that Internet name or address) and allows that
858   authority to determine which names are valid and how they might be used.
859</t>
860<t>
861   When an "http" URI is used within a context that calls for access to the
862   indicated resource, a client &MAY; attempt access by resolving
863   the host to an IP address, establishing a TCP connection to that address
864   on the indicated port, and sending an HTTP request message
865   (<xref target="http.message"/>) containing the URI's identifying data
866   (<xref target="message.routing"/>) to the server.
867   If the server responds to that request with a non-interim HTTP response
868   message, as described in &status-codes;, then that response
869   is considered an authoritative answer to the client's request.
870</t>
871<t>
872   Although HTTP is independent of the transport protocol, the "http"
873   scheme is specific to TCP-based services because the name delegation
874   process depends on TCP for establishing authority.
875   An HTTP service based on some other underlying connection protocol
876   would presumably be identified using a different URI scheme, just as
877   the "https" scheme (below) is used for resources that require an
878   end-to-end secured connection. Other protocols might also be used to
879   provide access to "http" identified resources &mdash; it is only the
880   authoritative interface used for mapping the namespace that is
881   specific to TCP.
882</t>
883<t>
884   The URI generic syntax for authority also includes a deprecated
885   userinfo subcomponent (<xref target="RFC3986" x:fmt="," x:sec="3.2.1"/>)
886   for including user authentication information in the URI.  Some
887   implementations make use of the userinfo component for internal
888   configuration of authentication information, such as within command
889   invocation options, configuration files, or bookmark lists, even
890   though such usage might expose a user identifier or password.
891   Senders &MUST; exclude the userinfo subcomponent (and its "@"
892   delimiter) when an "http" URI is transmitted within a message as a
893   request target or header field value.
894   Recipients of an "http" URI reference &SHOULD; parse for userinfo and
895   treat its presence as an error, since it is likely being used to obscure
896   the authority for the sake of phishing attacks.
897</t>
898</section>
899
900<section title="https URI scheme" anchor="https.uri">
901   <x:anchor-alias value="https-URI"/>
902   <iref item="https URI scheme"/>
903   <iref item="URI scheme" subitem="https"/>
904<t>
905   The "https" URI scheme is hereby defined for the purpose of minting
906   identifiers according to their association with the hierarchical
907   namespace governed by a potential HTTP origin server listening to a
908   given TCP port for TLS-secured connections <xref target="RFC5246"/>.
909</t>
910<t>
911   All of the requirements listed above for the "http" scheme are also
912   requirements for the "https" scheme, except that a default TCP port
913   of 443 is assumed if the port subcomponent is empty or not given,
914   and the TCP connection &MUST; be secured, end-to-end, through the
915   use of strong encryption prior to sending the first HTTP request.
916</t>
917<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="https-URI"><!--terminal production--></iref>
918  <x:ref>https-URI</x:ref> = "https:" "//" <x:ref>authority</x:ref> <x:ref>path-abempty</x:ref> [ "?" <x:ref>query</x:ref> ]
919</artwork></figure>
920<t>
921   Unlike the "http" scheme, responses to "https" identified requests
922   are never "public" and thus &MUST-NOT; be reused for shared caching.
923   They can, however, be reused in a private cache if the message is
924   cacheable by default in HTTP or specifically indicated as such by
925   the Cache-Control header field (&header-cache-control;).
926</t>
927<t>
928   Resources made available via the "https" scheme have no shared
929   identity with the "http" scheme even if their resource identifiers
930   indicate the same authority (the same host listening to the same
931   TCP port).  They are distinct name spaces and are considered to be
932   distinct origin servers.  However, an extension to HTTP that is
933   defined to apply to entire host domains, such as the Cookie protocol
934   <xref target="RFC6265"/>, can allow information
935   set by one service to impact communication with other services
936   within a matching group of host domains.
937</t>
938<t>
939   The process for authoritative access to an "https" identified
940   resource is defined in <xref target="RFC2818"/>.
941</t>
942</section>
943
944<section title="http and https URI Normalization and Comparison" anchor="uri.comparison">
945<t>
946   Since the "http" and "https" schemes conform to the URI generic syntax,
947   such URIs are normalized and compared according to the algorithm defined
948   in <xref target="RFC3986" x:fmt="," x:sec="6"/>, using the defaults
949   described above for each scheme.
950</t>
951<t>
952   If the port is equal to the default port for a scheme, the normal form is
953   to elide the port subcomponent. When not being used in absolute form as the
954   request target of an OPTIONS request, an empty path component is equivalent
955   to an absolute path of "/", so the normal form is to provide a path of "/"
956   instead. The scheme and host are case-insensitive and normally provided in
957   lowercase; all other components are compared in a case-sensitive manner.
958   Characters other than those in the "reserved" set are equivalent to their
959   percent-encoded octets (see <xref target="RFC3986" x:fmt=","
960   x:sec="2.1"/>): the normal form is to not encode them.
961</t>
962<t>
963   For example, the following three URIs are equivalent:
964</t>
965<figure><artwork type="example">
966   http://example.com:80/~smith/home.html
967   http://EXAMPLE.com/%7Esmith/home.html
968   http://EXAMPLE.com:/%7esmith/home.html
969</artwork></figure>
970</section>
971</section>
972</section>
973
974<section title="Message Format" anchor="http.message">
975<x:anchor-alias value="generic-message"/>
976<x:anchor-alias value="message.types"/>
977<x:anchor-alias value="HTTP-message"/>
978<x:anchor-alias value="start-line"/>
979<iref item="header section"/>
980<iref item="headers"/>
981<iref item="header field"/>
982<t>
983   All HTTP/1.1 messages consist of a start-line followed by a sequence of
984   octets in a format similar to the Internet Message Format
985   <xref target="RFC5322"/>: zero or more header fields (collectively
986   referred to as the "headers" or the "header section"), an empty line
987   indicating the end of the header section, and an optional message body.
988</t>
989<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-message"><!--terminal production--></iref>
990  <x:ref>HTTP-message</x:ref>   = <x:ref>start-line</x:ref>
991                   *( <x:ref>header-field</x:ref> <x:ref>CRLF</x:ref> )
992                   <x:ref>CRLF</x:ref>
993                   [ <x:ref>message-body</x:ref> ]
994</artwork></figure>
995<t>
996   The normal procedure for parsing an HTTP message is to read the
997   start-line into a structure, read each header field into a hash
998   table by field name until the empty line, and then use the parsed
999   data to determine if a message body is expected.  If a message body
1000   has been indicated, then it is read as a stream until an amount
1001   of octets equal to the message body length is read or the connection
1002   is closed.
1003</t>
1004<t>
1005   Recipients &MUST; parse an HTTP message as a sequence of octets in an
1006   encoding that is a superset of US-ASCII <xref target="USASCII"/>.
1007   Parsing an HTTP message as a stream of Unicode characters, without regard
1008   for the specific encoding, creates security vulnerabilities due to the
1009   varying ways that string processing libraries handle invalid multibyte
1010   character sequences that contain the octet LF (%x0A).  String-based
1011   parsers can only be safely used within protocol elements after the element
1012   has been extracted from the message, such as within a header field-value
1013   after message parsing has delineated the individual fields.
1014</t>
1015<t>
1016   An HTTP message can be parsed as a stream for incremental processing or
1017   forwarding downstream.  However, recipients cannot rely on incremental
1018   delivery of partial messages, since some implementations will buffer or
1019   delay message forwarding for the sake of network efficiency, security
1020   checks, or payload transformations.
1021</t>
1022
1023<section title="Start Line" anchor="start.line">
1024  <x:anchor-alias value="Start-Line"/>
1025<t>
1026   An HTTP message can either be a request from client to server or a
1027   response from server to client.  Syntactically, the two types of message
1028   differ only in the start-line, which is either a request-line (for requests)
1029   or a status-line (for responses), and in the algorithm for determining
1030   the length of the message body (<xref target="message.body"/>).
1031</t>
1032<t>
1033   In theory, a client could receive requests and a server could receive
1034   responses, distinguishing them by their different start-line formats,
1035   but in practice servers are implemented to only expect a request
1036   (a response is interpreted as an unknown or invalid request method)
1037   and clients are implemented to only expect a response.
1038</t>
1039<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="start-line"/>
1040  <x:ref>start-line</x:ref>     = <x:ref>request-line</x:ref> / <x:ref>status-line</x:ref>
1041</artwork></figure>
1042<t>
1043   A sender &MUST-NOT; send whitespace between the start-line and
1044   the first header field. The presence of such whitespace in a request
1045   might be an attempt to trick a server into ignoring that field or
1046   processing the line after it as a new request, either of which might
1047   result in a security vulnerability if other implementations within
1048   the request chain interpret the same message differently.
1049   Likewise, the presence of such whitespace in a response might be
1050   ignored by some clients or cause others to cease parsing.
1051</t>
1052<t>
1053   A recipient that receives whitespace between the start-line and
1054   the first header field &MUST; either reject the message as invalid or
1055   consume each whitespace-preceded line without further processing of it
1056   (i.e., ignore the entire line, along with any subsequent lines preceded
1057   by whitespace, until a properly formed header field is received or the
1058   header block is terminated).
1059</t>
1060
1061<section title="Request Line" anchor="request.line">
1062  <x:anchor-alias value="Request"/>
1063  <x:anchor-alias value="request-line"/>
1064<t>
1065   A request-line begins with a method token, followed by a single
1066   space (SP), the request-target, another single space (SP), the
1067   protocol version, and ending with CRLF.
1068</t>
1069<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="request-line"/>
1070  <x:ref>request-line</x:ref>   = <x:ref>method</x:ref> <x:ref>SP</x:ref> <x:ref>request-target</x:ref> <x:ref>SP</x:ref> <x:ref>HTTP-version</x:ref> <x:ref>CRLF</x:ref>
1071</artwork></figure>
1072<iref primary="true" item="method"/>
1073<t anchor="method">
1074   The method token indicates the request method to be performed on the
1075   target resource. The request method is case-sensitive.
1076</t>
1077<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="method"/>
1078  <x:ref>method</x:ref>         = <x:ref>token</x:ref>
1079</artwork></figure>
1080<t>
1081   The methods defined by this specification can be found in
1082   &methods;, along with information regarding the HTTP method registry
1083   and considerations for defining new methods.
1084</t>
1085<iref item="request-target"/>
1086<t>
1087   The request-target identifies the target resource upon which to apply
1088   the request, as defined in <xref target="request-target"/>.
1089</t>
1090<t>
1091   No whitespace is allowed inside the method, request-target, and
1092   protocol version.  Hence, recipients typically parse the request-line
1093   into its component parts by splitting on whitespace
1094   (see <xref target="message.robustness"/>).
1095</t>
1096<t>
1097   Unfortunately, some user agents fail to properly encode hypertext
1098   references that have embedded whitespace, sending the characters directly
1099   instead of properly encoding or excluding the disallowed characters.
1100   Recipients of an invalid request-line &SHOULD; respond with either a
1101   <x:ref>400 (Bad Request)</x:ref> error or a <x:ref>301 (Moved Permanently)</x:ref>
1102   redirect with the request-target properly encoded.  Recipients &SHOULD-NOT;
1103   attempt to autocorrect and then process the request without a redirect,
1104   since the invalid request-line might be deliberately crafted to bypass
1105   security filters along the request chain.
1106</t>
1107<t>
1108   HTTP does not place a pre-defined limit on the length of a request-line.
1109   A server that receives a method longer than any that it implements
1110   &SHOULD; respond with either a <x:ref>405 (Method Not Allowed)</x:ref>, if it is an origin
1111   server, or a <x:ref>501 (Not Implemented)</x:ref> status code.
1112   A server &MUST; be prepared to receive URIs of unbounded length and
1113   respond with the <x:ref>414 (URI Too Long)</x:ref> status code if the received
1114   request-target would be longer than the server wishes to handle
1115   (see &status-414;).
1116</t>
1117<t>
1118   Various ad-hoc limitations on request-line length are found in practice.
1119   It is &RECOMMENDED; that all HTTP senders and recipients support, at a
1120   minimum, request-line lengths of 8000 octets.
1121</t>
1122</section>
1123
1124<section title="Status Line" anchor="status.line">
1125  <x:anchor-alias value="response"/>
1126  <x:anchor-alias value="status-line"/>
1127  <x:anchor-alias value="status-code"/>
1128  <x:anchor-alias value="reason-phrase"/>
1129<t>
1130   The first line of a response message is the status-line, consisting
1131   of the protocol version, a space (SP), the status code, another space,
1132   a possibly-empty textual phrase describing the status code, and
1133   ending with CRLF.
1134</t>
1135<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="status-line"/>
1136  <x:ref>status-line</x:ref> = <x:ref>HTTP-version</x:ref> <x:ref>SP</x:ref> <x:ref>status-code</x:ref> <x:ref>SP</x:ref> <x:ref>reason-phrase</x:ref> <x:ref>CRLF</x:ref>
1137</artwork></figure>
1138<t>
1139   The status-code element is a 3-digit integer code describing the
1140   result of the server's attempt to understand and satisfy the client's
1141   corresponding request. The rest of the response message is to be
1142   interpreted in light of the semantics defined for that status code.
1143   See &status-codes; for information about the semantics of status codes,
1144   including the classes of status code (indicated by the first digit),
1145   the status codes defined by this specification, considerations for the
1146   definition of new status codes, and the IANA registry.
1147</t>
1148<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="status-code"/>
1149  <x:ref>status-code</x:ref>    = 3<x:ref>DIGIT</x:ref>
1150</artwork></figure>
1151<t>   
1152   The reason-phrase element exists for the sole purpose of providing a
1153   textual description associated with the numeric status code, mostly
1154   out of deference to earlier Internet application protocols that were more
1155   frequently used with interactive text clients. A client &SHOULD; ignore
1156   the reason-phrase content.
1157</t>
1158<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="reason-phrase"/>
1159  <x:ref>reason-phrase</x:ref>  = *( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1160</artwork></figure>
1161</section>
1162</section>
1163
1164<section title="Header Fields" anchor="header.fields">
1165  <x:anchor-alias value="header-field"/>
1166  <x:anchor-alias value="field-content"/>
1167  <x:anchor-alias value="field-name"/>
1168  <x:anchor-alias value="field-value"/>
1169  <x:anchor-alias value="obs-fold"/>
1170<t>
1171   Each HTTP header field consists of a case-insensitive field name
1172   followed by a colon (":"), optional whitespace, and the field value.
1173</t>
1174<figure><artwork type="abnf2616"><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"/>
1175  <x:ref>header-field</x:ref>   = <x:ref>field-name</x:ref> ":" <x:ref>OWS</x:ref> <x:ref>field-value</x:ref> <x:ref>BWS</x:ref>
1176  <x:ref>field-name</x:ref>     = <x:ref>token</x:ref>
1177  <x:ref>field-value</x:ref>    = *( <x:ref>field-content</x:ref> / <x:ref>obs-fold</x:ref> )
1178  <x:ref>field-content</x:ref>  = *( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1179  <x:ref>obs-fold</x:ref>       = <x:ref>CRLF</x:ref> ( <x:ref>SP</x:ref> / <x:ref>HTAB</x:ref> )
1180                 ; obsolete line folding
1181                 ; see <xref target="field.parsing"/>
1182</artwork></figure>
1183<t>
1184   The field-name token labels the corresponding field-value as having the
1185   semantics defined by that header field.  For example, the <x:ref>Date</x:ref>
1186   header field is defined in &header-date; as containing the origination
1187   timestamp for the message in which it appears.
1188</t>
1189
1190<section title="Field Extensibility" anchor="field.extensibility">
1191<t>
1192   HTTP header fields are fully extensible: there is no limit on the
1193   introduction of new field names, each presumably defining new semantics,
1194   nor on the number of header fields used in a given message.  Existing
1195   fields are defined in each part of this specification and in many other
1196   specifications outside the core standard.
1197   New header fields can be introduced without changing the protocol version
1198   if their defined semantics allow them to be safely ignored by recipients
1199   that do not recognize them.
1200</t>
1201<t>
1202   New HTTP header fields &SHOULD; be registered with IANA in the
1203   Message Header Field Registry, as described in &iana-header-registry;.
1204   Unrecognized header fields &MUST; be forwarded by a proxy unless the
1205   field-name is listed in the <x:ref>Connection</x:ref> header field
1206   (<xref target="header.connection"/>) or the proxy is specifically
1207   configured to block or otherwise transform such fields.
1208   Unrecognized header fields &SHOULD; be ignored by other recipients.
1209</t>
1210</section>
1211
1212<section title="Field Order" anchor="field.order">
1213<t>
1214   The order in which header fields with differing field names are
1215   received is not significant. However, it is "good practice" to send
1216   header fields that contain control data first, such as <x:ref>Host</x:ref>
1217   on requests and <x:ref>Date</x:ref> on responses, so that implementations
1218   can decide when not to handle a message as early as possible.  A server
1219   &MUST; wait until the entire header section is received before interpreting
1220   a request message, since later header fields might include conditionals,
1221   authentication credentials, or deliberately misleading duplicate
1222   header fields that would impact request processing.
1223</t>
1224<t>
1225   Multiple header fields with the same field name &MUST-NOT; be
1226   sent in a message unless the entire field value for that
1227   header field is defined as a comma-separated list [i.e., #(values)].
1228</t>
1229<t>
1230   Multiple header fields with the same field name can be combined into
1231   one "field-name: field-value" pair, without changing the semantics of the
1232   message, by appending each subsequent field value to the combined
1233   field value in order, separated by a comma. The order in which
1234   header fields with the same field name are received is therefore
1235   significant to the interpretation of the combined field value;
1236   a proxy &MUST-NOT; change the order of these field values when
1237   forwarding a message.
1238</t>
1239<x:note>
1240  <t>
1241   &Note; In practice, the "Set-Cookie" header field (<xref target="RFC6265"/>)
1242   often appears multiple times in a response message and does not use the
1243   list syntax, violating the above requirements on multiple header fields
1244   with the same name. Since it cannot be combined into a single field-value,
1245   recipients ought to handle "Set-Cookie" as a special case while processing
1246   header fields. (See Appendix A.2.3 of <xref target="Kri2001"/> for details.)
1247  </t>
1248</x:note>
1249</section>
1250
1251<section title="Whitespace" anchor="whitespace">
1252<t anchor="rule.LWS">
1253   This specification uses three rules to denote the use of linear
1254   whitespace: OWS (optional whitespace), RWS (required whitespace), and
1255   BWS ("bad" whitespace).
1256</t>
1257<t anchor="rule.OWS">
1258   The OWS rule is used where zero or more linear whitespace octets might
1259   appear. OWS &SHOULD; either not be generated or be generated as a single
1260   SP. Multiple OWS octets that occur within field-content &SHOULD; either
1261   be replaced with a single SP or transformed to all SP octets (each
1262   octet other than SP replaced with SP) before interpreting the field value
1263   or forwarding the message downstream.
1264</t>
1265<t anchor="rule.RWS">
1266   RWS is used when at least one linear whitespace octet is required to
1267   separate field tokens. RWS &SHOULD; be generated as a single SP.
1268   Multiple RWS octets that occur within field-content &SHOULD; either
1269   be replaced with a single SP or transformed to all SP octets before
1270   interpreting the field value or forwarding the message downstream.
1271</t>
1272<t anchor="rule.BWS">
1273   BWS is used where the grammar allows optional whitespace, for historical
1274   reasons, but senders &SHOULD-NOT; generate it in messages;
1275   recipients &MUST; accept such bad optional whitespace and remove it before
1276   interpreting the field value or forwarding the message downstream.
1277</t>
1278<t anchor="rule.whitespace">
1279  <x:anchor-alias value="BWS"/>
1280  <x:anchor-alias value="OWS"/>
1281  <x:anchor-alias value="RWS"/>
1282</t>
1283<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="OWS"/><iref primary="true" item="Grammar" subitem="RWS"/><iref primary="true" item="Grammar" subitem="BWS"/>
1284  <x:ref>OWS</x:ref>            = *( <x:ref>SP</x:ref> / <x:ref>HTAB</x:ref> )
1285                 ; optional whitespace
1286  <x:ref>RWS</x:ref>            = 1*( <x:ref>SP</x:ref> / <x:ref>HTAB</x:ref> )
1287                 ; required whitespace
1288  <x:ref>BWS</x:ref>            = <x:ref>OWS</x:ref>
1289                 ; "bad" whitespace
1290</artwork></figure>
1291</section>
1292
1293<section title="Field Parsing" anchor="field.parsing">
1294<t>
1295   No whitespace is allowed between the header field-name and colon.
1296   In the past, differences in the handling of such whitespace have led to
1297   security vulnerabilities in request routing and response handling.
1298   Any received request message that contains whitespace between a header
1299   field-name and colon &MUST; be rejected with a response code of 400
1300   (Bad Request).  A proxy &MUST; remove any such whitespace from a response
1301   message before forwarding the message downstream.
1302</t>
1303<t>
1304   A field value is preceded by optional whitespace (OWS); a single SP is
1305   preferred. The field value does not include any leading or trailing white
1306   space: OWS occurring before the first non-whitespace octet of the
1307   field value or after the last non-whitespace octet of the field value
1308   is ignored and &SHOULD; be removed before further processing (as this does
1309   not change the meaning of the header field).
1310</t>
1311<t>
1312   Historically, HTTP header field values could be extended over multiple
1313   lines by preceding each extra line with at least one space or horizontal
1314   tab (obs-fold). This specification deprecates such line
1315   folding except within the message/http media type
1316   (<xref target="internet.media.type.message.http"/>).
1317   HTTP senders &MUST-NOT; generate messages that include line folding
1318   (i.e., that contain any field-value that matches the obs-fold rule) unless
1319   the message is intended for packaging within the message/http media type.
1320   HTTP recipients &SHOULD; accept line folding and replace any embedded
1321   obs-fold whitespace with either a single SP or a matching number of SP
1322   octets (to avoid buffer copying) prior to interpreting the field value or
1323   forwarding the message downstream.
1324</t>
1325<t>
1326   Historically, HTTP has allowed field content with text in the ISO-8859-1
1327   <xref target="ISO-8859-1"/> charset, supporting other charsets only
1328   through use of <xref target="RFC2047"/> encoding.
1329   In practice, most HTTP header field values use only a subset of the
1330   US-ASCII charset <xref target="USASCII"/>. Newly defined
1331   header fields &SHOULD; limit their field values to US-ASCII octets.
1332   Recipients &SHOULD; treat other octets in field content (obs-text) as
1333   opaque data.
1334</t>
1335</section>
1336
1337<section title="Field Limits" anchor="field.limits">
1338<t>
1339   HTTP does not place a pre-defined limit on the length of each header field
1340   or on the length of the header block as a whole.  Various ad-hoc
1341   limitations on individual header field length are found in practice,
1342   often depending on the specific field semantics.
1343</t>
1344<t>
1345   A server &MUST; be prepared to receive request header fields of unbounded
1346   length and respond with an appropriate <x:ref>4xx (Client Error)</x:ref>
1347   status code if the received header field(s) are larger than the server
1348   wishes to process.
1349</t>
1350<t>
1351   A client &MUST; be prepared to receive response header fields of unbounded
1352   length. A client &MAY; discard or truncate received header fields that are
1353   larger than the client wishes to process if the field semantics are such
1354   that the dropped value(s) can be safely ignored without changing the
1355   response semantics.
1356</t>
1357</section>
1358
1359<section title="Field value components" anchor="field.components">
1360<t anchor="rule.token.separators">
1361  <x:anchor-alias value="tchar"/>
1362  <x:anchor-alias value="token"/>
1363  <x:anchor-alias value="special"/>
1364  <x:anchor-alias value="word"/>
1365   Many HTTP header field values consist of words (token or quoted-string)
1366   separated by whitespace or special characters. These special characters
1367   &MUST; be in a quoted string to be used within a parameter value (as defined
1368   in <xref target="transfer.codings"/>).
1369</t>
1370<figure><artwork type="abnf2616"><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>
1371  <x:ref>word</x:ref>           = <x:ref>token</x:ref> / <x:ref>quoted-string</x:ref>
1372
1373  <x:ref>token</x:ref>          = 1*<x:ref>tchar</x:ref>
1374<!--
1375  IMPORTANT: when editing "tchar" make sure that "special" is updated accordingly!!!
1376 -->
1377  <x:ref>tchar</x:ref>          = "!" / "#" / "$" / "%" / "&amp;" / "'" / "*"
1378                 / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
1379                 / <x:ref>DIGIT</x:ref> / <x:ref>ALPHA</x:ref>
1380                 ; any <x:ref>VCHAR</x:ref>, except <x:ref>special</x:ref>
1381
1382  <x:ref>special</x:ref>        = "(" / ")" / "&lt;" / ">" / "@" / ","
1383                 / ";" / ":" / "\" / DQUOTE / "/" / "["
1384                 / "]" / "?" / "=" / "{" / "}"
1385</artwork></figure>
1386<t anchor="rule.quoted-string">
1387  <x:anchor-alias value="quoted-string"/>
1388  <x:anchor-alias value="qdtext"/>
1389  <x:anchor-alias value="obs-text"/>
1390   A string of text is parsed as a single word if it is quoted using
1391   double-quote marks.
1392</t>
1393<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-string"/><iref primary="true" item="Grammar" subitem="qdtext"/><iref primary="true" item="Grammar" subitem="obs-text"/>
1394  <x:ref>quoted-string</x:ref>  = <x:ref>DQUOTE</x:ref> *( <x:ref>qdtext</x:ref> / <x:ref>quoted-pair</x:ref> ) <x:ref>DQUOTE</x:ref>
1395  <x:ref>qdtext</x:ref>         = <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> /%x21 / %x23-5B / %x5D-7E / <x:ref>obs-text</x:ref>
1396  <x:ref>obs-text</x:ref>       = %x80-FF
1397</artwork></figure>
1398<t anchor="rule.quoted-pair">
1399  <x:anchor-alias value="quoted-pair"/>
1400   The backslash octet ("\") can be used as a single-octet
1401   quoting mechanism within quoted-string constructs:
1402</t>
1403<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-pair"/>
1404  <x:ref>quoted-pair</x:ref>    = "\" ( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1405</artwork></figure>
1406<t>
1407   Recipients that process the value of a quoted-string &MUST; handle a
1408   quoted-pair as if it were replaced by the octet following the backslash.
1409</t>
1410<t>
1411   Senders &SHOULD-NOT; generate a quoted-pair in a quoted-string except where
1412   necessary to quote DQUOTE and backslash octets occurring within that string.
1413</t>
1414<t anchor="rule.comment">
1415  <x:anchor-alias value="comment"/>
1416  <x:anchor-alias value="ctext"/>
1417   Comments can be included in some HTTP header fields by surrounding
1418   the comment text with parentheses. Comments are only allowed in
1419   fields containing "comment" as part of their field value definition.
1420</t>
1421<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/>
1422  <x:ref>comment</x:ref>        = "(" *( <x:ref>ctext</x:ref> / <x:ref>quoted-cpair</x:ref> / <x:ref>comment</x:ref> ) ")"
1423  <x:ref>ctext</x:ref>          = <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / %x21-27 / %x2A-5B / %x5D-7E / <x:ref>obs-text</x:ref>
1424</artwork></figure>
1425<t anchor="rule.quoted-cpair">
1426  <x:anchor-alias value="quoted-cpair"/>
1427   The backslash octet ("\") can be used as a single-octet
1428   quoting mechanism within comment constructs:
1429</t>
1430<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-cpair"/>
1431  <x:ref>quoted-cpair</x:ref>   = "\" ( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1432</artwork></figure>
1433<t>
1434   Senders &SHOULD-NOT; escape octets in comments that do not require escaping
1435   (i.e., other than the backslash octet "\" and the parentheses "(" and ")").
1436</t>
1437</section>
1438
1439</section>
1440
1441<section title="Message Body" anchor="message.body">
1442  <x:anchor-alias value="message-body"/>
1443<t>
1444   The message body (if any) of an HTTP message is used to carry the
1445   payload body of that request or response.  The message body is
1446   identical to the payload body unless a transfer coding has been
1447   applied, as described in <xref target="header.transfer-encoding"/>.
1448</t>
1449<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="message-body"/>
1450  <x:ref>message-body</x:ref> = *OCTET
1451</artwork></figure>
1452<t>
1453   The rules for when a message body is allowed in a message differ for
1454   requests and responses.
1455</t>
1456<t>
1457   The presence of a message body in a request is signaled by a
1458   <x:ref>Content-Length</x:ref> or <x:ref>Transfer-Encoding</x:ref> header
1459   field. Request message framing is independent of method semantics,
1460   even if the method does not define any use for a message body.
1461</t>
1462<t>
1463   The presence of a message body in a response depends on both
1464   the request method to which it is responding and the response
1465   status code (<xref target="status.line"/>).
1466   Responses to the HEAD request method never include a message body
1467   because the associated response header fields (e.g.,
1468   <x:ref>Transfer-Encoding</x:ref>, <x:ref>Content-Length</x:ref>, etc.),
1469   if present, indicate only what their values would have been if the request
1470   method had been GET (&HEAD;).
1471   <x:ref>2xx (Successful)</x:ref> responses to CONNECT switch to tunnel
1472   mode instead of having a message body (&CONNECT;).
1473   All <x:ref>1xx (Informational)</x:ref>, <x:ref>204 (No Content)</x:ref>, and
1474   <x:ref>304 (Not Modified)</x:ref> responses &MUST-NOT; include a message body.
1475   All other responses do include a message body, although the body
1476   &MAY; be of zero length.
1477</t>
1478
1479<section title="Transfer-Encoding" anchor="header.transfer-encoding">
1480  <iref primary="true" item="Transfer-Encoding header field" x:for-anchor=""/>
1481  <iref item="chunked (Coding Format)"/>
1482  <x:anchor-alias value="Transfer-Encoding"/>
1483<t>
1484   The Transfer-Encoding header field lists the transfer coding names
1485   corresponding to the sequence of transfer codings that have been
1486   (or will be) applied to the payload body in order to form the message body.
1487   Transfer codings are defined in <xref target="transfer.codings"/>.
1488</t>
1489<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/>
1490  <x:ref>Transfer-Encoding</x:ref> = 1#<x:ref>transfer-coding</x:ref>
1491</artwork></figure>
1492<t>
1493   Transfer-Encoding is analogous to the Content-Transfer-Encoding field of
1494   MIME, which was designed to enable safe transport of binary data over a
1495   7-bit transport service (<xref target="RFC2045" x:fmt="," x:sec="6"/>).
1496   However, safe transport has a different focus for an 8bit-clean transfer
1497   protocol. In HTTP's case, Transfer-Encoding is primarily intended to
1498   accurately delimit a dynamically generated payload and to distinguish
1499   payload encodings that are only applied for transport efficiency or
1500   security from those that are characteristics of the selected resource.
1501</t>
1502<t>
1503   All HTTP/1.1 recipients &MUST; implement the chunked transfer coding
1504   (<xref target="chunked.encoding"/>) because it plays a crucial role in
1505   framing messages when the payload body size is not known in advance.
1506   If chunked is applied to a payload body, the sender &MUST-NOT; apply
1507   chunked more than once (i.e., chunking an already chunked message is not
1508   allowed).
1509   If any transfer coding is applied to a request payload body, the
1510   sender &MUST; apply chunked as the final transfer coding to ensure that
1511   the message is properly framed.
1512   If any transfer coding is applied to a response payload body, the
1513   sender &MUST; either apply chunked as the final transfer coding or
1514   terminate the message by closing the connection.
1515</t>
1516<figure><preamble>
1517   For example,
1518</preamble><artwork type="example">
1519  Transfer-Encoding: gzip, chunked
1520</artwork><postamble>
1521   indicates that the payload body has been compressed using the gzip
1522   coding and then chunked using the chunked coding while forming the
1523   message body.
1524</postamble></figure>
1525<t>
1526   Unlike <x:ref>Content-Encoding</x:ref> (&content-codings;),
1527   Transfer-Encoding is a property of the message, not of the payload, and
1528   any recipient along the request/response chain &MAY; decode the received
1529   transfer coding(s) or apply additional transfer coding(s) to the message
1530   body, assuming that corresponding changes are made to the Transfer-Encoding
1531   field-value. Additional information about the encoding parameters &MAY; be
1532   provided by other header fields not defined by this specification.
1533</t>
1534<t>
1535   Transfer-Encoding &MAY; be sent in a response to a HEAD request or in a
1536   <x:ref>304 (Not Modified)</x:ref> response (&status-304;) to a GET request,
1537   neither of which includes a message body,
1538   to indicate that the origin server would have applied a transfer coding
1539   to the message body if the request had been an unconditional GET.
1540   This indication is not required, however, because any recipient on
1541   the response chain (including the origin server) can remove transfer
1542   codings when they are not needed.
1543</t>
1544<t>
1545   Transfer-Encoding was added in HTTP/1.1.  It is generally assumed that
1546   implementations advertising only HTTP/1.0 support will not understand
1547   how to process a transfer-encoded payload.
1548   A client &MUST-NOT; send a request containing Transfer-Encoding unless it
1549   knows the server will handle HTTP/1.1 (or later) requests; such knowledge
1550   might be in the form of specific user configuration or by remembering the
1551   version of a prior received response.
1552   A server &MUST-NOT; send a response containing Transfer-Encoding unless
1553   the corresponding request indicates HTTP/1.1 (or later).
1554</t>
1555<t>
1556   A server that receives a request message with a transfer coding it does
1557   not understand &SHOULD; respond with <x:ref>501 (Not Implemented)</x:ref>.
1558</t>
1559</section>
1560
1561<section title="Content-Length" anchor="header.content-length">
1562  <iref primary="true" item="Content-Length header field" x:for-anchor=""/>
1563  <x:anchor-alias value="Content-Length"/>
1564<t>
1565   When a message does not have a <x:ref>Transfer-Encoding</x:ref> header
1566   field, a Content-Length header field can provide the anticipated size,
1567   as a decimal number of octets, for a potential payload body.
1568   For messages that do include a payload body, the Content-Length field-value
1569   provides the framing information necessary for determining where the body
1570   (and message) ends.  For messages that do not include a payload body, the
1571   Content-Length indicates the size of the selected representation
1572   (&selected-representation;).
1573</t>
1574<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Length"/>
1575  <x:ref>Content-Length</x:ref> = 1*<x:ref>DIGIT</x:ref>
1576</artwork></figure>
1577<t>
1578   An example is
1579</t>
1580<figure><artwork type="example">
1581  Content-Length: 3495
1582</artwork></figure>
1583<t>
1584   A sender &MUST-NOT; send a Content-Length header field in any message that
1585   contains a <x:ref>Transfer-Encoding</x:ref> header field.
1586</t>
1587<t>
1588   A user agent &SHOULD; send a Content-Length in a request message when no
1589   <x:ref>Transfer-Encoding</x:ref> is sent and the request method defines
1590   a meaning for an enclosed payload body. For example, a Content-Length
1591   header field is normally sent in a POST request even when the value is
1592   0 (indicating an empty payload body).  A user agent &SHOULD-NOT; send a
1593   Content-Length header field when the request message does not contain a
1594   payload body and the method semantics do not anticipate such a body.
1595</t>
1596<t>
1597   A server &MAY; send a Content-Length header field in a response to a HEAD
1598   request (&HEAD;); a server &MUST-NOT; send Content-Length in such a
1599   response unless its field-value equals the decimal number of octets that
1600   would have been sent in the payload body of a response if the same
1601   request had used the GET method.
1602</t>
1603<t>
1604   A server &MAY; send a Content-Length header field in a
1605   <x:ref>304 (Not Modified)</x:ref> response to a conditional GET request
1606   (&status-304;); a server &MUST-NOT; send Content-Length in such a
1607   response unless its field-value equals the decimal number of octets that
1608   would have been sent in the payload body of a <x:ref>200 (OK)</x:ref>
1609   response to the same request.
1610</t>
1611<t>
1612   A server &MUST-NOT; send a Content-Length header field in any response
1613   with a status code of
1614   <x:ref>1xx (Informational)</x:ref> or <x:ref>204 (No Content)</x:ref>.
1615   A server &SHOULD-NOT; send a Content-Length header field in any
1616   <x:ref>2xx (Successful)</x:ref> response to a CONNECT request (&CONNECT;).
1617</t>
1618<t>
1619   Aside from the cases defined above, in the absence of Transfer-Encoding,
1620   an origin server &SHOULD; send a Content-Length header field when the
1621   payload body size is known prior to sending the complete header block.
1622   This will allow downstream recipients to measure transfer progress,
1623   know when a received message is complete, and potentially reuse the
1624   connection for additional requests.
1625</t>
1626<t>
1627   Any Content-Length field value greater than or equal to zero is valid.
1628   Since there is no predefined limit to the length of an HTTP payload,
1629   recipients &SHOULD; anticipate potentially large decimal numerals and
1630   prevent parsing errors due to integer conversion overflows
1631   (<xref target="attack.protocol.element.size.overflows"/>).
1632</t>
1633<t>
1634   If a message is received that has multiple Content-Length header fields
1635   with field-values consisting of the same decimal value, or a single
1636   Content-Length header field with a field value containing a list of
1637   identical decimal values (e.g., "Content-Length: 42, 42"), indicating that
1638   duplicate Content-Length header fields have been generated or combined by an
1639   upstream message processor, then the recipient &MUST; either reject the
1640   message as invalid or replace the duplicated field-values with a single
1641   valid Content-Length field containing that decimal value prior to
1642   determining the message body length.
1643</t>
1644<x:note>
1645  <t>
1646   &Note; HTTP's use of Content-Length for message framing differs
1647   significantly from the same field's use in MIME, where it is an optional
1648   field used only within the "message/external-body" media-type.
1649  </t>
1650</x:note>
1651</section>
1652
1653<section title="Message Body Length" anchor="message.body.length">
1654  <iref item="chunked (Coding Format)"/>
1655<t>
1656   The length of a message body is determined by one of the following
1657   (in order of precedence):
1658</t>
1659<t>
1660  <list style="numbers">
1661    <x:lt><t>
1662     Any response to a HEAD request and any response with a
1663     <x:ref>1xx (Informational)</x:ref>, <x:ref>204 (No Content)</x:ref>, or
1664     <x:ref>304 (Not Modified)</x:ref> status code is always
1665     terminated by the first empty line after the header fields, regardless of
1666     the header fields present in the message, and thus cannot contain a
1667     message body.
1668    </t></x:lt>
1669    <x:lt><t>
1670     Any <x:ref>2xx (Successful)</x:ref> response to a CONNECT request implies that the
1671     connection will become a tunnel immediately after the empty line that
1672     concludes the header fields.  A client &MUST; ignore any
1673     <x:ref>Content-Length</x:ref> or <x:ref>Transfer-Encoding</x:ref> header
1674     fields received in such a message.
1675    </t></x:lt>
1676    <x:lt><t>
1677     If a <x:ref>Transfer-Encoding</x:ref> header field is present
1678     and the chunked transfer coding (<xref target="chunked.encoding"/>)
1679     is the final encoding, the message body length is determined by reading
1680     and decoding the chunked data until the transfer coding indicates the
1681     data is complete.
1682    </t>
1683    <t>
1684     If a <x:ref>Transfer-Encoding</x:ref> header field is present in a
1685     response and the chunked transfer coding is not the final encoding, the
1686     message body length is determined by reading the connection until it is
1687     closed by the server.
1688     If a <x:ref>Transfer-Encoding</x:ref> header field is present in a request and the
1689     chunked transfer coding is not the final encoding, the message body
1690     length cannot be determined reliably; the server &MUST; respond with
1691     the <x:ref>400 (Bad Request)</x:ref> status code and then close the connection.
1692    </t>
1693    <t>
1694     If a message is received with both a <x:ref>Transfer-Encoding</x:ref>
1695     and a <x:ref>Content-Length</x:ref> header field, the Transfer-Encoding
1696     overrides the Content-Length. Such a message might indicate an attempt
1697     to perform request or response smuggling (bypass of security-related
1698     checks on message routing or content) and thus ought to be handled as
1699     an error.  A sender &MUST; remove the received Content-Length field
1700     prior to forwarding such a message downstream.
1701    </t></x:lt>
1702    <x:lt><t>
1703     If a message is received without <x:ref>Transfer-Encoding</x:ref> and with
1704     either multiple <x:ref>Content-Length</x:ref> header fields having
1705     differing field-values or a single Content-Length header field having an
1706     invalid value, then the message framing is invalid and &MUST; be treated
1707     as an error to prevent request or response smuggling.
1708     If this is a request message, the server &MUST; respond with
1709     a <x:ref>400 (Bad Request)</x:ref> status code and then close the connection.
1710     If this is a response message received by a proxy, the proxy
1711     &MUST; discard the received response, send a <x:ref>502 (Bad Gateway)</x:ref>
1712     status code as its downstream response, and then close the connection.
1713     If this is a response message received by a user agent, it &MUST; be
1714     treated as an error by discarding the message and closing the connection.
1715    </t></x:lt>
1716    <x:lt><t>
1717     If a valid <x:ref>Content-Length</x:ref> header field is present without
1718     <x:ref>Transfer-Encoding</x:ref>, its decimal value defines the
1719     expected message body length in octets.
1720     If the sender closes the connection or the recipient times out before the
1721     indicated number of octets are received, the recipient &MUST; consider
1722     the message to be incomplete and close the connection.
1723    </t></x:lt>
1724    <x:lt><t>
1725     If this is a request message and none of the above are true, then the
1726     message body length is zero (no message body is present).
1727    </t></x:lt>
1728    <x:lt><t>
1729     Otherwise, this is a response message without a declared message body
1730     length, so the message body length is determined by the number of octets
1731     received prior to the server closing the connection.
1732    </t></x:lt>
1733  </list>
1734</t>
1735<t>
1736   Since there is no way to distinguish a successfully completed,
1737   close-delimited message from a partially-received message interrupted
1738   by network failure, a server &SHOULD; use encoding or
1739   length-delimited messages whenever possible.  The close-delimiting
1740   feature exists primarily for backwards compatibility with HTTP/1.0.
1741</t>
1742<t>
1743   A server &MAY; reject a request that contains a message body but
1744   not a <x:ref>Content-Length</x:ref> by responding with
1745   <x:ref>411 (Length Required)</x:ref>.
1746</t>
1747<t>
1748   Unless a transfer coding other than chunked has been applied,
1749   a client that sends a request containing a message body &SHOULD;
1750   use a valid <x:ref>Content-Length</x:ref> header field if the message body
1751   length is known in advance, rather than the chunked transfer coding, since some
1752   existing services respond to chunked with a <x:ref>411 (Length Required)</x:ref>
1753   status code even though they understand the chunked transfer coding.  This
1754   is typically because such services are implemented via a gateway that
1755   requires a content-length in advance of being called and the server
1756   is unable or unwilling to buffer the entire request before processing.
1757</t>
1758<t>
1759   A client that sends a request containing a message body &MUST; include a
1760   valid <x:ref>Content-Length</x:ref> header field if it does not know the
1761   server will handle HTTP/1.1 (or later) requests; such knowledge can be in
1762   the form of specific user configuration or by remembering the version of a
1763   prior received response.
1764</t>
1765<t>
1766   If the final response to the last request on a connection has been
1767   completely received and there remains additional data to read, a user agent
1768   &MAY; discard the remaining data or attempt to determine if that data
1769   belongs as part of the prior response body, which might be the case if the
1770   prior message's Content-Length value is incorrect. A client &MUST-NOT;
1771   process, cache, or forward such extra data as a separate response, since
1772   such behavior would be vulnerable to cache poisoning.
1773</t>
1774</section>
1775</section>
1776
1777<section anchor="incomplete.messages" title="Handling Incomplete Messages">
1778<t>
1779   A server that receives an incomplete request message, usually due to a
1780   canceled request or a triggered time-out exception, &MAY; send an error
1781   response prior to closing the connection.
1782</t>
1783<t>
1784   A client that receives an incomplete response message, which can occur
1785   when a connection is closed prematurely or when decoding a supposedly
1786   chunked transfer coding fails, &MUST; record the message as incomplete.
1787   Cache requirements for incomplete responses are defined in
1788   &cache-incomplete;.
1789</t>
1790<t>
1791   If a response terminates in the middle of the header block (before the
1792   empty line is received) and the status code might rely on header fields to
1793   convey the full meaning of the response, then the client cannot assume
1794   that meaning has been conveyed; the client might need to repeat the
1795   request in order to determine what action to take next.
1796</t>
1797<t>
1798   A message body that uses the chunked transfer coding is
1799   incomplete if the zero-sized chunk that terminates the encoding has not
1800   been received.  A message that uses a valid <x:ref>Content-Length</x:ref> is
1801   incomplete if the size of the message body received (in octets) is less than
1802   the value given by Content-Length.  A response that has neither chunked
1803   transfer coding nor Content-Length is terminated by closure of the
1804   connection, and thus is considered complete regardless of the number of
1805   message body octets received, provided that the header block was received
1806   intact.
1807</t>
1808</section>
1809
1810<section title="Message Parsing Robustness" anchor="message.robustness">
1811<t>
1812   Older HTTP/1.0 user agent implementations might send an extra CRLF
1813   after a POST request as a lame workaround for some early server
1814   applications that failed to read message body content that was
1815   not terminated by a line-ending. An HTTP/1.1 user agent &MUST-NOT;
1816   preface or follow a request with an extra CRLF.  If terminating
1817   the request message body with a line-ending is desired, then the
1818   user agent &MUST; include the terminating CRLF octets as part of the
1819   message body length.
1820</t>
1821<t>
1822   In the interest of robustness, servers &SHOULD; ignore at least one
1823   empty line received where a request-line is expected. In other words, if
1824   a server is reading the protocol stream at the beginning of a
1825   message and receives a CRLF first, the server &SHOULD; ignore the CRLF.
1826</t>
1827<t>
1828   Although the line terminator for the start-line and header
1829   fields is the sequence CRLF, recipients &MAY; recognize a
1830   single LF as a line terminator and ignore any preceding CR.
1831</t>
1832<t>
1833   Although the request-line and status-line grammar rules require that each
1834   of the component elements be separated by a single SP octet, recipients
1835   &MAY; instead parse on whitespace-delimited word boundaries and, aside
1836   from the CRLF terminator, treat any form of whitespace as the SP separator
1837   while ignoring preceding or trailing whitespace;
1838   such whitespace includes one or more of the following octets:
1839   SP, HTAB, VT (%x0B), FF (%x0C), or bare CR.
1840</t>
1841<t>
1842   When a server listening only for HTTP request messages, or processing
1843   what appears from the start-line to be an HTTP request message,
1844   receives a sequence of octets that does not match the HTTP-message
1845   grammar aside from the robustness exceptions listed above, the
1846   server &SHOULD; respond with a <x:ref>400 (Bad Request)</x:ref> response. 
1847</t>
1848</section>
1849</section>
1850
1851<section title="Transfer Codings" anchor="transfer.codings">
1852  <x:anchor-alias value="transfer-coding"/>
1853  <x:anchor-alias value="transfer-extension"/>
1854<t>
1855   Transfer coding names are used to indicate an encoding
1856   transformation that has been, can be, or might need to be applied to a
1857   payload body in order to ensure "safe transport" through the network.
1858   This differs from a content coding in that the transfer coding is a
1859   property of the message rather than a property of the representation
1860   that is being transferred.
1861</t>
1862<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/>
1863  <x:ref>transfer-coding</x:ref>    = "chunked" ; <xref target="chunked.encoding"/>
1864                     / "compress" ; <xref target="compress.coding"/>
1865                     / "deflate" ; <xref target="deflate.coding"/>
1866                     / "gzip" ; <xref target="gzip.coding"/>
1867                     / <x:ref>transfer-extension</x:ref>
1868  <x:ref>transfer-extension</x:ref> = <x:ref>token</x:ref> *( <x:ref>OWS</x:ref> ";" <x:ref>OWS</x:ref> <x:ref>transfer-parameter</x:ref> )
1869</artwork></figure>
1870<t anchor="rule.parameter">
1871  <x:anchor-alias value="attribute"/>
1872  <x:anchor-alias value="transfer-parameter"/>
1873  <x:anchor-alias value="value"/>
1874   Parameters are in the form of attribute/value pairs.
1875</t>
1876<figure><artwork type="abnf2616"><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"/>
1877  <x:ref>transfer-parameter</x:ref> = <x:ref>attribute</x:ref> <x:ref>BWS</x:ref> "=" <x:ref>BWS</x:ref> <x:ref>value</x:ref>
1878  <x:ref>attribute</x:ref>          = <x:ref>token</x:ref>
1879  <x:ref>value</x:ref>              = <x:ref>word</x:ref>
1880</artwork></figure>
1881<t>
1882   All transfer-coding names are case-insensitive and &SHOULD; be registered
1883   within the HTTP Transfer Coding registry, as defined in
1884   <xref target="transfer.coding.registry"/>.
1885   They are used in the <x:ref>TE</x:ref> (<xref target="header.te"/>) and
1886   <x:ref>Transfer-Encoding</x:ref> (<xref target="header.transfer-encoding"/>)
1887   header fields.
1888</t>
1889
1890<section title="Chunked Transfer Coding" anchor="chunked.encoding">
1891  <iref primary="true" item="chunked (Coding Format)"/>
1892  <x:anchor-alias value="chunk"/>
1893  <x:anchor-alias value="chunked-body"/>
1894  <x:anchor-alias value="chunk-data"/>
1895  <x:anchor-alias value="chunk-ext"/>
1896  <x:anchor-alias value="chunk-ext-name"/>
1897  <x:anchor-alias value="chunk-ext-val"/>
1898  <x:anchor-alias value="chunk-size"/>
1899  <x:anchor-alias value="last-chunk"/>
1900  <x:anchor-alias value="trailer-part"/>
1901  <x:anchor-alias value="quoted-str-nf"/>
1902  <x:anchor-alias value="qdtext-nf"/>
1903<t>
1904   The chunked transfer coding modifies the body of a message in order to
1905   transfer it as a series of chunks, each with its own size indicator,
1906   followed by an &OPTIONAL; trailer containing header fields. This
1907   allows dynamically generated content to be transferred along with the
1908   information necessary for the recipient to verify that it has
1909   received the full message.
1910</t>
1911<figure><artwork type="abnf2616"><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"/>
1912  <x:ref>chunked-body</x:ref>   = *<x:ref>chunk</x:ref>
1913                   <x:ref>last-chunk</x:ref>
1914                   <x:ref>trailer-part</x:ref>
1915                   <x:ref>CRLF</x:ref>
1916 
1917  <x:ref>chunk</x:ref>          = <x:ref>chunk-size</x:ref> [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>
1918                   <x:ref>chunk-data</x:ref> <x:ref>CRLF</x:ref>
1919  <x:ref>chunk-size</x:ref>     = 1*<x:ref>HEXDIG</x:ref>
1920  <x:ref>last-chunk</x:ref>     = 1*("0") [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>
1921 
1922  <x:ref>chunk-ext</x:ref>      = *( ";" <x:ref>chunk-ext-name</x:ref> [ "=" <x:ref>chunk-ext-val</x:ref> ] )
1923  <x:ref>chunk-ext-name</x:ref> = <x:ref>token</x:ref>
1924  <x:ref>chunk-ext-val</x:ref>  = <x:ref>token</x:ref> / <x:ref>quoted-str-nf</x:ref>
1925  <x:ref>chunk-data</x:ref>     = 1*<x:ref>OCTET</x:ref> ; a sequence of chunk-size octets
1926  <x:ref>trailer-part</x:ref>   = *( <x:ref>header-field</x:ref> <x:ref>CRLF</x:ref> )
1927 
1928  <x:ref>quoted-str-nf</x:ref>  = <x:ref>DQUOTE</x:ref> *( <x:ref>qdtext-nf</x:ref> / <x:ref>quoted-pair</x:ref> ) <x:ref>DQUOTE</x:ref>
1929                 ; like <x:ref>quoted-string</x:ref>, but disallowing line folding
1930  <x:ref>qdtext-nf</x:ref>      = <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / %x21 / %x23-5B / %x5D-7E / <x:ref>obs-text</x:ref>
1931</artwork></figure>
1932<t>
1933   Chunk extensions within the chunked transfer coding are deprecated.
1934   Senders &SHOULD-NOT; send chunk-ext.
1935   Definition of new chunk extensions is discouraged.
1936</t>
1937<t>
1938   The chunk-size field is a string of hex digits indicating the size of
1939   the chunk-data in octets. The chunked transfer coding is complete when a
1940   chunk with a chunk-size of zero is received, possibly followed by a
1941   trailer, and finally terminated by an empty line.
1942</t>
1943
1944<section title="Trailer" anchor="header.trailer">
1945  <iref primary="true" item="Trailer header field" x:for-anchor=""/>
1946  <x:anchor-alias value="Trailer"/>
1947<t>
1948   A trailer allows the sender to include additional fields at the end of a
1949   chunked message in order to supply metadata that might be dynamically
1950   generated while the message body is sent, such as a message integrity
1951   check, digital signature, or post-processing status.
1952   The trailer &MUST-NOT; contain fields that need to be known before a
1953   recipient processes the body, such as <x:ref>Transfer-Encoding</x:ref>,
1954   <x:ref>Content-Length</x:ref>, and <x:ref>Trailer</x:ref>.
1955</t>
1956<t>
1957   When a message includes a message body encoded with the chunked
1958   transfer coding and the sender desires to send metadata in the form of
1959   trailer fields at the end of the message, the sender &SHOULD; send a
1960   <x:ref>Trailer</x:ref> header field before the message body to indicate
1961   which fields will be present in the trailers. This allows the recipient
1962   to prepare for receipt of that metadata before it starts processing the body,
1963   which is useful if the message is being streamed and the recipient wishes
1964   to confirm an integrity check on the fly.
1965</t>
1966<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Trailer"/>
1967  <x:ref>Trailer</x:ref> = 1#<x:ref>field-name</x:ref>
1968</artwork></figure>
1969<t>
1970   If no <x:ref>Trailer</x:ref> header field is present, the sender of a
1971   chunked message body &SHOULD; send an empty trailer.
1972</t>
1973<t>
1974   A server &MUST; send an empty trailer with the chunked transfer coding
1975   unless at least one of the following is true:
1976  <list style="numbers">
1977    <t>the request included a <x:ref>TE</x:ref> header field that indicates
1978    "trailers" is acceptable in the transfer coding of the response, as
1979    described in <xref target="header.te"/>; or,</t>
1980     
1981    <t>the trailer fields consist entirely of optional metadata and the
1982    recipient could use the message (in a manner acceptable to the server where
1983    the field originated) without receiving that metadata. In other words,
1984    the server that generated the header field is willing to accept the
1985    possibility that the trailer fields might be silently discarded along
1986    the path to the client.</t>
1987  </list>
1988</t>
1989<t>
1990   The above requirement prevents the need for an infinite buffer when a
1991   message is being received by an HTTP/1.1 (or later) proxy and forwarded to
1992   an HTTP/1.0 recipient.
1993</t>
1994</section>
1995
1996<section title="Decoding chunked" anchor="decoding.chunked">
1997<t>
1998   A process for decoding the chunked transfer coding
1999   can be represented in pseudo-code as:
2000</t>
2001<figure><artwork type="code">
2002  length := 0
2003  read chunk-size, chunk-ext (if any) and CRLF
2004  while (chunk-size &gt; 0) {
2005     read chunk-data and CRLF
2006     append chunk-data to decoded-body
2007     length := length + chunk-size
2008     read chunk-size and CRLF
2009  }
2010  read header-field
2011  while (header-field not empty) {
2012     append header-field to existing header fields
2013     read header-field
2014  }
2015  Content-Length := length
2016  Remove "chunked" from Transfer-Encoding
2017  Remove Trailer from existing header fields
2018</artwork></figure>
2019<t>
2020   All recipients &MUST; be able to receive and decode the
2021   chunked transfer coding and &MUST; ignore chunk-ext extensions
2022   they do not understand.
2023</t>
2024</section>
2025</section>
2026
2027<section title="Compression Codings" anchor="compression.codings">
2028<t>
2029   The codings defined below can be used to compress the payload of a
2030   message.
2031</t>
2032
2033<section title="Compress Coding" anchor="compress.coding">
2034<iref item="compress (Coding Format)"/>
2035<t>
2036   The "compress" format is produced by the common UNIX file compression
2037   program "compress". This format is an adaptive Lempel-Ziv-Welch
2038   coding (LZW). Recipients &SHOULD; consider "x-compress" to be
2039   equivalent to "compress".
2040</t>
2041</section>
2042
2043<section title="Deflate Coding" anchor="deflate.coding">
2044<iref item="deflate (Coding Format)"/>
2045<t>
2046   The "deflate" format is defined as the "deflate" compression mechanism
2047   (described in <xref target="RFC1951"/>) used inside the "zlib"
2048   data format (<xref target="RFC1950"/>).
2049</t>
2050<x:note>
2051  <t>
2052    &Note; Some incorrect implementations send the "deflate"
2053    compressed data without the zlib wrapper.
2054   </t>
2055</x:note>
2056</section>
2057
2058<section title="Gzip Coding" anchor="gzip.coding">
2059<iref item="gzip (Coding Format)"/>
2060<t>
2061   The "gzip" format is produced by the file compression program
2062   "gzip" (GNU zip), as described in <xref target="RFC1952"/>. This format is a
2063   Lempel-Ziv coding (LZ77) with a 32 bit CRC.
2064   Recipients &SHOULD; consider "x-gzip" to be equivalent to "gzip".
2065</t>
2066</section>
2067
2068</section>
2069
2070<section title="TE" anchor="header.te">
2071  <iref primary="true" item="TE header field" x:for-anchor=""/>
2072  <x:anchor-alias value="TE"/>
2073  <x:anchor-alias value="t-codings"/>
2074  <x:anchor-alias value="t-ranking"/>
2075  <x:anchor-alias value="rank"/>
2076<t>
2077   The "TE" header field in a request indicates what transfer codings,
2078   besides chunked, the client is willing to accept in response, and
2079   whether or not the client is willing to accept trailer fields in a
2080   chunked transfer coding.
2081</t>
2082<t>
2083   The TE field-value consists of a comma-separated list of transfer coding
2084   names, each allowing for optional parameters (as described in
2085   <xref target="transfer.codings"/>), and/or the keyword "trailers".
2086   Clients &MUST-NOT; send the chunked transfer coding name in TE;
2087   chunked is always acceptable for HTTP/1.1 recipients.
2088</t>
2089<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="t-codings"/><iref primary="true" item="Grammar" subitem="t-ranking"/><iref primary="true" item="Grammar" subitem="rank"/>
2090  <x:ref>TE</x:ref>        = #<x:ref>t-codings</x:ref>
2091  <x:ref>t-codings</x:ref> = "trailers" / ( <x:ref>transfer-coding</x:ref> [ <x:ref>t-ranking</x:ref> ] )
2092  <x:ref>t-ranking</x:ref> = <x:ref>OWS</x:ref> ";" <x:ref>OWS</x:ref> "q=" <x:ref>rank</x:ref>
2093  <x:ref>rank</x:ref>      = ( "0" [ "." 0*3<x:ref>DIGIT</x:ref> ] )
2094             / ( "1" [ "." 0*3("0") ] )
2095</artwork></figure>
2096<t>
2097   Three examples of TE use are below.
2098</t>
2099<figure><artwork type="example">
2100  TE: deflate
2101  TE:
2102  TE: trailers, deflate;q=0.5
2103</artwork></figure>
2104<t>
2105   The presence of the keyword "trailers" indicates that the client is
2106   willing to accept trailer fields in a chunked transfer coding,
2107   as defined in <xref target="chunked.encoding"/>, on behalf of itself and
2108   any downstream clients. For chained requests, this implies that either:
2109   (a) all downstream clients are willing to accept trailer fields in the
2110   forwarded response; or,
2111   (b) the client will attempt to buffer the response on behalf of downstream
2112   recipients.
2113   Note that HTTP/1.1 does not define any means to limit the size of a
2114   chunked response such that a client can be assured of buffering the
2115   entire response.
2116</t>
2117<t>
2118   When multiple transfer codings are acceptable, the client &MAY; rank the
2119   codings by preference using a case-insensitive "q" parameter (similar to
2120   the qvalues used in content negotiation fields, &qvalue;). The rank value
2121   is a real number in the range 0 through 1, where 0.001 is the least
2122   preferred and 1 is the most preferred; a value of 0 means "not acceptable".
2123</t>
2124<t>
2125   If the TE field-value is empty or if no TE field is present, the only
2126   acceptable transfer coding is chunked. A message with no transfer coding
2127   is always acceptable.
2128</t>
2129<t>
2130   Since the TE header field only applies to the immediate connection,
2131   a sender of TE &MUST; also send a "TE" connection option within the
2132   <x:ref>Connection</x:ref> header field (<xref target="header.connection"/>)
2133   in order to prevent the TE field from being forwarded by intermediaries
2134   that do not support its semantics.
2135</t>
2136</section>
2137</section>
2138
2139<section title="Message Routing" anchor="message.routing">
2140<t>
2141   HTTP request message routing is determined by each client based on the
2142   target resource, the client's proxy configuration, and
2143   establishment or reuse of an inbound connection.  The corresponding
2144   response routing follows the same connection chain back to the client.
2145</t>
2146
2147<section title="Identifying a Target Resource" anchor="target-resource">
2148  <iref primary="true" item="target resource"/>
2149  <iref primary="true" item="target URI"/>
2150  <x:anchor-alias value="target resource"/>
2151  <x:anchor-alias value="target URI"/>
2152<t>
2153   HTTP is used in a wide variety of applications, ranging from
2154   general-purpose computers to home appliances.  In some cases,
2155   communication options are hard-coded in a client's configuration.
2156   However, most HTTP clients rely on the same resource identification
2157   mechanism and configuration techniques as general-purpose Web browsers.
2158</t>
2159<t>
2160   HTTP communication is initiated by a user agent for some purpose.
2161   The purpose is a combination of request semantics, which are defined in
2162   <xref target="Part2"/>, and a target resource upon which to apply those
2163   semantics.  A URI reference (<xref target="uri"/>) is typically used as
2164   an identifier for the "<x:dfn>target resource</x:dfn>", which a user agent
2165   would resolve to its absolute form in order to obtain the
2166   "<x:dfn>target URI</x:dfn>".  The target URI
2167   excludes the reference's fragment identifier component, if any,
2168   since fragment identifiers are reserved for client-side processing
2169   (<xref target="RFC3986" x:fmt="," x:sec="3.5"/>).
2170</t>
2171</section>
2172
2173<section title="Connecting Inbound" anchor="connecting.inbound">
2174<t>
2175   Once the target URI is determined, a client needs to decide whether
2176   a network request is necessary to accomplish the desired semantics and,
2177   if so, where that request is to be directed.
2178</t>
2179<t>
2180   If the client has a response cache and the request semantics can be
2181   satisfied by a cache (<xref target="Part6"/>), then the request is
2182   usually directed to the cache first.
2183</t>
2184<t>
2185   If the request is not satisfied by a cache, then a typical client will
2186   check its configuration to determine whether a proxy is to be used to
2187   satisfy the request.  Proxy configuration is implementation-dependent,
2188   but is often based on URI prefix matching, selective authority matching,
2189   or both, and the proxy itself is usually identified by an "http" or
2190   "https" URI.  If a proxy is applicable, the client connects inbound by
2191   establishing (or reusing) a connection to that proxy.
2192</t>
2193<t>
2194   If no proxy is applicable, a typical client will invoke a handler routine,
2195   usually specific to the target URI's scheme, to connect directly
2196   to an authority for the target resource.  How that is accomplished is
2197   dependent on the target URI scheme and defined by its associated
2198   specification, similar to how this specification defines origin server
2199   access for resolution of the "http" (<xref target="http.uri"/>) and
2200   "https" (<xref target="https.uri"/>) schemes.
2201</t>
2202<t>
2203   HTTP requirements regarding connection management are defined in
2204   <xref target="connection.management"/>.
2205</t>
2206</section>
2207
2208<section title="Request Target" anchor="request-target">
2209<t>
2210   Once an inbound connection is obtained,
2211   the client sends an HTTP request message (<xref target="http.message"/>)
2212   with a request-target derived from the target URI.
2213   There are four distinct formats for the request-target, depending on both
2214   the method being requested and whether the request is to a proxy.
2215</t>   
2216<figure><artwork type="abnf2616"><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"/>
2217  <x:ref>request-target</x:ref> = <x:ref>origin-form</x:ref>
2218                 / <x:ref>absolute-form</x:ref>
2219                 / <x:ref>authority-form</x:ref>
2220                 / <x:ref>asterisk-form</x:ref>
2221
2222  <x:ref>origin-form</x:ref>    = <x:ref>path-absolute</x:ref> [ "?" <x:ref>query</x:ref> ]
2223  <x:ref>absolute-form</x:ref>  = <x:ref>absolute-URI</x:ref>
2224  <x:ref>authority-form</x:ref> = <x:ref>authority</x:ref>
2225  <x:ref>asterisk-form</x:ref>  = "*"
2226</artwork></figure>
2227<t anchor="origin-form"><iref item="origin-form (of request-target)"/>
2228   The most common form of request-target is the origin-form.
2229   When making a request directly to an origin server, other than a CONNECT
2230   or server-wide OPTIONS request (as detailed below),
2231   a client &MUST; send only the absolute path and query components of
2232   the target URI as the request-target.
2233   If the target URI's path component is empty, then the client &MUST; send
2234   "/" as the path within the origin-form of request-target.
2235   A <x:ref>Host</x:ref> header field is also sent, as defined in
2236   <xref target="header.host"/>, containing the target URI's
2237   authority component (excluding any userinfo).
2238</t>
2239<t>
2240   For example, a client wishing to retrieve a representation of the resource
2241   identified as
2242</t>
2243<figure><artwork x:indent-with="  " type="example">
2244http://www.example.org/where?q=now
2245</artwork></figure>
2246<t>
2247   directly from the origin server would open (or reuse) a TCP connection
2248   to port 80 of the host "www.example.org" and send the lines:
2249</t>
2250<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2251GET /where?q=now HTTP/1.1
2252Host: www.example.org
2253</artwork></figure>
2254<t>
2255   followed by the remainder of the request message.
2256</t>
2257<t anchor="absolute-form"><iref item="absolute-form (of request-target)"/>
2258   When making a request to a proxy, other than a CONNECT or server-wide
2259   OPTIONS request (as detailed below), a client &MUST; send the target URI
2260   in absolute-form as the request-target.
2261   The proxy is requested to either service that request from a valid cache,
2262   if possible, or make the same request on the client's behalf to either
2263   the next inbound proxy server or directly to the origin server indicated
2264   by the request-target.  Requirements on such "forwarding" of messages are
2265   defined in <xref target="message.forwarding"/>.
2266</t>
2267<t>
2268   An example absolute-form of request-line would be:
2269</t>
2270<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2271GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1
2272</artwork></figure>
2273<t>
2274   To allow for transition to the absolute-form for all requests in some
2275   future version of HTTP, HTTP/1.1 servers &MUST; accept the absolute-form
2276   in requests, even though HTTP/1.1 clients will only send them in requests
2277   to proxies.
2278</t>
2279<t anchor="authority-form"><iref item="authority-form (of request-target)"/>
2280   The authority-form of request-target is only used for CONNECT requests
2281   (&CONNECT;).  When making a CONNECT request to establish a tunnel through
2282   one or more proxies, a client &MUST; send only the target URI's
2283   authority component (excluding any userinfo) as the request-target.
2284   For example,
2285</t>
2286<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2287CONNECT www.example.com:80 HTTP/1.1
2288</artwork></figure>
2289<t anchor="asterisk-form"><iref item="asterisk-form (of request-target)"/>
2290   The asterisk-form of request-target is only used for a server-wide
2291   OPTIONS request (&OPTIONS;).  When a client wishes to request OPTIONS
2292   for the server as a whole, as opposed to a specific named resource of
2293   that server, the client &MUST; send only "*" (%x2A) as the request-target.
2294   For example,
2295</t>
2296<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2297OPTIONS * HTTP/1.1
2298</artwork></figure>
2299<t>
2300   If a proxy receives an OPTIONS request with an absolute-form of
2301   request-target in which the URI has an empty path and no query component,
2302   then the last proxy on the request chain &MUST; send a request-target
2303   of "*" when it forwards the request to the indicated origin server.
2304</t>
2305<figure><preamble>   
2306   For example, the request
2307</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2308OPTIONS http://www.example.org:8001 HTTP/1.1
2309</artwork></figure>
2310<figure><preamble>   
2311  would be forwarded by the final proxy as
2312</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2313OPTIONS * HTTP/1.1
2314Host: www.example.org:8001
2315</artwork>
2316<postamble>
2317   after connecting to port 8001 of host "www.example.org".
2318</postamble>
2319</figure>
2320</section>
2321
2322<section title="Host" anchor="header.host">
2323  <iref primary="true" item="Host header field" x:for-anchor=""/>
2324  <x:anchor-alias value="Host"/>
2325<t>
2326   The "Host" header field in a request provides the host and port
2327   information from the target URI, enabling the origin
2328   server to distinguish among resources while servicing requests
2329   for multiple host names on a single IP address.  Since the Host
2330   field-value is critical information for handling a request, it
2331   &SHOULD; be sent as the first header field following the request-line.
2332</t>
2333<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Host"/>
2334  <x:ref>Host</x:ref> = <x:ref>uri-host</x:ref> [ ":" <x:ref>port</x:ref> ] ; <xref target="http.uri"/>
2335</artwork></figure>
2336<t>
2337   A client &MUST; send a Host header field in all HTTP/1.1 request
2338   messages.  If the target URI includes an authority component, then
2339   the Host field-value &MUST; be identical to that authority component
2340   after excluding any userinfo (<xref target="http.uri"/>).
2341   If the authority component is missing or undefined for the target URI,
2342   then the Host header field &MUST; be sent with an empty field-value.
2343</t>
2344<t>
2345   For example, a GET request to the origin server for
2346   &lt;http://www.example.org/pub/WWW/&gt; would begin with:
2347</t>
2348<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
2349GET /pub/WWW/ HTTP/1.1
2350Host: www.example.org
2351</artwork></figure>
2352<t>
2353   The Host header field &MUST; be sent in an HTTP/1.1 request even
2354   if the request-target is in the absolute-form, since this
2355   allows the Host information to be forwarded through ancient HTTP/1.0
2356   proxies that might not have implemented Host.
2357</t>
2358<t>
2359   When a proxy receives a request with an absolute-form of
2360   request-target, the proxy &MUST; ignore the received
2361   Host header field (if any) and instead replace it with the host
2362   information of the request-target.  If the proxy forwards the request,
2363   it &MUST; generate a new Host field-value based on the received
2364   request-target rather than forward the received Host field-value.
2365</t>
2366<t>
2367   Since the Host header field acts as an application-level routing
2368   mechanism, it is a frequent target for malware seeking to poison
2369   a shared cache or redirect a request to an unintended server.
2370   An interception proxy is particularly vulnerable if it relies on
2371   the Host field-value for redirecting requests to internal
2372   servers, or for use as a cache key in a shared cache, without
2373   first verifying that the intercepted connection is targeting a
2374   valid IP address for that host.
2375</t>
2376<t>
2377   A server &MUST; respond with a <x:ref>400 (Bad Request)</x:ref> status code
2378   to any HTTP/1.1 request message that lacks a Host header field and
2379   to any request message that contains more than one Host header field
2380   or a Host header field with an invalid field-value.
2381</t>
2382</section>
2383
2384<section title="Effective Request URI" anchor="effective.request.uri">
2385  <iref primary="true" item="effective request URI"/>
2386<t>
2387   A server that receives an HTTP request message &MUST; reconstruct
2388   the user agent's original target URI, based on the pieces of information
2389   learned from the request-target, <x:ref>Host</x:ref> header field, and
2390   connection context, in order to identify the intended target resource and
2391   properly service the request. The URI derived from this reconstruction
2392   process is referred to as the "<x:dfn>effective request URI</x:dfn>".
2393</t>
2394<t>
2395   For a user agent, the effective request URI is the target URI.
2396</t>
2397<t>
2398   If the request-target is in absolute-form, then the effective request URI
2399   is the same as the request-target.  Otherwise, the effective request URI
2400   is constructed as follows.
2401</t>
2402<t>
2403   If the request is received over a TLS-secured TCP connection,
2404   then the effective request URI's scheme is "https"; otherwise, the
2405   scheme is "http".
2406</t>
2407<t>
2408   If the request-target is in authority-form, then the effective
2409   request URI's authority component is the same as the request-target.
2410   Otherwise, if a <x:ref>Host</x:ref> header field is supplied with a
2411   non-empty field-value, then the authority component is the same as the
2412   Host field-value. Otherwise, the authority component is the concatenation of
2413   the default host name configured for the server, a colon (":"), and the
2414   connection's incoming TCP port number in decimal form.
2415</t>
2416<t>
2417   If the request-target is in authority-form or asterisk-form, then the
2418   effective request URI's combined path and query component is empty.
2419   Otherwise, the combined path and query component is the same as the
2420   request-target.
2421</t>
2422<t>
2423   The components of the effective request URI, once determined as above,
2424   can be combined into absolute-URI form by concatenating the scheme,
2425   "://", authority, and combined path and query component.
2426</t>
2427<figure>
2428<preamble>
2429   Example 1: the following message received over an insecure TCP connection
2430</preamble> 
2431<artwork type="example" x:indent-with="  ">
2432GET /pub/WWW/TheProject.html HTTP/1.1
2433Host: www.example.org:8080
2434</artwork>
2435</figure>
2436<figure>
2437<preamble>
2438  has an effective request URI of
2439</preamble>
2440<artwork type="example" x:indent-with="  ">
2441http://www.example.org:8080/pub/WWW/TheProject.html
2442</artwork>
2443</figure>
2444<figure>
2445<preamble>
2446   Example 2: the following message received over a TLS-secured TCP connection
2447</preamble> 
2448<artwork type="example" x:indent-with="  ">
2449OPTIONS * HTTP/1.1
2450Host: www.example.org
2451</artwork>
2452</figure>
2453<figure>
2454<preamble>
2455  has an effective request URI of
2456</preamble>
2457<artwork type="example" x:indent-with="  ">
2458https://www.example.org
2459</artwork>
2460</figure>
2461<t>
2462   An origin server that does not allow resources to differ by requested
2463   host &MAY; ignore the <x:ref>Host</x:ref> field-value and instead replace it
2464   with a configured server name when constructing the effective request URI.
2465</t>
2466<t>
2467   Recipients of an HTTP/1.0 request that lacks a <x:ref>Host</x:ref> header
2468   field &MAY; attempt to use heuristics (e.g., examination of the URI path for
2469   something unique to a particular host) in order to guess the
2470   effective request URI's authority component.
2471</t>
2472</section>
2473
2474<section title="Associating a Response to a Request" anchor="associating.response.to.request">
2475<t>
2476   HTTP does not include a request identifier for associating a given
2477   request message with its corresponding one or more response messages.
2478   Hence, it relies on the order of response arrival to correspond exactly
2479   to the order in which requests are made on the same connection.
2480   More than one response message per request only occurs when one or more
2481   informational responses (<x:ref>1xx</x:ref>, see &status-1xx;) precede a
2482   final response to the same request.
2483</t>
2484<t>
2485   A client that has more than one outstanding request on a connection &MUST;
2486   maintain a list of outstanding requests in the order sent and &MUST;
2487   associate each received response message on that connection to the highest
2488   ordered request that has not yet received a final (non-<x:ref>1xx</x:ref>)
2489   response.
2490</t>
2491</section>
2492
2493<section title="Message Forwarding" anchor="message.forwarding">
2494<t>
2495   As described in <xref target="intermediaries"/>, intermediaries can serve
2496   a variety of roles in the processing of HTTP requests and responses.
2497   Some intermediaries are used to improve performance or availability.
2498   Others are used for access control or to filter content.
2499   Since an HTTP stream has characteristics similar to a pipe-and-filter
2500   architecture, there are no inherent limits to the extent an intermediary
2501   can enhance (or interfere) with either direction of the stream.
2502</t>
2503<t>
2504   Intermediaries that forward a message &MUST; implement the
2505   <x:ref>Connection</x:ref> header field, as specified in
2506   <xref target="header.connection"/>, to exclude fields that are only
2507   intended for the incoming connection.
2508</t>
2509<t>
2510   In order to avoid request loops, a proxy that forwards requests to other
2511   proxies &MUST; be able to recognize and exclude all of its own server
2512   names, including any aliases, local variations, or literal IP addresses.
2513</t>
2514
2515<section title="Via" anchor="header.via">
2516  <iref primary="true" item="Via header field" x:for-anchor=""/>
2517  <x:anchor-alias value="pseudonym"/>
2518  <x:anchor-alias value="received-by"/>
2519  <x:anchor-alias value="received-protocol"/>
2520  <x:anchor-alias value="Via"/>
2521<t>
2522   The "Via" header field &MUST; be sent by a proxy or gateway in forwarded
2523   messages to indicate the intermediate protocols and recipients between the
2524   user agent and the server on requests, and between the origin server and
2525   the client on responses. It is analogous to the "Received" field
2526   used by email systems (<xref target="RFC5322" x:fmt="of" x:sec="3.6.7"/>).
2527   Via is used in HTTP for tracking message forwards,
2528   avoiding request loops, and identifying the protocol capabilities of
2529   all senders along the request/response chain.
2530</t>
2531<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Via"/><iref primary="true" item="Grammar" subitem="received-protocol"/><iref primary="true" item="Grammar" subitem="protocol-name"/><iref primary="true" item="Grammar" subitem="protocol-version"/><iref primary="true" item="Grammar" subitem="received-by"/><iref primary="true" item="Grammar" subitem="pseudonym"/>
2532  <x:ref>Via</x:ref>               = 1#( <x:ref>received-protocol</x:ref> <x:ref>RWS</x:ref> <x:ref>received-by</x:ref>
2533                          [ <x:ref>RWS</x:ref> <x:ref>comment</x:ref> ] )
2534  <x:ref>received-protocol</x:ref> = [ <x:ref>protocol-name</x:ref> "/" ] <x:ref>protocol-version</x:ref>
2535                      ; see <xref target="header.upgrade"/>
2536  <x:ref>received-by</x:ref>       = ( <x:ref>uri-host</x:ref> [ ":" <x:ref>port</x:ref> ] ) / <x:ref>pseudonym</x:ref>
2537  <x:ref>pseudonym</x:ref>         = <x:ref>token</x:ref>
2538</artwork></figure>
2539<t>
2540   The received-protocol indicates the protocol version of the message
2541   received by the server or client along each segment of the
2542   request/response chain. The received-protocol version is appended to
2543   the Via field value when the message is forwarded so that information
2544   about the protocol capabilities of upstream applications remains
2545   visible to all recipients.
2546</t>
2547<t>
2548   The protocol-name is excluded if and only if it would be "HTTP". The
2549   received-by field is normally the host and optional port number of a
2550   recipient server or client that subsequently forwarded the message.
2551   However, if the real host is considered to be sensitive information,
2552   it &MAY; be replaced by a pseudonym. If the port is not given, it &MAY;
2553   be assumed to be the default port of the received-protocol.
2554</t>
2555<t>
2556   Multiple Via field values represent each proxy or gateway that has
2557   forwarded the message. Each recipient &MUST; append its information
2558   such that the end result is ordered according to the sequence of
2559   forwarding applications.
2560</t>
2561<t>
2562   Comments &MAY; be used in the Via header field to identify the software
2563   of each recipient, analogous to the <x:ref>User-Agent</x:ref> and
2564   <x:ref>Server</x:ref> header fields. However, all comments in the Via field
2565   are optional and &MAY; be removed by any recipient prior to forwarding the
2566   message.
2567</t>
2568<t>
2569   For example, a request message could be sent from an HTTP/1.0 user
2570   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
2571   forward the request to a public proxy at p.example.net, which completes
2572   the request by forwarding it to the origin server at www.example.com.
2573   The request received by www.example.com would then have the following
2574   Via header field:
2575</t>
2576<figure><artwork type="example">
2577  Via: 1.0 fred, 1.1 p.example.net (Apache/1.1)
2578</artwork></figure>
2579<t>
2580   A proxy or gateway used as a portal through a network firewall
2581   &SHOULD-NOT; forward the names and ports of hosts within the firewall
2582   region unless it is explicitly enabled to do so. If not enabled, the
2583   received-by host of any host behind the firewall &SHOULD; be replaced
2584   by an appropriate pseudonym for that host.
2585</t>
2586<t>
2587   A proxy or gateway &MAY; combine an ordered subsequence of Via header
2588   field entries into a single such entry if the entries have identical
2589   received-protocol values. For example,
2590</t>
2591<figure><artwork type="example">
2592  Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
2593</artwork></figure>
2594<t>
2595  could be collapsed to
2596</t>
2597<figure><artwork type="example">
2598  Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
2599</artwork></figure>
2600<t>
2601   Senders &SHOULD-NOT; combine multiple entries unless they are all
2602   under the same organizational control and the hosts have already been
2603   replaced by pseudonyms. Senders &MUST-NOT; combine entries which
2604   have different received-protocol values.
2605</t>
2606</section>
2607
2608<section title="Transformation" anchor="message.transformation">
2609<t>
2610   If a proxy receives a request-target with a host name that is not a
2611   fully qualified domain name, it &MAY; add its own domain to the host name
2612   it received when forwarding the request.  A proxy &MUST-NOT; change the
2613   host name if it is a fully qualified domain name.
2614</t>
2615<t>
2616   A non-transforming proxy &MUST-NOT; modify the "path-absolute" and "query"
2617   parts of the received request-target when forwarding it to the next inbound
2618   server, except as noted above to replace an empty path with "/" or "*".
2619</t>
2620<t>
2621   A non-transforming proxy &MUST; preserve the message payload (&payload;),
2622   though it &MAY; change the message body through application or removal
2623   of a transfer coding (<xref target="transfer.codings"/>).
2624</t>
2625<t>
2626   A non-transforming proxy &SHOULD-NOT; modify header fields that provide
2627   information about the end points of the communication chain, the resource
2628   state, or the selected representation.
2629</t>
2630<t>
2631   A non-transforming proxy &MUST-NOT; modify any of the following fields in a
2632   request or response, and it &MUST-NOT; add any of these fields if not
2633   already present:
2634  <list style="symbols">
2635    <t><x:ref>Allow</x:ref> (&header-allow;)</t>
2636    <t><x:ref>Content-Location</x:ref> (&header-content-location;)</t>
2637    <t>Content-MD5 (<xref target="RFC2616" x:fmt="of" x:sec="14.15"/>)</t>
2638    <t><x:ref>ETag</x:ref> (&header-etag;)</t>
2639    <t><x:ref>Last-Modified</x:ref> (&header-last-modified;)</t>
2640    <t><x:ref>Server</x:ref> (&header-server;)</t>
2641  </list>
2642</t>
2643<t>
2644   A non-transforming proxy &MUST-NOT; modify an <x:ref>Expires</x:ref>
2645   header field (&header-expires;) if already present in a response, but
2646   it &MAY; add an <x:ref>Expires</x:ref> header field with a field-value
2647   identical to that of the <x:ref>Date</x:ref> header field.
2648</t>
2649<t>
2650   A proxy &MUST-NOT; modify or add any of the following fields in a
2651   message that contains the no-transform cache-control directive:
2652  <list style="symbols">
2653    <t><x:ref>Content-Encoding</x:ref> (&header-content-encoding;)</t>
2654    <t><x:ref>Content-Range</x:ref> (&header-content-range;)</t>
2655    <t><x:ref>Content-Type</x:ref> (&header-content-type;)</t>
2656  </list>
2657</t>
2658<t>
2659   A transforming proxy &MAY; modify or add these fields to a message
2660   that does not include no-transform, but if it does so, it &MUST; add a
2661   Warning 214 (Transformation applied) if one does not already appear
2662   in the message (see &header-warning;).
2663</t>
2664<x:note>
2665  <t>
2666    <x:h>Warning:</x:h> Unnecessary modification of header fields might
2667    cause authentication failures if stronger authentication
2668    mechanisms are introduced in later versions of HTTP. Such
2669    authentication mechanisms &MAY; rely on the values of header fields
2670    not listed here.
2671  </t>
2672</x:note>
2673</section>
2674</section>
2675</section>
2676
2677<section title="Connection Management" anchor="connection.management">
2678<t>
2679   HTTP messaging is independent of the underlying transport or
2680   session-layer connection protocol(s).  HTTP only presumes a reliable
2681   transport with in-order delivery of requests and the corresponding
2682   in-order delivery of responses.  The mapping of HTTP request and
2683   response structures onto the data units of an underlying transport
2684   protocol is outside the scope of this specification.
2685</t>
2686<t>
2687   As described in <xref target="connecting.inbound"/>, the specific
2688   connection protocols to be used for an HTTP interaction are determined by
2689   client configuration and the <x:ref>target URI</x:ref>.
2690   For example, the "http" URI scheme
2691   (<xref target="http.uri"/>) indicates a default connection of TCP
2692   over IP, with a default TCP port of 80, but the client might be
2693   configured to use a proxy via some other connection, port, or protocol.
2694</t>
2695<t>
2696   HTTP implementations are expected to engage in connection management,
2697   which includes maintaining the state of current connections,
2698   establishing a new connection or reusing an existing connection,
2699   processing messages received on a connection, detecting connection
2700   failures, and closing each connection.
2701   Most clients maintain multiple connections in parallel, including
2702   more than one connection per server endpoint.
2703   Most servers are designed to maintain thousands of concurrent connections,
2704   while controlling request queues to enable fair use and detect
2705   denial of service attacks.
2706</t>
2707
2708<section title="Connection" anchor="header.connection">
2709  <iref primary="true" item="Connection header field" x:for-anchor=""/>
2710  <iref primary="true" item="close" x:for-anchor=""/>
2711  <x:anchor-alias value="Connection"/>
2712  <x:anchor-alias value="connection-option"/>
2713  <x:anchor-alias value="close"/>
2714<t>
2715   The "Connection" header field allows the sender to indicate desired
2716   control options for the current connection.  In order to avoid confusing
2717   downstream recipients, a proxy or gateway &MUST; remove or replace any
2718   received connection options before forwarding the message.
2719</t>
2720<t>
2721   When a header field aside from Connection is used to supply control
2722   information for or about the current connection, the sender &MUST; list
2723   the corresponding field-name within the "Connection" header field.
2724   A proxy or gateway &MUST; parse a received Connection
2725   header field before a message is forwarded and, for each
2726   connection-option in this field, remove any header field(s) from
2727   the message with the same name as the connection-option, and then
2728   remove the Connection header field itself (or replace it with the
2729   intermediary's own connection options for the forwarded message).
2730</t>
2731<t>
2732   Hence, the Connection header field provides a declarative way of
2733   distinguishing header fields that are only intended for the
2734   immediate recipient ("hop-by-hop") from those fields that are
2735   intended for all recipients on the chain ("end-to-end"), enabling the
2736   message to be self-descriptive and allowing future connection-specific
2737   extensions to be deployed without fear that they will be blindly
2738   forwarded by older intermediaries.
2739</t>
2740<t>
2741   The Connection header field's value has the following grammar:
2742</t>
2743<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="connection-option"/>
2744  <x:ref>Connection</x:ref>        = 1#<x:ref>connection-option</x:ref>
2745  <x:ref>connection-option</x:ref> = <x:ref>token</x:ref>
2746</artwork></figure>
2747<t>
2748   Connection options are case-insensitive.
2749</t>
2750<t>
2751   A sender &MUST-NOT; include field-names in the Connection header
2752   field-value for fields that are defined as expressing constraints
2753   for all recipients in the request or response chain, such as the
2754   Cache-Control header field (&header-cache-control;).
2755</t>
2756<t>
2757   The connection options do not have to correspond to a header field
2758   present in the message, since a connection-specific header field
2759   might not be needed if there are no parameters associated with that
2760   connection option.  Recipients that trigger certain connection
2761   behavior based on the presence of connection options &MUST; do so
2762   based on the presence of the connection-option rather than only the
2763   presence of the optional header field.  In other words, if the
2764   connection option is received as a header field but not indicated
2765   within the Connection field-value, then the recipient &MUST; ignore
2766   the connection-specific header field because it has likely been
2767   forwarded by an intermediary that is only partially conformant.
2768</t>
2769<t>
2770   When defining new connection options, specifications ought to
2771   carefully consider existing deployed header fields and ensure
2772   that the new connection option does not share the same name as
2773   an unrelated header field that might already be deployed.
2774   Defining a new connection option essentially reserves that potential
2775   field-name for carrying additional information related to the
2776   connection option, since it would be unwise for senders to use
2777   that field-name for anything else.
2778</t>
2779<t>
2780   The "<x:dfn>close</x:dfn>" connection option is defined for a
2781   sender to signal that this connection will be closed after completion of
2782   the response. For example,
2783</t>
2784<figure><artwork type="example">
2785  Connection: close
2786</artwork></figure>
2787<t>
2788   in either the request or the response header fields indicates that
2789   the connection &MUST; be closed after the current request/response
2790   is complete (<xref target="persistent.tear-down"/>).
2791</t>
2792<t>
2793   A client that does not support <x:ref>persistent connections</x:ref> &MUST;
2794   send the "close" connection option in every request message.
2795</t>
2796<t>
2797   A server that does not support <x:ref>persistent connections</x:ref> &MUST;
2798   send the "close" connection option in every response message that
2799   does not have a <x:ref>1xx (Informational)</x:ref> status code.
2800</t>
2801</section>
2802
2803<section title="Establishment" anchor="persistent.establishment">
2804<t>
2805   It is beyond the scope of this specification to describe how connections
2806   are established via various transport or session-layer protocols.
2807   Each connection applies to only one transport link.
2808</t>
2809</section>
2810
2811<section title="Persistence" anchor="persistent.connections">
2812   <x:anchor-alias value="persistent connections"/>
2813<t>
2814   HTTP/1.1 defaults to the use of "<x:dfn>persistent connections</x:dfn>",
2815   which allow multiple requests and responses to be carried over a single
2816   connection. The "<x:ref>close</x:ref>" connection-option is used to signal
2817   that a connection will not persist after the current request/response.
2818   HTTP implementations &SHOULD; support persistent connections.
2819</t>
2820<t>
2821   A recipient determines whether a connection is persistent or not based on
2822   the most recently received message's protocol version and
2823   <x:ref>Connection</x:ref> header field (if any):
2824   <list style="symbols">
2825     <t>If the <x:ref>close</x:ref> connection option is present, the
2826        connection will not persist after the current response; else,</t>
2827     <t>If the received protocol is HTTP/1.1 (or later), the connection will
2828        persist after the current response; else,</t>
2829     <t>If the received protocol is HTTP/1.0, the "keep-alive"
2830        connection option is present, the recipient is not a proxy, and
2831        the recipient wishes to honor the HTTP/1.0 "keep-alive" mechanism,
2832        the connection will persist after the current response; otherwise,</t>
2833     <t>The connection will close after the current response.</t>
2834   </list>
2835</t>
2836<t>
2837   A server &MAY; assume that an HTTP/1.1 client intends to maintain a
2838   persistent connection until a <x:ref>close</x:ref> connection option
2839   is received in a request.
2840</t>
2841<t>
2842   A client &MAY; reuse a persistent connection until it sends or receives
2843   a <x:ref>close</x:ref> connection option or receives an HTTP/1.0 response
2844   without a "keep-alive" connection option.
2845</t>
2846<t>
2847   In order to remain persistent, all messages on a connection &MUST;
2848   have a self-defined message length (i.e., one not defined by closure
2849   of the connection), as described in <xref target="message.body"/>.
2850   A server &MUST; read the entire request message body or close
2851   the connection after sending its response, since otherwise the
2852   remaining data on a persistent connection would be misinterpreted
2853   as the next request.  Likewise,
2854   a client &MUST; read the entire response message body if it intends
2855   to reuse the same connection for a subsequent request.
2856</t>
2857<t>
2858   A proxy server &MUST-NOT; maintain a persistent connection with an
2859   HTTP/1.0 client (see <xref x:sec="19.7.1" x:fmt="of" target="RFC2068"/> for
2860   information and discussion of the problems with the Keep-Alive header field
2861   implemented by many HTTP/1.0 clients).
2862</t>
2863<t>
2864   Clients and servers &SHOULD-NOT; assume that a persistent connection is
2865   maintained for HTTP versions less than 1.1 unless it is explicitly
2866   signaled.
2867   See <xref target="compatibility.with.http.1.0.persistent.connections"/>
2868   for more information on backward compatibility with HTTP/1.0 clients.
2869</t>
2870
2871<section title="Pipelining" anchor="pipelining">
2872<t>
2873   A client that supports persistent connections &MAY; "pipeline" its
2874   requests (i.e., send multiple requests without waiting for each
2875   response). A server &MUST; send its responses to those requests in the
2876   same order that the requests were received.
2877</t>
2878<t>
2879   Clients which assume persistent connections and pipeline immediately
2880   after connection establishment &SHOULD; be prepared to retry their
2881   connection if the first pipelined attempt fails. If a client does
2882   such a retry, it &MUST-NOT; pipeline before it knows the connection is
2883   persistent. Clients &MUST; also be prepared to resend their requests if
2884   the server closes the connection before sending all of the
2885   corresponding responses.
2886</t>
2887<t>
2888   Clients &SHOULD-NOT; pipeline requests using non-idempotent request methods
2889   or non-idempotent sequences of request methods (see &idempotent-methods;).
2890   Otherwise, a premature termination of the transport connection could lead
2891   to indeterminate results. A client wishing to send a non-idempotent
2892   request &SHOULD; wait to send that request until it has received the
2893   response status line for the previous request.
2894</t>
2895</section>
2896
2897<section title="Retrying Requests" anchor="persistent.retrying.requests">
2898<t>
2899   Connections can be closed at any time, with or without intention.
2900   Implementations ought to anticipate the need to recover
2901   from asynchronous close events.
2902   A client &MAY; open a new connection and retransmit an aborted sequence
2903   of requests without user interaction so long as the request sequence is
2904   idempotent (see &idempotent-methods;).
2905   A client &MUST-NOT; automatically retry non-idempotent request sequences,
2906   although user agents &MAY; offer a human operator the choice of retrying
2907   the request(s). Confirmation by
2908   user agent software with semantic understanding of the application
2909   &MAY; substitute for user confirmation. An automatic retry &SHOULD-NOT; 
2910   be repeated if a second sequence of requests fails.
2911</t>
2912</section>
2913</section>
2914   
2915<section title="Concurrency" anchor="persistent.concurrency">
2916<t>
2917   Clients &SHOULD; limit the number of simultaneous
2918   connections that they maintain to a given server.
2919</t>
2920<t>
2921   Previous revisions of HTTP gave a specific number of connections as a
2922   ceiling, but this was found to be impractical for many applications. As a
2923   result, this specification does not mandate a particular maximum number of
2924   connections, but instead encourages clients to be conservative when opening
2925   multiple connections.
2926</t>
2927<t>
2928   Multiple connections are typically used to avoid the "head-of-line
2929   blocking" problem, wherein a request that takes significant server-side
2930   processing and/or has a large payload blocks subsequent requests on the
2931   same connection. However, each connection consumes server resources.
2932   Furthermore, using multiple connections can cause undesirable side effects
2933   in congested networks.
2934</t>
2935<t>
2936   Note that servers might reject traffic that they deem abusive, including an
2937   excessive number of connections from a client.
2938</t>
2939</section>
2940
2941<section title="Failures and Time-outs" anchor="persistent.failures">
2942<t>
2943   Servers will usually have some time-out value beyond which they will
2944   no longer maintain an inactive connection. Proxy servers might make
2945   this a higher value since it is likely that the client will be making
2946   more connections through the same server. The use of persistent
2947   connections places no requirements on the length (or existence) of
2948   this time-out for either the client or the server.
2949</t>
2950<t>
2951   When a client or server wishes to time-out it &SHOULD; issue a graceful
2952   close on the transport connection. Clients and servers &SHOULD; both
2953   constantly watch for the other side of the transport close, and
2954   respond to it as appropriate. If a client or server does not detect
2955   the other side's close promptly it could cause unnecessary resource
2956   drain on the network.
2957</t>
2958<t>
2959   A client, server, or proxy &MAY; close the transport connection at any
2960   time. For example, a client might have started to send a new request
2961   at the same time that the server has decided to close the "idle"
2962   connection. From the server's point of view, the connection is being
2963   closed while it was idle, but from the client's point of view, a
2964   request is in progress.
2965</t>
2966<t>
2967   Servers &SHOULD; maintain persistent connections and allow the underlying
2968   transport's flow control mechanisms to resolve temporary overloads, rather
2969   than terminate connections with the expectation that clients will retry.
2970   The latter technique can exacerbate network congestion.
2971</t>
2972<t>
2973   A client sending a message body &SHOULD; monitor
2974   the network connection for an error status code while it is transmitting
2975   the request. If the client sees an error status code, it &SHOULD;
2976   immediately cease transmitting the body and close the connection.
2977</t>
2978</section>
2979   
2980<section title="Tear-down" anchor="persistent.tear-down">
2981  <iref primary="false" item="Connection header field" x:for-anchor=""/>
2982  <iref primary="false" item="close" x:for-anchor=""/>
2983<t>
2984   The <x:ref>Connection</x:ref> header field
2985   (<xref target="header.connection"/>) provides a "<x:ref>close</x:ref>"
2986   connection option that a sender &SHOULD; send when it wishes to close
2987   the connection after the current request/response pair.
2988</t>
2989<t>
2990   A client that sends a <x:ref>close</x:ref> connection option &MUST-NOT;
2991   send further requests on that connection (after the one containing
2992   <x:ref>close</x:ref>) and &MUST; close the connection after reading the
2993   final response message corresponding to this request.
2994</t>
2995<t>
2996   A server that receives a <x:ref>close</x:ref> connection option &MUST;
2997   initiate a lingering close (see below) of the connection after it sends the
2998   final response to the request that contained <x:ref>close</x:ref>.
2999   The server &SHOULD; include a <x:ref>close</x:ref> connection option
3000   in its final response on that connection. The server &MUST-NOT; process
3001   any further requests received on that connection.
3002</t>
3003<t>
3004   A server that sends a <x:ref>close</x:ref> connection option &MUST;
3005   initiate a lingering close of the connection after it sends the
3006   response containing <x:ref>close</x:ref>. The server &MUST-NOT; process
3007   any further requests received on that connection.
3008</t>
3009<t>
3010   A client that receives a <x:ref>close</x:ref> connection option &MUST;
3011   cease sending requests on that connection and close the connection
3012   after reading the response message containing the close; if additional
3013   pipelined requests had been sent on the connection, the client &SHOULD;
3014   assume that they will not be processed by the server.
3015</t>
3016<t>
3017   If a server performs an immediate close of a TCP connection, there is a
3018   significant risk that the client will not be able to read the last HTTP
3019   response.  If the server receives additional data from the client on a
3020   fully-closed connection, such as another request that was sent by the
3021   client before receiving the server's response, the server's TCP stack will
3022   send a reset packet to the client; unfortunately, the reset packet might
3023   erase the client's unacknowledged input buffers before they can be read
3024   and interpreted by the client's HTTP parser.
3025</t>
3026<t>
3027   To avoid the TCP reset problem, a server can perform a lingering close on a
3028   connection by closing only the write side of the read/write connection
3029   (a half-close) and continuing to read from the connection until the
3030   connection is closed by the client or the server is reasonably certain
3031   that its own TCP stack has received the client's acknowledgement of the
3032   packet(s) containing the server's last response. It is then safe for the
3033   server to fully close the connection.
3034</t>
3035<t>
3036   It is unknown whether the reset problem is exclusive to TCP or might also
3037   be found in other transport connection protocols.
3038</t>
3039</section>
3040
3041<section title="Upgrade" anchor="header.upgrade">
3042  <iref primary="true" item="Upgrade header field" x:for-anchor=""/>
3043  <x:anchor-alias value="Upgrade"/>
3044  <x:anchor-alias value="protocol"/>
3045  <x:anchor-alias value="protocol-name"/>
3046  <x:anchor-alias value="protocol-version"/>
3047<t>
3048   The "Upgrade" header field is intended to provide a simple mechanism
3049   for transitioning from HTTP/1.1 to some other protocol on the same
3050   connection.  A client &MAY; send a list of protocols in the Upgrade
3051   header field of a request to invite the server to switch to one or
3052   more of those protocols before sending the final response.
3053   A server &MUST; send an Upgrade header field in <x:ref>101 (Switching
3054   Protocols)</x:ref> responses to indicate which protocol(s) are being
3055   switched to, and &MUST; send it in <x:ref>426 (Upgrade Required)</x:ref>
3056   responses to indicate acceptable protocols.
3057   A server &MAY; send an Upgrade header field in any other response to
3058   indicate that they might be willing to upgrade to one of the
3059   specified protocols for a future request.
3060</t>
3061<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Upgrade"/>
3062  <x:ref>Upgrade</x:ref>          = 1#<x:ref>protocol</x:ref>
3063
3064  <x:ref>protocol</x:ref>         = <x:ref>protocol-name</x:ref> ["/" <x:ref>protocol-version</x:ref>]
3065  <x:ref>protocol-name</x:ref>    = <x:ref>token</x:ref>
3066  <x:ref>protocol-version</x:ref> = <x:ref>token</x:ref>
3067</artwork></figure>
3068<t>
3069   For example,
3070</t>
3071<figure><artwork type="example">
3072  Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
3073</artwork></figure>
3074<t>
3075   Upgrade eases the difficult transition between incompatible protocols by
3076   allowing the client to initiate a request in the more commonly
3077   supported protocol while indicating to the server that it would like
3078   to use a "better" protocol if available (where "better" is determined
3079   by the server, possibly according to the nature of the request method
3080   or target resource).
3081</t>
3082<t>
3083   Upgrade cannot be used to insist on a protocol change; its acceptance and
3084   use by the server is optional. The capabilities and nature of the
3085   application-level communication after the protocol change is entirely
3086   dependent upon the new protocol chosen, although the first action
3087   after changing the protocol &MUST; be a response to the initial HTTP
3088   request that contained the Upgrade header field.
3089</t>
3090<t>
3091   For example, if the Upgrade header field is received in a GET request
3092   and the server decides to switch protocols, then it &MUST; first respond
3093   with a <x:ref>101 (Switching Protocols)</x:ref> message in HTTP/1.1 and
3094   then immediately follow that with the new protocol's equivalent of a
3095   response to a GET on the target resource.  This allows a connection to be
3096   upgraded to protocols with the same semantics as HTTP without the
3097   latency cost of an additional round-trip.  A server &MUST-NOT; switch
3098   protocols unless the received message semantics can be honored by the new
3099   protocol; an OPTIONS request can be honored by any protocol.
3100</t>
3101<t>
3102   When Upgrade is sent, a sender &MUST; also send a
3103   <x:ref>Connection</x:ref> header field (<xref target="header.connection"/>)
3104   that contains the "upgrade" connection option, in order to prevent Upgrade
3105   from being accidentally forwarded by intermediaries that might not implement
3106   the listed protocols.  A server &MUST; ignore an Upgrade header field that
3107   is received in an HTTP/1.0 request.
3108</t>
3109<t>
3110   The Upgrade header field only applies to switching application-level
3111   protocols on the existing connection; it cannot be used
3112   to switch to a protocol on a different connection. For that purpose, it is
3113   more appropriate to use a <x:ref>3xx (Redirection)</x:ref> response
3114   (&status-3xx;).
3115</t>
3116<t>
3117   This specification only defines the protocol name "HTTP" for use by
3118   the family of Hypertext Transfer Protocols, as defined by the HTTP
3119   version rules of <xref target="http.version"/> and future updates to this
3120   specification. Additional tokens can be registered with IANA using the
3121   registration procedure defined in <xref target="upgrade.token.registry"/>.
3122</t>
3123</section>
3124</section>
3125
3126<section title="IANA Considerations" anchor="IANA.considerations">
3127
3128<section title="Header Field Registration" anchor="header.field.registration">
3129<t>
3130   HTTP header fields are registered within the Message Header Field Registry
3131   <xref target="RFC3864"/> maintained by IANA at
3132   <eref target="http://www.iana.org/assignments/message-headers/message-header-index.html"/>.
3133</t>
3134<t>
3135   This document defines the following HTTP header fields, so their
3136   associated registry entries shall be updated according to the permanent
3137   registrations below:
3138</t>
3139<?BEGININC p1-messaging.iana-headers ?>
3140<!--AUTOGENERATED FROM extract-header-defs.xslt, do not edit manually-->
3141<texttable align="left" suppress-title="true" anchor="iana.header.registration.table">
3142   <ttcol>Header Field Name</ttcol>
3143   <ttcol>Protocol</ttcol>
3144   <ttcol>Status</ttcol>
3145   <ttcol>Reference</ttcol>
3146
3147   <c>Connection</c>
3148   <c>http</c>
3149   <c>standard</c>
3150   <c>
3151      <xref target="header.connection"/>
3152   </c>
3153   <c>Content-Length</c>
3154   <c>http</c>
3155   <c>standard</c>
3156   <c>
3157      <xref target="header.content-length"/>
3158   </c>
3159   <c>Host</c>
3160   <c>http</c>
3161   <c>standard</c>
3162   <c>
3163      <xref target="header.host"/>
3164   </c>
3165   <c>TE</c>
3166   <c>http</c>
3167   <c>standard</c>
3168   <c>
3169      <xref target="header.te"/>
3170   </c>
3171   <c>Trailer</c>
3172   <c>http</c>
3173   <c>standard</c>
3174   <c>
3175      <xref target="header.trailer"/>
3176   </c>
3177   <c>Transfer-Encoding</c>
3178   <c>http</c>
3179   <c>standard</c>
3180   <c>
3181      <xref target="header.transfer-encoding"/>
3182   </c>
3183   <c>Upgrade</c>
3184   <c>http</c>
3185   <c>standard</c>
3186   <c>
3187      <xref target="header.upgrade"/>
3188   </c>
3189   <c>Via</c>
3190   <c>http</c>
3191   <c>standard</c>
3192   <c>
3193      <xref target="header.via"/>
3194   </c>
3195</texttable>
3196<!--(END)-->
3197<?ENDINC p1-messaging.iana-headers ?>
3198<t>
3199   Furthermore, the header field-name "Close" shall be registered as
3200   "reserved", since using that name as an HTTP header field might
3201   conflict with the "close" connection option of the "<x:ref>Connection</x:ref>"
3202   header field (<xref target="header.connection"/>).
3203</t>
3204<texttable align="left" suppress-title="true">
3205   <ttcol>Header Field Name</ttcol>
3206   <ttcol>Protocol</ttcol>
3207   <ttcol>Status</ttcol>
3208   <ttcol>Reference</ttcol>
3209
3210   <c>Close</c>
3211   <c>http</c>
3212   <c>reserved</c>
3213   <c>
3214      <xref target="header.field.registration"/>
3215   </c>
3216</texttable>
3217<t>
3218   The change controller is: "IETF (iesg@ietf.org) - Internet Engineering Task Force".
3219</t>
3220</section>
3221
3222<section title="URI Scheme Registration" anchor="uri.scheme.registration">
3223<t>
3224   IANA maintains the registry of URI Schemes <xref target="RFC4395"/> at
3225   <eref target="http://www.iana.org/assignments/uri-schemes.html"/>.
3226</t>
3227<t>
3228   This document defines the following URI schemes, so their
3229   associated registry entries shall be updated according to the permanent
3230   registrations below:
3231</t>
3232<texttable align="left" suppress-title="true">
3233   <ttcol>URI Scheme</ttcol>
3234   <ttcol>Description</ttcol>
3235   <ttcol>Reference</ttcol>
3236
3237   <c>http</c>
3238   <c>Hypertext Transfer Protocol</c>
3239   <c><xref target="http.uri"/></c>
3240
3241   <c>https</c>
3242   <c>Hypertext Transfer Protocol Secure</c>
3243   <c><xref target="https.uri"/></c>
3244</texttable>
3245</section>
3246
3247<section title="Internet Media Type Registrations" anchor="internet.media.type.http">
3248<t>
3249   This document serves as the specification for the Internet media types
3250   "message/http" and "application/http". The following is to be registered with
3251   IANA (see <xref target="RFC4288"/>).
3252</t>
3253<section title="Internet Media Type message/http" anchor="internet.media.type.message.http">
3254<iref item="Media Type" subitem="message/http" primary="true"/>
3255<iref item="message/http Media Type" primary="true"/>
3256<t>
3257   The message/http type can be used to enclose a single HTTP request or
3258   response message, provided that it obeys the MIME restrictions for all
3259   "message" types regarding line length and encodings.
3260</t>
3261<t>
3262  <list style="hanging" x:indent="12em">
3263    <t hangText="Type name:">
3264      message
3265    </t>
3266    <t hangText="Subtype name:">
3267      http
3268    </t>
3269    <t hangText="Required parameters:">
3270      none
3271    </t>
3272    <t hangText="Optional parameters:">
3273      version, msgtype
3274      <list style="hanging">
3275        <t hangText="version:">
3276          The HTTP-version number of the enclosed message
3277          (e.g., "1.1"). If not present, the version can be
3278          determined from the first line of the body.
3279        </t>
3280        <t hangText="msgtype:">
3281          The message type &mdash; "request" or "response". If not
3282          present, the type can be determined from the first
3283          line of the body.
3284        </t>
3285      </list>
3286    </t>
3287    <t hangText="Encoding considerations:">
3288      only "7bit", "8bit", or "binary" are permitted
3289    </t>
3290    <t hangText="Security considerations:">
3291      none
3292    </t>
3293    <t hangText="Interoperability considerations:">
3294      none
3295    </t>
3296    <t hangText="Published specification:">
3297      This specification (see <xref target="internet.media.type.message.http"/>).
3298    </t>
3299    <t hangText="Applications that use this media type:">
3300    </t>
3301    <t hangText="Additional information:">
3302      <list style="hanging">
3303        <t hangText="Magic number(s):">none</t>
3304        <t hangText="File extension(s):">none</t>
3305        <t hangText="Macintosh file type code(s):">none</t>
3306      </list>
3307    </t>
3308    <t hangText="Person and email address to contact for further information:">
3309      See Authors Section.
3310    </t>
3311    <t hangText="Intended usage:">
3312      COMMON
3313    </t>
3314    <t hangText="Restrictions on usage:">
3315      none
3316    </t>
3317    <t hangText="Author/Change controller:">
3318      IESG
3319    </t>
3320  </list>
3321</t>
3322</section>
3323<section title="Internet Media Type application/http" anchor="internet.media.type.application.http">
3324<iref item="Media Type" subitem="application/http" primary="true"/>
3325<iref item="application/http Media Type" primary="true"/>
3326<t>
3327   The application/http type can be used to enclose a pipeline of one or more
3328   HTTP request or response messages (not intermixed).
3329</t>
3330<t>
3331  <list style="hanging" x:indent="12em">
3332    <t hangText="Type name:">
3333      application
3334    </t>
3335    <t hangText="Subtype name:">
3336      http
3337    </t>
3338    <t hangText="Required parameters:">
3339      none
3340    </t>
3341    <t hangText="Optional parameters:">
3342      version, msgtype
3343      <list style="hanging">
3344        <t hangText="version:">
3345          The HTTP-version number of the enclosed messages
3346          (e.g., "1.1"). If not present, the version can be
3347          determined from the first line of the body.
3348        </t>
3349        <t hangText="msgtype:">
3350          The message type &mdash; "request" or "response". If not
3351          present, the type can be determined from the first
3352          line of the body.
3353        </t>
3354      </list>
3355    </t>
3356    <t hangText="Encoding considerations:">
3357      HTTP messages enclosed by this type
3358      are in "binary" format; use of an appropriate
3359      Content-Transfer-Encoding is required when
3360      transmitted via E-mail.
3361    </t>
3362    <t hangText="Security considerations:">
3363      none
3364    </t>
3365    <t hangText="Interoperability considerations:">
3366      none
3367    </t>
3368    <t hangText="Published specification:">
3369      This specification (see <xref target="internet.media.type.application.http"/>).
3370    </t>
3371    <t hangText="Applications that use this media type:">
3372    </t>
3373    <t hangText="Additional information:">
3374      <list style="hanging">
3375        <t hangText="Magic number(s):">none</t>
3376        <t hangText="File extension(s):">none</t>
3377        <t hangText="Macintosh file type code(s):">none</t>
3378      </list>
3379    </t>
3380    <t hangText="Person and email address to contact for further information:">
3381      See Authors Section.
3382    </t>
3383    <t hangText="Intended usage:">
3384      COMMON
3385    </t>
3386    <t hangText="Restrictions on usage:">
3387      none
3388    </t>
3389    <t hangText="Author/Change controller:">
3390      IESG
3391    </t>
3392  </list>
3393</t>
3394</section>
3395</section>
3396
3397<section title="Transfer Coding Registry" anchor="transfer.coding.registry">
3398<t>
3399   The HTTP Transfer Coding Registry defines the name space for transfer
3400   coding names.
3401</t>
3402<t>
3403   Registrations &MUST; include the following fields:
3404   <list style="symbols">
3405     <t>Name</t>
3406     <t>Description</t>
3407     <t>Pointer to specification text</t>
3408   </list>
3409</t>
3410<t>
3411   Names of transfer codings &MUST-NOT; overlap with names of content codings
3412   (&content-codings;) unless the encoding transformation is identical, as
3413   is the case for the compression codings defined in
3414   <xref target="compression.codings"/>.
3415</t>
3416<t>
3417   Values to be added to this name space require IETF Review (see
3418   <xref target="RFC5226" x:fmt="of" x:sec="4.1"/>), and &MUST;
3419   conform to the purpose of transfer coding defined in this section.
3420   Use of program names for the identification of encoding formats
3421   is not desirable and is discouraged for future encodings.
3422</t>
3423<t>
3424   The registry itself is maintained at
3425   <eref target="http://www.iana.org/assignments/http-parameters"/>.
3426</t>
3427</section>
3428
3429<section title="Transfer Coding Registrations" anchor="transfer.coding.registration">
3430<t>
3431   The HTTP Transfer Coding Registry shall be updated with the registrations
3432   below:
3433</t>
3434<texttable align="left" suppress-title="true" anchor="iana.transfer.coding.registration.table">
3435   <ttcol>Name</ttcol>
3436   <ttcol>Description</ttcol>
3437   <ttcol>Reference</ttcol>
3438   <c>chunked</c>
3439   <c>Transfer in a series of chunks</c>
3440   <c>
3441      <xref target="chunked.encoding"/>
3442   </c>
3443   <c>compress</c>
3444   <c>UNIX "compress" program method</c>
3445   <c>
3446      <xref target="compress.coding"/>
3447   </c>
3448   <c>deflate</c>
3449   <c>"deflate" compression mechanism (<xref target="RFC1951"/>) used inside
3450   the "zlib" data format (<xref target="RFC1950"/>)
3451   </c>
3452   <c>
3453      <xref target="deflate.coding"/>
3454   </c>
3455   <c>gzip</c>
3456   <c>Same as GNU zip <xref target="RFC1952"/></c>
3457   <c>
3458      <xref target="gzip.coding"/>
3459   </c>
3460</texttable>
3461</section>
3462
3463<section title="Upgrade Token Registry" anchor="upgrade.token.registry">
3464<t>
3465   The HTTP Upgrade Token Registry defines the name space for protocol-name
3466   tokens used to identify protocols in the <x:ref>Upgrade</x:ref> header
3467   field. Each registered protocol name is associated with contact information
3468   and an optional set of specifications that details how the connection
3469   will be processed after it has been upgraded.
3470</t>
3471<t>
3472   Registrations happen on a "First Come First Served" basis (see
3473   <xref target="RFC5226" x:sec="4.1" x:fmt="of"/>) and are subject to the
3474   following rules:
3475  <list style="numbers">
3476    <t>A protocol-name token, once registered, stays registered forever.</t>
3477    <t>The registration &MUST; name a responsible party for the
3478       registration.</t>
3479    <t>The registration &MUST; name a point of contact.</t>
3480    <t>The registration &MAY; name a set of specifications associated with
3481       that token. Such specifications need not be publicly available.</t>
3482    <t>The registration &SHOULD; name a set of expected "protocol-version"
3483       tokens associated with that token at the time of registration.</t>
3484    <t>The responsible party &MAY; change the registration at any time.
3485       The IANA will keep a record of all such changes, and make them
3486       available upon request.</t>
3487    <t>The IESG &MAY; reassign responsibility for a protocol token.
3488       This will normally only be used in the case when a
3489       responsible party cannot be contacted.</t>
3490  </list>
3491</t>
3492<t>
3493   This registration procedure for HTTP Upgrade Tokens replaces that
3494   previously defined in <xref target="RFC2817" x:fmt="of" x:sec="7.2"/>.
3495</t>
3496</section>
3497
3498<section title="Upgrade Token Registration" anchor="upgrade.token.registration">
3499<t>
3500   The HTTP Upgrade Token Registry shall be updated with the registration
3501   below:
3502</t>
3503<texttable align="left" suppress-title="true">
3504   <ttcol>Value</ttcol>
3505   <ttcol>Description</ttcol>
3506   <ttcol>Expected Version Tokens</ttcol>
3507   <ttcol>Reference</ttcol>
3508
3509   <c>HTTP</c>
3510   <c>Hypertext Transfer Protocol</c>
3511   <c>any DIGIT.DIGIT (e.g, "2.0")</c>
3512   <c><xref target="http.version"/></c>
3513</texttable>
3514<t>
3515   The responsible party is: "IETF (iesg@ietf.org) - Internet Engineering Task Force".
3516</t>
3517</section>
3518
3519</section>
3520
3521<section title="Security Considerations" anchor="security.considerations">
3522<t>
3523   This section is meant to inform application developers, information
3524   providers, and users of the security limitations in HTTP/1.1 as
3525   described by this document. The discussion does not include
3526   definitive solutions to the problems revealed, though it does make
3527   some suggestions for reducing security risks.
3528</t>
3529
3530<section title="Personal Information" anchor="personal.information">
3531<t>
3532   HTTP clients are often privy to large amounts of personal information,
3533   including both information provided by the user to interact with resources
3534   (e.g., the user's name, location, mail address, passwords, encryption
3535   keys, etc.) and information about the user's browsing activity over
3536   time (e.g., history, bookmarks, etc.). HTTP implementations need to
3537   prevent unintentional leakage of this information.
3538</t>
3539</section>
3540
3541<section title="Abuse of Server Log Information" anchor="abuse.of.server.log.information">
3542<t>
3543   A server is in the position to save personal data about a user's
3544   requests which might identify their reading patterns or subjects of
3545   interest.  In particular, log information gathered at an intermediary
3546   often contains a history of user agent interaction, across a multitude
3547   of sites, that can be traced to individual users.
3548</t>
3549<t>
3550   HTTP log information is confidential in nature; its handling is often
3551   constrained by laws and regulations.  Log information needs to be securely
3552   stored and appropriate guidelines followed for its analysis.
3553   Anonymization of personal information within individual entries helps,
3554   but is generally not sufficient to prevent real log traces from being
3555   re-identified based on correlation with other access characteristics.
3556   As such, access traces that are keyed to a specific client should not
3557   be published even if the key is pseudonymous.
3558</t>
3559<t>
3560   To minimize the risk of theft or accidental publication, log information
3561   should be purged of personally identifiable information, including
3562   user identifiers, IP addresses, and user-provided query parameters,
3563   as soon as that information is no longer necessary to support operational
3564   needs for security, auditing, or fraud control.
3565</t>
3566</section>
3567
3568<section title="Attacks Based On File and Path Names" anchor="attack.pathname">
3569<t>
3570   Origin servers &SHOULD; be careful to restrict
3571   the documents returned by HTTP requests to be only those that were
3572   intended by the server administrators. If an HTTP server translates
3573   HTTP URIs directly into file system calls, the server &MUST; take
3574   special care not to serve files that were not intended to be
3575   delivered to HTTP clients. For example, UNIX, Microsoft Windows, and
3576   other operating systems use ".." as a path component to indicate a
3577   directory level above the current one. On such a system, an HTTP
3578   server &MUST; disallow any such construct in the request-target if it
3579   would otherwise allow access to a resource outside those intended to
3580   be accessible via the HTTP server. Similarly, files intended for
3581   reference only internally to the server (such as access control
3582   files, configuration files, and script code) &MUST; be protected from
3583   inappropriate retrieval, since they might contain sensitive
3584   information.
3585</t>
3586</section>
3587
3588<section title="DNS-related Attacks" anchor="dns.related.attacks">
3589<t>
3590   HTTP clients rely heavily on the Domain Name Service (DNS), and are thus
3591   generally prone to security attacks based on the deliberate misassociation
3592   of IP addresses and DNS names not protected by DNSSEC. Clients need to be
3593   cautious in assuming the validity of an IP number/DNS name association unless
3594   the response is protected by DNSSEC (<xref target="RFC4033"/>).
3595</t>
3596</section>
3597
3598<section title="Intermediaries and Caching" anchor="attack.intermediaries">
3599<t>
3600   By their very nature, HTTP intermediaries are men-in-the-middle, and
3601   represent an opportunity for man-in-the-middle attacks. Compromise of
3602   the systems on which the intermediaries run can result in serious security
3603   and privacy problems. Intermediaries have access to security-related
3604   information, personal information about individual users and
3605   organizations, and proprietary information belonging to users and
3606   content providers. A compromised intermediary, or an intermediary
3607   implemented or configured without regard to security and privacy
3608   considerations, might be used in the commission of a wide range of
3609   potential attacks.
3610</t>
3611<t>
3612   Intermediaries that contain a shared cache are especially vulnerable
3613   to cache poisoning attacks.
3614</t>
3615<t>
3616   Implementers need to consider the privacy and security
3617   implications of their design and coding decisions, and of the
3618   configuration options they provide to operators (especially the
3619   default configuration).
3620</t>
3621<t>
3622   Users need to be aware that intermediaries are no more trustworthy than
3623   the people who run them; HTTP itself cannot solve this problem.
3624</t>
3625</section>
3626
3627<section title="Buffer Overflows" anchor="attack.protocol.element.size.overflows">
3628<t>
3629   Because HTTP uses mostly textual, character-delimited fields, attackers can
3630   overflow buffers in implementations, and/or perform a Denial of Service
3631   against implementations that accept fields with unlimited lengths.
3632</t>
3633<t>
3634   To promote interoperability, this specification makes specific
3635   recommendations for minimum size limits on request-line
3636   (<xref target="request.line"/>)
3637   and blocks of header fields (<xref target="header.fields"/>). These are
3638   minimum recommendations, chosen to be supportable even by implementations
3639   with limited resources; it is expected that most implementations will
3640   choose substantially higher limits.
3641</t>
3642<t>
3643   This specification also provides a way for servers to reject messages that
3644   have request-targets that are too long (&status-414;) or request entities
3645   that are too large (&status-4xx;).
3646</t>
3647<t>
3648   Recipients &SHOULD; carefully limit the extent to which they read other
3649   fields, including (but not limited to) request methods, response status
3650   phrases, header field-names, and body chunks, so as to avoid denial of
3651   service attacks without impeding interoperability.
3652</t>
3653</section>
3654
3655<section title="Message Integrity" anchor="message.integrity">
3656<t>
3657   HTTP does not define a specific mechanism for ensuring message integrity,
3658   instead relying on the error-detection ability of underlying transport
3659   protocols and the use of length or chunk-delimited framing to detect
3660   completeness. Additional integrity mechanisms, such as hash functions or
3661   digital signatures applied to the content, can be selectively added to
3662   messages via extensible metadata header fields. Historically, the lack of
3663   a single integrity mechanism has been justified by the informal nature of
3664   most HTTP communication.  However, the prevalence of HTTP as an information
3665   access mechanism has resulted in its increasing use within environments
3666   where verification of message integrity is crucial.
3667</t>
3668<t>
3669   User agents are encouraged to implement configurable means for detecting
3670   and reporting failures of message integrity such that those means can be
3671   enabled within environments for which integrity is necessary. For example,
3672   a browser being used to view medical history or drug interaction
3673   information needs to indicate to the user when such information is detected
3674   by the protocol to be incomplete, expired, or corrupted during transfer.
3675   Such mechanisms might be selectively enabled via user agent extensions or
3676   the presence of message integrity metadata in a response.
3677   At a minimum, user agents ought to provide some indication that allows a
3678   user to distinguish between a complete and incomplete response message
3679   (<xref target="incomplete.messages"/>) when such verification is desired.
3680</t>
3681</section>
3682</section>
3683
3684<section title="Acknowledgments" anchor="acks">
3685<t>
3686   This edition of HTTP/1.1 builds on the many contributions that went into
3687   <xref target="RFC1945" format="none">RFC 1945</xref>,
3688   <xref target="RFC2068" format="none">RFC 2068</xref>,
3689   <xref target="RFC2145" format="none">RFC 2145</xref>, and
3690   <xref target="RFC2616" format="none">RFC 2616</xref>, including
3691   substantial contributions made by the previous authors, editors, and
3692   working group chairs: Tim Berners-Lee, Ari Luotonen, Roy T. Fielding,
3693   Henrik Frystyk Nielsen, Jim Gettys, Jeffrey C. Mogul, Larry Masinter,
3694   and Paul J. Leach. Mark Nottingham oversaw this effort as working group chair.
3695</t>
3696<t>
3697   Since 1999, the following contributors have helped improve the HTTP
3698   specification by reporting bugs, asking smart questions, drafting or
3699   reviewing text, and evaluating open issues:
3700</t>
3701<?BEGININC acks ?>
3702<t>Adam Barth,
3703Adam Roach,
3704Addison Phillips,
3705Adrian Chadd,
3706Adrien W. de Croy,
3707Alan Ford,
3708Alan Ruttenberg,
3709Albert Lunde,
3710Alek Storm,
3711Alex Rousskov,
3712Alexandre Morgaut,
3713Alexey Melnikov,
3714Alisha Smith,
3715Amichai Rothman,
3716Amit Klein,
3717Amos Jeffries,
3718Andreas Maier,
3719Andreas Petersson,
3720Anil Sharma,
3721Anne van Kesteren,
3722Anthony Bryan,
3723Asbjorn Ulsberg,
3724Ashok Kumar,
3725Balachander Krishnamurthy,
3726Barry Leiba,
3727Ben Laurie,
3728Benjamin Niven-Jenkins,
3729Bil Corry,
3730Bill Burke,
3731Bjoern Hoehrmann,
3732Bob Scheifler,
3733Boris Zbarsky,
3734Brett Slatkin,
3735Brian Kell,
3736Brian McBarron,
3737Brian Pane,
3738Brian Smith,
3739Bryce Nesbitt,
3740Cameron Heavon-Jones,
3741Carl Kugler,
3742Carsten Bormann,
3743Charles Fry,
3744Chris Newman,
3745Cyrus Daboo,
3746Dale Robert Anderson,
3747Dan Wing,
3748Dan Winship,
3749Daniel Stenberg,
3750Dave Cridland,
3751Dave Crocker,
3752Dave Kristol,
3753David Booth,
3754David Singer,
3755David W. Morris,
3756Diwakar Shetty,
3757Dmitry Kurochkin,
3758Drummond Reed,
3759Duane Wessels,
3760Edward Lee,
3761Eliot Lear,
3762Eran Hammer-Lahav,
3763Eric D. Williams,
3764Eric J. Bowman,
3765Eric Lawrence,
3766Eric Rescorla,
3767Erik Aronesty,
3768Evan Prodromou,
3769Florian Weimer,
3770Frank Ellermann,
3771Fred Bohle,
3772Gabriel Montenegro,
3773Geoffrey Sneddon,
3774Gervase Markham,
3775Grahame Grieve,
3776Greg Wilkins,
3777Harald Tveit Alvestrand,
3778Harry Halpin,
3779Helge Hess,
3780Henrik Nordstrom,
3781Henry S. Thompson,
3782Henry Story,
3783Herbert van de Sompel,
3784Howard Melman,
3785Hugo Haas,
3786Ian Fette,
3787Ian Hickson,
3788Ido Safruti,
3789Ilya Grigorik,
3790Ingo Struck,
3791J. Ross Nicoll,
3792James H. Manger,
3793James Lacey,
3794James M. Snell,
3795Jamie Lokier,
3796Jan Algermissen,
3797Jeff Hodges (who came up with the term 'effective Request-URI'),
3798Jeff Walden,
3799Jim Luther,
3800Joe D. Williams,
3801Joe Gregorio,
3802Joe Orton,
3803John C. Klensin,
3804John C. Mallery,
3805John Cowan,
3806John Kemp,
3807John Panzer,
3808John Schneider,
3809John Stracke,
3810John Sullivan,
3811Jonas Sicking,
3812Jonathan Billington,
3813Jonathan Moore,
3814Jonathan Rees,
3815Jonathan Silvera,
3816Jordi Ros,
3817Joris Dobbelsteen,
3818Josh Cohen,
3819Julien Pierre,
3820Jungshik Shin,
3821Justin Chapweske,
3822Justin Erenkrantz,
3823Justin James,
3824Kalvinder Singh,
3825Karl Dubost,
3826Keith Hoffman,
3827Keith Moore,
3828Ken Murchison,
3829Koen Holtman,
3830Konstantin Voronkov,
3831Kris Zyp,
3832Lisa Dusseault,
3833Maciej Stachowiak,
3834Marc Schneider,
3835Marc Slemko,
3836Mark Baker,
3837Mark Pauley,
3838Mark Watson,
3839Markus Isomaki,
3840Markus Lanthaler,
3841Martin J. Duerst,
3842Martin Musatov,
3843Martin Nilsson,
3844Martin Thomson,
3845Matt Lynch,
3846Matthew Cox,
3847Max Clark,
3848Michael Burrows,
3849Michael Hausenblas,
3850Mike Amundsen,
3851Mike Belshe,
3852Mike Kelly,
3853Mike Schinkel,
3854Miles Sabin,
3855Murray S. Kucherawy,
3856Mykyta Yevstifeyev,
3857Nathan Rixham,
3858Nicholas Shanks,
3859Nico Williams,
3860Nicolas Alvarez,
3861Nicolas Mailhot,
3862Noah Slater,
3863Pablo Castro,
3864Pat Hayes,
3865Patrick R. McManus,
3866Paul E. Jones,
3867Paul Hoffman,
3868Paul Marquess,
3869Peter Lepeska,
3870Peter Saint-Andre,
3871Peter Watkins,
3872Phil Archer,
3873Philippe Mougin,
3874Phillip Hallam-Baker,
3875Poul-Henning Kamp,
3876Preethi Natarajan,
3877Rajeev Bector,
3878Ray Polk,
3879Reto Bachmann-Gmuer,
3880Richard Cyganiak,
3881Robert Brewer,
3882Robert Collins,
3883Robert O'Callahan,
3884Robert Olofsson,
3885Robert Sayre,
3886Robert Siemer,
3887Robert de Wilde,
3888Roberto Javier Godoy,
3889Roberto Peon,
3890Roland Zink,
3891Ronny Widjaja,
3892S. Mike Dierken,
3893Salvatore Loreto,
3894Sam Johnston,
3895Sam Ruby,
3896Scott Lawrence (who maintained the original issues list),
3897Sean B. Palmer,
3898Shane McCarron,
3899Stefan Eissing,
3900Stefan Tilkov,
3901Stefanos Harhalakis,
3902Stephane Bortzmeyer,
3903Stephen Farrell,
3904Stephen Ludin,
3905Stuart Williams,
3906Subbu Allamaraju,
3907Sylvain Hellegouarch,
3908Tapan Divekar,
3909Tatsuya Hayashi,
3910Ted Hardie,
3911Thomas Broyer,
3912Thomas Fossati,
3913Thomas Nordin,
3914Thomas Roessler,
3915Tim Bray,
3916Tim Morgan,
3917Tim Olsen,
3918Tom Zhou,
3919Travis Snoozy,
3920Tyler Close,
3921Vincent Murphy,
3922Wenbo Zhu,
3923Werner Baumann,
3924Wilbur Streett,
3925Wilfredo Sanchez Vega,
3926William A. Rowe Jr.,
3927William Chan,
3928Willy Tarreau,
3929Xiaoshu Wang,
3930Yaron Goland,
3931Yngve Nysaeter Pettersen,
3932Yoav Nir,
3933Yogesh Bang,
3934Yutaka Oiwa,
3935Yves Lafon (long-time member of the editor team),
3936Zed A. Shaw, and
3937Zhong Yu.
3938</t>
3939<?ENDINC acks ?>
3940<t>
3941   See <xref target="RFC2616" x:fmt="of" x:sec="16"/> for additional
3942   acknowledgements from prior revisions.
3943</t>
3944</section>
3945
3946</middle>
3947<back>
3948
3949<references title="Normative References">
3950
3951<reference anchor="Part2">
3952  <front>
3953    <title>Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content</title>
3954    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
3955      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
3956      <address><email>fielding@gbiv.com</email></address>
3957    </author>
3958    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
3959      <organization abbrev="greenbytes">greenbytes GmbH</organization>
3960      <address><email>julian.reschke@greenbytes.de</email></address>
3961    </author>
3962    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
3963  </front>
3964  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-&ID-VERSION;"/>
3965  <x:source href="p2-semantics.xml" basename="p2-semantics">
3966    <x:defines>1xx (Informational)</x:defines>
3967    <x:defines>1xx</x:defines>
3968    <x:defines>100 (Continue)</x:defines>
3969    <x:defines>101 (Switching Protocols)</x:defines>
3970    <x:defines>2xx (Successful)</x:defines>
3971    <x:defines>2xx</x:defines>
3972    <x:defines>200 (OK)</x:defines>
3973    <x:defines>204 (No Content)</x:defines>
3974    <x:defines>3xx (Redirection)</x:defines>
3975    <x:defines>3xx</x:defines>
3976    <x:defines>301 (Moved Permanently)</x:defines>
3977    <x:defines>4xx (Client Error)</x:defines>
3978    <x:defines>4xx</x:defines>
3979    <x:defines>400 (Bad Request)</x:defines>
3980    <x:defines>405 (Method Not Allowed)</x:defines>
3981    <x:defines>411 (Length Required)</x:defines>
3982    <x:defines>414 (URI Too Long)</x:defines>
3983    <x:defines>417 (Expectation Failed)</x:defines>
3984    <x:defines>426 (Upgrade Required)</x:defines>
3985    <x:defines>501 (Not Implemented)</x:defines>
3986    <x:defines>502 (Bad Gateway)</x:defines>
3987    <x:defines>505 (HTTP Version Not Supported)</x:defines>
3988    <x:defines>Allow</x:defines>
3989    <x:defines>Content-Encoding</x:defines>
3990    <x:defines>Content-Location</x:defines>
3991    <x:defines>Content-Type</x:defines>
3992    <x:defines>Date</x:defines>
3993    <x:defines>Expect</x:defines>
3994    <x:defines>Location</x:defines>
3995    <x:defines>Server</x:defines>
3996    <x:defines>User-Agent</x:defines>
3997  </x:source>
3998</reference>
3999
4000<reference anchor="Part4">
4001  <front>
4002    <title>Hypertext Transfer Protocol (HTTP/1.1): Conditional Requests</title>
4003    <author fullname="Roy T. Fielding" initials="R." role="editor" surname="Fielding">
4004      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4005      <address><email>fielding@gbiv.com</email></address>
4006    </author>
4007    <author fullname="Julian F. Reschke" initials="J. F." role="editor" surname="Reschke">
4008      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4009      <address><email>julian.reschke@greenbytes.de</email></address>
4010    </author>
4011    <date month="&ID-MONTH;" year="&ID-YEAR;" />
4012  </front>
4013  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p4-conditional-&ID-VERSION;" />
4014  <x:source basename="p4-conditional" href="p4-conditional.xml">
4015    <x:defines>304 (Not Modified)</x:defines>
4016    <x:defines>ETag</x:defines>
4017    <x:defines>Last-Modified</x:defines>
4018  </x:source>
4019</reference>
4020
4021<reference anchor="Part5">
4022  <front>
4023    <title>Hypertext Transfer Protocol (HTTP/1.1): Range Requests</title>
4024    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4025      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4026      <address><email>fielding@gbiv.com</email></address>
4027    </author>
4028    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
4029      <organization abbrev="W3C">World Wide Web Consortium</organization>
4030      <address><email>ylafon@w3.org</email></address>
4031    </author>
4032    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4033      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4034      <address><email>julian.reschke@greenbytes.de</email></address>
4035    </author>
4036    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
4037  </front>
4038  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-&ID-VERSION;"/>
4039  <x:source href="p5-range.xml" basename="p5-range">
4040    <x:defines>Content-Range</x:defines>
4041  </x:source>
4042</reference>
4043
4044<reference anchor="Part6">
4045  <front>
4046    <title>Hypertext Transfer Protocol (HTTP/1.1): Caching</title>
4047    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4048      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4049      <address><email>fielding@gbiv.com</email></address>
4050    </author>
4051    <author initials="M." surname="Nottingham" fullname="Mark Nottingham" role="editor">
4052      <organization>Akamai</organization>
4053      <address><email>mnot@mnot.net</email></address>
4054    </author>
4055    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4056      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4057      <address><email>julian.reschke@greenbytes.de</email></address>
4058    </author>
4059    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
4060  </front>
4061  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-&ID-VERSION;"/>
4062  <x:source href="p6-cache.xml" basename="p6-cache">
4063    <x:defines>Expires</x:defines>
4064  </x:source>
4065</reference>
4066
4067<reference anchor="Part7">
4068  <front>
4069    <title>Hypertext Transfer Protocol (HTTP/1.1): Authentication</title>
4070    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
4071      <organization abbrev="Adobe">Adobe Systems Incorporated</organization>
4072      <address><email>fielding@gbiv.com</email></address>
4073    </author>
4074    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
4075      <organization abbrev="greenbytes">greenbytes GmbH</organization>
4076      <address><email>julian.reschke@greenbytes.de</email></address>
4077    </author>
4078    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
4079  </front>
4080  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p7-auth-&ID-VERSION;"/>
4081  <x:source href="p7-auth.xml" basename="p7-auth">
4082    <x:defines>Proxy-Authenticate</x:defines>
4083    <x:defines>Proxy-Authorization</x:defines>
4084  </x:source>
4085</reference>
4086
4087<reference anchor="RFC5234">
4088  <front>
4089    <title abbrev="ABNF for Syntax Specifications">Augmented BNF for Syntax Specifications: ABNF</title>
4090    <author initials="D." surname="Crocker" fullname="Dave Crocker" role="editor">
4091      <organization>Brandenburg InternetWorking</organization>
4092      <address>
4093        <email>dcrocker@bbiw.net</email>
4094      </address> 
4095    </author>
4096    <author initials="P." surname="Overell" fullname="Paul Overell">
4097      <organization>THUS plc.</organization>
4098      <address>
4099        <email>paul.overell@thus.net</email>
4100      </address>
4101    </author>
4102    <date month="January" year="2008"/>
4103  </front>
4104  <seriesInfo name="STD" value="68"/>
4105  <seriesInfo name="RFC" value="5234"/>
4106</reference>
4107
4108<reference anchor="RFC2119">
4109  <front>
4110    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
4111    <author initials="S." surname="Bradner" fullname="Scott Bradner">
4112      <organization>Harvard University</organization>
4113      <address><email>sob@harvard.edu</email></address>
4114    </author>
4115    <date month="March" year="1997"/>
4116  </front>
4117  <seriesInfo name="BCP" value="14"/>
4118  <seriesInfo name="RFC" value="2119"/>
4119</reference>
4120
4121<reference anchor="RFC3986">
4122 <front>
4123  <title abbrev='URI Generic Syntax'>Uniform Resource Identifier (URI): Generic Syntax</title>
4124  <author initials='T.' surname='Berners-Lee' fullname='Tim Berners-Lee'>
4125    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
4126    <address>
4127       <email>timbl@w3.org</email>
4128       <uri>http://www.w3.org/People/Berners-Lee/</uri>
4129    </address>
4130  </author>
4131  <author initials='R.' surname='Fielding' fullname='Roy T. Fielding'>
4132    <organization abbrev="Day Software">Day Software</organization>
4133    <address>
4134      <email>fielding@gbiv.com</email>
4135      <uri>http://roy.gbiv.com/</uri>
4136    </address>
4137  </author>
4138  <author initials='L.' surname='Masinter' fullname='Larry Masinter'>
4139    <organization abbrev="Adobe Systems">Adobe Systems Incorporated</organization>
4140    <address>
4141      <email>LMM@acm.org</email>
4142      <uri>http://larry.masinter.net/</uri>
4143    </address>
4144  </author>
4145  <date month='January' year='2005'></date>
4146 </front>
4147 <seriesInfo name="STD" value="66"/>
4148 <seriesInfo name="RFC" value="3986"/>
4149</reference>
4150
4151<reference anchor="USASCII">
4152  <front>
4153    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
4154    <author>
4155      <organization>American National Standards Institute</organization>
4156    </author>
4157    <date year="1986"/>
4158  </front>
4159  <seriesInfo name="ANSI" value="X3.4"/>
4160</reference>
4161
4162<reference anchor="RFC1950">
4163  <front>
4164    <title>ZLIB Compressed Data Format Specification version 3.3</title>
4165    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4166      <organization>Aladdin Enterprises</organization>
4167      <address><email>ghost@aladdin.com</email></address>
4168    </author>
4169    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly"/>
4170    <date month="May" year="1996"/>
4171  </front>
4172  <seriesInfo name="RFC" value="1950"/>
4173  <!--<annotation>
4174    RFC 1950 is an Informational RFC, thus it might be less stable than
4175    this specification. On the other hand, this downward reference was
4176    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4177    therefore it is unlikely to cause problems in practice. See also
4178    <xref target="BCP97"/>.
4179  </annotation>-->
4180</reference>
4181
4182<reference anchor="RFC1951">
4183  <front>
4184    <title>DEFLATE Compressed Data Format Specification version 1.3</title>
4185    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4186      <organization>Aladdin Enterprises</organization>
4187      <address><email>ghost@aladdin.com</email></address>
4188    </author>
4189    <date month="May" year="1996"/>
4190  </front>
4191  <seriesInfo name="RFC" value="1951"/>
4192  <!--<annotation>
4193    RFC 1951 is an Informational RFC, thus it might be less stable than
4194    this specification. On the other hand, this downward reference was
4195    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4196    therefore it is unlikely to cause problems in practice. See also
4197    <xref target="BCP97"/>.
4198  </annotation>-->
4199</reference>
4200
4201<reference anchor="RFC1952">
4202  <front>
4203    <title>GZIP file format specification version 4.3</title>
4204    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
4205      <organization>Aladdin Enterprises</organization>
4206      <address><email>ghost@aladdin.com</email></address>
4207    </author>
4208    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
4209      <address><email>gzip@prep.ai.mit.edu</email></address>
4210    </author>
4211    <author initials="M." surname="Adler" fullname="Mark Adler">
4212      <address><email>madler@alumni.caltech.edu</email></address>
4213    </author>
4214    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
4215      <address><email>ghost@aladdin.com</email></address>
4216    </author>
4217    <author initials="G." surname="Randers-Pehrson" fullname="Glenn Randers-Pehrson">
4218      <address><email>randeg@alumni.rpi.edu</email></address>
4219    </author>
4220    <date month="May" year="1996"/>
4221  </front>
4222  <seriesInfo name="RFC" value="1952"/>
4223  <!--<annotation>
4224    RFC 1952 is an Informational RFC, thus it might be less stable than
4225    this specification. On the other hand, this downward reference was
4226    present since the publication of <xref target="RFC2068" x:fmt="none">RFC 2068</xref> in 1997,
4227    therefore it is unlikely to cause problems in practice. See also
4228    <xref target="BCP97"/>.
4229  </annotation>-->
4230</reference>
4231
4232</references>
4233
4234<references title="Informative References">
4235
4236<reference anchor="ISO-8859-1">
4237  <front>
4238    <title>
4239     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
4240    </title>
4241    <author>
4242      <organization>International Organization for Standardization</organization>
4243    </author>
4244    <date year="1998"/>
4245  </front>
4246  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
4247</reference>
4248
4249<reference anchor='RFC1919'>
4250  <front>
4251    <title>Classical versus Transparent IP Proxies</title>
4252    <author initials='M.' surname='Chatel' fullname='Marc Chatel'>
4253      <address><email>mchatel@pax.eunet.ch</email></address>
4254    </author>
4255    <date year='1996' month='March' />
4256  </front>
4257  <seriesInfo name='RFC' value='1919' />
4258</reference>
4259
4260<reference anchor="RFC1945">
4261  <front>
4262    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
4263    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4264      <organization>MIT, Laboratory for Computer Science</organization>
4265      <address><email>timbl@w3.org</email></address>
4266    </author>
4267    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4268      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4269      <address><email>fielding@ics.uci.edu</email></address>
4270    </author>
4271    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4272      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
4273      <address><email>frystyk@w3.org</email></address>
4274    </author>
4275    <date month="May" year="1996"/>
4276  </front>
4277  <seriesInfo name="RFC" value="1945"/>
4278</reference>
4279
4280<reference anchor="RFC2045">
4281  <front>
4282    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
4283    <author initials="N." surname="Freed" fullname="Ned Freed">
4284      <organization>Innosoft International, Inc.</organization>
4285      <address><email>ned@innosoft.com</email></address>
4286    </author>
4287    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
4288      <organization>First Virtual Holdings</organization>
4289      <address><email>nsb@nsb.fv.com</email></address>
4290    </author>
4291    <date month="November" year="1996"/>
4292  </front>
4293  <seriesInfo name="RFC" value="2045"/>
4294</reference>
4295
4296<reference anchor="RFC2047">
4297  <front>
4298    <title abbrev="Message Header Extensions">MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text</title>
4299    <author initials="K." surname="Moore" fullname="Keith Moore">
4300      <organization>University of Tennessee</organization>
4301      <address><email>moore@cs.utk.edu</email></address>
4302    </author>
4303    <date month="November" year="1996"/>
4304  </front>
4305  <seriesInfo name="RFC" value="2047"/>
4306</reference>
4307
4308<reference anchor="RFC2068">
4309  <front>
4310    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4311    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
4312      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
4313      <address><email>fielding@ics.uci.edu</email></address>
4314    </author>
4315    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4316      <organization>MIT Laboratory for Computer Science</organization>
4317      <address><email>jg@w3.org</email></address>
4318    </author>
4319    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
4320      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
4321      <address><email>mogul@wrl.dec.com</email></address>
4322    </author>
4323    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4324      <organization>MIT Laboratory for Computer Science</organization>
4325      <address><email>frystyk@w3.org</email></address>
4326    </author>
4327    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
4328      <organization>MIT Laboratory for Computer Science</organization>
4329      <address><email>timbl@w3.org</email></address>
4330    </author>
4331    <date month="January" year="1997"/>
4332  </front>
4333  <seriesInfo name="RFC" value="2068"/>
4334</reference>
4335
4336<reference anchor="RFC2145">
4337  <front>
4338    <title abbrev="HTTP Version Numbers">Use and Interpretation of HTTP Version Numbers</title>
4339    <author initials="J.C." surname="Mogul" fullname="Jeffrey C. Mogul">
4340      <organization>Western Research Laboratory</organization>
4341      <address><email>mogul@wrl.dec.com</email></address>
4342    </author>
4343    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
4344      <organization>Department of Information and Computer Science</organization>
4345      <address><email>fielding@ics.uci.edu</email></address>
4346    </author>
4347    <author initials="J." surname="Gettys" fullname="Jim Gettys">
4348      <organization>MIT Laboratory for Computer Science</organization>
4349      <address><email>jg@w3.org</email></address>
4350    </author>
4351    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
4352      <organization>W3 Consortium</organization>
4353      <address><email>frystyk@w3.org</email></address>
4354    </author>
4355    <date month="May" year="1997"/>
4356  </front>
4357  <seriesInfo name="RFC" value="2145"/>
4358</reference>
4359
4360<reference anchor="RFC2616">
4361  <front>
4362    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
4363    <author initials="R." surname="Fielding" fullname="R. Fielding">
4364      <organization>University of California, Irvine</organization>
4365      <address><email>fielding@ics.uci.edu</email></address>
4366    </author>
4367    <author initials="J." surname="Gettys" fullname="J. Gettys">
4368      <organization>W3C</organization>
4369      <address><email>jg@w3.org</email></address>
4370    </author>
4371    <author initials="J." surname="Mogul" fullname="J. Mogul">
4372      <organization>Compaq Computer Corporation</organization>
4373      <address><email>mogul@wrl.dec.com</email></address>
4374    </author>
4375    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
4376      <organization>MIT Laboratory for Computer Science</organization>
4377      <address><email>frystyk@w3.org</email></address>
4378    </author>
4379    <author initials="L." surname="Masinter" fullname="L. Masinter">
4380      <organization>Xerox Corporation</organization>
4381      <address><email>masinter@parc.xerox.com</email></address>
4382    </author>
4383    <author initials="P." surname="Leach" fullname="P. Leach">
4384      <organization>Microsoft Corporation</organization>
4385      <address><email>paulle@microsoft.com</email></address>
4386    </author>
4387    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
4388      <organization>W3C</organization>
4389      <address><email>timbl@w3.org</email></address>
4390    </author>
4391    <date month="June" year="1999"/>
4392  </front>
4393  <seriesInfo name="RFC" value="2616"/>
4394</reference>
4395
4396<reference anchor='RFC2817'>
4397  <front>
4398    <title>Upgrading to TLS Within HTTP/1.1</title>
4399    <author initials='R.' surname='Khare' fullname='R. Khare'>
4400      <organization>4K Associates / UC Irvine</organization>
4401      <address><email>rohit@4K-associates.com</email></address>
4402    </author>
4403    <author initials='S.' surname='Lawrence' fullname='S. Lawrence'>
4404      <organization>Agranat Systems, Inc.</organization>
4405      <address><email>lawrence@agranat.com</email></address>
4406    </author>
4407    <date year='2000' month='May' />
4408  </front>
4409  <seriesInfo name='RFC' value='2817' />
4410</reference>
4411
4412<reference anchor='RFC2818'>
4413  <front>
4414    <title>HTTP Over TLS</title>
4415    <author initials='E.' surname='Rescorla' fullname='Eric Rescorla'>
4416      <organization>RTFM, Inc.</organization>
4417      <address><email>ekr@rtfm.com</email></address>
4418    </author>
4419    <date year='2000' month='May' />
4420  </front>
4421  <seriesInfo name='RFC' value='2818' />
4422</reference>
4423
4424<reference anchor='RFC3040'>
4425  <front>
4426    <title>Internet Web Replication and Caching Taxonomy</title>
4427    <author initials='I.' surname='Cooper' fullname='I. Cooper'>
4428      <organization>Equinix, Inc.</organization>
4429    </author>
4430    <author initials='I.' surname='Melve' fullname='I. Melve'>
4431      <organization>UNINETT</organization>
4432    </author>
4433    <author initials='G.' surname='Tomlinson' fullname='G. Tomlinson'>
4434      <organization>CacheFlow Inc.</organization>
4435    </author>
4436    <date year='2001' month='January' />
4437  </front>
4438  <seriesInfo name='RFC' value='3040' />
4439</reference>
4440
4441<reference anchor='RFC3864'>
4442  <front>
4443    <title>Registration Procedures for Message Header Fields</title>
4444    <author initials='G.' surname='Klyne' fullname='G. Klyne'>
4445      <organization>Nine by Nine</organization>
4446      <address><email>GK-IETF@ninebynine.org</email></address>
4447    </author>
4448    <author initials='M.' surname='Nottingham' fullname='M. Nottingham'>
4449      <organization>BEA Systems</organization>
4450      <address><email>mnot@pobox.com</email></address>
4451    </author>
4452    <author initials='J.' surname='Mogul' fullname='J. Mogul'>
4453      <organization>HP Labs</organization>
4454      <address><email>JeffMogul@acm.org</email></address>
4455    </author>
4456    <date year='2004' month='September' />
4457  </front>
4458  <seriesInfo name='BCP' value='90' />
4459  <seriesInfo name='RFC' value='3864' />
4460</reference>
4461
4462<reference anchor='RFC4033'>
4463  <front>
4464    <title>DNS Security Introduction and Requirements</title>
4465    <author initials='R.' surname='Arends' fullname='R. Arends'/>
4466    <author initials='R.' surname='Austein' fullname='R. Austein'/>
4467    <author initials='M.' surname='Larson' fullname='M. Larson'/>
4468    <author initials='D.' surname='Massey' fullname='D. Massey'/>
4469    <author initials='S.' surname='Rose' fullname='S. Rose'/>
4470    <date year='2005' month='March' />
4471  </front>
4472  <seriesInfo name='RFC' value='4033' />
4473</reference>
4474
4475<reference anchor="RFC4288">
4476  <front>
4477    <title>Media Type Specifications and Registration Procedures</title>
4478    <author initials="N." surname="Freed" fullname="N. Freed">
4479      <organization>Sun Microsystems</organization>
4480      <address>
4481        <email>ned.freed@mrochek.com</email>
4482      </address>
4483    </author>
4484    <author initials="J." surname="Klensin" fullname="J. Klensin">
4485      <address>
4486        <email>klensin+ietf@jck.com</email>
4487      </address>
4488    </author>
4489    <date year="2005" month="December"/>
4490  </front>
4491  <seriesInfo name="BCP" value="13"/>
4492  <seriesInfo name="RFC" value="4288"/>
4493</reference>
4494
4495<reference anchor='RFC4395'>
4496  <front>
4497    <title>Guidelines and Registration Procedures for New URI Schemes</title>
4498    <author initials='T.' surname='Hansen' fullname='T. Hansen'>
4499      <organization>AT&amp;T Laboratories</organization>
4500      <address>
4501        <email>tony+urireg@maillennium.att.com</email>
4502      </address>
4503    </author>
4504    <author initials='T.' surname='Hardie' fullname='T. Hardie'>
4505      <organization>Qualcomm, Inc.</organization>
4506      <address>
4507        <email>hardie@qualcomm.com</email>
4508      </address>
4509    </author>
4510    <author initials='L.' surname='Masinter' fullname='L. Masinter'>
4511      <organization>Adobe Systems</organization>
4512      <address>
4513        <email>LMM@acm.org</email>
4514      </address>
4515    </author>
4516    <date year='2006' month='February' />
4517  </front>
4518  <seriesInfo name='BCP' value='115' />
4519  <seriesInfo name='RFC' value='4395' />
4520</reference>
4521
4522<reference anchor='RFC4559'>
4523  <front>
4524    <title>SPNEGO-based Kerberos and NTLM HTTP Authentication in Microsoft Windows</title>
4525    <author initials='K.' surname='Jaganathan' fullname='K. Jaganathan'/>
4526    <author initials='L.' surname='Zhu' fullname='L. Zhu'/>
4527    <author initials='J.' surname='Brezak' fullname='J. Brezak'/>
4528    <date year='2006' month='June' />
4529  </front>
4530  <seriesInfo name='RFC' value='4559' />
4531</reference>
4532
4533<reference anchor='RFC5226'>
4534  <front>
4535    <title>Guidelines for Writing an IANA Considerations Section in RFCs</title>
4536    <author initials='T.' surname='Narten' fullname='T. Narten'>
4537      <organization>IBM</organization>
4538      <address><email>narten@us.ibm.com</email></address>
4539    </author>
4540    <author initials='H.' surname='Alvestrand' fullname='H. Alvestrand'>
4541      <organization>Google</organization>
4542      <address><email>Harald@Alvestrand.no</email></address>
4543    </author>
4544    <date year='2008' month='May' />
4545  </front>
4546  <seriesInfo name='BCP' value='26' />
4547  <seriesInfo name='RFC' value='5226' />
4548</reference>
4549
4550<reference anchor='RFC5246'>
4551   <front>
4552      <title>The Transport Layer Security (TLS) Protocol Version 1.2</title>
4553      <author initials='T.' surname='Dierks' fullname='T. Dierks'>
4554         <organization />
4555      </author>
4556      <author initials='E.' surname='Rescorla' fullname='E. Rescorla'>
4557         <organization>RTFM, Inc.</organization>
4558      </author>
4559      <date year='2008' month='August' />
4560   </front>
4561   <seriesInfo name='RFC' value='5246' />
4562</reference>
4563
4564<reference anchor="RFC5322">
4565  <front>
4566    <title>Internet Message Format</title>
4567    <author initials="P." surname="Resnick" fullname="P. Resnick">
4568      <organization>Qualcomm Incorporated</organization>
4569    </author>
4570    <date year="2008" month="October"/>
4571  </front> 
4572  <seriesInfo name="RFC" value="5322"/>
4573</reference>
4574
4575<reference anchor="RFC6265">
4576  <front>
4577    <title>HTTP State Management Mechanism</title>
4578    <author initials="A." surname="Barth" fullname="Adam Barth">
4579      <organization abbrev="U.C. Berkeley">
4580        University of California, Berkeley
4581      </organization>
4582      <address><email>abarth@eecs.berkeley.edu</email></address>
4583    </author>
4584    <date year="2011" month="April" />
4585  </front>
4586  <seriesInfo name="RFC" value="6265"/>
4587</reference>
4588
4589<!--<reference anchor='BCP97'>
4590  <front>
4591    <title>Handling Normative References to Standards-Track Documents</title>
4592    <author initials='J.' surname='Klensin' fullname='J. Klensin'>
4593      <address>
4594        <email>klensin+ietf@jck.com</email>
4595      </address>
4596    </author>
4597    <author initials='S.' surname='Hartman' fullname='S. Hartman'>
4598      <organization>MIT</organization>
4599      <address>
4600        <email>hartmans-ietf@mit.edu</email>
4601      </address>
4602    </author>
4603    <date year='2007' month='June' />
4604  </front>
4605  <seriesInfo name='BCP' value='97' />
4606  <seriesInfo name='RFC' value='4897' />
4607</reference>-->
4608
4609<reference anchor="Kri2001" target="http://arxiv.org/abs/cs.SE/0105018">
4610  <front>
4611    <title>HTTP Cookies: Standards, Privacy, and Politics</title>
4612    <author initials="D." surname="Kristol" fullname="David M. Kristol"/>
4613    <date year="2001" month="November"/>
4614  </front>
4615  <seriesInfo name="ACM Transactions on Internet Technology" value="Vol. 1, #2"/>
4616</reference>
4617
4618</references>
4619
4620
4621<section title="HTTP Version History" anchor="compatibility">
4622<t>
4623   HTTP has been in use by the World-Wide Web global information initiative
4624   since 1990. The first version of HTTP, later referred to as HTTP/0.9,
4625   was a simple protocol for hypertext data transfer across the Internet
4626   with only a single request method (GET) and no metadata.
4627   HTTP/1.0, as defined by <xref target="RFC1945"/>, added a range of request
4628   methods and MIME-like messaging that could include metadata about the data
4629   transferred and modifiers on the request/response semantics. However,
4630   HTTP/1.0 did not sufficiently take into consideration the effects of
4631   hierarchical proxies, caching, the need for persistent connections, or
4632   name-based virtual hosts. The proliferation of incompletely-implemented
4633   applications calling themselves "HTTP/1.0" further necessitated a
4634   protocol version change in order for two communicating applications
4635   to determine each other's true capabilities.
4636</t>
4637<t>
4638   HTTP/1.1 remains compatible with HTTP/1.0 by including more stringent
4639   requirements that enable reliable implementations, adding only
4640   those new features that will either be safely ignored by an HTTP/1.0
4641   recipient or only sent when communicating with a party advertising
4642   conformance with HTTP/1.1.
4643</t>
4644<t>
4645   It is beyond the scope of a protocol specification to mandate
4646   conformance with previous versions. HTTP/1.1 was deliberately
4647   designed, however, to make supporting previous versions easy.
4648   We would expect a general-purpose HTTP/1.1 server to understand
4649   any valid request in the format of HTTP/1.0 and respond appropriately
4650   with an HTTP/1.1 message that only uses features understood (or
4651   safely ignored) by HTTP/1.0 clients.  Likewise, we would expect
4652   an HTTP/1.1 client to understand any valid HTTP/1.0 response.
4653</t>
4654<t>
4655   Since HTTP/0.9 did not support header fields in a request,
4656   there is no mechanism for it to support name-based virtual
4657   hosts (selection of resource by inspection of the <x:ref>Host</x:ref> header
4658   field).  Any server that implements name-based virtual hosts
4659   ought to disable support for HTTP/0.9.  Most requests that
4660   appear to be HTTP/0.9 are, in fact, badly constructed HTTP/1.x
4661   requests wherein a buggy client failed to properly encode
4662   linear whitespace found in a URI reference and placed in
4663   the request-target.
4664</t>
4665
4666<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
4667<t>
4668   This section summarizes major differences between versions HTTP/1.0
4669   and HTTP/1.1.
4670</t>
4671
4672<section title="Multi-homed Web Servers" anchor="changes.to.simplify.multi-homed.web.servers.and.conserve.ip.addresses">
4673<t>
4674   The requirements that clients and servers support the <x:ref>Host</x:ref>
4675   header field (<xref target="header.host"/>), report an error if it is
4676   missing from an HTTP/1.1 request, and accept absolute URIs (<xref target="request-target"/>)
4677   are among the most important changes defined by HTTP/1.1.
4678</t>
4679<t>
4680   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
4681   addresses and servers; there was no other established mechanism for
4682   distinguishing the intended server of a request than the IP address
4683   to which that request was directed. The <x:ref>Host</x:ref> header field was
4684   introduced during the development of HTTP/1.1 and, though it was
4685   quickly implemented by most HTTP/1.0 browsers, additional requirements
4686   were placed on all HTTP/1.1 requests in order to ensure complete
4687   adoption.  At the time of this writing, most HTTP-based services
4688   are dependent upon the Host header field for targeting requests.
4689</t>
4690</section>
4691
4692<section title="Keep-Alive Connections" anchor="compatibility.with.http.1.0.persistent.connections">
4693<t>
4694   In HTTP/1.0, each connection is established by the client prior to the
4695   request and closed by the server after sending the response. However, some
4696   implementations implement the explicitly negotiated ("Keep-Alive") version
4697   of persistent connections described in <xref x:sec="19.7.1" x:fmt="of"
4698   target="RFC2068"/>.
4699</t>
4700<t>
4701   Some clients and servers might wish to be compatible with these previous
4702   approaches to persistent connections, by explicitly negotiating for them
4703   with a "Connection: keep-alive" request header field. However, some
4704   experimental implementations of HTTP/1.0 persistent connections are faulty;
4705   for example, if an HTTP/1.0 proxy server doesn't understand
4706   <x:ref>Connection</x:ref>, it will erroneously forward that header field
4707   to the next inbound server, which would result in a hung connection.
4708</t>
4709<t>
4710   One attempted solution was the introduction of a Proxy-Connection header
4711   field, targeted specifically at proxies. In practice, this was also
4712   unworkable, because proxies are often deployed in multiple layers, bringing
4713   about the same problem discussed above.
4714</t>
4715<t>
4716   As a result, clients are encouraged not to send the Proxy-Connection header
4717   field in any requests.
4718</t>
4719<t>
4720   Clients are also encouraged to consider the use of Connection: keep-alive
4721   in requests carefully; while they can enable persistent connections with
4722   HTTP/1.0 servers, clients using them need will need to monitor the
4723   connection for "hung" requests (which indicate that the client ought stop
4724   sending the header field), and this mechanism ought not be used by clients
4725   at all when a proxy is being used.
4726</t>
4727</section>
4728
4729<section title="Introduction of Transfer-Encoding" anchor="introduction.of.transfer-encoding">
4730<t>
4731   HTTP/1.1 introduces the <x:ref>Transfer-Encoding</x:ref> header field
4732   (<xref target="header.transfer-encoding"/>). Proxies/gateways &MUST; remove
4733   any transfer coding prior to forwarding a message via a MIME-compliant
4734   protocol.
4735</t>
4736</section>
4737
4738</section>
4739
4740<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
4741<t>
4742  HTTP's approach to error handling has been explained.
4743  (<xref target="conformance"/>)
4744</t>
4745<t>
4746  The expectation to support HTTP/0.9 requests has been removed.
4747</t>
4748<t>
4749  The term "Effective Request URI" has been introduced.
4750  (<xref target="effective.request.uri" />)
4751</t>
4752<t>
4753  HTTP messages can be (and often are) buffered by implementations; despite
4754  it sometimes being available as a stream, HTTP is fundamentally a
4755  message-oriented protocol.
4756  (<xref target="http.message" />)
4757</t>
4758<t>
4759  Minimum supported sizes for various protocol elements have been
4760  suggested, to improve interoperability.
4761</t>
4762<t>
4763  Header fields that span multiple lines ("line folding") are deprecated.
4764  (<xref target="field.parsing" />)
4765</t>
4766<t>
4767  The HTTP-version ABNF production has been clarified to be case-sensitive.
4768  Additionally, version numbers has been restricted to single digits, due
4769  to the fact that implementations are known to handle multi-digit version
4770  numbers incorrectly.
4771  (<xref target="http.version"/>)
4772</t>
4773<t>
4774  The HTTPS URI scheme is now defined by this specification; previously,
4775  it was done in  <xref target="RFC2818" x:fmt="of" x:sec="2.4"/>.
4776  (<xref target="https.uri"/>)
4777</t>
4778<t>
4779  The HTTPS URI scheme implies end-to-end security.
4780  (<xref target="https.uri"/>)
4781</t>
4782<t>
4783  Userinfo (i.e., username and password) are now disallowed in HTTP and
4784  HTTPS URIs, because of security issues related to their transmission on the
4785  wire.
4786  (<xref target="http.uri" />)
4787</t>
4788<t>
4789  Invalid whitespace around field-names is now required to be rejected,
4790  because accepting it represents a security vulnerability.
4791  (<xref target="header.fields"/>)
4792</t>
4793<t>
4794  The ABNF productions defining header fields now only list the field value.
4795  (<xref target="header.fields"/>)
4796</t>
4797<t>
4798  Rules about implicit linear whitespace between certain grammar productions
4799  have been removed; now whitespace is only allowed where specifically
4800  defined in the ABNF.
4801  (<xref target="whitespace"/>)
4802</t>
4803<t> 
4804  The NUL octet is no longer allowed in comment and quoted-string text, and
4805  handling of backslash-escaping in them has been clarified.
4806  (<xref target="field.components"/>)
4807</t> 
4808<t>
4809  The quoted-pair rule no longer allows escaping control characters other than
4810  HTAB.
4811  (<xref target="field.components"/>)
4812</t>
4813<t>
4814  Non-ASCII content in header fields and the reason phrase has been obsoleted
4815  and made opaque (the TEXT rule was removed).
4816  (<xref target="field.components"/>)
4817</t>
4818<t>
4819  Bogus "<x:ref>Content-Length</x:ref>" header fields are now required to be
4820  handled as errors by recipients.
4821  (<xref target="header.content-length"/>)
4822</t>
4823<t>
4824  The "identity" transfer coding token has been removed.
4825  (Sections <xref format="counter" target="message.body"/> and
4826  <xref format="counter" target="transfer.codings"/>)
4827</t>
4828<t>
4829  The algorithm for determining the message body length has been clarified
4830  to indicate all of the special cases (e.g., driven by methods or status
4831  codes) that affect it, and that new protocol elements cannot define such
4832  special cases.
4833  (<xref target="message.body.length"/>)
4834</t>
4835<t>
4836  "multipart/byteranges" is no longer a way of determining message body length
4837  detection.
4838  (<xref target="message.body.length"/>)
4839</t>
4840<t>
4841  CONNECT is a new, special case in determining message body length.
4842  (<xref target="message.body.length"/>)
4843</t>
4844<t>
4845  Chunk length does not include the count of the octets in the
4846  chunk header and trailer.
4847  (<xref target="chunked.encoding"/>)
4848</t>
4849<t>
4850  Use of chunk extensions is deprecated, and line folding in them is
4851  disallowed.
4852  (<xref target="chunked.encoding"/>)
4853</t>
4854<t>
4855  The path-absolute + query components of RFC3986 have been used to define the
4856  request-target, instead of abs_path from RFC 1808.
4857  (<xref target="request-target"/>)
4858</t>
4859<t>
4860  The asterisk form of the request-target is only allowed in the OPTIONS
4861  method.
4862  (<xref target="request-target"/>)
4863</t>
4864<t>
4865  Exactly when "close" connection options have to be sent has been clarified.
4866  (<xref target="header.connection"/>)
4867</t>
4868<t> 
4869  "hop-by-hop" header fields are required to appear in the Connection header
4870  field; just because they're defined as hop-by-hop in this specification
4871  doesn't exempt them.
4872  (<xref target="header.connection"/>)
4873</t>
4874<t>
4875  The limit of two connections per server has been removed.
4876  (<xref target="persistent.connections"/>)
4877</t>
4878<t>
4879  An idempotent sequence of requests is no longer required to be retried.
4880  (<xref target="persistent.connections"/>)
4881</t>
4882<t>
4883  The requirement to retry requests under certain circumstances when the
4884  server prematurely closes the connection has been removed.
4885  (<xref target="persistent.connections"/>)
4886</t>
4887<t>
4888  Some extraneous requirements about when servers are allowed to close
4889  connections prematurely have been removed.
4890  (<xref target="persistent.connections"/>)
4891</t>
4892<t>
4893  The semantics of the <x:ref>Upgrade</x:ref> header field is now defined in
4894  responses other than 101 (this was incorporated from <xref
4895  target="RFC2817"/>).
4896  (<xref target="header.upgrade"/>)
4897</t>
4898<t>
4899  Registration of Transfer Codings now requires IETF Review
4900  (<xref target="transfer.coding.registry"/>)
4901</t>
4902<t>
4903  The meaning of the "deflate" content coding has been clarified.
4904  (<xref target="deflate.coding" />)
4905</t>
4906<t>
4907  This specification now defines the Upgrade Token Registry, previously
4908  defined in <xref target="RFC2817" x:fmt="of" x:sec="7.2"/>.
4909  (<xref target="upgrade.token.registry"/>)
4910</t>
4911<t>
4912  Empty list elements in list productions (e.g., a list header containing
4913  ", ,") have been deprecated.
4914  (<xref target="abnf.extension"/>)
4915</t>
4916<t>
4917  Issues with the Keep-Alive and Proxy-Connection headers in requests
4918  are pointed out, with use of the latter being discouraged altogether.
4919  (<xref target="compatibility.with.http.1.0.persistent.connections" />)
4920</t>
4921</section>
4922</section>
4923
4924<section title="ABNF list extension: #rule" anchor="abnf.extension">
4925<t>
4926  A #rule extension to the ABNF rules of <xref target="RFC5234"/> is used to
4927  improve readability in the definitions of some header field values.
4928</t>
4929<t>
4930  A construct "#" is defined, similar to "*", for defining comma-delimited
4931  lists of elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating
4932  at least &lt;n&gt; and at most &lt;m&gt; elements, each separated by a single
4933  comma (",") and optional whitespace (OWS).   
4934</t>
4935<figure><preamble>
4936  Thus,
4937</preamble><artwork type="example">
4938  1#element =&gt; element *( OWS "," OWS element )
4939</artwork></figure>
4940<figure><preamble>
4941  and:
4942</preamble><artwork type="example">
4943  #element =&gt; [ 1#element ]
4944</artwork></figure>
4945<figure><preamble>
4946  and for n &gt;= 1 and m &gt; 1:
4947</preamble><artwork type="example">
4948  &lt;n&gt;#&lt;m&gt;element =&gt; element &lt;n-1&gt;*&lt;m-1&gt;( OWS "," OWS element )
4949</artwork></figure>
4950<t>
4951  For compatibility with legacy list rules, recipients &SHOULD; accept empty
4952  list elements. In other words, consumers would follow the list productions:
4953</t>
4954<figure><artwork type="example">
4955  #element =&gt; [ ( "," / element ) *( OWS "," [ OWS element ] ) ]
4956 
4957  1#element =&gt; *( "," OWS ) element *( OWS "," [ OWS element ] )
4958</artwork></figure>
4959<t>
4960  Note that empty elements do not contribute to the count of elements present,
4961  though.
4962</t>
4963<t>
4964  For example, given these ABNF productions:
4965</t>
4966<figure><artwork type="example">
4967  example-list      = 1#example-list-elmt
4968  example-list-elmt = token ; see <xref target="field.components"/> 
4969</artwork></figure>
4970<t>
4971  Then these are valid values for example-list (not including the double
4972  quotes, which are present for delimitation only):
4973</t>
4974<figure><artwork type="example">
4975  "foo,bar"
4976  "foo ,bar,"
4977  "foo , ,bar,charlie   "
4978</artwork></figure>
4979<t>
4980  But these values would be invalid, as at least one non-empty element is
4981  required:
4982</t>
4983<figure><artwork type="example">
4984  ""
4985  ","
4986  ",   ,"
4987</artwork></figure>
4988<t>
4989  <xref target="collected.abnf"/> shows the collected ABNF, with the list rules
4990  expanded as explained above.
4991</t>
4992</section>
4993
4994<?BEGININC p1-messaging.abnf-appendix ?>
4995<section xmlns:x="http://purl.org/net/xml2rfc/ext" title="Collected ABNF" anchor="collected.abnf">
4996<figure>
4997<artwork type="abnf" name="p1-messaging.parsed-abnf">
4998<x:ref>BWS</x:ref> = OWS
4999
5000<x:ref>Connection</x:ref> = *( "," OWS ) connection-option *( OWS "," [ OWS
5001 connection-option ] )
5002<x:ref>Content-Length</x:ref> = 1*DIGIT
5003
5004<x:ref>HTTP-message</x:ref> = start-line *( header-field CRLF ) CRLF [ message-body
5005 ]
5006<x:ref>HTTP-name</x:ref> = %x48.54.54.50 ; HTTP
5007<x:ref>HTTP-version</x:ref> = HTTP-name "/" DIGIT "." DIGIT
5008<x:ref>Host</x:ref> = uri-host [ ":" port ]
5009
5010<x:ref>OWS</x:ref> = *( SP / HTAB )
5011
5012<x:ref>RWS</x:ref> = 1*( SP / HTAB )
5013
5014<x:ref>TE</x:ref> = [ ( "," / t-codings ) *( OWS "," [ OWS t-codings ] ) ]
5015<x:ref>Trailer</x:ref> = *( "," OWS ) field-name *( OWS "," [ OWS field-name ] )
5016<x:ref>Transfer-Encoding</x:ref> = *( "," OWS ) transfer-coding *( OWS "," [ OWS
5017 transfer-coding ] )
5018
5019<x:ref>URI-reference</x:ref> = &lt;URI-reference, defined in [RFC3986], Section 4.1&gt;
5020<x:ref>Upgrade</x:ref> = *( "," OWS ) protocol *( OWS "," [ OWS protocol ] )
5021
5022<x:ref>Via</x:ref> = *( "," OWS ) ( received-protocol RWS received-by [ RWS comment
5023 ] ) *( OWS "," [ OWS ( received-protocol RWS received-by [ RWS
5024 comment ] ) ] )
5025
5026<x:ref>absolute-URI</x:ref> = &lt;absolute-URI, defined in [RFC3986], Section 4.3&gt;
5027<x:ref>absolute-form</x:ref> = absolute-URI
5028<x:ref>asterisk-form</x:ref> = "*"
5029<x:ref>attribute</x:ref> = token
5030<x:ref>authority</x:ref> = &lt;authority, defined in [RFC3986], Section 3.2&gt;
5031<x:ref>authority-form</x:ref> = authority
5032
5033<x:ref>chunk</x:ref> = chunk-size [ chunk-ext ] CRLF chunk-data CRLF
5034<x:ref>chunk-data</x:ref> = 1*OCTET
5035<x:ref>chunk-ext</x:ref> = *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
5036<x:ref>chunk-ext-name</x:ref> = token
5037<x:ref>chunk-ext-val</x:ref> = token / quoted-str-nf
5038<x:ref>chunk-size</x:ref> = 1*HEXDIG
5039<x:ref>chunked-body</x:ref> = *chunk last-chunk trailer-part CRLF
5040<x:ref>comment</x:ref> = "(" *( ctext / quoted-cpair / comment ) ")"
5041<x:ref>connection-option</x:ref> = token
5042<x:ref>ctext</x:ref> = HTAB / SP / %x21-27 ; '!'-'''
5043 / %x2A-5B ; '*'-'['
5044 / %x5D-7E ; ']'-'~'
5045 / obs-text
5046
5047<x:ref>field-content</x:ref> = *( HTAB / SP / VCHAR / obs-text )
5048<x:ref>field-name</x:ref> = token
5049<x:ref>field-value</x:ref> = *( field-content / obs-fold )
5050
5051<x:ref>header-field</x:ref> = field-name ":" OWS field-value BWS
5052<x:ref>http-URI</x:ref> = "http://" authority path-abempty [ "?" query ]
5053<x:ref>https-URI</x:ref> = "https://" authority path-abempty [ "?" query ]
5054
5055<x:ref>last-chunk</x:ref> = 1*"0" [ chunk-ext ] CRLF
5056
5057<x:ref>message-body</x:ref> = *OCTET
5058<x:ref>method</x:ref> = token
5059
5060<x:ref>obs-fold</x:ref> = CRLF ( SP / HTAB )
5061<x:ref>obs-text</x:ref> = %x80-FF
5062<x:ref>origin-form</x:ref> = path-absolute [ "?" query ]
5063
5064<x:ref>partial-URI</x:ref> = relative-part [ "?" query ]
5065<x:ref>path-abempty</x:ref> = &lt;path-abempty, defined in [RFC3986], Section 3.3&gt;
5066<x:ref>path-absolute</x:ref> = &lt;path-absolute, defined in [RFC3986], Section 3.3&gt;
5067<x:ref>port</x:ref> = &lt;port, defined in [RFC3986], Section 3.2.3&gt;
5068<x:ref>protocol</x:ref> = protocol-name [ "/" protocol-version ]
5069<x:ref>protocol-name</x:ref> = token
5070<x:ref>protocol-version</x:ref> = token
5071<x:ref>pseudonym</x:ref> = token
5072
5073<x:ref>qdtext</x:ref> = HTAB / SP / "!" / %x23-5B ; '#'-'['
5074 / %x5D-7E ; ']'-'~'
5075 / obs-text
5076<x:ref>qdtext-nf</x:ref> = HTAB / SP / "!" / %x23-5B ; '#'-'['
5077 / %x5D-7E ; ']'-'~'
5078 / obs-text
5079<x:ref>query</x:ref> = &lt;query, defined in [RFC3986], Section 3.4&gt;
5080<x:ref>quoted-cpair</x:ref> = "\" ( HTAB / SP / VCHAR / obs-text )
5081<x:ref>quoted-pair</x:ref> = "\" ( HTAB / SP / VCHAR / obs-text )
5082<x:ref>quoted-str-nf</x:ref> = DQUOTE *( qdtext-nf / quoted-pair ) DQUOTE
5083<x:ref>quoted-string</x:ref> = DQUOTE *( qdtext / quoted-pair ) DQUOTE
5084
5085<x:ref>rank</x:ref> = ( "0" [ "." *3DIGIT ] ) / ( "1" [ "." *3"0" ] )
5086<x:ref>reason-phrase</x:ref> = *( HTAB / SP / VCHAR / obs-text )
5087<x:ref>received-by</x:ref> = ( uri-host [ ":" port ] ) / pseudonym
5088<x:ref>received-protocol</x:ref> = [ protocol-name "/" ] protocol-version
5089<x:ref>relative-part</x:ref> = &lt;relative-part, defined in [RFC3986], Section 4.2&gt;
5090<x:ref>request-line</x:ref> = method SP request-target SP HTTP-version CRLF
5091<x:ref>request-target</x:ref> = origin-form / absolute-form / authority-form /
5092 asterisk-form
5093
5094<x:ref>special</x:ref> = "(" / ")" / "&lt;" / "&gt;" / "@" / "," / ";" / ":" / "\" /
5095 DQUOTE / "/" / "[" / "]" / "?" / "=" / "{" / "}"
5096<x:ref>start-line</x:ref> = request-line / status-line
5097<x:ref>status-code</x:ref> = 3DIGIT
5098<x:ref>status-line</x:ref> = HTTP-version SP status-code SP reason-phrase CRLF
5099
5100<x:ref>t-codings</x:ref> = "trailers" / ( transfer-coding [ t-ranking ] )
5101<x:ref>t-ranking</x:ref> = OWS ";" OWS "q=" rank
5102<x:ref>tchar</x:ref> = "!" / "#" / "$" / "%" / "&amp;" / "'" / "*" / "+" / "-" / "." /
5103 "^" / "_" / "`" / "|" / "~" / DIGIT / ALPHA
5104<x:ref>token</x:ref> = 1*tchar
5105<x:ref>trailer-part</x:ref> = *( header-field CRLF )
5106<x:ref>transfer-coding</x:ref> = "chunked" / "compress" / "deflate" / "gzip" /
5107 transfer-extension
5108<x:ref>transfer-extension</x:ref> = token *( OWS ";" OWS transfer-parameter )
5109<x:ref>transfer-parameter</x:ref> = attribute BWS "=" BWS value
5110
5111<x:ref>uri-host</x:ref> = &lt;host, defined in [RFC3986], Section 3.2.2&gt;
5112
5113<x:ref>value</x:ref> = word
5114
5115<x:ref>word</x:ref> = token / quoted-string
5116</artwork>
5117</figure>
5118</section>
5119<?ENDINC p1-messaging.abnf-appendix ?>
5120
5121<section title="Change Log (to be removed by RFC Editor before publication)" anchor="change.log">
5122
5123<section title="Since RFC 2616">
5124<t>
5125  Changes up to the first Working Group Last Call draft are summarized
5126  in <eref target="http://trac.tools.ietf.org/html/draft-ietf-httpbis-p1-messaging-21#appendix-D"/>.
5127</t>
5128</section>
5129
5130<section title="Since draft-ietf-httpbis-p1-messaging-21" anchor="changes.since.21">
5131<t>
5132  Closed issues:
5133  <list style="symbols">
5134    <t>
5135      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/128"/>:
5136      "Cite HTTPS URI scheme definition" (the spec now includes the HTTPs
5137      scheme definition and thus updates RFC 2818)
5138    </t>
5139    <t>
5140      <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/389"/>:
5141      "mention of 'proxies' in section about caches"
5142    </t>
5143  </list>
5144</t>
5145</section>
5146
5147</section>
5148
5149</back>
5150</rfc>
Note: See TracBrowser for help on using the repository browser.