source: draft-ietf-httpbis/10/p1-messaging.xml @ 845

Last change on this file since 845 was 845, checked in by fielding@…, 9 years ago

update organization for Jim Gettys

  • Property svn:eol-style set to native
File size: 239.6 KB
[29]1<?xml version="1.0" encoding="utf-8"?>
[101]2<?xml-stylesheet type='text/xsl' href='../myxml2rfc.xslt'?>
[8]3<!DOCTYPE rfc [
4  <!ENTITY MAY "<bcp14 xmlns=''>MAY</bcp14>">
5  <!ENTITY MUST "<bcp14 xmlns=''>MUST</bcp14>">
6  <!ENTITY MUST-NOT "<bcp14 xmlns=''>MUST NOT</bcp14>">
7  <!ENTITY OPTIONAL "<bcp14 xmlns=''>OPTIONAL</bcp14>">
8  <!ENTITY RECOMMENDED "<bcp14 xmlns=''>RECOMMENDED</bcp14>">
9  <!ENTITY REQUIRED "<bcp14 xmlns=''>REQUIRED</bcp14>">
10  <!ENTITY SHALL "<bcp14 xmlns=''>SHALL</bcp14>">
11  <!ENTITY SHALL-NOT "<bcp14 xmlns=''>SHALL NOT</bcp14>">
12  <!ENTITY SHOULD "<bcp14 xmlns=''>SHOULD</bcp14>">
13  <!ENTITY SHOULD-NOT "<bcp14 xmlns=''>SHOULD NOT</bcp14>">
[841]14  <!ENTITY ID-VERSION "10">
[832]15  <!ENTITY ID-MONTH "July">
[741]16  <!ENTITY ID-YEAR "2010">
[640]17  <!ENTITY caching-overview       "<xref target='Part6' x:rel='#caching.overview' xmlns:x=''/>">
[31]18  <!ENTITY payload                "<xref target='Part3' xmlns:x=''/>">
[115]19  <!ENTITY media-types            "<xref target='Part3' x:rel='#media.types' xmlns:x=''/>">
20  <!ENTITY content-codings        "<xref target='Part3' x:rel='#content.codings' xmlns:x=''/>">
[31]21  <!ENTITY CONNECT                "<xref target='Part2' x:rel='#CONNECT' xmlns:x=''/>">
22  <!ENTITY content.negotiation    "<xref target='Part3' x:rel='#content.negotiation' xmlns:x=''/>">
23  <!ENTITY diff2045entity         "<xref target='Part3' x:rel='#differences.between.http.entities.and.rfc.2045.entities' xmlns:x=''/>">
24  <!ENTITY entity                 "<xref target='Part3' x:rel='#entity' xmlns:x=''/>">
[207]25  <!ENTITY entity-body            "<xref target='Part3' x:rel='#entity.body' xmlns:x=''/>">
[31]26  <!ENTITY entity-header-fields   "<xref target='Part3' x:rel='#entity.header.fields' xmlns:x=''/>">
[769]27  <!ENTITY entity-length          "<xref target='Part3' x:rel='#entity.length' xmlns:x=''/>">
[31]28  <!ENTITY header-cache-control   "<xref target='Part6' x:rel='#header.cache-control' xmlns:x=''/>">
29  <!ENTITY header-expect          "<xref target='Part2' x:rel='#header.expect' xmlns:x=''/>">
30  <!ENTITY header-pragma          "<xref target='Part6' x:rel='#header.pragma' xmlns:x=''/>">
31  <!ENTITY header-warning         "<xref target='Part6' x:rel='#header.warning' xmlns:x=''/>">
32  <!ENTITY idempotent-methods     "<xref target='Part2' x:rel='#idempotent.methods' xmlns:x=''/>">
33  <!ENTITY request-header-fields  "<xref target='Part2' x:rel='#request.header.fields' xmlns:x=''/>">
34  <!ENTITY response-header-fields "<xref target='Part2' x:rel='#response.header.fields' xmlns:x=''/>">
35  <!ENTITY status-codes           "<xref target='Part2' x:rel='' xmlns:x=''/>">
36  <!ENTITY status-100             "<xref target='Part2' x:rel='#status.100' xmlns:x=''/>">
37  <!ENTITY status-1xx             "<xref target='Part2' x:rel='#status.1xx' xmlns:x=''/>">
38  <!ENTITY status-414             "<xref target='Part2' x:rel='#status.414' xmlns:x=''/>">
40<?rfc toc="yes" ?>
[29]41<?rfc symrefs="yes" ?>
42<?rfc sortrefs="yes" ?>
[8]43<?rfc compact="yes"?>
44<?rfc subcompact="no" ?>
45<?rfc linkmailto="no" ?>
46<?rfc editing="no" ?>
[203]47<?rfc comments="yes"?>
48<?rfc inline="yes"?>
[799]49<?rfc rfcedstyle="yes"?>
[8]50<?rfc-ext allow-markup-in-artwork="yes" ?>
51<?rfc-ext include-references-in-index="yes" ?>
[684]52<rfc obsoletes="2616" updates="2817" category="std" x:maturity-level="draft"
[446]53     ipr="pre5378Trust200902" docName="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"
[153]54     xmlns:x=''>
[120]57  <title abbrev="HTTP/1.1, Part 1">HTTP/1.1, part 1: URIs, Connections, and Message Parsing</title>
[29]59  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
60    <organization abbrev="Day Software">Day Software</organization>
[8]61    <address>
62      <postal>
[29]63        <street>23 Corporate Plaza DR, Suite 280</street>
64        <city>Newport Beach</city>
[8]65        <region>CA</region>
[29]66        <code>92660</code>
67        <country>USA</country>
[8]68      </postal>
[29]69      <phone>+1-949-706-5300</phone>
70      <facsimile>+1-949-706-5305</facsimile>
71      <email></email>
72      <uri></uri>
[8]73    </address>
74  </author>
[29]76  <author initials="J." surname="Gettys" fullname="Jim Gettys">
[845]77    <organization abbrev="Alcatel-Lucent">Alcatel-Lucent Bell Labs</organization>
[8]78    <address>
79      <postal>
[29]80        <street>21 Oak Knoll Road</street>
81        <city>Carlisle</city>
[8]82        <region>MA</region>
[29]83        <code>01741</code>
84        <country>USA</country>
[8]85      </postal>
[845]86      <email></email>
87      <uri></uri>
[8]88    </address>
89  </author>
91  <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
[29]92    <organization abbrev="HP">Hewlett-Packard Company</organization>
[8]93    <address>
94      <postal>
[29]95        <street>HP Labs, Large Scale Systems Group</street>
96        <street>1501 Page Mill Road, MS 1177</street>
[8]97        <city>Palo Alto</city>
98        <region>CA</region>
[29]99        <code>94304</code>
100        <country>USA</country>
[8]101      </postal>
[29]102      <email></email>
[8]103    </address>
104  </author>
106  <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
[29]107    <organization abbrev="Microsoft">Microsoft Corporation</organization>
[8]108    <address>
109      <postal>
[29]110        <street>1 Microsoft Way</street>
111        <city>Redmond</city>
112        <region>WA</region>
113        <code>98052</code>
114        <country>USA</country>
[8]115      </postal>
[29]116      <email></email>
[8]117    </address>
118  </author>
120  <author initials="L." surname="Masinter" fullname="Larry Masinter">
[29]121    <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
[8]122    <address>
123      <postal>
[29]124        <street>345 Park Ave</street>
125        <city>San Jose</city>
[8]126        <region>CA</region>
[29]127        <code>95110</code>
128        <country>USA</country>
[8]129      </postal>
[29]130      <email></email>
131      <uri></uri>
[8]132    </address>
133  </author>
135  <author initials="P." surname="Leach" fullname="Paul J. Leach">
136    <organization abbrev="Microsoft">Microsoft Corporation</organization>
137    <address>
138      <postal>
139        <street>1 Microsoft Way</street>
140        <city>Redmond</city>
141        <region>WA</region>
142        <code>98052</code>
143      </postal>
144      <email></email>
145    </address>
146  </author>
148  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
149    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
150    <address>
151      <postal>
[34]152        <street>MIT Computer Science and Artificial Intelligence Laboratory</street>
153        <street>The Stata Center, Building 32</street>
154        <street>32 Vassar Street</street>
[8]155        <city>Cambridge</city>
156        <region>MA</region>
157        <code>02139</code>
[29]158        <country>USA</country>
[8]159      </postal>
160      <email></email>
[34]161      <uri></uri>
[8]162    </address>
163  </author>
[95]165  <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
[94]166    <organization abbrev="W3C">World Wide Web Consortium</organization>
167    <address>
168      <postal>
169        <street>W3C / ERCIM</street>
170        <street>2004, rte des Lucioles</street>
171        <city>Sophia-Antipolis</city>
172        <region>AM</region>
173        <code>06902</code>
174        <country>France</country>
175      </postal>
176      <email></email>
177      <uri></uri>
178    </address>
179  </author>
[95]181  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
182    <organization abbrev="greenbytes">greenbytes GmbH</organization>
183    <address>
184      <postal>
185        <street>Hafenweg 16</street>
186        <city>Muenster</city><region>NW</region><code>48155</code>
187        <country>Germany</country>
188      </postal>
[609]189      <phone>+49 251 2807760</phone>
190      <facsimile>+49 251 2807761</facsimile>
191      <email></email>
192      <uri></uri>
[95]193    </address>
194  </author>
[841]196  <date month="&ID-MONTH;" year="&ID-YEAR;" day="12"/>
[440]197  <workgroup>HTTPbis Working Group</workgroup>
201   The Hypertext Transfer Protocol (HTTP) is an application-level
[451]202   protocol for distributed, collaborative, hypertext information
[29]203   systems. HTTP has been in use by the World Wide Web global information
[35]204   initiative since 1990. This document is Part 1 of the seven-part specification
[29]205   that defines the protocol referred to as "HTTP/1.1" and, taken together,
[51]206   obsoletes RFC 2616.  Part 1 provides an overview of HTTP and
[29]207   its associated terminology, defines the "http" and "https" Uniform
208   Resource Identifier (URI) schemes, defines the generic message syntax
209   and parsing requirements for HTTP message frames, and describes
210   general security concerns for implementations.
214<note title="Editorial Note (To be removed by RFC Editor)">
215  <t>
216    Discussion of this draft should take place on the HTTPBIS working group
217    mailing list ( The current issues list is
[324]218    at <eref target=""/>
[36]219    and related documents (including fancy diffs) can be found at
[324]220    <eref target=""/>.
[36]221  </t>
[153]222  <t>
[773]223    The changes in this draft are summarized in <xref target="changes.since.09"/>.
[153]224  </t>
228<section title="Introduction" anchor="introduction">
[8]230   The Hypertext Transfer Protocol (HTTP) is an application-level
[374]231   request/response protocol that uses extensible semantics and MIME-like
[391]232   message payloads for flexible interaction with network-based hypertext
[374]233   information systems. HTTP relies upon the Uniform Resource Identifier (URI)
[544]234   standard <xref target="RFC3986"/> to indicate request targets and
[391]235   relationships between resources.
[374]236   Messages are passed in a format similar to that used by Internet mail
237   <xref target="RFC5322"/> and the Multipurpose Internet Mail Extensions
238   (MIME) <xref target="RFC2045"/> (see &diff2045entity; for the differences
239   between HTTP and MIME messages).
[544]242   HTTP is a generic interface protocol for information systems. It is
[391]243   designed to hide the details of how a service is implemented by presenting
244   a uniform interface to clients that is independent of the types of
245   resources provided. Likewise, servers do not need to be aware of each
246   client's purpose: an HTTP request can be considered in isolation rather
247   than being associated with a specific type of client or a predetermined
248   sequence of application steps. The result is a protocol that can be used
249   effectively in many different contexts and for which implementations can
250   evolve independently over time.
[374]253   HTTP is also designed for use as a generic protocol for translating
[544]254   communication to and from other Internet information systems.
[374]255   HTTP proxies and gateways provide access to alternative information
[451]256   services by translating their diverse protocols into a hypertext
[374]257   format that can be viewed and manipulated by clients in the same way
258   as HTTP services.
[544]261   One consequence of HTTP flexibility is that the protocol cannot be
262   defined in terms of what occurs behind the interface. Instead, we
263   are limited to defining the syntax of communication, the intent
264   of received communication, and the expected behavior of recipients.
265   If the communication is considered in isolation, then successful
266   actions should be reflected in corresponding changes to the
267   observable interface provided by servers. However, since multiple
268   clients may act in parallel and perhaps at cross-purposes, we
269   cannot require that such changes be observable beyond the scope
270   of a single response.
[374]273   This document is Part 1 of the seven-part specification of HTTP,
274   defining the protocol referred to as "HTTP/1.1" and obsoleting
275   <xref target="RFC2616"/>.
[544]276   Part 1 describes the architectural elements that are used or
[621]277   referred to in HTTP, defines the "http" and "https" URI schemes,
278   describes overall network operation and connection management,
279   and defines HTTP message framing and forwarding requirements.
[374]280   Our goal is to define all of the mechanisms necessary for HTTP message
281   handling that are independent of message semantics, thereby defining the
[544]282   complete set of requirements for message parsers and
[391]283   message-forwarding intermediaries.
[8]286<section title="Requirements" anchor="intro.requirements">
288   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
289   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
[96]290   document are to be interpreted as described in <xref target="RFC2119"/>.
293   An implementation is not compliant if it fails to satisfy one or more
[802]294   of the "MUST" or "REQUIRED" level requirements for the protocols it
295   implements. An implementation that satisfies all the "MUST" or "REQUIRED"
296   level and all the "SHOULD" level requirements for its protocols is said
297   to be "unconditionally compliant"; one that satisfies all the "MUST"
298   level requirements but not all the "SHOULD" level requirements for its
[8]299   protocols is said to be "conditionally compliant."
[390]303<section title="Syntax Notation" anchor="notation">
304<iref primary="true" item="Grammar" subitem="ALPHA"/>
305<iref primary="true" item="Grammar" subitem="CR"/>
306<iref primary="true" item="Grammar" subitem="CRLF"/>
307<iref primary="true" item="Grammar" subitem="CTL"/>
308<iref primary="true" item="Grammar" subitem="DIGIT"/>
309<iref primary="true" item="Grammar" subitem="DQUOTE"/>
310<iref primary="true" item="Grammar" subitem="HEXDIG"/>
311<iref primary="true" item="Grammar" subitem="LF"/>
312<iref primary="true" item="Grammar" subitem="OCTET"/>
313<iref primary="true" item="Grammar" subitem="SP"/>
[395]314<iref primary="true" item="Grammar" subitem="VCHAR"/>
[390]315<iref primary="true" item="Grammar" subitem="WSP"/>
317   This specification uses the Augmented Backus-Naur Form (ABNF) notation
318   of <xref target="RFC5234"/>.
[390]320<t anchor="core.rules">
321  <x:anchor-alias value="ALPHA"/>
322  <x:anchor-alias value="CTL"/>
323  <x:anchor-alias value="CR"/>
324  <x:anchor-alias value="CRLF"/>
325  <x:anchor-alias value="DIGIT"/>
326  <x:anchor-alias value="DQUOTE"/>
327  <x:anchor-alias value="HEXDIG"/>
328  <x:anchor-alias value="LF"/>
329  <x:anchor-alias value="OCTET"/>
330  <x:anchor-alias value="SP"/>
[395]331  <x:anchor-alias value="VCHAR"/>
[390]332  <x:anchor-alias value="WSP"/>
[543]333   The following core rules are included by
[390]334   reference, as defined in <xref target="RFC5234" x:fmt="," x:sec="B.1"/>:
[395]335   ALPHA (letters), CR (carriage return), CRLF (CR LF), CTL (controls),
[390]336   DIGIT (decimal 0-9), DQUOTE (double quote),
[395]337   HEXDIG (hexadecimal 0-9/A-F/a-f), LF (line feed),
338   OCTET (any 8-bit sequence of data), SP (space),
339   VCHAR (any visible <xref target="USASCII"/> character),
[401]340   and WSP (whitespace).
343   As a syntactical convention, ABNF rule names prefixed with "obs-" denote
344   "obsolete" grammar rules that appear for historical reasons.
[368]347<section title="ABNF Extension: #rule" anchor="notation.abnf">
349  The #rule extension to the ABNF rules of <xref target="RFC5234"/> is used to
350  improve readability.
353  A construct "#" is defined, similar to "*", for defining comma-delimited
354  lists of elements. The full form is "&lt;n&gt;#&lt;m&gt;element" indicating
355  at least &lt;n&gt; and at most &lt;m&gt; elements, each separated by a single
356  comma (",") and optional whitespace (OWS,
357  <xref target="basic.rules"/>).   
360  Thus,
[400]361</preamble><artwork type="example">
362  1#element =&gt; element *( OWS "," OWS element )
365  and:
[400]366</preamble><artwork type="example">
367  #element =&gt; [ 1#element ]
370  and for n &gt;= 1 and m &gt; 1:
[400]371</preamble><artwork type="example">
372  &lt;n&gt;#&lt;m&gt;element =&gt; element &lt;n-1&gt;*&lt;m-1&gt;( OWS "," OWS element )
375  For compatibility with legacy list rules, recipients &SHOULD; accept empty
376  list elements. In other words, consumers would follow the list productions:
[400]378<figure><artwork type="example">
[458]379  #element =&gt; [ ( "," / element ) *( OWS "," [ OWS element ] ) ]
381  1#element =&gt; *( "," OWS ) element *( OWS "," [ OWS element ] )
[738]384  Note that empty elements do not contribute to the count of elements present,
385  though.
388  For example, given these ABNF productions:
390<figure><artwork type="example">
391  example-list      = 1#example-list-elmt
392  example-list-elmt = token ; see <xref target="basic.rules"/> 
395  Then these are valid values for example-list (not including the double
396  quotes, which are present for delimitation only):
398<figure><artwork type="example">
399  "foo,bar"
400  " foo ,bar,"
401  "  foo , ,bar,charlie   "
402  "foo ,bar,   charlie "
405  But these values would be invalid, as at least one non-empty element is
406  required:
408<figure><artwork type="example">
409  ""
410  ","
411  ",   ,"
[421]414  <xref target="collected.abnf"/> shows the collected ABNF, with the list rules
415  expanded as explained above.
[8]419<section title="Basic Rules" anchor="basic.rules">
[229]420<t anchor="rule.CRLF">
421  <x:anchor-alias value="CRLF"/>
[8]422   HTTP/1.1 defines the sequence CR LF as the end-of-line marker for all
423   protocol elements except the entity-body (see <xref target="tolerant.applications"/> for
424   tolerant applications). The end-of-line marker within an entity-body
[115]425   is defined by its associated media type, as described in &media-types;.
[229]427<t anchor="rule.LWS">
[395]428   This specification uses three rules to denote the use of linear
429   whitespace: OWS (optional whitespace), RWS (required whitespace), and
430   BWS ("bad" whitespace).
[401]433   The OWS rule is used where zero or more linear whitespace characters may
[395]434   appear. OWS &SHOULD; either not be produced or be produced as a single SP
435   character. Multiple OWS characters that occur within field-content &SHOULD;
436   be replaced with a single SP before interpreting the field value or
437   forwarding the message downstream.
[401]440   RWS is used when at least one linear whitespace character is required to
[395]441   separate field tokens. RWS &SHOULD; be produced as a single SP character.
442   Multiple RWS characters that occur within field-content &SHOULD; be
443   replaced with a single SP before interpreting the field value or
444   forwarding the message downstream.
[395]447   BWS is used where the grammar allows optional whitespace for historical
448   reasons but senders &SHOULD-NOT; produce it in messages. HTTP/1.1
449   recipients &MUST; accept such bad optional whitespace and remove it before
450   interpreting the field value or forwarding the message downstream.
[351]452<t anchor="rule.whitespace">
453  <x:anchor-alias value="BWS"/>
454  <x:anchor-alias value="OWS"/>
455  <x:anchor-alias value="RWS"/>
456  <x:anchor-alias value="obs-fold"/>
[351]458<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"/>
[367]459  <x:ref>OWS</x:ref>            = *( [ obs-fold ] <x:ref>WSP</x:ref> )
[401]460                 ; "optional" whitespace
[351]461  <x:ref>RWS</x:ref>            = 1*( [ obs-fold ] <x:ref>WSP</x:ref> )
[401]462                 ; "required" whitespace
[351]463  <x:ref>BWS</x:ref>            = <x:ref>OWS</x:ref>
[401]464                 ; "bad" whitespace
[351]465  <x:ref>obs-fold</x:ref>       = <x:ref>CRLF</x:ref>
[647]466                 ; see <xref target="header.fields"/>
[229]468<t anchor="rule.token.separators">
469  <x:anchor-alias value="tchar"/>
470  <x:anchor-alias value="token"/>
[744]471  <x:anchor-alias value="special"/>
[810]472  <x:anchor-alias value="word"/>
[747]473   Many HTTP/1.1 header field values consist of words (token or quoted-string)
474   separated by whitespace or special characters. These special characters
475   &MUST; be in a quoted string to be used within a parameter value (as defined
476   in <xref target="transfer.codings"/>).
[810]478<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"/>
479  <x:ref>word</x:ref>           = <x:ref>token</x:ref> / <x:ref>quoted-string</x:ref>
[744]481  <x:ref>token</x:ref>          = 1*<x:ref>tchar</x:ref>
483  IMPORTANT: when editing "tchar" make sure that "special" is updated accordingly!!!
484 -->
[334]485  <x:ref>tchar</x:ref>          = "!" / "#" / "$" / "%" / "&amp;" / "'" / "*"
486                 / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
487                 / <x:ref>DIGIT</x:ref> / <x:ref>ALPHA</x:ref>
[744]488                 ; any <x:ref>VCHAR</x:ref>, except <x:ref>special</x:ref>
490  <x:ref>special</x:ref>        = "(" / ")" / "&lt;" / ">" / "@" / ","
491                 / ";" / ":" / "\" / DQUOTE / "/" / "["
492                 / "]" / "?" / "=" / "{" / "}"
[229]494<t anchor="rule.quoted-string">
495  <x:anchor-alias value="quoted-string"/>
496  <x:anchor-alias value="qdtext"/>
[395]497  <x:anchor-alias value="obs-text"/>
[8]498   A string of text is parsed as a single word if it is quoted using
499   double-quote marks.
[395]501<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"/>
[429]502  <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>
[687]503  <x:ref>qdtext</x:ref>         = <x:ref>OWS</x:ref> / %x21 / %x23-5B / %x5D-7E / <x:ref>obs-text</x:ref>
504                 ; <x:ref>OWS</x:ref> / &lt;<x:ref>VCHAR</x:ref> except <x:ref>DQUOTE</x:ref> and "\"&gt; / <x:ref>obs-text</x:ref> 
[395]505  <x:ref>obs-text</x:ref>       = %x80-FF
[229]507<t anchor="rule.quoted-pair">
508  <x:anchor-alias value="quoted-pair"/>
[696]509   The backslash character ("\") can be used as a single-character
[703]510   quoting mechanism within quoted-string constructs:
[696]512<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-pair"/>
513  <x:ref>quoted-pair</x:ref>    = "\" ( <x:ref>WSP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
516   Producers &SHOULD-NOT; escape characters that do not require escaping
[703]517   (i.e., other than DQUOTE and the backslash character).
521<section title="ABNF Rules defined in other Parts of the Specification" anchor="abnf.dependencies">
[229]522  <x:anchor-alias value="request-header"/>
523  <x:anchor-alias value="response-header"/>
524  <x:anchor-alias value="entity-body"/>
525  <x:anchor-alias value="entity-header"/>
526  <x:anchor-alias value="Cache-Control"/>
527  <x:anchor-alias value="Pragma"/>
528  <x:anchor-alias value="Warning"/>
530  The ABNF rules below are defined in other parts:
532<figure><!-- Part2--><artwork type="abnf2616">
[229]533  <x:ref>request-header</x:ref>  = &lt;request-header, defined in &request-header-fields;&gt;
534  <x:ref>response-header</x:ref> = &lt;response-header, defined in &response-header-fields;&gt;
536<figure><!-- Part3--><artwork type="abnf2616">
[229]537  <x:ref>entity-body</x:ref>     = &lt;entity-body, defined in &entity-body;&gt;
538  <x:ref>entity-header</x:ref>   = &lt;entity-header, defined in &entity-header-fields;&gt;
540<figure><!-- Part6--><artwork type="abnf2616">
[229]541  <x:ref>Cache-Control</x:ref>   = &lt;Cache-Control, defined in &header-pragma;&gt;
542  <x:ref>Pragma</x:ref>          = &lt;Pragma, defined in &header-pragma;&gt;
543  <x:ref>Warning</x:ref>         = &lt;Warning, defined in &header-warning;&gt;
[391]550<section title="HTTP architecture" anchor="architecture">
[621]552   HTTP was created for the World Wide Web architecture
[391]553   and has evolved over time to support the scalability needs of a worldwide
554   hypertext system. Much of that architecture is reflected in the terminology
555   and syntax productions used to define HTTP.
[630]558<section title="Client/Server Operation" anchor="operation">
559<iref item="client"/>
560<iref item="server"/>
561<iref item="connection"/>
[630]563   HTTP is a request/response protocol that operates by exchanging messages
564   across a reliable transport or session-layer connection. An HTTP client
565   is a program that establishes a connection to a server for the purpose
566   of sending one or more HTTP requests.  An HTTP server is a program that
567   accepts connections in order to service HTTP requests by sending HTTP
568   responses.
[630]570<iref item="user agent"/>
571<iref item="origin server"/>
[630]573   Note that the terms "client" and "server" refer only to the roles that
574   these programs perform for a particular connection.  The same program
575   may act as a client on some connections and a server on others.  We use
576   the term "user agent" to refer to the program that initiates a request,
577   such as a WWW browser, editor, or spider (web-traversing robot), and
578   the term "origin server" to refer to the program that can originate
579   authoritative responses to a request.
582   Most HTTP communication consists of a retrieval request (GET) for
583   a representation of some resource identified by a URI.  In the
[624]584   simplest case, this may be accomplished via a single connection (v)
585   between the user agent (UA) and the origin server (O).
587<figure><artwork type="drawing">
588       request chain ------------------------&gt;
589    UA -------------------v------------------- O
590       &lt;----------------------- response chain
[630]592<iref item="message"/>
593<iref item="request"/>
594<iref item="response"/>
[630]596   A client sends an HTTP request to the server in the form of a request
597   message (<xref target="request"/>), beginning with a method, URI, and
598   protocol version, followed by MIME-like header fields containing
599   request modifiers, client information, and payload metadata, an empty
[677]600   line to indicate the end of the header section, and finally the payload
601   body (if any).
604   A server responds to the client's request by sending an HTTP response
605   message (<xref target="response"/>), beginning with a status line that
606   includes the protocol version, a success or error code, and textual
[630]607   reason phrase, followed by MIME-like header fields containing server
[677]608   information, resource metadata, and payload metadata, an empty line to
609   indicate the end of the header section, and finally the payload body (if any).
[630]612   The following example illustrates a typical message exchange for a
613   GET request on the URI "":
[630]616client request:
[803]617</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
[633]618GET /hello.txt HTTP/1.1
619User-Agent: curl/7.16.3 libcurl/7.16.3 OpenSSL/0.9.7l zlib/1.2.3
621Accept: */*
[630]625server response:
[633]626</preamble><artwork type="message/http; msgtype=&#34;response&#34;" x:indent-with="  ">
627HTTP/1.1 200 OK
628Date: Mon, 27 Jul 2009 12:28:53 GMT
629Server: Apache
630Last-Modified: Wed, 22 Jul 2009 19:15:56 GMT
631ETag: "34aa387-d-1568eb00"
632Accept-Ranges: bytes
633Content-Length: <x:length-of target="exbody"/>
634Vary: Accept-Encoding
635Content-Type: text/plain
[633]637<x:span anchor="exbody">Hello World!
641<section title="Intermediaries" anchor="intermediaries">
[624]643   A more complicated situation occurs when one or more intermediaries
644   are present in the request/response chain. There are three common
[630]645   forms of intermediary: proxy, gateway, and tunnel.  In some cases,
646   a single intermediary may act as an origin server, proxy, gateway,
647   or tunnel, switching behavior based on the nature of each request.
649<figure><artwork type="drawing">
650       request chain --------------------------------------&gt;
651    UA -----v----- A -----v----- B -----v----- C -----v----- O
652       &lt;------------------------------------- response chain
655   The figure above shows three intermediaries (A, B, and C) between the
656   user agent and origin server. A request or response message that
657   travels the whole chain will pass through four separate connections.
[630]658   Some HTTP communication options
[624]659   may apply only to the connection with the nearest, non-tunnel
660   neighbor, only to the end-points of the chain, or to all connections
661   along the chain. Although the diagram is linear, each participant may
662   be engaged in multiple, simultaneous communications. For example, B
663   may be receiving requests from many clients other than A, and/or
664   forwarding requests to servers other than C, at the same time that it
665   is handling A's request.
[630]668<iref item="upstream"/><iref item="downstream"/>
669<iref item="inbound"/><iref item="outbound"/>
670   We use the terms "upstream" and "downstream" to describe various
671   requirements in relation to the directional flow of a message:
672   all messages flow from upstream to downstream.
673   Likewise, we use the terms "inbound" and "outbound" to refer to
674   directions in relation to the request path: "inbound" means toward
675   the origin server and "outbound" means toward the user agent.
[630]677<t><iref item="proxy"/>
678   A proxy is a message forwarding agent that is selected by the
679   client, usually via local configuration rules, to receive requests
680   for some type(s) of absolute URI and attempt to satisfy those
681   requests via translation through the HTTP interface.  Some translations
682   are minimal, such as for proxy requests for "http" URIs, whereas
683   other requests may require translation to and from entirely different
684   application-layer protocols. Proxies are often used to group an
685   organization's HTTP requests through a common intermediary for the
686   sake of security, annotation services, or shared caching.
688<t><iref item="gateway"/><iref item="reverse proxy"/>
689   A gateway (a.k.a., reverse proxy) is a receiving agent that acts
690   as a layer above some other server(s) and translates the received
691   requests to the underlying server's protocol.  Gateways are often
692   used for load balancing or partitioning HTTP services across
693   multiple machines.
694   Unlike a proxy, a gateway receives requests as if it were the
695   origin server for the requested resource; the requesting client
696   will not be aware that it is communicating with a gateway.
697   A gateway communicates with the client as if the gateway is the
698   origin server and thus is subject to all of the requirements on
699   origin servers for that connection.  A gateway communicates
700   with inbound servers using any protocol it desires, including
701   private extensions to HTTP that are outside the scope of this
702   specification.
704<t><iref item="tunnel"/>
705   A tunnel acts as a blind relay between two connections
706   without changing the messages. Once active, a tunnel is not
707   considered a party to the HTTP communication, though the tunnel may
708   have been initiated by an HTTP request. A tunnel ceases to exist when
709   both ends of the relayed connection are closed. Tunnels are used to
710   extend a virtual connection through an intermediary, such as when
711   transport-layer security is used to establish private communication
712   through a shared firewall proxy.
716<section title="Caches" anchor="caches">
717<iref item="cache"/>
719   Any party to HTTP communication that is not acting as a tunnel may
720   employ an internal cache for handling requests.
721   A cache is a local store of previous response messages and the
722   subsystem that controls its message storage, retrieval, and deletion.
723   A cache stores cacheable responses in order to reduce the response
724   time and network bandwidth consumption on future, equivalent
725   requests. Any client or server may include a cache, though a cache
726   cannot be used by a server while it is acting as a tunnel.
729   The effect of a cache is that the request/response chain is shortened
730   if one of the participants along the chain has a cached response
731   applicable to that request. The following illustrates the resulting
732   chain if B has a cached copy of an earlier response from O (via C)
733   for a request which has not been cached by UA or A.
[624]735<figure><artwork type="drawing">
736          request chain ----------&gt;
737       UA -----v----- A -----v----- B - - - - - - C - - - - - - O
738          &lt;--------- response chain
[630]740<t><iref item="cacheable"/>
741   A response is cacheable if a cache is allowed to store a copy of
742   the response message for use in answering subsequent requests.
743   Even when a response is cacheable, there may be additional
744   constraints placed by the client or by the origin server on when
745   that cached response can be used for a particular request. HTTP
746   requirements for cache behavior and cacheable responses are
[640]747   defined in &caching-overview;
[630]750   There are a wide variety of architectures and configurations
751   of caches and proxies deployed across the World Wide Web and
752   inside large organizations. These systems include national hierarchies
[624]753   of proxy caches to save transoceanic bandwidth, systems that
754   broadcast or multicast cache entries, organizations that distribute
[639]755   subsets of cached data via optical media, and so on.
759<section title="Transport Independence" anchor="transport-independence">
[630]761  HTTP systems are used in a wide variety of environments, from
762  corporate intranets with high-bandwidth links to long-distance
763  communication over low-power radio links and intermittent connectivity.
[624]766   HTTP communication usually takes place over TCP/IP connections. The
767   default port is TCP 80 (<eref target=""/>), but other ports can be used. This does
768   not preclude HTTP from being implemented on top of any other protocol
769   on the Internet, or on other networks. HTTP only presumes a reliable
770   transport; any protocol that provides such guarantees can be used;
771   the mapping of the HTTP/1.1 request and response structures onto the
772   transport data units of the protocol in question is outside the scope
773   of this specification.
776   In HTTP/1.0, most implementations used a new connection for each
777   request/response exchange. In HTTP/1.1, a connection may be used for
778   one or more request/response exchanges, although connections may be
779   closed for a variety of reasons (see <xref target="persistent.connections"/>).
[625]783<section title="HTTP Version" anchor="http.version">
784  <x:anchor-alias value="HTTP-Version"/>
785  <x:anchor-alias value="HTTP-Prot-Name"/>
787   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate versions
788   of the protocol. The protocol versioning policy is intended to allow
789   the sender to indicate the format of a message and its capacity for
790   understanding further HTTP communication, rather than the features
791   obtained via that communication. No change is made to the version
792   number for the addition of message components which do not affect
793   communication behavior or which only add to extensible field values.
794   The &lt;minor&gt; number is incremented when the changes made to the
795   protocol add features which do not change the general message parsing
796   algorithm, but which may add to the message semantics and imply
797   additional capabilities of the sender. The &lt;major&gt; number is
798   incremented when the format of a message within the protocol is
799   changed. See <xref target="RFC2145"/> for a fuller explanation.
802   The version of an HTTP message is indicated by an HTTP-Version field
803   in the first line of the message. HTTP-Version is case-sensitive.
805<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-Version"/><iref primary="true" item="Grammar" subitem="HTTP-Prot-Name"/>
806  <x:ref>HTTP-Version</x:ref>   = <x:ref>HTTP-Prot-Name</x:ref> "/" 1*<x:ref>DIGIT</x:ref> "." 1*<x:ref>DIGIT</x:ref>
807  <x:ref>HTTP-Prot-Name</x:ref> = <x:abnf-char-sequence>"HTTP"</x:abnf-char-sequence> ; "HTTP", case-sensitive
810   Note that the major and minor numbers &MUST; be treated as separate
811   integers and that each &MAY; be incremented higher than a single digit.
812   Thus, HTTP/2.4 is a lower version than HTTP/2.13, which in turn is
813   lower than HTTP/12.3. Leading zeros &MUST; be ignored by recipients and
814   &MUST-NOT; be sent.
817   An application that sends a request or response message that includes
818   HTTP-Version of "HTTP/1.1" &MUST; be at least conditionally compliant
819   with this specification. Applications that are at least conditionally
820   compliant with this specification &SHOULD; use an HTTP-Version of
821   "HTTP/1.1" in their messages, and &MUST; do so for any message that is
822   not compatible with HTTP/1.0. For more details on when to send
823   specific HTTP-Version values, see <xref target="RFC2145"/>.
826   The HTTP version of an application is the highest HTTP version for
827   which the application is at least conditionally compliant.
830   Proxy and gateway applications need to be careful when forwarding
831   messages in protocol versions different from that of the application.
832   Since the protocol version indicates the protocol capability of the
833   sender, a proxy/gateway &MUST-NOT; send a message with a version
834   indicator which is greater than its actual version. If a higher
835   version request is received, the proxy/gateway &MUST; either downgrade
836   the request version, or respond with an error, or switch to tunnel
837   behavior.
840   Due to interoperability problems with HTTP/1.0 proxies discovered
841   since the publication of <xref target="RFC2068"/>, caching proxies &MUST;, gateways
842   &MAY;, and tunnels &MUST-NOT; upgrade the request to the highest version
843   they support. The proxy/gateway's response to that request &MUST; be in
844   the same major version as the request.
847  <t>
848    <x:h>Note:</x:h> Converting between versions of HTTP may involve modification
849    of header fields required or forbidden by the versions involved.
850  </t>
[391]854<section title="Uniform Resource Identifiers" anchor="uri">
[621]855<iref primary="true" item="resource"/>
857   Uniform Resource Identifiers (URIs) <xref target="RFC3986"/> are used
858   throughout HTTP as the means for identifying resources. URI references
[621]859   are used to target requests, indicate redirects, and define relationships.
[391]860   HTTP does not limit what a resource may be; it merely defines an interface
861   that can be used to interact with a resource via HTTP. More information on
862   the scope of URIs and resources can be found in <xref target="RFC3986"/>.
864  <x:anchor-alias value="URI-reference"/>
865  <x:anchor-alias value="absolute-URI"/>
866  <x:anchor-alias value="relative-part"/>
867  <x:anchor-alias value="authority"/>
868  <x:anchor-alias value="path-abempty"/>
869  <x:anchor-alias value="path-absolute"/>
870  <x:anchor-alias value="port"/>
871  <x:anchor-alias value="query"/>
872  <x:anchor-alias value="uri-host"/>
873  <x:anchor-alias value="partial-URI"/>
875   This specification adopts the definitions of "URI-reference",
[649]876   "absolute-URI", "relative-part", "port", "host",
[391]877   "path-abempty", "path-absolute", "query", and "authority" from
878   <xref target="RFC3986"/>. In addition, we define a partial-URI rule for
879   protocol elements that allow a relative URI without a fragment.
881<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="URI-reference"/><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"/>
[395]882  <x:ref>URI-reference</x:ref> = &lt;URI-reference, defined in <xref target="RFC3986" x:fmt="," x:sec="4.1"/>&gt;
883  <x:ref>absolute-URI</x:ref>  = &lt;absolute-URI, defined in <xref target="RFC3986" x:fmt="," x:sec="4.3"/>&gt;
884  <x:ref>relative-part</x:ref> = &lt;relative-part, defined in <xref target="RFC3986" x:fmt="," x:sec="4.2"/>&gt;
885  <x:ref>authority</x:ref>     = &lt;authority, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2"/>&gt;
886  <x:ref>path-abempty</x:ref>  = &lt;path-abempty, defined in <xref target="RFC3986" x:fmt="," x:sec="3.3"/>&gt;
887  <x:ref>path-absolute</x:ref> = &lt;path-absolute, defined in <xref target="RFC3986" x:fmt="," x:sec="3.3"/>&gt;
888  <x:ref>port</x:ref>          = &lt;port, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2.3"/>&gt;
889  <x:ref>query</x:ref>         = &lt;query, defined in <xref target="RFC3986" x:fmt="," x:sec="3.4"/>&gt;
890  <x:ref>uri-host</x:ref>      = &lt;host, defined in <xref target="RFC3986" x:fmt="," x:sec="3.2.2"/>&gt;
892  <x:ref>partial-URI</x:ref>   = relative-part [ "?" query ]
895   Each protocol element in HTTP that allows a URI reference will indicate in
896   its ABNF production whether the element allows only a URI in absolute form
897   (absolute-URI), any relative reference (relative-ref), or some other subset
898   of the URI-reference grammar. Unless otherwise indicated, URI references
899   are parsed relative to the request target (the default base URI for both
900   the request and its corresponding response).
903<section title="http URI scheme" anchor="http.uri">
904  <x:anchor-alias value="http-URI"/>
905  <iref item="http URI scheme" primary="true"/>
906  <iref item="URI scheme" subitem="http" primary="true"/>
[621]908   The "http" URI scheme is hereby defined for the purpose of minting
909   identifiers according to their association with the hierarchical
910   namespace governed by a potential HTTP origin server listening for
911   TCP connections on a given port.
912   The HTTP server is identified via the generic syntax's
913   <x:ref>authority</x:ref> component, which includes a host
914   identifier and optional TCP port, and the remainder of the URI is
915   considered to be identifying data corresponding to a resource for
916   which that server might provide an HTTP interface.
918<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="http-URI"/>
919  <x:ref>http-URI</x:ref> = "http:" "//" <x:ref>authority</x:ref> <x:ref>path-abempty</x:ref> [ "?" <x:ref>query</x:ref> ]
[621]922   The host identifier within an <x:ref>authority</x:ref> component is
923   defined in <xref target="RFC3986" x:fmt="," x:sec="3.2.2"/>.  If host is
924   provided as an IP literal or IPv4 address, then the HTTP server is any
925   listener on the indicated TCP port at that IP address. If host is a
926   registered name, then that name is considered an indirect identifier
927   and the recipient might use a name resolution service, such as DNS,
928   to find the address of a listener for that host.
929   The host &MUST-NOT; be empty; if an "http" URI is received with an
930   empty host, then it &MUST; be rejected as invalid.
931   If the port subcomponent is empty or not given, then TCP port 80 is
932   assumed (the default reserved port for WWW services).
935   Regardless of the form of host identifier, access to that host is not
936   implied by the mere presence of its name or address. The host may or may
937   not exist and, even when it does exist, may or may not be running an
938   HTTP server or listening to the indicated port. The "http" URI scheme
939   makes use of the delegated nature of Internet names and addresses to
940   establish a naming authority (whatever entity has the ability to place
941   an HTTP server at that Internet name or address) and allows that
942   authority to determine which names are valid and how they might be used.
945   When an "http" URI is used within a context that calls for access to the
946   indicated resource, a client &MAY; attempt access by resolving
947   the host to an IP address, establishing a TCP connection to that address
948   on the indicated port, and sending an HTTP request message to the server
949   containing the URI's identifying data as described in <xref target="request"/>.
950   If the server responds to that request with a non-interim HTTP response
951   message, as described in <xref target="response"/>, then that response
952   is considered an authoritative answer to the client's request.
955   Although HTTP is independent of the transport protocol, the "http"
956   scheme is specific to TCP-based services because the name delegation
957   process depends on TCP for establishing authority.
958   An HTTP service based on some other underlying connection protocol
959   would presumably be identified using a different URI scheme, just as
960   the "https" scheme (below) is used for servers that require an SSL/TLS
961   transport layer on a connection. Other protocols may also be used to
962   provide access to "http" identified resources --- it is only the
963   authoritative interface used for mapping the namespace that is
964   specific to TCP.
968<section title="https URI scheme" anchor="https.uri">
[622]969   <x:anchor-alias value="https-URI"/>
[452]970   <iref item="https URI scheme"/>
971   <iref item="URI scheme" subitem="https"/>
[621]973   The "https" URI scheme is hereby defined for the purpose of minting
974   identifiers according to their association with the hierarchical
975   namespace governed by a potential HTTP origin server listening for
976   SSL/TLS-secured connections on a given TCP port.
977   The host and port are determined in the same way
978   as for the "http" scheme, except that a default TCP port of 443
979   is assumed if the port subcomponent is empty or not given.
[621]981<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="https-URI"/>
982  <x:ref>https-URI</x:ref> = "https:" "//" <x:ref>authority</x:ref> <x:ref>path-abempty</x:ref> [ "?" <x:ref>query</x:ref> ]
985   The primary difference between the "http" and "https" schemes is
986   that interaction with the latter is required to be secured for
987   privacy through the use of strong encryption. The URI cannot be
988   sent in a request until the connection is secure. Likewise, the
989   default for caching is that each response that would be considered
990   "public" under the "http" scheme is instead treated as "private"
991   and thus not eligible for shared caching.
994   The process for authoritative access to an "https" identified
995   resource is defined in <xref target="RFC2818"/>.
[621]999<section title="http and https URI Normalization and Comparison" anchor="uri.comparison">
[621]1001   Since the "http" and "https" schemes conform to the URI generic syntax,
1002   such URIs are normalized and compared according to the algorithm defined
1003   in <xref target="RFC3986" x:fmt="," x:sec="6"/>, using the defaults
1004   described above for each scheme.
[621]1007   If the port is equal to the default port for a scheme, the normal
1008   form is to elide the port subcomponent. Likewise, an empty path
1009   component is equivalent to an absolute path of "/", so the normal
1010   form is to provide a path of "/" instead. The scheme and host
1011   are case-insensitive and normally provided in lowercase; all
1012   other components are compared in a case-sensitive manner.
1013   Characters other than those in the "reserved" set are equivalent
1014   to their percent-encoded octets (see <xref target="RFC3986"
1015   x:fmt="," x:sec="2.1"/>): the normal form is to not encode them.
[391]1018   For example, the following three URIs are equivalent:
1020<figure><artwork type="example">
[767]1026   <cref anchor="TODO-not-here" source="roy">This paragraph does not belong here.</cref>
[621]1027   If path-abempty is the empty string (i.e., there is no slash "/"
1028   path separator following the authority), then the "http" URI
1029   &MUST; be given as "/" when
1030   used as a request-target (<xref target="request-target"/>). If a proxy
1031   receives a host name which is not a fully qualified domain name, it
1032   &MAY; add its domain to the host name it received. If a proxy receives
1033   a fully qualified domain name, the proxy &MUST-NOT; change the host
1034   name.
[8]1040<section title="HTTP Message" anchor="http.message">
[647]1041<x:anchor-alias value="generic-message"/>
1042<x:anchor-alias value="message.types"/>
1043<x:anchor-alias value="HTTP-message"/>
1044<x:anchor-alias value="start-line"/>
1045<iref item="header section"/>
1046<iref item="headers"/>
1047<iref item="header field"/>
[647]1049   All HTTP/1.1 messages consist of a start-line followed by a sequence of
1050   characters in a format similar to the Internet Message Format
1051   <xref target="RFC5322"/>: zero or more header fields (collectively
1052   referred to as the "headers" or the "header section"), an empty line
1053   indicating the end of the header section, and an optional message-body.
[647]1056   An HTTP message can either be a request from client to server or a
1057   response from server to client.  Syntactically, the two types of message
1058   differ only in the start-line, which is either a Request-Line (for requests)
1059   or a Status-Line (for responses), and in the algorithm for determining
1060   the length of the message-body (<xref target="message.length"/>).
1061   In theory, a client could receive requests and a server could receive
1062   responses, distinguishing them by their different start-line formats,
1063   but in practice servers are implemented to only expect a request
1064   (a response is interpreted as an unknown or invalid request method)
1065   and clients are implemented to only expect a response.
[647]1067<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-message"/>
1068  <x:ref>HTTP-message</x:ref>    = <x:ref>start-line</x:ref>
1069                    *( <x:ref>header-field</x:ref> <x:ref>CRLF</x:ref> )
[229]1070                    <x:ref>CRLF</x:ref>
1071                    [ <x:ref>message-body</x:ref> ]
[334]1072  <x:ref>start-line</x:ref>      = <x:ref>Request-Line</x:ref> / <x:ref>Status-Line</x:ref>
[395]1075   Whitespace (WSP) &MUST-NOT; be sent between the start-line and the first
1076   header field. The presence of whitespace might be an attempt to trick a
1077   noncompliant implementation of HTTP into ignoring that field or processing
1078   the next line as a new request, either of which may result in security
1079   issues when implementations within the request chain interpret the
1080   same message differently. HTTP/1.1 servers &MUST; reject such a message
1081   with a 400 (Bad Request) response.
1084<section title="Message Parsing Robustness" anchor="message.robustness">
1086   In the interest of robustness, servers &SHOULD; ignore at least one
1087   empty line received where a Request-Line is expected. In other words, if
1088   the server is reading the protocol stream at the beginning of a
1089   message and receives a CRLF first, it should ignore the CRLF.
1092   Some old HTTP/1.0 client implementations generate an extra CRLF
1093   after a POST request as a lame workaround for some early server
1094   applications that failed to read message-body content that was
1095   not terminated by a line-ending. An HTTP/1.1 client &MUST-NOT;
1096   preface or follow a request with an extra CRLF.  If terminating
1097   the request message-body with a line-ending is desired, then the
1098   client &MUST; include the terminating CRLF octets as part of the
1099   message-body length.
1102   The normal procedure for parsing an HTTP message is to read the
1103   start-line into a structure, read each header field into a hash
1104   table by field name until the empty line, and then use the parsed
1105   data to determine if a message-body is expected.  If a message-body
1106   has been indicated, then it is read as a stream until an amount
1107   of OCTETs equal to the message-length is read or the connection
1108   is closed.  Care must be taken to parse an HTTP message as a sequence
1109   of OCTETs in an encoding that is a superset of US-ASCII.  Attempting
1110   to parse HTTP as a stream of Unicode characters in a character encoding
1111   like UTF-16 may introduce security flaws due to the differing ways
1112   that such parsers interpret invalid characters.
[647]1116<section title="Header Fields" anchor="header.fields">
1117  <x:anchor-alias value="header-field"/>
[229]1118  <x:anchor-alias value="field-content"/>
1119  <x:anchor-alias value="field-name"/>
1120  <x:anchor-alias value="field-value"/>
[647]1121  <x:anchor-alias value="OWS"/>
[647]1123   Each HTTP header field consists of a case-insensitive field name
1124   followed by a colon (":"), optional whitespace, and the field value.
[647]1126<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"/>
[693]1127  <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>OWS</x:ref>
[229]1128  <x:ref>field-name</x:ref>     = <x:ref>token</x:ref>
[369]1129  <x:ref>field-value</x:ref>    = *( <x:ref>field-content</x:ref> / <x:ref>OWS</x:ref> )
[395]1130  <x:ref>field-content</x:ref>  = *( <x:ref>WSP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
[647]1133   No whitespace is allowed between the header field name and colon. For
[395]1134   security reasons, any request message received containing such whitespace
[647]1135   &MUST; be rejected with a response code of 400 (Bad Request). A proxy
1136   &MUST; remove any such whitespace from a response message before
1137   forwarding the message downstream.
[647]1140   A field value &MAY; be preceded by optional whitespace (OWS); a single SP is
1141   preferred. The field value does not include any leading or trailing white
[395]1142   space: OWS occurring before the first non-whitespace character of the
[647]1143   field value or after the last non-whitespace character of the field value
[748]1144   is ignored and &SHOULD; be removed before further processing (as this does
1145   not change the meaning of the header field).
[647]1148   The order in which header fields with differing field names are
1149   received is not significant. However, it is "good practice" to send
1150   header fields that contain control data first, such as Host on
1151   requests and Date on responses, so that implementations can decide
1152   when not to handle a message as early as possible.  A server &MUST;
1153   wait until the entire header section is received before interpreting
1154   a request message, since later header fields might include conditionals,
1155   authentication credentials, or deliberately misleading duplicate
1156   header fields that would impact request processing.
[651]1159   Multiple header fields with the same field name &MUST-NOT; be
1160   sent in a message unless the entire field value for that
[647]1161   header field is defined as a comma-separated list [i.e., #(values)].
1162   Multiple header fields with the same field name can be combined into
1163   one "field-name: field-value" pair, without changing the semantics of the
1164   message, by appending each subsequent field value to the combined
1165   field value in order, separated by a comma. The order in which
1166   header fields with the same field name are received is therefore
1167   significant to the interpretation of the combined field value;
1168   a proxy &MUST-NOT; change the order of these field values when
1169   forwarding a message.
1172  <t>
[756]1173   <x:h>Note:</x:h> The "Set-Cookie" header as implemented in
[647]1174   practice (as opposed to how it is specified in <xref target="RFC2109"/>)
1175   can occur multiple times, but does not use the list syntax, and thus cannot
1176   be combined into a single line. (See Appendix A.2.3 of <xref target="Kri2001"/>
1177   for details.) Also note that the Set-Cookie2 header specified in
1178   <xref target="RFC2965"/> does not share this problem.
1179  </t>
[395]1182   Historically, HTTP header field values could be extended over multiple
1183   lines by preceding each extra line with at least one space or horizontal
1184   tab character (line folding). This specification deprecates such line
1185   folding except within the message/http media type
1186   (<xref target=""/>).
1187   HTTP/1.1 senders &MUST-NOT; produce messages that include line folding
1188   (i.e., that contain any field-content that matches the obs-fold rule) unless
1189   the message is intended for packaging within the message/http media type.
1190   HTTP/1.1 recipients &SHOULD; accept line folding and replace any embedded
1191   obs-fold whitespace with a single SP prior to interpreting the field value
1192   or forwarding the message downstream.
1195   Historically, HTTP has allowed field content with text in the ISO-8859-1
1196   <xref target="ISO-8859-1"/> character encoding and supported other
1197   character sets only through use of <xref target="RFC2047"/> encoding.
1198   In practice, most HTTP header field values use only a subset of the
1199   US-ASCII character encoding <xref target="USASCII"/>. Newly defined
1200   header fields &SHOULD; limit their field values to US-ASCII characters.
1201   Recipients &SHOULD; treat other (obs-text) octets in field content as
1202   opaque data.
[395]1204<t anchor="rule.comment">
1205  <x:anchor-alias value="comment"/>
1206  <x:anchor-alias value="ctext"/>
1207   Comments can be included in some HTTP header fields by surrounding
1208   the comment text with parentheses. Comments are only allowed in
1209   fields containing "comment" as part of their field value definition.
1211<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="comment"/><iref primary="true" item="Grammar" subitem="ctext"/>
[702]1212  <x:ref>comment</x:ref>        = "(" *( <x:ref>ctext</x:ref> / <x:ref>quoted-cpair</x:ref> / <x:ref>comment</x:ref> ) ")"
[687]1213  <x:ref>ctext</x:ref>          = <x:ref>OWS</x:ref> / %x21-27 / %x2A-5B / %x5D-7E / <x:ref>obs-text</x:ref>
1214                 ; <x:ref>OWS</x:ref> / &lt;<x:ref>VCHAR</x:ref> except "(", ")", and "\"&gt; / <x:ref>obs-text</x:ref>
[702]1216<t anchor="rule.quoted-cpair">
1217  <x:anchor-alias value="quoted-cpair"/>
1218   The backslash character ("\") can be used as a single-character
[703]1219   quoting mechanism within comment constructs:
1221<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="quoted-cpair"/>
1222  <x:ref>quoted-cpair</x:ref>    = "\" ( <x:ref>WSP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
1225   Producers &SHOULD-NOT; escape characters that do not require escaping
[703]1226   (i.e., other than the backslash character "\" and the parentheses "(" and
1227   ")").
1231<section title="Message Body" anchor="message.body">
[229]1232  <x:anchor-alias value="message-body"/>
1234   The message-body (if any) of an HTTP message is used to carry the
1235   entity-body associated with the request or response. The message-body
1236   differs from the entity-body only when a transfer-coding has been
1237   applied, as indicated by the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1239<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="message-body"/>
[229]1240  <x:ref>message-body</x:ref> = <x:ref>entity-body</x:ref>
[334]1241               / &lt;entity-body encoded as per <x:ref>Transfer-Encoding</x:ref>&gt;
1244   Transfer-Encoding &MUST; be used to indicate any transfer-codings
1245   applied by an application to ensure safe and proper transfer of the
1246   message. Transfer-Encoding is a property of the message, not of the
1247   entity, and thus &MAY; be added or removed by any application along the
1248   request/response chain. (However, <xref target="transfer.codings"/> places restrictions on
1249   when certain transfer-codings may be used.)
1252   The rules for when a message-body is allowed in a message differ for
1253   requests and responses.
1256   The presence of a message-body in a request is signaled by the
1257   inclusion of a Content-Length or Transfer-Encoding header field in
[647]1258   the request's header fields.
[171]1259   When a request message contains both a message-body of non-zero
1260   length and a method that does not define any semantics for that
1261   request message-body, then an origin server &SHOULD; either ignore
1262   the message-body or respond with an appropriate error message
1263   (e.g., 413).  A proxy or gateway, when presented the same request,
1264   &SHOULD; either forward the request inbound with the message-body or
1265   ignore the message-body when determining a response.
1268   For response messages, whether or not a message-body is included with
1269   a message is dependent on both the request method and the response
1270   status code (<xref target="status.code.and.reason.phrase"/>). All responses to the HEAD request method
1271   &MUST-NOT; include a message-body, even though the presence of entity-header
1272   fields might lead one to believe they do. All 1xx
[753]1273   (Informational), 204 (No Content), and 304 (Not Modified) responses
[8]1274   &MUST-NOT; include a message-body. All other responses do include a
1275   message-body, although it &MAY; be of zero length.
1279<section title="Message Length" anchor="message.length">
1281   The transfer-length of a message is the length of the message-body as
1282   it appears in the message; that is, after any transfer-codings have
1283   been applied. When a message-body is included with a message, the
1284   transfer-length of that body is determined by one of the following
1285   (in order of precedence):
1288  <list style="numbers">
1289    <x:lt><t>
1290     Any response message which "&MUST-NOT;" include a message-body (such
1291     as the 1xx, 204, and 304 responses and any response to a HEAD
1292     request) is always terminated by the first empty line after the
1293     header fields, regardless of the entity-header fields present in
1294     the message.
1295    </t></x:lt>
1296    <x:lt><t>
[85]1297     If a Transfer-Encoding header field (<xref target="header.transfer-encoding"/>)
[276]1298     is present and the "chunked" transfer-coding (<xref target="transfer.codings"/>)
1299     is used, the transfer-length is defined by the use of this transfer-coding.
1300     If a Transfer-Encoding header field is present and the "chunked" transfer-coding
1301     is not present, the transfer-length is defined by the sender closing the connection.
[8]1302    </t></x:lt>
1303    <x:lt><t>
1304     If a Content-Length header field (<xref target="header.content-length"/>) is present, its
[576]1305     value in OCTETs represents both the entity-length and the
[8]1306     transfer-length. The Content-Length header field &MUST-NOT; be sent
1307     if these two lengths are different (i.e., if a Transfer-Encoding
1308     header field is present). If a message is received with both a
1309     Transfer-Encoding header field and a Content-Length header field,
1310     the latter &MUST; be ignored.
1311    </t></x:lt>
1312    <x:lt><t>
1313     If the message uses the media type "multipart/byteranges", and the
[71]1314     transfer-length is not otherwise specified, then this self-delimiting
[8]1315     media type defines the transfer-length. This media type
[71]1316     &MUST-NOT; be used unless the sender knows that the recipient can parse
1317     it; the presence in a request of a Range header with multiple byte-range
[761]1318     specifiers from a HTTP/1.1 client implies that the client can parse
[8]1319     multipart/byteranges responses.
1320    <list style="empty"><t>
[761]1321       A range header might be forwarded by a HTTP/1.0 proxy that does not
[8]1322       understand multipart/byteranges; in this case the server &MUST;
1323       delimit the message using methods defined in items 1, 3 or 5 of
1324       this section.
1325    </t></list>
1326    </t></x:lt>
1327    <x:lt><t>
1328     By the server closing the connection. (Closing the connection
1329     cannot be used to indicate the end of a request body, since that
1330     would leave no possibility for the server to send back a response.)
1331    </t></x:lt>
1332  </list>
1335   For compatibility with HTTP/1.0 applications, HTTP/1.1 requests
1336   containing a message-body &MUST; include a valid Content-Length header
1337   field unless the server is known to be HTTP/1.1 compliant. If a
1338   request contains a message-body and a Content-Length is not given,
[137]1339   the server &SHOULD; respond with 400 (Bad Request) if it cannot
1340   determine the length of the message, or with 411 (Length Required) if
[8]1341   it wishes to insist on receiving a valid Content-Length.
1344   All HTTP/1.1 applications that receive entities &MUST; accept the
1345   "chunked" transfer-coding (<xref target="transfer.codings"/>), thus allowing this mechanism
1346   to be used for messages when the message length cannot be determined
1347   in advance.
1350   Messages &MUST-NOT; include both a Content-Length header field and a
[85]1351   transfer-coding. If the message does include a
[8]1352   transfer-coding, the Content-Length &MUST; be ignored.
1355   When a Content-Length is given in a message where a message-body is
1356   allowed, its field value &MUST; exactly match the number of OCTETs in
1357   the message-body. HTTP/1.1 user agents &MUST; notify the user when an
1358   invalid length is received and detected.
1362<section title="General Header Fields" anchor="general.header.fields">
[229]1363  <x:anchor-alias value="general-header"/>
1365   There are a few header fields which have general applicability for
1366   both request and response messages, but which do not apply to the
1367   entity being transferred. These header fields apply only to the
1368   message being transmitted.
1370<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="general-header"/>
[229]1371  <x:ref>general-header</x:ref> = <x:ref>Cache-Control</x:ref>            ; &header-cache-control;
[334]1372                 / <x:ref>Connection</x:ref>               ; <xref target="header.connection"/>
1373                 / <x:ref>Date</x:ref>                     ; <xref target=""/>
1374                 / <x:ref>Pragma</x:ref>                   ; &header-pragma;
1375                 / <x:ref>Trailer</x:ref>                  ; <xref target="header.trailer"/>
1376                 / <x:ref>Transfer-Encoding</x:ref>        ; <xref target="header.transfer-encoding"/>
1377                 / <x:ref>Upgrade</x:ref>                  ; <xref target="header.upgrade"/>
1378                 / <x:ref>Via</x:ref>                      ; <xref target="header.via"/>
1379                 / <x:ref>Warning</x:ref>                  ; &header-warning;
1382   General-header field names can be extended reliably only in
1383   combination with a change in the protocol version. However, new or
1384   experimental header fields may be given the semantics of general
1385   header fields if all parties in the communication recognize them to
1386   be general-header fields. Unrecognized header fields are treated as
1387   entity-header fields.
1392<section title="Request" anchor="request">
[229]1393  <x:anchor-alias value="Request"/>
1395   A request message from a client to a server includes, within the
1396   first line of that message, the method to be applied to the resource,
1397   the identifier of the resource, and the protocol version in use.
[29]1399<!--                 Host                      ; should be moved here eventually -->
[8]1400<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request"/>
[229]1401  <x:ref>Request</x:ref>       = <x:ref>Request-Line</x:ref>              ; <xref target="request-line"/>
1402                  *(( <x:ref>general-header</x:ref>        ; <xref target="general.header.fields"/>
[334]1403                   / <x:ref>request-header</x:ref>         ; &request-header-fields;
[636]1404                   / <x:ref>entity-header</x:ref> ) <x:ref>CRLF</x:ref> ) ; &entity-header-fields;
[229]1405                  <x:ref>CRLF</x:ref>
1406                  [ <x:ref>message-body</x:ref> ]          ; <xref target="message.body"/>
1409<section title="Request-Line" anchor="request-line">
[229]1410  <x:anchor-alias value="Request-Line"/>
1412   The Request-Line begins with a method token, followed by the
[391]1413   request-target and the protocol version, and ending with CRLF. The
[8]1414   elements are separated by SP characters. No CR or LF is allowed
1415   except in the final CRLF sequence.
1417<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Request-Line"/>
[391]1418  <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>
1421<section title="Method" anchor="method">
[229]1422  <x:anchor-alias value="Method"/>
1424   The Method  token indicates the method to be performed on the
[391]1425   resource identified by the request-target. The method is case-sensitive.
1427<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Method"/><iref primary="true" item="Grammar" subitem="extension-method"/>
[229]1428  <x:ref>Method</x:ref>         = <x:ref>token</x:ref>
[391]1432<section title="request-target" anchor="request-target">
1433  <x:anchor-alias value="request-target"/>
[452]1435   The request-target
[8]1436   identifies the resource upon which to apply the request.
[391]1438<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="request-target"/>
[404]1439  <x:ref>request-target</x:ref> = "*"
[374]1440                 / <x:ref>absolute-URI</x:ref>
[334]1441                 / ( <x:ref>path-absolute</x:ref> [ "?" <x:ref>query</x:ref> ] )
1442                 / <x:ref>authority</x:ref>
[391]1445   The four options for request-target are dependent on the nature of the
[809]1446   request.
1449   The asterisk "*" means that the request does not apply to a
[8]1450   particular resource, but to the server itself, and is only allowed
1451   when the method used does not necessarily apply to a resource. One
1452   example would be
[803]1454<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
1455OPTIONS * HTTP/1.1
[374]1458   The absolute-URI form is &REQUIRED; when the request is being made to a
[8]1459   proxy. The proxy is requested to forward the request or service it
1460   from a valid cache, and return the response. Note that the proxy &MAY;
1461   forward the request on to another proxy or directly to the server
[374]1462   specified by the absolute-URI. In order to avoid request loops, a
[8]1463   proxy &MUST; be able to recognize all of its server names, including
1464   any aliases, local variations, and the numeric IP address. An example
1465   Request-Line would be:
[803]1467<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
1468GET HTTP/1.1
[374]1471   To allow for transition to absolute-URIs in all requests in future
1472   versions of HTTP, all HTTP/1.1 servers &MUST; accept the absolute-URI
[8]1473   form in requests, even though HTTP/1.1 clients will only generate
1474   them in requests to proxies.
[29]1477   The authority form is only used by the CONNECT method (&CONNECT;).
[391]1480   The most common form of request-target is that used to identify a
[8]1481   resource on an origin server or gateway. In this case the absolute
[374]1482   path of the URI &MUST; be transmitted (see <xref target="http.uri"/>, path-absolute) as
[391]1483   the request-target, and the network location of the URI (authority) &MUST;
[8]1484   be transmitted in a Host header field. For example, a client wishing
1485   to retrieve the resource above directly from the origin server would
[90]1486   create a TCP connection to port 80 of the host "" and send
[8]1487   the lines:
[803]1489<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
1490GET /pub/WWW/TheProject.html HTTP/1.1
1494   followed by the remainder of the Request. Note that the absolute path
1495   cannot be empty; if none is present in the original URI, it &MUST; be
1496   given as "/" (the server root).
[403]1499   If a proxy receives a request without any path in the request-target and
1500   the method specified is capable of supporting the asterisk form of
1501   request-target, then the last proxy on the request chain &MUST; forward the
1502   request with "*" as the final request-target.
1505   For example, the request
[803]1506</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
1510  would be forwarded by the proxy as
[803]1511</preamble><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
1512OPTIONS * HTTP/1.1
1516   after connecting to port 8001 of host "".
[391]1520   The request-target is transmitted in the format specified in
[452]1521   <xref target="http.uri"/>. If the request-target is percent-encoded
1522   (<xref target="RFC3986" x:fmt="," x:sec="2.1"/>), the origin server
[391]1523   &MUST; decode the request-target in order to
[8]1524   properly interpret the request. Servers &SHOULD; respond to invalid
[391]1525   request-targets with an appropriate status code.
[185]1528   A transparent proxy &MUST-NOT; rewrite the "path-absolute" part of the
[391]1529   received request-target when forwarding it to the next inbound server,
[821]1530   except as noted above to replace a null path-absolute with "/" or "*".
1533  <t>
1534    <x:h>Note:</x:h> The "no rewrite" rule prevents the proxy from changing the
1535    meaning of the request when the origin server is improperly using
1536    a non-reserved URI character for a reserved purpose.  Implementors
1537    should be aware that some pre-HTTP/1.1 proxies have been known to
1538    rewrite the request-target.
1539  </t>
[391]1542   HTTP does not place a pre-defined limit on the length of a request-target.
1543   A server &MUST; be prepared to receive URIs of unbounded length and
[452]1544   respond with the 414 (URI Too Long) status if the received
[391]1545   request-target would be longer than the server wishes to handle
1546   (see &status-414;).
1549   Various ad-hoc limitations on request-target length are found in practice.
1550   It is &RECOMMENDED; that all HTTP senders and recipients support
1551   request-target lengths of 8000 or more OCTETs.
1554  <t>
1555    <x:h>Note:</x:h> Fragments (<xref target="RFC3986" x:fmt="," x:sec="3.5"/>)
1556    are not part of the request-target and thus will not be transmitted
1557    in an HTTP request.
1558  </t>
1563<section title="The Resource Identified by a Request" anchor="">
1565   The exact resource identified by an Internet request is determined by
[391]1566   examining both the request-target and the Host header field.
1569   An origin server that does not allow resources to differ by the
1570   requested host &MAY; ignore the Host header field value when
1571   determining the resource identified by an HTTP/1.1 request. (But see
1572   <xref target=""/>
1573   for other requirements on Host support in HTTP/1.1.)
1576   An origin server that does differentiate resources based on the host
1577   requested (sometimes referred to as virtual hosts or vanity host
1578   names) &MUST; use the following rules for determining the requested
1579   resource on an HTTP/1.1 request:
1580  <list style="numbers">
[391]1581    <t>If request-target is an absolute-URI, the host is part of the
1582     request-target. Any Host header field value in the request &MUST; be
[8]1583     ignored.</t>
[391]1584    <t>If the request-target is not an absolute-URI, and the request includes
[8]1585     a Host header field, the host is determined by the Host header
1586     field value.</t>
1587    <t>If the host as determined by rule 1 or 2 is not a valid host on
1588     the server, the response &MUST; be a 400 (Bad Request) error message.</t>
1589  </list>
1592   Recipients of an HTTP/1.0 request that lacks a Host header field &MAY;
1593   attempt to use heuristics (e.g., examination of the URI path for
1594   something unique to a particular host) in order to determine what
1595   exact resource is being requested.
[823]1599<section title="Effective Request URI" anchor="effective.request.uri">
1600  <iref primary="true" item="Effective Request URI"/>
1602   HTTP requests often do not carry the absolute URI (<xref target="RFC3986" x:fmt="," x:sec="4.3"/>)
1603   for the resource they are intended for; instead, the value needs to be inferred from the
1604   request-target, Host header and other context. The result of this process is
1605   the "Effective Request URI".
1608   If the request-target is an absolute-URI, then the Effective Request URI is
1609   the request-target.
1612   If the request-target uses the path-absolute (plus optional query) syntax
1613   or if it is just the asterisk "*", then the Effective Request URI is
1614   constructed by concatenating
1617  <list style="symbols">
1618    <t>
1619      the scheme name: "http" if the request was received over an insecure
1620      TCP connection, or "https" when received over SSL/TLS-secured TCP
1621      connection,
1622    </t>
1623    <t>
1624      the character sequence "://",
1625    </t>
1626    <t>
1627      the authority component, as specified in the Host header
1628      (<xref target=""/>) and determined by the rules in
1629      <xref target=""/>,
1630      <cref anchor="effrequri-nohost" source="jre">Do we need to include the handling of missing hosts in HTTP/1.0 messages, as
1631      described in <xref target=""/>?</cref>
1632      and
1633    </t>
1634    <t>
1635      the request-target obtained from the Request-Line, unless the
1636      request-target is just the asterisk "*".
1637    </t>
1638  </list>
1641   Otherwise, when request-target uses the authority form, the Effective
1642   Request URI is undefined.
1646   Example 1: the Effective Request URI for the message
1648<artwork type="example" x:indent-with="  ">
1649GET /pub/WWW/TheProject.html HTTP/1.1
1653  (received over an insecure TCP connection) is "http", plus "://", plus the
1654  authority component "", plus the request-target
1655  "/pub/WWW/TheProject.html", thus
1656  "".
1661   Example 2: the Effective Request URI for the message
1663<artwork type="example" x:indent-with="  ">
1664GET * HTTP/1.1
1668  (received over an SSL/TLS secured TCP connection) is "https", plus "://", plus the
1669  authority component "", thus "".
1673   Effective Request URIs are compared using the rules described in
1674   <xref target="uri.comparison"/>, except that empty path components &MUST-NOT;
1675   be treated as equivalent to an absolute path of "/".
[8]1682<section title="Response" anchor="response">
[229]1683  <x:anchor-alias value="Response"/>
1685   After receiving and interpreting a request message, a server responds
1686   with an HTTP response message.
1688<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Response"/>
[229]1689  <x:ref>Response</x:ref>      = <x:ref>Status-Line</x:ref>               ; <xref target="status-line"/>
1690                  *(( <x:ref>general-header</x:ref>        ; <xref target="general.header.fields"/>
[334]1691                   / <x:ref>response-header</x:ref>        ; &response-header-fields;
[692]1692                   / <x:ref>entity-header</x:ref> ) <x:ref>CRLF</x:ref> ) ; &entity-header-fields;
[229]1693                  <x:ref>CRLF</x:ref>
1694                  [ <x:ref>message-body</x:ref> ]          ; <xref target="message.body"/>
1697<section title="Status-Line" anchor="status-line">
[229]1698  <x:anchor-alias value="Status-Line"/>
1700   The first line of a Response message is the Status-Line, consisting
1701   of the protocol version followed by a numeric status code and its
1702   associated textual phrase, with each element separated by SP
1703   characters. No CR or LF is allowed except in the final CRLF sequence.
1705<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Status-Line"/>
[229]1706  <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>
1709<section title="Status Code and Reason Phrase" anchor="status.code.and.reason.phrase">
[229]1710  <x:anchor-alias value="Reason-Phrase"/>
1711  <x:anchor-alias value="Status-Code"/>
1713   The Status-Code element is a 3-digit integer result code of the
1714   attempt to understand and satisfy the request. These codes are fully
[198]1715   defined in &status-codes;.  The Reason Phrase exists for the sole
1716   purpose of providing a textual description associated with the numeric
1717   status code, out of deference to earlier Internet application protocols
1718   that were more frequently used with interactive text clients.
1719   A client &SHOULD; ignore the content of the Reason Phrase.
1722   The first digit of the Status-Code defines the class of response. The
1723   last two digits do not have any categorization role. There are 5
1724   values for the first digit:
1725  <list style="symbols">
1726    <t>
1727      1xx: Informational - Request received, continuing process
1728    </t>
1729    <t>
1730      2xx: Success - The action was successfully received,
1731        understood, and accepted
1732    </t>
1733    <t>
1734      3xx: Redirection - Further action must be taken in order to
1735        complete the request
1736    </t>
1737    <t>
1738      4xx: Client Error - The request contains bad syntax or cannot
1739        be fulfilled
1740    </t>
1741    <t>
1742      5xx: Server Error - The server failed to fulfill an apparently
1743        valid request
1744    </t>
1745  </list>
1747<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Status-Code"/><iref primary="true" item="Grammar" subitem="extension-code"/><iref primary="true" item="Grammar" subitem="Reason-Phrase"/>
[229]1748  <x:ref>Status-Code</x:ref>    = 3<x:ref>DIGIT</x:ref>
[395]1749  <x:ref>Reason-Phrase</x:ref>  = *( <x:ref>WSP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
[623]1757<section title="Protocol Parameters" anchor="protocol.parameters">
1759<section title="Date/Time Formats: Full Date" anchor="">
1760  <x:anchor-alias value="HTTP-date"/>
1762   HTTP applications have historically allowed three different formats
[804]1763   for the representation of date/time stamps.
1766   The first format is preferred as an Internet standard and represents
1767   a fixed-length subset of that defined by <xref target="RFC1123"/>:
1769<figure><artwork type="example" x:indent-with="  ">
1770Sun, 06 Nov 1994 08:49:37 GMT  ; RFC 1123
[804]1773   The other formats are described here only for compatibility with obsolete
1774   implementations.
1776<figure><artwork type="example" x:indent-with="  ">
1777Sunday, 06-Nov-94 08:49:37 GMT ; obsolete RFC 850 format
1778Sun Nov  6 08:49:37 1994       ; ANSI C's asctime() format
[623]1781   HTTP/1.1 clients and servers that parse the date value &MUST; accept
1782   all three formats (for compatibility with HTTP/1.0), though they &MUST;
1783   only generate the RFC 1123 format for representing HTTP-date values
1784   in header fields. See <xref target="tolerant.applications"/> for further information.
1787   All HTTP date/time stamps &MUST; be represented in Greenwich Mean Time
1788   (GMT), without exception. For the purposes of HTTP, GMT is exactly
1789   equal to UTC (Coordinated Universal Time). This is indicated in the
1790   first two formats by the inclusion of "GMT" as the three-letter
1791   abbreviation for time zone, and &MUST; be assumed when reading the
1792   asctime format. HTTP-date is case sensitive and &MUST-NOT; include
1793   additional whitespace beyond that specifically included as SP in the
1794   grammar.
1796<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="HTTP-date"/>
1797  <x:ref>HTTP-date</x:ref>    = <x:ref>rfc1123-date</x:ref> / <x:ref>obs-date</x:ref>
1799<t anchor="">
1800  <x:anchor-alias value="rfc1123-date"/>
1801  <x:anchor-alias value="time-of-day"/>
1802  <x:anchor-alias value="hour"/>
1803  <x:anchor-alias value="minute"/>
1804  <x:anchor-alias value="second"/>
1805  <x:anchor-alias value="day-name"/>
1806  <x:anchor-alias value="day"/>
1807  <x:anchor-alias value="month"/>
1808  <x:anchor-alias value="year"/>
1809  <x:anchor-alias value="GMT"/>
1810  Preferred format:
1812<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="rfc1123-date"/><iref primary="true" item="Grammar" subitem="date1"/><iref primary="true" item="Grammar" subitem="time-of-day"/><iref primary="true" item="Grammar" subitem="hour"/><iref primary="true" item="Grammar" subitem="minute"/><iref primary="true" item="Grammar" subitem="second"/><iref primary="true" item="Grammar" subitem="day-name"/><iref primary="true" item="Grammar" subitem="day-name-l"/><iref primary="true" item="Grammar" subitem="day"/><iref primary="true" item="Grammar" subitem="month"/><iref primary="true" item="Grammar" subitem="year"/><iref primary="true" item="Grammar" subitem="GMT"/>
1813  <x:ref>rfc1123-date</x:ref> = <x:ref>day-name</x:ref> "," <x:ref>SP</x:ref> date1 <x:ref>SP</x:ref> <x:ref>time-of-day</x:ref> <x:ref>SP</x:ref> <x:ref>GMT</x:ref>
1815  <x:ref>day-name</x:ref>     = <x:abnf-char-sequence>"Mon"</x:abnf-char-sequence> ; "Mon", case-sensitive
1816               / <x:abnf-char-sequence>"Tue"</x:abnf-char-sequence> ; "Tue", case-sensitive
1817               / <x:abnf-char-sequence>"Wed"</x:abnf-char-sequence> ; "Wed", case-sensitive
1818               / <x:abnf-char-sequence>"Thu"</x:abnf-char-sequence> ; "Thu", case-sensitive
1819               / <x:abnf-char-sequence>"Fri"</x:abnf-char-sequence> ; "Fri", case-sensitive
1820               / <x:abnf-char-sequence>"Sat"</x:abnf-char-sequence> ; "Sat", case-sensitive
1821               / <x:abnf-char-sequence>"Sun"</x:abnf-char-sequence> ; "Sun", case-sensitive
1823  <x:ref>date1</x:ref>        = <x:ref>day</x:ref> <x:ref>SP</x:ref> <x:ref>month</x:ref> <x:ref>SP</x:ref> <x:ref>year</x:ref>
1824               ; e.g., 02 Jun 1982
1826  <x:ref>day</x:ref>          = 2<x:ref>DIGIT</x:ref>
1827  <x:ref>month</x:ref>        = <x:abnf-char-sequence>"Jan"</x:abnf-char-sequence> ; "Jan", case-sensitive
1828               / <x:abnf-char-sequence>"Feb"</x:abnf-char-sequence> ; "Feb", case-sensitive
1829               / <x:abnf-char-sequence>"Mar"</x:abnf-char-sequence> ; "Mar", case-sensitive
1830               / <x:abnf-char-sequence>"Apr"</x:abnf-char-sequence> ; "Apr", case-sensitive
1831               / <x:abnf-char-sequence>"May"</x:abnf-char-sequence> ; "May", case-sensitive
1832               / <x:abnf-char-sequence>"Jun"</x:abnf-char-sequence> ; "Jun", case-sensitive
1833               / <x:abnf-char-sequence>"Jul"</x:abnf-char-sequence> ; "Jul", case-sensitive
1834               / <x:abnf-char-sequence>"Aug"</x:abnf-char-sequence> ; "Aug", case-sensitive
1835               / <x:abnf-char-sequence>"Sep"</x:abnf-char-sequence> ; "Sep", case-sensitive
1836               / <x:abnf-char-sequence>"Oct"</x:abnf-char-sequence> ; "Oct", case-sensitive
1837               / <x:abnf-char-sequence>"Nov"</x:abnf-char-sequence> ; "Nov", case-sensitive
1838               / <x:abnf-char-sequence>"Dec"</x:abnf-char-sequence> ; "Dec", case-sensitive
1839  <x:ref>year</x:ref>         = 4<x:ref>DIGIT</x:ref>
1841  <x:ref>GMT</x:ref>   = <x:abnf-char-sequence>"GMT"</x:abnf-char-sequence> ; "GMT", case-sensitive
1843  <x:ref>time-of-day</x:ref>  = <x:ref>hour</x:ref> ":" <x:ref>minute</x:ref> ":" <x:ref>second</x:ref>
1844                 ; 00:00:00 - 23:59:59
1846  <x:ref>hour</x:ref>         = 2<x:ref>DIGIT</x:ref>               
1847  <x:ref>minute</x:ref>       = 2<x:ref>DIGIT</x:ref>               
1848  <x:ref>second</x:ref>       = 2<x:ref>DIGIT</x:ref>               
1851  The semantics of <x:ref>day-name</x:ref>, <x:ref>day</x:ref>,
1852  <x:ref>month</x:ref>, <x:ref>year</x:ref>, and <x:ref>time-of-day</x:ref> are the
1853  same as those defined for the RFC 5322 constructs
1854  with the corresponding name (<xref target="RFC5322" x:fmt="," x:sec="3.3"/>).
1856<t anchor="">
1857  <x:anchor-alias value="obs-date"/>
1858  <x:anchor-alias value="rfc850-date"/>
1859  <x:anchor-alias value="asctime-date"/>
1860  <x:anchor-alias value="date1"/>
1861  <x:anchor-alias value="date2"/>
1862  <x:anchor-alias value="date3"/>
1863  <x:anchor-alias value="rfc1123-date"/>
1864  <x:anchor-alias value="day-name-l"/>
1865  Obsolete formats:
1867<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="obs-date"/>
1868  <x:ref>obs-date</x:ref>     = <x:ref>rfc850-date</x:ref> / <x:ref>asctime-date</x:ref> 
1870<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="rfc850-date"/>
1871  <x:ref>rfc850-date</x:ref>  = <x:ref>day-name-l</x:ref> "," <x:ref>SP</x:ref> <x:ref>date2</x:ref> <x:ref>SP</x:ref> <x:ref>time-of-day</x:ref> <x:ref>SP</x:ref> <x:ref>GMT</x:ref>
1872  <x:ref>date2</x:ref>        = <x:ref>day</x:ref> "-" <x:ref>month</x:ref> "-" 2<x:ref>DIGIT</x:ref>
1873                 ; day-month-year (e.g., 02-Jun-82)
1875  <x:ref>day-name-l</x:ref>   = <x:abnf-char-sequence>"Monday"</x:abnf-char-sequence> ; "Monday", case-sensitive
1876         / <x:abnf-char-sequence>"Tuesday"</x:abnf-char-sequence> ; "Tuesday", case-sensitive
1877         / <x:abnf-char-sequence>"Wednesday"</x:abnf-char-sequence> ; "Wednesday", case-sensitive
1878         / <x:abnf-char-sequence>"Thursday"</x:abnf-char-sequence> ; "Thursday", case-sensitive
1879         / <x:abnf-char-sequence>"Friday"</x:abnf-char-sequence> ; "Friday", case-sensitive
1880         / <x:abnf-char-sequence>"Saturday"</x:abnf-char-sequence> ; "Saturday", case-sensitive
1881         / <x:abnf-char-sequence>"Sunday"</x:abnf-char-sequence> ; "Sunday", case-sensitive
1883<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="asctime-date"/>
1884  <x:ref>asctime-date</x:ref> = <x:ref>day-name</x:ref> <x:ref>SP</x:ref> <x:ref>date3</x:ref> <x:ref>SP</x:ref> <x:ref>time-of-day</x:ref> <x:ref>SP</x:ref> <x:ref>year</x:ref>
1885  <x:ref>date3</x:ref>        = <x:ref>month</x:ref> <x:ref>SP</x:ref> ( 2<x:ref>DIGIT</x:ref> / ( <x:ref>SP</x:ref> 1<x:ref>DIGIT</x:ref> ))
1886                 ; month day (e.g., Jun  2)
1889  <t>
1890    <x:h>Note:</x:h> Recipients of date values are encouraged to be robust in
1891    accepting date values that may have been sent by non-HTTP
1892    applications, as is sometimes the case when retrieving or posting
1893    messages via proxies/gateways to SMTP or NNTP.
1894  </t>
1897  <t>
1898    <x:h>Note:</x:h> HTTP requirements for the date/time stamp format apply only
1899    to their usage within the protocol stream. Clients and servers are
1900    not required to use these formats for user presentation, request
1901    logging, etc.
1902  </t>
1906<section title="Transfer Codings" anchor="transfer.codings">
1907  <x:anchor-alias value="transfer-coding"/>
1908  <x:anchor-alias value="transfer-extension"/>
1910   Transfer-coding values are used to indicate an encoding
1911   transformation that has been, can be, or may need to be applied to an
1912   entity-body in order to ensure "safe transport" through the network.
1913   This differs from a content coding in that the transfer-coding is a
1914   property of the message, not of the original entity.
1916<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="transfer-coding"/><iref primary="true" item="Grammar" subitem="transfer-extension"/>
[673]1917  <x:ref>transfer-coding</x:ref>         = "chunked" ; <xref target="chunked.encoding"/>
1918                          / "compress" ; <xref target="compress.coding"/>
1919                          / "deflate" ; <xref target="deflate.coding"/>
1920                          / "gzip" ; <xref target="gzip.coding"/>
1921                          / <x:ref>transfer-extension</x:ref>
[623]1922  <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> )
1924<t anchor="rule.parameter">
1925  <x:anchor-alias value="attribute"/>
1926  <x:anchor-alias value="transfer-parameter"/>
1927  <x:anchor-alias value="value"/>
1928   Parameters are in  the form of attribute/value pairs.
1930<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"/>
1931  <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>
1932  <x:ref>attribute</x:ref>               = <x:ref>token</x:ref>
[810]1933  <x:ref>value</x:ref>                   = <x:ref>word</x:ref>
1936   All transfer-coding values are case-insensitive. HTTP/1.1 uses
1937   transfer-coding values in the TE header field (<xref target="header.te"/>) and in
1938   the Transfer-Encoding header field (<xref target="header.transfer-encoding"/>).
1941   Whenever a transfer-coding is applied to a message-body, the set of
1942   transfer-codings &MUST; include "chunked", unless the message indicates it
1943   is terminated by closing the connection. When the "chunked" transfer-coding
1944   is used, it &MUST; be the last transfer-coding applied to the
1945   message-body. The "chunked" transfer-coding &MUST-NOT; be applied more
1946   than once to a message-body. These rules allow the recipient to
1947   determine the transfer-length of the message (<xref target="message.length"/>).
[641]1950   Transfer-codings are analogous to the Content-Transfer-Encoding values of
1951   MIME, which were designed to enable safe transport of binary data over a
1952   7-bit transport service (<xref target="RFC2045" x:fmt="," x:sec="6"/>).
1953   However, safe transport
[623]1954   has a different focus for an 8bit-clean transfer protocol. In HTTP,
1955   the only unsafe characteristic of message-bodies is the difficulty in
1956   determining the exact body length (<xref target="message.length"/>), or the desire to
1957   encrypt data over a shared transport.
1960   A server which receives an entity-body with a transfer-coding it does
1961   not understand &SHOULD; return 501 (Not Implemented), and close the
1962   connection. A server &MUST-NOT; send transfer-codings to an HTTP/1.0
1963   client.
[673]1966<section title="Chunked Transfer Coding" anchor="chunked.encoding">
1967  <iref item="chunked (Coding Format)"/>
1968  <iref item="Coding Format" subitem="chunked"/>
[623]1969  <x:anchor-alias value="chunk"/>
1970  <x:anchor-alias value="Chunked-Body"/>
1971  <x:anchor-alias value="chunk-data"/>
1972  <x:anchor-alias value="chunk-ext"/>
1973  <x:anchor-alias value="chunk-ext-name"/>
1974  <x:anchor-alias value="chunk-ext-val"/>
1975  <x:anchor-alias value="chunk-size"/>
1976  <x:anchor-alias value="last-chunk"/>
1977  <x:anchor-alias value="trailer-part"/>
[707]1978  <x:anchor-alias value="quoted-str-nf"/>
1979  <x:anchor-alias value="qdtext-nf"/>
1981   The chunked encoding modifies the body of a message in order to
1982   transfer it as a series of chunks, each with its own size indicator,
1983   followed by an &OPTIONAL; trailer containing entity-header fields. This
1984   allows dynamically produced content to be transferred along with the
1985   information necessary for the recipient to verify that it has
1986   received the full message.
[707]1988<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Chunked-Body"/><iref primary="true" item="Grammar" subitem="chunk"/><iref primary="true" item="Grammar" subitem="chunk-size"/><iref primary="true" item="Grammar" subitem="last-chunk"/><iref primary="true" item="Grammar" subitem="chunk-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"/>
[623]1989  <x:ref>Chunked-Body</x:ref>   = *<x:ref>chunk</x:ref>
1990                   <x:ref>last-chunk</x:ref>
1991                   <x:ref>trailer-part</x:ref>
1992                   <x:ref>CRLF</x:ref>
1994  <x:ref>chunk</x:ref>          = <x:ref>chunk-size</x:ref> *WSP [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>
1995                   <x:ref>chunk-data</x:ref> <x:ref>CRLF</x:ref>
1996  <x:ref>chunk-size</x:ref>     = 1*<x:ref>HEXDIG</x:ref>
1997  <x:ref>last-chunk</x:ref>     = 1*("0") *WSP [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>
1999  <x:ref>chunk-ext</x:ref>      = *( ";" *WSP <x:ref>chunk-ext-name</x:ref>
2000                      [ "=" <x:ref>chunk-ext-val</x:ref> ] *WSP )
2001  <x:ref>chunk-ext-name</x:ref> = <x:ref>token</x:ref>
[707]2002  <x:ref>chunk-ext-val</x:ref>  = <x:ref>token</x:ref> / <x:ref>quoted-str-nf</x:ref>
[623]2003  <x:ref>chunk-data</x:ref>     = 1*<x:ref>OCTET</x:ref> ; a sequence of chunk-size octets
2004  <x:ref>trailer-part</x:ref>   = *( <x:ref>entity-header</x:ref> <x:ref>CRLF</x:ref> )
2006  <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>
2007                 ; like <x:ref>quoted-string</x:ref>, but disallowing line folding
2008  <x:ref>qdtext-nf</x:ref>      = <x:ref>WSP</x:ref> / %x21 / %x23-5B / %x5D-7E / <x:ref>obs-text</x:ref>
2009                 ; <x:ref>WSP</x:ref> / &lt;<x:ref>VCHAR</x:ref> except <x:ref>DQUOTE</x:ref> and "\"&gt; / <x:ref>obs-text</x:ref> 
2012   The chunk-size field is a string of hex digits indicating the size of
2013   the chunk-data in octets. The chunked encoding is ended by any chunk whose size is
2014   zero, followed by the trailer, which is terminated by an empty line.
2017   The trailer allows the sender to include additional HTTP header
2018   fields at the end of the message. The Trailer header field can be
2019   used to indicate which header fields are included in a trailer (see
2020   <xref target="header.trailer"/>).
2023   A server using chunked transfer-coding in a response &MUST-NOT; use the
2024   trailer for any header fields unless at least one of the following is
2025   true:
2026  <list style="numbers">
2027    <t>the request included a TE header field that indicates "trailers" is
2028     acceptable in the transfer-coding of the  response, as described in
2029     <xref target="header.te"/>; or,</t>
2031    <t>the server is the origin server for the response, the trailer
2032     fields consist entirely of optional metadata, and the recipient
2033     could use the message (in a manner acceptable to the origin server)
2034     without receiving this metadata.  In other words, the origin server
2035     is willing to accept the possibility that the trailer fields might
2036     be silently discarded along the path to the client.</t>
2037  </list>
2040   This requirement prevents an interoperability failure when the
2041   message is being received by an HTTP/1.1 (or later) proxy and
2042   forwarded to an HTTP/1.0 recipient. It avoids a situation where
2043   compliance with the protocol would have necessitated a possibly
2044   infinite buffer on the proxy.
2047   A process for decoding the "chunked" transfer-coding
2048   can be represented in pseudo-code as:
2050<figure><artwork type="code">
2051  length := 0
2052  read chunk-size, chunk-ext (if any) and CRLF
2053  while (chunk-size &gt; 0) {
2054     read chunk-data and CRLF
2055     append chunk-data to entity-body
2056     length := length + chunk-size
2057     read chunk-size and CRLF
2058  }
2059  read entity-header
2060  while (entity-header not empty) {
2061     append entity-header to existing header fields
2062     read entity-header
2063  }
2064  Content-Length := length
2065  Remove "chunked" from Transfer-Encoding
2068   All HTTP/1.1 applications &MUST; be able to receive and decode the
2069   "chunked" transfer-coding, and &MUST; ignore chunk-ext extensions
2070   they do not understand.
[673]2074<section title="Compression Codings" anchor="compression.codings">
2076   The codings defined below can be used to compress the payload of a
2077   message.
2080   <x:h>Note:</x:h> Use of program names for the identification of encoding formats
2081   is not desirable and is discouraged for future encodings. Their
2082   use here is representative of historical practice, not good
2083   design.
2086   <x:h>Note:</x:h> For compatibility with previous implementations of HTTP,
2087   applications &SHOULD; consider "x-gzip" and "x-compress" to be
2088   equivalent to "gzip" and "compress" respectively.
2091<section title="Compress Coding" anchor="compress.coding">
2092<iref item="compress (Coding Format)"/>
2093<iref item="Coding Format" subitem="compress"/>
2095   The "compress" format is produced by the common UNIX file compression
2096   program "compress". This format is an adaptive Lempel-Ziv-Welch
2097   coding (LZW).
2101<section title="Deflate Coding" anchor="deflate.coding">
2102<iref item="deflate (Coding Format)"/>
2103<iref item="Coding Format" subitem="deflate"/>
[801]2105   The "deflate" format is defined as the "deflate" compression mechanism
2106   (described in <xref target="RFC1951"/>) used inside the "zlib"
2107   data format (<xref target="RFC1950"/>).
2110  <t>
2111    <x:h>Note:</x:h> Some incorrect implementations send the "deflate"
2112    compressed data without the zlib wrapper.
2113   </t>
2117<section title="Gzip Coding" anchor="gzip.coding">
2118<iref item="gzip (Coding Format)"/>
2119<iref item="Coding Format" subitem="gzip"/>
2121   The "gzip" format is produced by the file compression program
2122   "gzip" (GNU zip), as described in <xref target="RFC1952"/>. This format is a
2123   Lempel-Ziv coding (LZ77) with a 32 bit CRC.
[670]2129<section title="Transfer Coding Registry" anchor="transfer.coding.registry">
2131   The HTTP Transfer Coding Registry defines the name space for the transfer
2132   coding names.
2135   Registrations &MUST; include the following fields:
2136   <list style="symbols">
2137     <t>Name</t>
2138     <t>Description</t>
2139     <t>Pointer to specification text</t>
2140   </list>
[808]2143   Names of transfer codings &MUST-NOT; overlap with names of content codings
2144   (&content-codings;), unless the encoding transformation is identical (as it
2145   is the case for the compression codings defined in
2146   <xref target="compression.codings"/>).
[670]2149   Values to be added to this name space require expert review and a specification
2150   (see "Expert Review" and "Specification Required" in
2151   <xref target="RFC5226" x:fmt="of" x:sec="4.1"/>), and &MUST;
2152   conform to the purpose of transfer coding defined in this section.
2155   The registry itself is maintained at
2156   <eref target=""/>.
2161<section title="Product Tokens" anchor="product.tokens">
2162  <x:anchor-alias value="product"/>
2163  <x:anchor-alias value="product-version"/>
2165   Product tokens are used to allow communicating applications to
2166   identify themselves by software name and version. Most fields using
2167   product tokens also allow sub-products which form a significant part
2168   of the application to be listed, separated by whitespace. By
2169   convention, the products are listed in order of their significance
2170   for identifying the application.
2172<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="product"/><iref primary="true" item="Grammar" subitem="product-version"/>
2173  <x:ref>product</x:ref>         = <x:ref>token</x:ref> ["/" <x:ref>product-version</x:ref>]
2174  <x:ref>product-version</x:ref> = <x:ref>token</x:ref>
2177   Examples:
2179<figure><artwork type="example">
2180  User-Agent: CERN-LineMode/2.15 libwww/2.17b3
2181  Server: Apache/0.8.4
2184   Product tokens &SHOULD; be short and to the point. They &MUST-NOT; be
2185   used for advertising or other non-essential information. Although any
2186   token character &MAY; appear in a product-version, this token &SHOULD;
2187   only be used for a version identifier (i.e., successive versions of
2188   the same product &SHOULD; only differ in the product-version portion of
2189   the product value).
2193<section title="Quality Values" anchor="quality.values">
2194  <x:anchor-alias value="qvalue"/>
2196   Both transfer codings (TE request header, <xref target="header.te"/>)
2197   and content negotiation (&content.negotiation;) use short "floating point"
2198   numbers to indicate the relative importance ("weight") of various
2199   negotiable parameters.  A weight is normalized to a real number in
2200   the range 0 through 1, where 0 is the minimum and 1 the maximum
2201   value. If a parameter has a quality value of 0, then content with
[746]2202   this parameter is "not acceptable" for the client. HTTP/1.1
[623]2203   applications &MUST-NOT; generate more than three digits after the
2204   decimal point. User configuration of these values &SHOULD; also be
2205   limited in this fashion.
2207<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="qvalue"/>
2208  <x:ref>qvalue</x:ref>         = ( "0" [ "." 0*3<x:ref>DIGIT</x:ref> ] )
2209                 / ( "1" [ "." 0*3("0") ] )
2212  <t>
2213     <x:h>Note:</x:h> "Quality values" is a misnomer, since these values merely represent
2214     relative degradation in desired quality.
2215  </t>
[8]2221<section title="Connections" anchor="connections">
2223<section title="Persistent Connections" anchor="persistent.connections">
2225<section title="Purpose" anchor="persistent.purpose">
2227   Prior to persistent connections, a separate TCP connection was
2228   established to fetch each URL, increasing the load on HTTP servers
2229   and causing congestion on the Internet. The use of inline images and
[761]2230   other associated data often requires a client to make multiple
[8]2231   requests of the same server in a short amount of time. Analysis of
2232   these performance problems and results from a prototype
2233   implementation are available <xref target="Pad1995"/> <xref target="Spe"/>. Implementation experience and
[578]2234   measurements of actual HTTP/1.1 implementations show good
[8]2235   results <xref target="Nie1997"/>. Alternatives have also been explored, for example,
2236   T/TCP <xref target="Tou1998"/>.
2239   Persistent HTTP connections have a number of advantages:
2240  <list style="symbols">
2241      <t>
2242        By opening and closing fewer TCP connections, CPU time is saved
2243        in routers and hosts (clients, servers, proxies, gateways,
2244        tunnels, or caches), and memory used for TCP protocol control
2245        blocks can be saved in hosts.
2246      </t>
2247      <t>
2248        HTTP requests and responses can be pipelined on a connection.
2249        Pipelining allows a client to make multiple requests without
2250        waiting for each response, allowing a single TCP connection to
2251        be used much more efficiently, with much lower elapsed time.
2252      </t>
2253      <t>
2254        Network congestion is reduced by reducing the number of packets
2255        caused by TCP opens, and by allowing TCP sufficient time to
2256        determine the congestion state of the network.
2257      </t>
2258      <t>
2259        Latency on subsequent requests is reduced since there is no time
2260        spent in TCP's connection opening handshake.
2261      </t>
2262      <t>
2263        HTTP can evolve more gracefully, since errors can be reported
2264        without the penalty of closing the TCP connection. Clients using
2265        future versions of HTTP might optimistically try a new feature,
2266        but if communicating with an older server, retry with old
2267        semantics after an error is reported.
2268      </t>
2269    </list>
2272   HTTP implementations &SHOULD; implement persistent connections.
2276<section title="Overall Operation" anchor="persistent.overall">
2278   A significant difference between HTTP/1.1 and earlier versions of
2279   HTTP is that persistent connections are the default behavior of any
2280   HTTP connection. That is, unless otherwise indicated, the client
2281   &SHOULD; assume that the server will maintain a persistent connection,
2282   even after error responses from the server.
2285   Persistent connections provide a mechanism by which a client and a
2286   server can signal the close of a TCP connection. This signaling takes
2287   place using the Connection header field (<xref target="header.connection"/>). Once a close
2288   has been signaled, the client &MUST-NOT; send any more requests on that
2289   connection.
2292<section title="Negotiation" anchor="persistent.negotiation">
2294   An HTTP/1.1 server &MAY; assume that a HTTP/1.1 client intends to
2295   maintain a persistent connection unless a Connection header including
2296   the connection-token "close" was sent in the request. If the server
2297   chooses to close the connection immediately after sending the
2298   response, it &SHOULD; send a Connection header including the
[761]2299   connection-token "close".
2302   An HTTP/1.1 client &MAY; expect a connection to remain open, but would
2303   decide to keep it open based on whether the response from a server
2304   contains a Connection header with the connection-token close. In case
2305   the client does not want to maintain a connection for more than that
2306   request, it &SHOULD; send a Connection header including the
2307   connection-token close.
2310   If either the client or the server sends the close token in the
2311   Connection header, that request becomes the last one for the
2312   connection.
2315   Clients and servers &SHOULD-NOT;  assume that a persistent connection is
2316   maintained for HTTP versions less than 1.1 unless it is explicitly
2317   signaled. See <xref target="compatibility.with.http.1.0.persistent.connections"/> for more information on backward
2318   compatibility with HTTP/1.0 clients.
2321   In order to remain persistent, all messages on the connection &MUST;
2322   have a self-defined message length (i.e., one not defined by closure
2323   of the connection), as described in <xref target="message.length"/>.
2327<section title="Pipelining" anchor="pipelining">
2329   A client that supports persistent connections &MAY; "pipeline" its
2330   requests (i.e., send multiple requests without waiting for each
2331   response). A server &MUST; send its responses to those requests in the
2332   same order that the requests were received.
2335   Clients which assume persistent connections and pipeline immediately
2336   after connection establishment &SHOULD; be prepared to retry their
2337   connection if the first pipelined attempt fails. If a client does
2338   such a retry, it &MUST-NOT; pipeline before it knows the connection is
2339   persistent. Clients &MUST; also be prepared to resend their requests if
2340   the server closes the connection before sending all of the
2341   corresponding responses.
2344   Clients &SHOULD-NOT;  pipeline requests using non-idempotent methods or
[29]2345   non-idempotent sequences of methods (see &idempotent-methods;). Otherwise, a
[8]2346   premature termination of the transport connection could lead to
2347   indeterminate results. A client wishing to send a non-idempotent
2348   request &SHOULD; wait to send that request until it has received the
2349   response status for the previous request.
2354<section title="Proxy Servers" anchor="persistent.proxy">
2356   It is especially important that proxies correctly implement the
2357   properties of the Connection header field as specified in <xref target="header.connection"/>.
2360   The proxy server &MUST; signal persistent connections separately with
2361   its clients and the origin servers (or other proxy servers) that it
2362   connects to. Each persistent connection applies to only one transport
2363   link.
2366   A proxy server &MUST-NOT; establish a HTTP/1.1 persistent connection
[578]2367   with an HTTP/1.0 client (but see <xref x:sec="19.7.1" x:fmt="of" target="RFC2068"/>
2368   for information and discussion of the problems with the Keep-Alive header
2369   implemented by many HTTP/1.0 clients).
2372<section title="End-to-end and Hop-by-hop Headers" anchor="end-to-end.and.hop-by-hop.headers">
[769]2374  <cref anchor="TODO-end-to-end" source="jre">
2375    Restored from <eref target=""/>.
[839]2376    See also <eref target=""/>.
[769]2377  </cref>
2380   For the purpose of defining the behavior of caches and non-caching
2381   proxies, we divide HTTP headers into two categories:
2382  <list style="symbols">
2383      <t>End-to-end headers, which are  transmitted to the ultimate
2384        recipient of a request or response. End-to-end headers in
2385        responses MUST be stored as part of a cache entry and &MUST; be
2386        transmitted in any response formed from a cache entry.</t>
2388      <t>Hop-by-hop headers, which are meaningful only for a single
2389        transport-level connection, and are not stored by caches or
2390        forwarded by proxies.</t>
2391  </list>
2394   The following HTTP/1.1 headers are hop-by-hop headers:
2395  <list style="symbols">
2396      <t>Connection</t>
2397      <t>Keep-Alive</t>
2398      <t>Proxy-Authenticate</t>
2399      <t>Proxy-Authorization</t>
2400      <t>TE</t>
2401      <t>Trailer</t>
2402      <t>Transfer-Encoding</t>
2403      <t>Upgrade</t>
2404  </list>
2407   All other headers defined by HTTP/1.1 are end-to-end headers.
2410   Other hop-by-hop headers &MUST; be listed in a Connection header
2411   (<xref target="header.connection"/>).
[769]2415<section title="Non-modifiable Headers" anchor="non-modifiable.headers">
[769]2417  <cref anchor="TODO-non-mod-headers" source="jre">
2418    Restored from <eref target=""/>.
[839]2419    See also <eref target=""/>.
[769]2420  </cref>
2423   Some features of HTTP/1.1, such as Digest Authentication, depend on the
2424   value of certain end-to-end headers. A transparent proxy &SHOULD-NOT;
2425   modify an end-to-end header unless the definition of that header requires
2426   or specifically allows that.
2429   A transparent proxy &MUST-NOT; modify any of the following fields in a
2430   request or response, and it &MUST-NOT; add any of these fields if not
2431   already present:
2432  <list style="symbols">
2433      <t>Content-Location</t>
2434      <t>Content-MD5</t>
2435      <t>ETag</t>
2436      <t>Last-Modified</t>
2437  </list>
2440   A transparent proxy &MUST-NOT; modify any of the following fields in a
2441   response:
2442  <list style="symbols">
2443    <t>Expires</t>
2444  </list>
2447   but it &MAY; add any of these fields if not already present. If an
2448   Expires header is added, it &MUST; be given a field-value identical to
2449   that of the Date header in that response.
2452   A proxy &MUST-NOT; modify or add any of the following fields in a
2453   message that contains the no-transform cache-control directive, or in
2454   any request:
2455  <list style="symbols">
2456    <t>Content-Encoding</t>
2457    <t>Content-Range</t>
2458    <t>Content-Type</t>
2459  </list>
2462   A non-transparent proxy &MAY; modify or add these fields to a message
2463   that does not include no-transform, but if it does so, it &MUST; add a
2464   Warning 214 (Transformation applied) if one does not already appear
2465   in the message (see &header-warning;).
2468  <t>
2469    <x:h>Warning:</x:h> Unnecessary modification of end-to-end headers might
2470    cause authentication failures if stronger authentication
2471    mechanisms are introduced in later versions of HTTP. Such
2472    authentication mechanisms &MAY; rely on the values of header fields
2473    not listed here.
2474  </t>
2477   The Content-Length field of a request or response is added or deleted
2478   according to the rules in <xref target="message.length"/>. A transparent proxy &MUST;
2479   preserve the entity-length (&entity-length;) of the entity-body,
2480   although it &MAY; change the transfer-length (<xref target="message.length"/>).
[8]2486<section title="Practical Considerations" anchor="persistent.practical">
2488   Servers will usually have some time-out value beyond which they will
2489   no longer maintain an inactive connection. Proxy servers might make
2490   this a higher value since it is likely that the client will be making
2491   more connections through the same server. The use of persistent
2492   connections places no requirements on the length (or existence) of
2493   this time-out for either the client or the server.
2496   When a client or server wishes to time-out it &SHOULD; issue a graceful
2497   close on the transport connection. Clients and servers &SHOULD; both
2498   constantly watch for the other side of the transport close, and
2499   respond to it as appropriate. If a client or server does not detect
2500   the other side's close promptly it could cause unnecessary resource
2501   drain on the network.
2504   A client, server, or proxy &MAY; close the transport connection at any
2505   time. For example, a client might have started to send a new request
2506   at the same time that the server has decided to close the "idle"
2507   connection. From the server's point of view, the connection is being
2508   closed while it was idle, but from the client's point of view, a
2509   request is in progress.
2512   This means that clients, servers, and proxies &MUST; be able to recover
2513   from asynchronous close events. Client software &SHOULD; reopen the
2514   transport connection and retransmit the aborted sequence of requests
2515   without user interaction so long as the request sequence is
[29]2516   idempotent (see &idempotent-methods;). Non-idempotent methods or sequences
[8]2517   &MUST-NOT; be automatically retried, although user agents &MAY; offer a
2518   human operator the choice of retrying the request(s). Confirmation by
2519   user-agent software with semantic understanding of the application
2520   &MAY; substitute for user confirmation. The automatic retry &SHOULD-NOT; 
2521   be repeated if the second sequence of requests fails.
2524   Servers &SHOULD; always respond to at least one request per connection,
2525   if at all possible. Servers &SHOULD-NOT;  close a connection in the
2526   middle of transmitting a response, unless a network or client failure
2527   is suspected.
[715]2530   Clients (including proxies) &SHOULD; limit the number of simultaneous
2531   connections that they maintain to a given server (including proxies).
2534   Previous revisions of HTTP gave a specific number of connections as a
2535   ceiling, but this was found to be impractical for many applications. As a
2536   result, this specification does not mandate a particular maximum number of
2537   connections, but instead encourages clients to be conservative when opening
2538   multiple connections.
2541   In particular, while using multiple connections avoids the "head-of-line
2542   blocking" problem (whereby a request that takes significant server-side
2543   processing and/or has a large payload can block subsequent requests on the
2544   same connection), each connection used consumes server resources (sometimes
2545   significantly), and furthermore using multiple connections can cause
2546   undesirable side effects in congested networks.
2549   Note that servers might reject traffic that they deem abusive, including an
2550   excessive number of connections from a client.
2555<section title="Message Transmission Requirements" anchor="message.transmission.requirements">
2557<section title="Persistent Connections and Flow Control" anchor="persistent.flow">
2559   HTTP/1.1 servers &SHOULD; maintain persistent connections and use TCP's
2560   flow control mechanisms to resolve temporary overloads, rather than
2561   terminating connections with the expectation that clients will retry.
2562   The latter technique can exacerbate network congestion.
2566<section title="Monitoring Connections for Error Status Messages" anchor="persistent.monitor">
2568   An HTTP/1.1 (or later) client sending a message-body &SHOULD; monitor
2569   the network connection for an error status while it is transmitting
2570   the request. If the client sees an error status, it &SHOULD;
2571   immediately cease transmitting the body. If the body is being sent
2572   using a "chunked" encoding (<xref target="transfer.codings"/>), a zero length chunk and
2573   empty trailer &MAY; be used to prematurely mark the end of the message.
2574   If the body was preceded by a Content-Length header, the client &MUST;
2575   close the connection.
2579<section title="Use of the 100 (Continue) Status" anchor="use.of.the.100.status">
[29]2581   The purpose of the 100 (Continue) status (see &status-100;) is to
[8]2582   allow a client that is sending a request message with a request body
2583   to determine if the origin server is willing to accept the request
2584   (based on the request headers) before the client sends the request
2585   body. In some cases, it might either be inappropriate or highly
2586   inefficient for the client to send the body if the server will reject
2587   the message without looking at the body.
2590   Requirements for HTTP/1.1 clients:
2591  <list style="symbols">
2592    <t>
2593        If a client will wait for a 100 (Continue) response before
2594        sending the request body, it &MUST; send an Expect request-header
[29]2595        field (&header-expect;) with the "100-continue" expectation.
[8]2596    </t>
2597    <t>
[29]2598        A client &MUST-NOT; send an Expect request-header field (&header-expect;)
[8]2599        with the "100-continue" expectation if it does not intend
2600        to send a request body.
2601    </t>
2602  </list>
2605   Because of the presence of older implementations, the protocol allows
2606   ambiguous situations in which a client may send "Expect: 100-continue"
2607   without receiving either a 417 (Expectation Failed) status
2608   or a 100 (Continue) status. Therefore, when a client sends this
2609   header field to an origin server (possibly via a proxy) from which it
2610   has never seen a 100 (Continue) status, the client &SHOULD-NOT;  wait
2611   for an indefinite period before sending the request body.
2614   Requirements for HTTP/1.1 origin servers:
2615  <list style="symbols">
2616    <t> Upon receiving a request which includes an Expect request-header
2617        field with the "100-continue" expectation, an origin server &MUST;
2618        either respond with 100 (Continue) status and continue to read
2619        from the input stream, or respond with a final status code. The
2620        origin server &MUST-NOT; wait for the request body before sending
2621        the 100 (Continue) response. If it responds with a final status
2622        code, it &MAY; close the transport connection or it &MAY; continue
2623        to read and discard the rest of the request.  It &MUST-NOT;
2624        perform the requested method if it returns a final status code.
2625    </t>
2626    <t> An origin server &SHOULD-NOT;  send a 100 (Continue) response if
2627        the request message does not include an Expect request-header
2628        field with the "100-continue" expectation, and &MUST-NOT; send a
2629        100 (Continue) response if such a request comes from an HTTP/1.0
2630        (or earlier) client. There is an exception to this rule: for
[97]2631        compatibility with <xref target="RFC2068"/>, a server &MAY; send a 100 (Continue)
[8]2632        status in response to an HTTP/1.1 PUT or POST request that does
2633        not include an Expect request-header field with the "100-continue"
2634        expectation. This exception, the purpose of which is
2635        to minimize any client processing delays associated with an
2636        undeclared wait for 100 (Continue) status, applies only to
2637        HTTP/1.1 requests, and not to requests with any other HTTP-version
2638        value.
2639    </t>
2640    <t> An origin server &MAY; omit a 100 (Continue) response if it has
2641        already received some or all of the request body for the
2642        corresponding request.
2643    </t>
2644    <t> An origin server that sends a 100 (Continue) response &MUST;
2645    ultimately send a final status code, once the request body is
2646        received and processed, unless it terminates the transport
2647        connection prematurely.
2648    </t>
2649    <t> If an origin server receives a request that does not include an
2650        Expect request-header field with the "100-continue" expectation,
2651        the request includes a request body, and the server responds
2652        with a final status code before reading the entire request body
2653        from the transport connection, then the server &SHOULD-NOT;  close
2654        the transport connection until it has read the entire request,
2655        or until the client closes the connection. Otherwise, the client
2656        might not reliably receive the response message. However, this
2657        requirement is not be construed as preventing a server from
2658        defending itself against denial-of-service attacks, or from
2659        badly broken client implementations.
2660      </t>
2661    </list>
2664   Requirements for HTTP/1.1 proxies:
2665  <list style="symbols">
2666    <t> If a proxy receives a request that includes an Expect request-header
2667        field with the "100-continue" expectation, and the proxy
2668        either knows that the next-hop server complies with HTTP/1.1 or
2669        higher, or does not know the HTTP version of the next-hop
2670        server, it &MUST; forward the request, including the Expect header
2671        field.
2672    </t>
2673    <t> If the proxy knows that the version of the next-hop server is
2674        HTTP/1.0 or lower, it &MUST-NOT; forward the request, and it &MUST;
2675        respond with a 417 (Expectation Failed) status.
2676    </t>
2677    <t> Proxies &SHOULD; maintain a cache recording the HTTP version
2678        numbers received from recently-referenced next-hop servers.
2679    </t>
2680    <t> A proxy &MUST-NOT; forward a 100 (Continue) response if the
2681        request message was received from an HTTP/1.0 (or earlier)
2682        client and did not include an Expect request-header field with
2683        the "100-continue" expectation. This requirement overrides the
[29]2684        general rule for forwarding of 1xx responses (see &status-1xx;).
[8]2685    </t>
2686  </list>
2690<section title="Client Behavior if Server Prematurely Closes Connection" anchor="connection.premature">
2692   If an HTTP/1.1 client sends a request which includes a request body,
2693   but which does not include an Expect request-header field with the
2694   "100-continue" expectation, and if the client is not directly
2695   connected to an HTTP/1.1 origin server, and if the client sees the
2696   connection close before receiving any status from the server, the
2697   client &SHOULD; retry the request.  If the client does retry this
2698   request, it &MAY; use the following "binary exponential backoff"
2699   algorithm to be assured of obtaining a reliable response:
2700  <list style="numbers">
2701    <t>
2702      Initiate a new connection to the server
2703    </t>
2704    <t>
2705      Transmit the request-headers
2706    </t>
2707    <t>
2708      Initialize a variable R to the estimated round-trip time to the
2709         server (e.g., based on the time it took to establish the
2710         connection), or to a constant value of 5 seconds if the round-trip
2711         time is not available.
2712    </t>
2713    <t>
2714       Compute T = R * (2**N), where N is the number of previous
2715         retries of this request.
2716    </t>
2717    <t>
2718       Wait either for an error response from the server, or for T
2719         seconds (whichever comes first)
2720    </t>
2721    <t>
2722       If no error response is received, after T seconds transmit the
2723         body of the request.
2724    </t>
2725    <t>
2726       If client sees that the connection is closed prematurely,
2727         repeat from step 1 until the request is accepted, an error
2728         response is received, or the user becomes impatient and
2729         terminates the retry process.
2730    </t>
2731  </list>
2734   If at any point an error status is received, the client
2735  <list style="symbols">
2736      <t>&SHOULD-NOT;  continue and</t>
2738      <t>&SHOULD; close the connection if it has not completed sending the
2739        request message.</t>
2740    </list>
[651]2747<section title="Miscellaneous notes that may disappear" anchor="misc">
2748<section title="Scheme aliases considered harmful" anchor="scheme.aliases">
[767]2750   <cref anchor="TBD-aliases-harmful">describe why aliases like webcal are harmful.</cref>
2754<section title="Use of HTTP for proxy communication" anchor="http.proxy">
[767]2756   <cref anchor="TBD-proxy-other">Configured to use HTTP to proxy HTTP or other protocols.</cref>
[651]2760<section title="Interception of HTTP for access control" anchor="http.intercept">
[767]2762   <cref anchor="TBD-intercept">Interception of HTTP traffic for initiating access control.</cref>
[651]2766<section title="Use of HTTP by other protocols" anchor="http.others">
[767]2768   <cref anchor="TBD-profiles">Profiles of HTTP defined by other protocol.
[651]2769   Extensions of HTTP like WebDAV.</cref>
2773<section title="Use of HTTP by media type specification" anchor="">
[767]2775   <cref anchor="TBD-hypertext">Instructions on composing HTTP requests via hypertext formats.</cref>
[647]2780<section title="Header Field Definitions" anchor="header.field.definitions">
[117]2782   This section defines the syntax and semantics of HTTP/1.1 header fields
2783   related to message framing and transport protocols.
2786   For entity-header fields, both sender and recipient refer to either the
2787   client or the server, depending on who sends and who receives the entity.
2790<section title="Connection" anchor="header.connection">
2791  <iref primary="true" item="Connection header" x:for-anchor=""/>
2792  <iref primary="true" item="Headers" subitem="Connection" x:for-anchor=""/>
[229]2793  <x:anchor-alias value="Connection"/>
2794  <x:anchor-alias value="connection-token"/>
[354]2795  <x:anchor-alias value="Connection-v"/>
[697]2797   The "Connection" general-header field allows the sender to specify
[8]2798   options that are desired for that particular connection and &MUST-NOT;
2799   be communicated by proxies over further connections.
[354]2802   The Connection header's value has the following grammar:
[354]2804<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Connection"/><iref primary="true" item="Grammar" subitem="Connection-v"/><iref primary="true" item="Grammar" subitem="connection-token"/>
[366]2805  <x:ref>Connection</x:ref>       = "Connection" ":" <x:ref>OWS</x:ref> <x:ref>Connection-v</x:ref>
[354]2806  <x:ref>Connection-v</x:ref>     = 1#<x:ref>connection-token</x:ref>
2807  <x:ref>connection-token</x:ref> = <x:ref>token</x:ref>
2810   HTTP/1.1 proxies &MUST; parse the Connection header field before a
2811   message is forwarded and, for each connection-token in this field,
2812   remove any header field(s) from the message with the same name as the
2813   connection-token. Connection options are signaled by the presence of
2814   a connection-token in the Connection header field, not by any
2815   corresponding additional header field(s), since the additional header
2816   field may not be sent if there are no parameters associated with that
2817   connection option.
2820   Message headers listed in the Connection header &MUST-NOT; include
2821   end-to-end headers, such as Cache-Control.
2824   HTTP/1.1 defines the "close" connection option for the sender to
2825   signal that the connection will be closed after completion of the
2826   response. For example,
2828<figure><artwork type="example">
[354]2829  Connection: close
2832   in either the request or the response header fields indicates that
[746]2833   the connection &SHOULD-NOT;  be considered "persistent" (<xref target="persistent.connections"/>)
[8]2834   after the current request/response is complete.
[86]2837   An HTTP/1.1 client that does not support persistent connections &MUST;
2838   include the "close" connection option in every request message.
[86]2841   An HTTP/1.1 server that does not support persistent connections &MUST;
2842   include the "close" connection option in every response message that
[753]2843   does not have a 1xx (Informational) status code.
[8]2846   A system receiving an HTTP/1.0 (or lower-version) message that
[96]2847   includes a Connection header &MUST;, for each connection-token in this
[8]2848   field, remove and ignore any header field(s) from the message with
2849   the same name as the connection-token. This protects against mistaken
2850   forwarding of such header fields by pre-HTTP/1.1 proxies. See <xref target="compatibility.with.http.1.0.persistent.connections"/>.
2854<section title="Content-Length" anchor="header.content-length">
2855  <iref primary="true" item="Content-Length header" x:for-anchor=""/>
2856  <iref primary="true" item="Headers" subitem="Content-Length" x:for-anchor=""/>
[229]2857  <x:anchor-alias value="Content-Length"/>
[354]2858  <x:anchor-alias value="Content-Length-v"/>
[697]2860   The "Content-Length" entity-header field indicates the size of the
[698]2861   entity-body, in number of OCTETs. In the case of responses to the HEAD
2862   method, it indicates the size of the entity-body that would have been sent
2863   had the request been a GET.
[354]2865<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Length"/><iref primary="true" item="Grammar" subitem="Content-Length-v"/>
[366]2866  <x:ref>Content-Length</x:ref>   = "Content-Length" ":" <x:ref>OWS</x:ref> 1*<x:ref>Content-Length-v</x:ref>
[354]2867  <x:ref>Content-Length-v</x:ref> = 1*<x:ref>DIGIT</x:ref>
2870   An example is
2872<figure><artwork type="example">
[354]2873  Content-Length: 3495
2876   Applications &SHOULD; use this field to indicate the transfer-length of
2877   the message-body, unless this is prohibited by the rules in <xref target="message.length"/>.
2880   Any Content-Length greater than or equal to zero is a valid value.
2881   <xref target="message.length"/> describes how to determine the length of a message-body
2882   if a Content-Length is not given.
2885   Note that the meaning of this field is significantly different from
2886   the corresponding definition in MIME, where it is an optional field
2887   used within the "message/external-body" content-type. In HTTP, it
2888   &SHOULD; be sent whenever the message's length can be determined prior
2889   to being transferred, unless this is prohibited by the rules in
2890   <xref target="message.length"/>.
2894<section title="Date" anchor="">
2895  <iref primary="true" item="Date header" x:for-anchor=""/>
2896  <iref primary="true" item="Headers" subitem="Date" x:for-anchor=""/>
[229]2897  <x:anchor-alias value="Date"/>
[354]2898  <x:anchor-alias value="Date-v"/>
[697]2900   The "Date" general-header field represents the date and time at which
[727]2901   the message was originated, having the same semantics as the Origination
2902   Date Field (orig-date) defined in <xref target="RFC5322" x:fmt="of" x:sec="3.6.1"/>.
2903   The field value is an HTTP-date, as described in <xref target=""/>;
[84]2904   it &MUST; be sent in rfc1123-date format.
[354]2906<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Date"/><iref primary="true" item="Grammar" subitem="Date-v"/>
[366]2907  <x:ref>Date</x:ref>   = "Date" ":" <x:ref>OWS</x:ref> <x:ref>Date-v</x:ref>
[354]2908  <x:ref>Date-v</x:ref> = <x:ref>HTTP-date</x:ref>
2911   An example is
2913<figure><artwork type="example">
[354]2914  Date: Tue, 15 Nov 1994 08:12:31 GMT
2917   Origin servers &MUST; include a Date header field in all responses,
2918   except in these cases:
2919  <list style="numbers">
2920      <t>If the response status code is 100 (Continue) or 101 (Switching
2921         Protocols), the response &MAY; include a Date header field, at
2922         the server's option.</t>
[763]2924      <t>If the response status code conveys a server error, e.g., 500
[8]2925         (Internal Server Error) or 503 (Service Unavailable), and it is
2926         inconvenient or impossible to generate a valid Date.</t>
2928      <t>If the server does not have a clock that can provide a
2929         reasonable approximation of the current time, its responses
2930         &MUST-NOT; include a Date header field. In this case, the rules
2931         in <xref target="clockless.origin.server.operation"/> &MUST; be followed.</t>
2932  </list>
2935   A received message that does not have a Date header field &MUST; be
2936   assigned one by the recipient if the message will be cached by that
2937   recipient or gatewayed via a protocol which requires a Date. An HTTP
2938   implementation without a clock &MUST-NOT; cache responses without
2939   revalidating them on every use. An HTTP cache, especially a shared
2940   cache, &SHOULD; use a mechanism, such as NTP <xref target="RFC1305"/>, to synchronize its
2941   clock with a reliable external standard.
2944   Clients &SHOULD; only send a Date header field in messages that include
2945   an entity-body, as in the case of the PUT and POST requests, and even
2946   then it is optional. A client without a clock &MUST-NOT; send a Date
2947   header field in a request.
2950   The HTTP-date sent in a Date header &SHOULD-NOT;  represent a date and
2951   time subsequent to the generation of the message. It &SHOULD; represent
2952   the best available approximation of the date and time of message
2953   generation, unless the implementation has no means of generating a
2954   reasonably accurate date and time. In theory, the date ought to
2955   represent the moment just before the entity is generated. In
2956   practice, the date can be generated at any time during the message
2957   origination without affecting its semantic value.
2960<section title="Clockless Origin Server Operation" anchor="clockless.origin.server.operation">
2962   Some origin server implementations might not have a clock available.
2963   An origin server without a clock &MUST-NOT; assign Expires or Last-Modified
2964   values to a response, unless these values were associated
2965   with the resource by a system or user with a reliable clock. It &MAY;
2966   assign an Expires value that is known, at or before server
2967   configuration time, to be in the past (this allows "pre-expiration"
2968   of responses without storing separate Expires values for each
2969   resource).
2974<section title="Host" anchor="">
2975  <iref primary="true" item="Host header" x:for-anchor=""/>
2976  <iref primary="true" item="Headers" subitem="Host" x:for-anchor=""/>
[229]2977  <x:anchor-alias value="Host"/>
[354]2978  <x:anchor-alias value="Host-v"/>
[697]2980   The "Host" request-header field specifies the Internet host and port
[698]2981   number of the resource being requested, allowing the origin server or
2982   gateway to differentiate between internally-ambiguous URLs, such as the root
2983   "/" URL of a server for multiple host names on a single IP address.
2986   The Host field value &MUST; represent the naming authority of the origin
2987   server or gateway given by the original URL obtained from the user or
2988   referring resource (generally an http URI, as described in
2989   <xref target="http.uri"/>).
[354]2991<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Host"/><iref primary="true" item="Grammar" subitem="Host-v"/>
[366]2992  <x:ref>Host</x:ref>   = "Host" ":" <x:ref>OWS</x:ref> <x:ref>Host-v</x:ref>
[374]2993  <x:ref>Host-v</x:ref> = <x:ref>uri-host</x:ref> [ ":" <x:ref>port</x:ref> ] ; <xref target="http.uri"/>
2996   A "host" without any trailing port information implies the default
2997   port for the service requested (e.g., "80" for an HTTP URL). For
2998   example, a request on the origin server for
[90]2999   &lt;; would properly include:
[803]3001<figure><artwork type="message/http; msgtype=&#34;request&#34;" x:indent-with="  ">
3002GET /pub/WWW/ HTTP/1.1
3006   A client &MUST; include a Host header field in all HTTP/1.1 request
[148]3007   messages. If the requested URI does not include an Internet host
[8]3008   name for the service being requested, then the Host header field &MUST;
3009   be given with an empty value. An HTTP/1.1 proxy &MUST; ensure that any
3010   request message it forwards does contain an appropriate Host header
3011   field that identifies the service being requested by the proxy. All
3012   Internet-based HTTP/1.1 servers &MUST; respond with a 400 (Bad Request)
3013   status code to any HTTP/1.1 request message which lacks a Host header
3014   field.
[97]3017   See Sections <xref target="" format="counter"/>
[8]3018   and <xref target="" format="counter"/>
3019   for other requirements relating to Host.
3023<section title="TE" anchor="header.te">
3024  <iref primary="true" item="TE header" x:for-anchor=""/>
3025  <iref primary="true" item="Headers" subitem="TE" x:for-anchor=""/>
[229]3026  <x:anchor-alias value="TE"/>
[354]3027  <x:anchor-alias value="TE-v"/>
[229]3028  <x:anchor-alias value="t-codings"/>
[457]3029  <x:anchor-alias value="te-params"/>
3030  <x:anchor-alias value="te-ext"/>
[697]3032   The "TE" request-header field indicates what extension transfer-codings
[698]3033   it is willing to accept in the response, and whether or not it is
3034   willing to accept trailer fields in a chunked transfer-coding.
3037   Its value may consist of the keyword "trailers" and/or a comma-separated
[8]3038   list of extension transfer-coding names with optional accept
3039   parameters (as described in <xref target="transfer.codings"/>).
[457]3041<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="TE"/><iref primary="true" item="Grammar" subitem="TE-v"/><iref primary="true" item="Grammar" subitem="t-codings"/><iref primary="true" item="Grammar" subitem="te-params"/><iref primary="true" item="Grammar" subitem="te-ext"/>
[366]3042  <x:ref>TE</x:ref>        = "TE" ":" <x:ref>OWS</x:ref> <x:ref>TE-v</x:ref>
[354]3043  <x:ref>TE-v</x:ref>      = #<x:ref>t-codings</x:ref>
[457]3044  <x:ref>t-codings</x:ref> = "trailers" / ( <x:ref>transfer-extension</x:ref> [ <x:ref>te-params</x:ref> ] )
3045  <x:ref>te-params</x:ref> = <x:ref>OWS</x:ref> ";" <x:ref>OWS</x:ref> "q=" <x:ref>qvalue</x:ref> *( <x:ref>te-ext</x:ref> )
[810]3046  <x:ref>te-ext</x:ref>    = <x:ref>OWS</x:ref> ";" <x:ref>OWS</x:ref> <x:ref>token</x:ref> [ "=" <x:ref>word</x:ref> ]
3049   The presence of the keyword "trailers" indicates that the client is
3050   willing to accept trailer fields in a chunked transfer-coding, as
[673]3051   defined in <xref target="chunked.encoding"/>. This keyword is reserved for use with
[8]3052   transfer-coding values even though it does not itself represent a
3053   transfer-coding.
3056   Examples of its use are:
3058<figure><artwork type="example">
[354]3059  TE: deflate
3060  TE:
3061  TE: trailers, deflate;q=0.5
3064   The TE header field only applies to the immediate connection.
3065   Therefore, the keyword &MUST; be supplied within a Connection header
3066   field (<xref target="header.connection"/>) whenever TE is present in an HTTP/1.1 message.
3069   A server tests whether a transfer-coding is acceptable, according to
3070   a TE field, using these rules:
3071  <list style="numbers">
3072    <x:lt>
3073      <t>The "chunked" transfer-coding is always acceptable. If the
3074         keyword "trailers" is listed, the client indicates that it is
3075         willing to accept trailer fields in the chunked response on
3076         behalf of itself and any downstream clients. The implication is
3077         that, if given, the client is stating that either all
3078         downstream clients are willing to accept trailer fields in the
3079         forwarded response, or that it will attempt to buffer the
3080         response on behalf of downstream recipients.
3081      </t><t>
3082         <x:h>Note:</x:h> HTTP/1.1 does not define any means to limit the size of a
3083         chunked response such that a client can be assured of buffering
3084         the entire response.</t>
3085    </x:lt>
3086    <x:lt>
3087      <t>If the transfer-coding being tested is one of the transfer-codings
3088         listed in the TE field, then it is acceptable unless it
[457]3089         is accompanied by a qvalue of 0. (As defined in <xref target="quality.values"/>, a
[8]3090         qvalue of 0 means "not acceptable.")</t>
3091    </x:lt>
3092    <x:lt>
3093      <t>If multiple transfer-codings are acceptable, then the
3094         acceptable transfer-coding with the highest non-zero qvalue is
3095         preferred.  The "chunked" transfer-coding always has a qvalue
3096         of 1.</t>
3097    </x:lt>
3098  </list>
3101   If the TE field-value is empty or if no TE field is present, the only
[457]3102   transfer-coding is "chunked". A message with no transfer-coding is
[8]3103   always acceptable.
3107<section title="Trailer" anchor="header.trailer">
3108  <iref primary="true" item="Trailer header" x:for-anchor=""/>
3109  <iref primary="true" item="Headers" subitem="Trailer" x:for-anchor=""/>
[229]3110  <x:anchor-alias value="Trailer"/>
[354]3111  <x:anchor-alias value="Trailer-v"/>
[697]3113   The "Trailer" general-header field indicates that the given set of
[8]3114   header fields is present in the trailer of a message encoded with
3115   chunked transfer-coding.
[354]3117<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Trailer"/><iref primary="true" item="Grammar" subitem="Trailer-v"/>
[366]3118  <x:ref>Trailer</x:ref>   = "Trailer" ":" <x:ref>OWS</x:ref> <x:ref>Trailer-v</x:ref>
[354]3119  <x:ref>Trailer-v</x:ref> = 1#<x:ref>field-name</x:ref>
3122   An HTTP/1.1 message &SHOULD; include a Trailer header field in a
3123   message using chunked transfer-coding with a non-empty trailer. Doing
3124   so allows the recipient to know which header fields to expect in the
3125   trailer.
3128   If no Trailer header field is present, the trailer &SHOULD-NOT;  include
[673]3129   any header fields. See <xref target="chunked.encoding"/> for restrictions on the use of
[8]3130   trailer fields in a "chunked" transfer-coding.
3133   Message header fields listed in the Trailer header field &MUST-NOT;
3134   include the following header fields:
3135  <list style="symbols">
3136    <t>Transfer-Encoding</t>
3137    <t>Content-Length</t>
3138    <t>Trailer</t>
3139  </list>
3143<section title="Transfer-Encoding" anchor="header.transfer-encoding">
3144  <iref primary="true" item="Transfer-Encoding header" x:for-anchor=""/>
3145  <iref primary="true" item="Headers" subitem="Transfer-Encoding" x:for-anchor=""/>
[229]3146  <x:anchor-alias value="Transfer-Encoding"/>
[354]3147  <x:anchor-alias value="Transfer-Encoding-v"/>
[698]3149   The "Transfer-Encoding" general-header field indicates what transfer-codings
3150   (if any) have been applied to the message body. It differs from
3151   Content-Encoding (&content-codings;) in that transfer-codings are a property
3152   of the message (and therefore are removed by intermediaries), whereas
3153   content-codings are not.
[354]3155<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/><iref primary="true" item="Grammar" subitem="Transfer-Encoding-v"/>
[376]3156  <x:ref>Transfer-Encoding</x:ref>   = "Transfer-Encoding" ":" <x:ref>OWS</x:ref>
3157                        <x:ref>Transfer-Encoding-v</x:ref>
[354]3158  <x:ref>Transfer-Encoding-v</x:ref> = 1#<x:ref>transfer-coding</x:ref>
3161   Transfer-codings are defined in <xref target="transfer.codings"/>. An example is:
3163<figure><artwork type="example">
3164  Transfer-Encoding: chunked
3167   If multiple encodings have been applied to an entity, the transfer-codings
3168   &MUST; be listed in the order in which they were applied.
3169   Additional information about the encoding parameters &MAY; be provided
3170   by other entity-header fields not defined by this specification.
3173   Many older HTTP/1.0 applications do not understand the Transfer-Encoding
3174   header.
3178<section title="Upgrade" anchor="header.upgrade">
3179  <iref primary="true" item="Upgrade header" x:for-anchor=""/>
3180  <iref primary="true" item="Headers" subitem="Upgrade" x:for-anchor=""/>
[229]3181  <x:anchor-alias value="Upgrade"/>
[354]3182  <x:anchor-alias value="Upgrade-v"/>
[697]3184   The "Upgrade" general-header field allows the client to specify what
[698]3185   additional communication protocols it would like to use, if the server
3186   chooses to switch protocols. Additionally, the server &MUST; use the Upgrade
3187   header field within a 101 (Switching Protocols) response to indicate which
3188   protocol(s) are being switched to.
[354]3190<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Upgrade"/><iref primary="true" item="Grammar" subitem="Upgrade-v"/>
[366]3191  <x:ref>Upgrade</x:ref>   = "Upgrade" ":" <x:ref>OWS</x:ref> <x:ref>Upgrade-v</x:ref>
[354]3192  <x:ref>Upgrade-v</x:ref> = 1#<x:ref>product</x:ref>
3195   For example,
3197<figure><artwork type="example">
[354]3198  Upgrade: HTTP/2.0, SHTTP/1.3, IRC/6.9, RTA/x11
3201   The Upgrade header field is intended to provide a simple mechanism
3202   for transition from HTTP/1.1 to some other, incompatible protocol. It
3203   does so by allowing the client to advertise its desire to use another
3204   protocol, such as a later version of HTTP with a higher major version
3205   number, even though the current request has been made using HTTP/1.1.
3206   This eases the difficult transition between incompatible protocols by
3207   allowing the client to initiate a request in the more commonly
3208   supported protocol while indicating to the server that it would like
3209   to use a "better" protocol if available (where "better" is determined
3210   by the server, possibly according to the nature of the method and/or
3211   resource being requested).
3214   The Upgrade header field only applies to switching application-layer
3215   protocols upon the existing transport-layer connection. Upgrade
3216   cannot be used to insist on a protocol change; its acceptance and use
3217   by the server is optional. The capabilities and nature of the
3218   application-layer communication after the protocol change is entirely
3219   dependent upon the new protocol chosen, although the first action
3220   after changing the protocol &MUST; be a response to the initial HTTP
3221   request containing the Upgrade header field.
3224   The Upgrade header field only applies to the immediate connection.
3225   Therefore, the upgrade keyword &MUST; be supplied within a Connection
3226   header field (<xref target="header.connection"/>) whenever Upgrade is present in an
3227   HTTP/1.1 message.
3230   The Upgrade header field cannot be used to indicate a switch to a
3231   protocol on a different connection. For that purpose, it is more
3232   appropriate to use a 301, 302, 303, or 305 redirection response.
3235   This specification only defines the protocol name "HTTP" for use by
3236   the family of Hypertext Transfer Protocols, as defined by the HTTP
3237   version rules of <xref target="http.version"/> and future updates to this
[684]3238   specification. Additional tokens can be registered with IANA using the
3239   registration procedure defined below. 
3242<section title="Upgrade Token Registry" anchor="upgrade.token.registry">
3244   The HTTP Upgrade Token Registry defines the name space for product
3245   tokens used to identify protocols in the Upgrade header field.
3246   Each registered token should be associated with one or a set of
3247   specifications, and with contact information.
3250   Registrations should be allowed on a First Come First Served basis as
3251   described in <xref target="RFC5226" x:sec="4.1" x:fmt="of"/>. These
3252   specifications need not be IETF documents or be subject to IESG review, but
3253   should obey the following rules:
3254  <list style="numbers">
3255    <t>A token, once registered, stays registered forever.</t>
3256    <t>The registration &MUST; name a responsible party for the
3257       registration.</t>
3258    <t>The registration &MUST; name a point of contact.</t>
3259    <t>The registration &MAY; name the documentation required for the
3260       token.</t>
3261    <t>The responsible party &MAY; change the registration at any time.
3262       The IANA will keep a record of all such changes, and make them
3263       available upon request.</t>
3264    <t>The responsible party for the first registration of a "product"
3265       token &MUST; approve later registrations of a "version" token
3266       together with that "product" token before they can be registered.</t>
3267    <t>If absolutely required, the IESG &MAY; reassign the responsibility
3268       for a token. This will normally only be used in the case when a
3269       responsible party cannot be contacted.</t>
3270  </list>
3273   It is not required that specifications for upgrade tokens be made
3274   publicly available, but the contact information for the registration
3275   should be.
[8]3282<section title="Via" anchor="header.via">
3283  <iref primary="true" item="Via header" x:for-anchor=""/>
3284  <iref primary="true" item="Headers" subitem="Via" x:for-anchor=""/>
[229]3285  <x:anchor-alias value="protocol-name"/>
3286  <x:anchor-alias value="protocol-version"/>
3287  <x:anchor-alias value="pseudonym"/>
3288  <x:anchor-alias value="received-by"/>
3289  <x:anchor-alias value="received-protocol"/>
3290  <x:anchor-alias value="Via"/>
[354]3291  <x:anchor-alias value="Via-v"/>
[697]3293   The "Via" general-header field &MUST; be used by gateways and proxies to
[8]3294   indicate the intermediate protocols and recipients between the user
3295   agent and the server on requests, and between the origin server and
[257]3296   the client on responses. It is analogous to the "Received" field defined in
[327]3297   <xref target="RFC5322" x:fmt="of" x:sec="3.6.7"/> and is intended to be used for tracking message forwards,
[8]3298   avoiding request loops, and identifying the protocol capabilities of
3299   all senders along the request/response chain.
[354]3301<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Via"/><iref primary="true" item="Grammar" subitem="Via-v"/><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"/>
[366]3302  <x:ref>Via</x:ref>               = "Via" ":" <x:ref>OWS</x:ref> <x:ref>Via-v</x:ref>
[376]3303  <x:ref>Via-v</x:ref>             = 1#( <x:ref>received-protocol</x:ref> <x:ref>RWS</x:ref> <x:ref>received-by</x:ref>
3304                          [ <x:ref>RWS</x:ref> <x:ref>comment</x:ref> ] )
[229]3305  <x:ref>received-protocol</x:ref> = [ <x:ref>protocol-name</x:ref> "/" ] <x:ref>protocol-version</x:ref>
3306  <x:ref>protocol-name</x:ref>     = <x:ref>token</x:ref>
3307  <x:ref>protocol-version</x:ref>  = <x:ref>token</x:ref>
[334]3308  <x:ref>received-by</x:ref>       = ( <x:ref>uri-host</x:ref> [ ":" <x:ref>port</x:ref> ] ) / <x:ref>pseudonym</x:ref>
[229]3309  <x:ref>pseudonym</x:ref>         = <x:ref>token</x:ref>
3312   The received-protocol indicates the protocol version of the message
3313   received by the server or client along each segment of the
3314   request/response chain. The received-protocol version is appended to
3315   the Via field value when the message is forwarded so that information
3316   about the protocol capabilities of upstream applications remains
3317   visible to all recipients.
3320   The protocol-name is optional if and only if it would be "HTTP". The
3321   received-by field is normally the host and optional port number of a
3322   recipient server or client that subsequently forwarded the message.
3323   However, if the real host is considered to be sensitive information,
3324   it &MAY; be replaced by a pseudonym. If the port is not given, it &MAY;
3325   be assumed to be the default port of the received-protocol.
[761]3328   Multiple Via field values represent each proxy or gateway that has
[8]3329   forwarded the message. Each recipient &MUST; append its information
3330   such that the end result is ordered according to the sequence of
3331   forwarding applications.
3334   Comments &MAY; be used in the Via header field to identify the software
3335   of the recipient proxy or gateway, analogous to the User-Agent and
3336   Server header fields. However, all comments in the Via field are
3337   optional and &MAY; be removed by any recipient prior to forwarding the
3338   message.
3341   For example, a request message could be sent from an HTTP/1.0 user
3342   agent to an internal proxy code-named "fred", which uses HTTP/1.1 to
[90]3343   forward the request to a public proxy at, which completes
3344   the request by forwarding it to the origin server at
3345   The request received by would then have the following
[8]3346   Via header field:
3348<figure><artwork type="example">
[354]3349  Via: 1.0 fred, 1.1 (Apache/1.1)
3352   Proxies and gateways used as a portal through a network firewall
3353   &SHOULD-NOT;, by default, forward the names and ports of hosts within
3354   the firewall region. This information &SHOULD; only be propagated if
3355   explicitly enabled. If not enabled, the received-by host of any host
3356   behind the firewall &SHOULD; be replaced by an appropriate pseudonym
3357   for that host.
3360   For organizations that have strong privacy requirements for hiding
3361   internal structures, a proxy &MAY; combine an ordered subsequence of
3362   Via header field entries with identical received-protocol values into
3363   a single such entry. For example,
3365<figure><artwork type="example">
[354]3366  Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
[803]3369  could be collapsed to
3371<figure><artwork type="example">
[354]3372  Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
3375   Applications &SHOULD-NOT;  combine multiple entries unless they are all
3376   under the same organizational control and the hosts have already been
3377   replaced by pseudonyms. Applications &MUST-NOT; combine entries which
3378   have different received-protocol values.
[29]3384<section title="IANA Considerations" anchor="IANA.considerations">
[253]3386<section title="Message Header Registration" anchor="message.header.registration">
[290]3388   The Message Header Registry located at <eref target=""/> should be updated
3389   with the permanent registrations below (see <xref target="RFC3864"/>):
[680]3391<?BEGININC p1-messaging.iana-headers ?>
[290]3392<!--AUTOGENERATED FROM extract-header-defs.xslt, do not edit manually-->
3393<texttable align="left" suppress-title="true" anchor="iana.header.registration.table">
[253]3394   <ttcol>Header Field Name</ttcol>
3395   <ttcol>Protocol</ttcol>
3396   <ttcol>Status</ttcol>
3397   <ttcol>Reference</ttcol>
3399   <c>Connection</c>
3400   <c>http</c>
3401   <c>standard</c>
3402   <c>
3403      <xref target="header.connection"/>
3404   </c>
3405   <c>Content-Length</c>
3406   <c>http</c>
3407   <c>standard</c>
3408   <c>
3409      <xref target="header.content-length"/>
3410   </c>
3411   <c>Date</c>
3412   <c>http</c>
3413   <c>standard</c>
3414   <c>
3415      <xref target=""/>
3416   </c>
3417   <c>Host</c>
3418   <c>http</c>
3419   <c>standard</c>
3420   <c>
3421      <xref target=""/>
3422   </c>
3423   <c>TE</c>
3424   <c>http</c>
3425   <c>standard</c>
3426   <c>
3427      <xref target="header.te"/>
3428   </c>
3429   <c>Trailer</c>
3430   <c>http</c>
3431   <c>standard</c>
3432   <c>
3433      <xref target="header.trailer"/>
3434   </c>
3435   <c>Transfer-Encoding</c>
3436   <c>http</c>
3437   <c>standard</c>
3438   <c>
3439      <xref target="header.transfer-encoding"/>
3440   </c>
3441   <c>Upgrade</c>
3442   <c>http</c>
3443   <c>standard</c>
3444   <c>
3445      <xref target="header.upgrade"/>
3446   </c>
3447   <c>Via</c>
3448   <c>http</c>
3449   <c>standard</c>
3450   <c>
3451      <xref target="header.via"/>
3452   </c>
[680]3455<?ENDINC p1-messaging.iana-headers ?>
[290]3457   The change controller is: "IETF ( - Internet Engineering Task Force".
3461<section title="URI Scheme Registration" anchor="uri.scheme.registration">
[646]3463   The entries for the "http" and "https" URI Schemes in the registry located at
[307]3464   <eref target=""/>
[646]3465   should be updated to point to Sections <xref target="http.uri" format="counter"/>
3466   and <xref target="https.uri" format="counter"/> of this document
[307]3467   (see <xref target="RFC4395"/>).
[296]3471<section title="Internet Media Type Registrations" anchor="">
3473   This document serves as the specification for the Internet media types
3474   "message/http" and "application/http". The following is to be registered with
3475   IANA (see <xref target="RFC4288"/>).
3477<section title="Internet Media Type message/http" anchor="">
3478<iref item="Media Type" subitem="message/http" primary="true"/>
3479<iref item="message/http Media Type" primary="true"/>
3481   The message/http type can be used to enclose a single HTTP request or
3482   response message, provided that it obeys the MIME restrictions for all
3483   "message" types regarding line length and encodings.
3486  <list style="hanging" x:indent="12em">
3487    <t hangText="Type name:">
3488      message
3489    </t>
3490    <t hangText="Subtype name:">
3491      http
3492    </t>
3493    <t hangText="Required parameters:">
3494      none
3495    </t>
3496    <t hangText="Optional parameters:">
3497      version, msgtype
3498      <list style="hanging">
3499        <t hangText="version:">
3500          The HTTP-Version number of the enclosed message
3501          (e.g., "1.1"). If not present, the version can be
3502          determined from the first line of the body.
3503        </t>
3504        <t hangText="msgtype:">
3505          The message type -- "request" or "response". If not
3506          present, the type can be determined from the first
3507          line of the body.
3508        </t>
3509      </list>
3510    </t>
3511    <t hangText="Encoding considerations:">
3512      only "7bit", "8bit", or "binary" are permitted
3513    </t>
3514    <t hangText="Security considerations:">
3515      none
3516    </t>
3517    <t hangText="Interoperability considerations:">
3518      none
3519    </t>
3520    <t hangText="Published specification:">
3521      This specification (see <xref target=""/>).
3522    </t>
3523    <t hangText="Applications that use this media type:">
3524    </t>
3525    <t hangText="Additional information:">
3526      <list style="hanging">
3527        <t hangText="Magic number(s):">none</t>
3528        <t hangText="File extension(s):">none</t>
3529        <t hangText="Macintosh file type code(s):">none</t>
3530      </list>
3531    </t>
3532    <t hangText="Person and email address to contact for further information:">
3533      See Authors Section.
3534    </t>
[609]3535    <t hangText="Intended usage:">
3536      COMMON
[296]3537    </t>
[609]3538    <t hangText="Restrictions on usage:">
3539      none
[296]3540    </t>
3541    <t hangText="Author/Change controller:">
3542      IESG
3543    </t>
3544  </list>
[296]3547<section title="Internet Media Type application/http" anchor="">
3548<iref item="Media Type" subitem="application/http" primary="true"/>
3549<iref item="application/http Media Type" primary="true"/>
3551   The application/http type can be used to enclose a pipeline of one or more
3552   HTTP request or response messages (not intermixed).
3555  <list style="hanging" x:indent="12em">
3556    <t hangText="Type name:">
3557      application
3558    </t>
3559    <t hangText="Subtype name:">
3560      http
3561    </t>
3562    <t hangText="Required parameters:">
3563      none
3564    </t>
3565    <t hangText="Optional parameters:">
3566      version, msgtype
3567      <list style="hanging">
3568        <t hangText="version:">
3569          The HTTP-Version number of the enclosed messages
3570          (e.g., "1.1"). If not present, the version can be
3571          determined from the first line of the body.
3572        </t>
3573        <t hangText="msgtype:">
3574          The message type -- "request" or "response". If not
3575          present, the type can be determined from the first
3576          line of the body.
3577        </t>
3578      </list>
3579    </t>
3580    <t hangText="Encoding considerations:">
3581      HTTP messages enclosed by this type
3582      are in "binary" format; use of an appropriate
3583      Content-Transfer-Encoding is required when
3584      transmitted via E-mail.
3585    </t>
3586    <t hangText="Security considerations:">
3587      none
3588    </t>
3589    <t hangText="Interoperability considerations:">
3590      none
3591    </t>
3592    <t hangText="Published specification:">
3593      This specification (see <xref target=""/>).
3594    </t>
3595    <t hangText="Applications that use this media type:">
3596    </t>
3597    <t hangText="Additional information:">
3598      <list style="hanging">
3599        <t hangText="Magic number(s):">none</t>
3600        <t hangText="File extension(s):">none</t>
3601        <t hangText="Macintosh file type code(s):">none</t>
3602      </list>
3603    </t>
3604    <t hangText="Person and email address to contact for further information:">
3605      See Authors Section.
3606    </t>
[609]3607    <t hangText="Intended usage:">
3608      COMMON
[296]3609    </t>
[609]3610    <t hangText="Restrictions on usage:">
3611      none
[296]3612    </t>
3613    <t hangText="Author/Change controller:">
3614      IESG
3615    </t>
3616  </list>
[650]3621<section title="Transfer Coding Registry" anchor="transfer.coding.registration">
[673]3623   The registration procedure for HTTP Transfer Codings is now defined by
3624   <xref target="transfer.coding.registry"/> of this document.
3627   The HTTP Transfer Codings Registry located at <eref target=""/>
[673]3628   should be updated with the registrations below:
3630<texttable align="left" suppress-title="true" anchor="iana.transfer.coding.registration.table">
[670]3631   <ttcol>Name</ttcol>
[650]3632   <ttcol>Description</ttcol>
3633   <ttcol>Reference</ttcol>
[673]3634   <c>chunked</c>
[650]3635   <c>Transfer in a series of chunks</c>
3636   <c>
[673]3637      <xref target="chunked.encoding"/>
[650]3638   </c>
[673]3639   <c>compress</c>
3640   <c>UNIX "compress" program method</c>
3641   <c>
3642      <xref target="compress.coding"/>
3643   </c>
3644   <c>deflate</c>
[805]3645   <c>"deflate" compression mechanism (<xref target="RFC1951"/>) used inside
3646   the "zlib" data format (<xref target="RFC1950"/>)
3647   </c>
[673]3648   <c>
3649      <xref target="deflate.coding"/>
3650   </c>
3651   <c>gzip</c>
3652   <c>Same as GNU zip <xref target="RFC1952"/></c>
3653   <c>
3654      <xref target="gzip.coding"/>
3655   </c>
[684]3659<section title="Upgrade Token Registration" anchor="upgrade.token.registration">
3661   The registration procedure for HTTP Upgrade Tokens -- previously defined
3662   in <xref target="RFC2817" x:fmt="of" x:sec="7.2"/> -- is now defined
3663   by <xref target="upgrade.token.registry"/> of this document.
3666   The HTTP Status Code Registry located at <eref target=""/>
3667   should be updated with the registration below:
3669<texttable align="left" suppress-title="true">
3670   <ttcol>Value</ttcol>
3671   <ttcol>Description</ttcol>
3672   <ttcol>Reference</ttcol>
3674   <c>HTTP</c>
3675   <c>Hypertext Transfer Protocol</c> 
3676   <c><xref target="http.version"/> of this specification</c>
3677<!-- IANA should add this without our instructions; emailed on June 05, 2009