source: draft-ietf-httpbis/latest/p3-payload.xml @ 206

Last change on this file since 206 was 206, checked in by julian.reschke@…, 12 years ago

Work on referencing ABNF rules adopted from other parts (done for P2 and P3); relates to #36.

  • Property svn:eol-style set to native
File size: 108.1 KB
Line 
1<?xml version="1.0" encoding="utf-8"?>
2<?xml-stylesheet type='text/xsl' href='../myxml2rfc.xslt'?>
3<!DOCTYPE rfc [
4  <!ENTITY MAY "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MAY</bcp14>">
5  <!ENTITY MUST "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST</bcp14>">
6  <!ENTITY MUST-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>MUST NOT</bcp14>">
7  <!ENTITY OPTIONAL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>OPTIONAL</bcp14>">
8  <!ENTITY RECOMMENDED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>RECOMMENDED</bcp14>">
9  <!ENTITY REQUIRED "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>REQUIRED</bcp14>">
10  <!ENTITY SHALL "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL</bcp14>">
11  <!ENTITY SHALL-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHALL NOT</bcp14>">
12  <!ENTITY SHOULD "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD</bcp14>">
13  <!ENTITY SHOULD-NOT "<bcp14 xmlns='http://purl.org/net/xml2rfc/ext'>SHOULD NOT</bcp14>">
14  <!ENTITY ID-VERSION "latest">
15  <!ENTITY ID-MONTH "February">
16  <!ENTITY ID-YEAR "2008">
17  <!ENTITY notation-abnf            "<xref target='Part1' x:rel='#notation.abnf' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
18  <!ENTITY basic-rules              "<xref target='Part1' x:rel='#basic.rules' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
19  <!ENTITY caching-neg-resp         "<xref target='Part6' x:rel='#caching.negotiated.responses' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
20  <!ENTITY header-transfer-encoding "<xref target='Part1' x:rel='#header.transfer-encoding' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
21  <!ENTITY header-allow             "<xref target='Part2' x:rel='#header.allow' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
22  <!ENTITY header-content-length    "<xref target='Part1' x:rel='#header.content-length' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
23  <!ENTITY header-content-range     "<xref target='Part5' x:rel='#header.content-range' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
24  <!ENTITY header-expires           "<xref target='Part6' x:rel='#header.expires' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
25  <!ENTITY header-last-modified     "<xref target='Part4' x:rel='#header.last-modified' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
26  <!ENTITY header-user-agent        "<xref target='Part2' x:rel='#header.user-agent' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
27  <!ENTITY header-vary              "<xref target='Part6' x:rel='#header.vary' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
28  <!ENTITY message-body             "<xref target='Part1' x:rel='#message.body' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
29  <!ENTITY message-length           "<xref target='Part1' x:rel='#message.length' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
30  <!ENTITY message-headers          "<xref target='Part1' x:rel='#message.headers' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
31  <!ENTITY general-syntax           "<xref target='Part1' x:rel='#general.syntax' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
32  <!ENTITY multipart-byteranges     "<xref target='Part5' x:rel='#internet.media.type.multipart.byteranges' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
33]>
34<?rfc toc="yes" ?>
35<?rfc symrefs="yes" ?>
36<?rfc sortrefs="yes" ?>
37<?rfc compact="yes"?>
38<?rfc subcompact="no" ?>
39<?rfc linkmailto="no" ?>
40<?rfc editing="no" ?>
41<?rfc comments="yes"?>
42<?rfc inline="yes"?>
43<?rfc-ext allow-markup-in-artwork="yes" ?>
44<?rfc-ext include-references-in-index="yes" ?>
45<rfc obsoletes="2616" category="std"
46     ipr="full3978" docName="draft-ietf-httpbis-p3-payload-&ID-VERSION;"
47     xmlns:x='http://purl.org/net/xml2rfc/ext'>
48<front>
49
50  <title abbrev="HTTP/1.1, Part 3">HTTP/1.1, part 3: Message Payload and Content Negotiation</title>
51
52  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
53    <organization abbrev="Day Software">Day Software</organization>
54    <address>
55      <postal>
56        <street>23 Corporate Plaza DR, Suite 280</street>
57        <city>Newport Beach</city>
58        <region>CA</region>
59        <code>92660</code>
60        <country>USA</country>
61      </postal>
62      <phone>+1-949-706-5300</phone>
63      <facsimile>+1-949-706-5305</facsimile>
64      <email>fielding@gbiv.com</email>
65      <uri>http://roy.gbiv.com/</uri>
66    </address>
67  </author>
68
69  <author initials="J." surname="Gettys" fullname="Jim Gettys">
70    <organization>One Laptop per Child</organization>
71    <address>
72      <postal>
73        <street>21 Oak Knoll Road</street>
74        <city>Carlisle</city>
75        <region>MA</region>
76        <code>01741</code>
77        <country>USA</country>
78      </postal>
79      <email>jg@laptop.org</email>
80      <uri>http://www.laptop.org/</uri>
81    </address>
82  </author>
83 
84  <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
85    <organization abbrev="HP">Hewlett-Packard Company</organization>
86    <address>
87      <postal>
88        <street>HP Labs, Large Scale Systems Group</street>
89        <street>1501 Page Mill Road, MS 1177</street>
90        <city>Palo Alto</city>
91        <region>CA</region>
92        <code>94304</code>
93        <country>USA</country>
94      </postal>
95      <email>JeffMogul@acm.org</email>
96    </address>
97  </author>
98
99  <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
100    <organization abbrev="Microsoft">Microsoft Corporation</organization>
101    <address>
102      <postal>
103        <street>1 Microsoft Way</street>
104        <city>Redmond</city>
105        <region>WA</region>
106        <code>98052</code>
107        <country>USA</country>
108      </postal>
109      <email>henrikn@microsoft.com</email>
110    </address>
111  </author>
112
113  <author initials="L." surname="Masinter" fullname="Larry Masinter">
114    <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
115    <address>
116      <postal>
117        <street>345 Park Ave</street>
118        <city>San Jose</city>
119        <region>CA</region>
120        <code>95110</code>
121        <country>USA</country>
122      </postal>
123      <email>LMM@acm.org</email>
124      <uri>http://larry.masinter.net/</uri>
125    </address>
126  </author>
127 
128  <author initials="P." surname="Leach" fullname="Paul J. Leach">
129    <organization abbrev="Microsoft">Microsoft Corporation</organization>
130    <address>
131      <postal>
132        <street>1 Microsoft Way</street>
133        <city>Redmond</city>
134        <region>WA</region>
135        <code>98052</code>
136      </postal>
137      <email>paulle@microsoft.com</email>
138    </address>
139  </author>
140   
141  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
142    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
143    <address>
144      <postal>
145        <street>MIT Computer Science and Artificial Intelligence Laboratory</street>
146        <street>The Stata Center, Building 32</street>
147        <street>32 Vassar Street</street>
148        <city>Cambridge</city>
149        <region>MA</region>
150        <code>02139</code>
151        <country>USA</country>
152      </postal>
153      <email>timbl@w3.org</email>
154      <uri>http://www.w3.org/People/Berners-Lee/</uri>
155    </address>
156  </author>
157
158  <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
159    <organization abbrev="W3C">World Wide Web Consortium</organization>
160    <address>
161      <postal>
162        <street>W3C / ERCIM</street>
163        <street>2004, rte des Lucioles</street>
164        <city>Sophia-Antipolis</city>
165        <region>AM</region>
166        <code>06902</code>
167        <country>France</country>
168      </postal>
169      <email>ylafon@w3.org</email>
170      <uri>http://www.raubacapeu.net/people/yves/</uri>
171    </address>
172  </author>
173
174  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
175    <organization abbrev="greenbytes">greenbytes GmbH</organization>
176    <address>
177      <postal>
178        <street>Hafenweg 16</street>
179        <city>Muenster</city><region>NW</region><code>48155</code>
180        <country>Germany</country>
181      </postal>
182      <phone>+49 251 2807760</phone>   
183      <facsimile>+49 251 2807761</facsimile>   
184      <email>julian.reschke@greenbytes.de</email>       
185      <uri>http://greenbytes.de/tech/webdav/</uri>     
186    </address>
187  </author>
188
189  <date month="&ID-MONTH;" year="&ID-YEAR;"/>
190
191<abstract>
192<t>
193   The Hypertext Transfer Protocol (HTTP) is an application-level
194   protocol for distributed, collaborative, hypermedia information
195   systems. HTTP has been in use by the World Wide Web global information
196   initiative since 1990. This document is Part 3 of the seven-part specification
197   that defines the protocol referred to as "HTTP/1.1" and, taken together,
198   obsoletes RFC 2616.  Part 3 defines HTTP message content,
199   metadata, and content negotiation.
200</t>
201</abstract>
202
203<note title="Editorial Note (To be removed by RFC Editor)">
204  <t>
205    Discussion of this draft should take place on the HTTPBIS working group
206    mailing list (ietf-http-wg@w3.org). The current issues list is
207    at <eref target="http://www.tools.ietf.org/wg/httpbis/trac/report/11"/>
208    and related documents (including fancy diffs) can be found at
209    <eref target="http://www.tools.ietf.org/wg/httpbis/"/>.
210  </t>
211  <t>
212    This draft incorporates those issue resolutions that were either
213    collected in the original RFC2616 errata list (<eref target="http://purl.org/NET/http-errata"/>),
214    or which were agreed upon on the mailing list between October 2006 and
215    November 2007 (as published in "draft-lafon-rfc2616bis-03").
216  </t>
217</note>
218</front>
219<middle>
220<section title="Introduction" anchor="introduction">
221<t>
222   This document defines HTTP/1.1 message payloads (a.k.a., content), the
223   associated metadata header fields that define how the payload is intended
224   to be interpreted by a recipient, the request header fields that
225   may influence content selection, and the various selection algorithms
226   that are collectively referred to as HTTP content negotiation.
227</t>
228<t>
229   This document is currently disorganized in order to minimize the changes
230   between drafts and enable reviewers to see the smaller errata changes.
231   The next draft will reorganize the sections to better reflect the content.
232   In particular, the sections on entities will be renamed payload and moved
233   to the first half of the document, while the sections on content negotiation
234   and associated request header fields will be moved to the second half.  The
235   current mess reflects how widely dispersed these topics and associated
236   requirements had become in <xref target="RFC2616"/>.
237</t>
238
239<section title="Requirements" anchor="intro.requirements">
240<t>
241   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
242   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
243   document are to be interpreted as described in <xref target="RFC2119"/>.
244</t>
245<t>
246   An implementation is not compliant if it fails to satisfy one or more
247   of the &MUST; or &REQUIRED; level requirements for the protocols it
248   implements. An implementation that satisfies all the &MUST; or &REQUIRED;
249   level and all the &SHOULD; level requirements for its protocols is said
250   to be "unconditionally compliant"; one that satisfies all the &MUST;
251   level requirements but not all the &SHOULD; level requirements for its
252   protocols is said to be "conditionally compliant."
253</t>
254</section>
255</section>
256
257<section title="Notational Conventions and Generic Grammar" anchor="notation">
258<t>
259  This specification uses the ABNF syntax defined in &notation-abnf; and
260  the core rules defined in &basic-rules;:
261  <cref anchor="abnf.dep">ABNF syntax and basic rules will be adopted from RFC 5234, see
262  <eref target="http://tools.ietf.org/wg/httpbis/trac/ticket/36"/>.</cref>
263</t>
264<figure><artwork type="abnf2616">
265  ALPHA          = &lt;ALPHA, defined in &basic-rules;&gt;
266  DIGIT          = &lt;DIGIT, defined in &basic-rules;&gt;
267  OCTET          = &lt;OCTET, defined in &basic-rules;&gt;
268</artwork></figure>
269<figure><artwork type="abnf2616">
270  quoted-string  = &lt;quoted-string, defined in &basic-rules;&gt;
271  token          = &lt;token, defined in &basic-rules;&gt;
272</artwork></figure>
273<t>
274  The ABNF rules below are defined in other parts:
275</t>
276<figure><artwork type="abnf2616">
277  absoluteURI    = &lt;absoluteURI, defined in &general-syntax;&gt;
278  Allow          = &lt;Allow, defined in &header-allow;&gt;
279  Content-Length = &lt;Content-Length, defined in &header-content-length;&gt;
280  Content-Range  = &lt;Content-Range, defined in &header-content-range;&gt;
281  Expires        = &lt;Expires, defined in &header-expires;&gt;
282  Last-Modified  = &lt;Last-Modified, defined in &header-last-modified;&gt;
283  message-header = &lt;message-header, defined in &message-headers;&gt;
284  relativeURI    = &lt;relativeURI, defined in &general-syntax;&gt;
285</artwork></figure>
286</section>
287
288<section title="Protocol Parameters" anchor="protocol.parameters">
289
290<section title="Character Sets" anchor="character.sets">
291<t>
292   HTTP uses the same definition of the term "character set" as that
293   described for MIME:
294</t>
295<t>
296   The term "character set" is used in this document to refer to a
297   method used with one or more tables to convert a sequence of octets
298   into a sequence of characters. Note that unconditional conversion in
299   the other direction is not required, in that not all characters may
300   be available in a given character set and a character set may provide
301   more than one sequence of octets to represent a particular character.
302   This definition is intended to allow various kinds of character
303   encoding, from simple single-table mappings such as US-ASCII to
304   complex table switching methods such as those that use ISO-2022's
305   techniques. However, the definition associated with a MIME character
306   set name &MUST; fully specify the mapping to be performed from octets
307   to characters. In particular, use of external profiling information
308   to determine the exact mapping is not permitted.
309</t>
310<t><list><t>
311      <x:h>Note:</x:h> This use of the term "character set" is more commonly
312      referred to as a "character encoding." However, since HTTP and
313      MIME share the same registry, it is important that the terminology
314      also be shared.
315</t></list></t>
316<t>
317   HTTP character sets are identified by case-insensitive tokens. The
318   complete set of tokens is defined by the IANA Character Set registry
319   (<eref target="http://www.iana.org/assignments/character-sets"/>).
320</t>
321<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="charset"/>
322  charset = token
323</artwork></figure>
324<t>
325   Although HTTP allows an arbitrary token to be used as a charset
326   value, any token that has a predefined value within the IANA
327   Character Set registry &MUST; represent the character set defined
328   by that registry. Applications &SHOULD; limit their use of character
329   sets to those defined by the IANA registry.
330</t>
331<t>
332   HTTP uses charset in two contexts: within an Accept-Charset request
333   header (in which the charset value is an unquoted token) and as the
334   value of a parameter in a Content-Type header (within a request or
335   response), in which case the parameter value of the charset parameter
336   may be quoted.
337</t>
338<t>
339   Implementors should be aware of IETF character set requirements <xref target="RFC3629"/>
340   <xref target="RFC2277"/>.
341</t>
342
343<section title="Missing Charset" anchor="missing.charset">
344<t>
345   Some HTTP/1.0 software has interpreted a Content-Type header without
346   charset parameter incorrectly to mean "recipient should guess."
347   Senders wishing to defeat this behavior &MAY; include a charset
348   parameter even when the charset is ISO-8859-1 (<xref target="ISO-8859-1"/>) and &SHOULD; do so when
349   it is known that it will not confuse the recipient.
350</t>
351<t>
352   Unfortunately, some older HTTP/1.0 clients did not deal properly with
353   an explicit charset parameter. HTTP/1.1 recipients &MUST; respect the
354   charset label provided by the sender; and those user agents that have
355   a provision to "guess" a charset &MUST; use the charset from the
356   content-type field if they support that charset, rather than the
357   recipient's preference, when initially displaying a document. See
358   <xref target="canonicalization.and.text.defaults"/>.
359</t>
360</section>
361</section>
362
363<section title="Content Codings" anchor="content.codings">
364<t>
365   Content coding values indicate an encoding transformation that has
366   been or can be applied to an entity. Content codings are primarily
367   used to allow a document to be compressed or otherwise usefully
368   transformed without losing the identity of its underlying media type
369   and without loss of information. Frequently, the entity is stored in
370   coded form, transmitted directly, and only decoded by the recipient.
371</t>
372<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="content-coding"/>
373  content-coding   = token
374</artwork></figure>
375<t>
376   All content-coding values are case-insensitive. HTTP/1.1 uses
377   content-coding values in the Accept-Encoding (<xref target="header.accept-encoding"/>) and
378   Content-Encoding (<xref target="header.content-encoding"/>) header fields. Although the value
379   describes the content-coding, what is more important is that it
380   indicates what decoding mechanism will be required to remove the
381   encoding.
382</t>
383<t>
384   The Internet Assigned Numbers Authority (IANA) acts as a registry for
385   content-coding value tokens. Initially, the registry contains the
386   following tokens:
387</t>
388<t>
389   gzip<iref item="gzip"/>
390  <list>
391    <t>
392        An encoding format produced by the file compression program
393        "gzip" (GNU zip) as described in <xref target="RFC1952"/>. This format is a
394        Lempel-Ziv coding (LZ77) with a 32 bit CRC.
395    </t>
396  </list>
397</t>
398<t>
399   compress<iref item="compress"/>
400  <list><t>
401        The encoding format produced by the common UNIX file compression
402        program "compress". This format is an adaptive Lempel-Ziv-Welch
403        coding (LZW).
404</t><t>
405        Use of program names for the identification of encoding formats
406        is not desirable and is discouraged for future encodings. Their
407        use here is representative of historical practice, not good
408        design. For compatibility with previous implementations of HTTP,
409        applications &SHOULD; consider "x-gzip" and "x-compress" to be
410        equivalent to "gzip" and "compress" respectively.
411  </t></list>
412</t>
413<t>
414   deflate<iref item="deflate"/>
415  <list><t>
416        The "zlib" format defined in <xref target="RFC1950"/> in combination with
417        the "deflate" compression mechanism described in <xref target="RFC1951"/>.
418  </t></list>
419</t>
420<t>
421   identity<iref item="identity"/>
422  <list><t>
423        The default (identity) encoding; the use of no transformation
424        whatsoever. This content-coding is used only in the Accept-Encoding
425        header, and &SHOULD-NOT;  be used in the Content-Encoding
426        header.
427  </t></list>
428</t>
429<t>
430   New content-coding value tokens &SHOULD; be registered; to allow
431   interoperability between clients and servers, specifications of the
432   content coding algorithms needed to implement a new value &SHOULD; be
433   publicly available and adequate for independent implementation, and
434   conform to the purpose of content coding defined in this section.
435</t>
436</section>
437
438<section title="Media Types" anchor="media.types">
439<t>
440   HTTP uses Internet Media Types <xref target="RFC2046"/> in the Content-Type (<xref target="header.content-type"/>)
441   and Accept (<xref target="header.accept"/>) header fields in order to provide
442   open and extensible data typing and type negotiation.
443</t>
444<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="media-type"/><iref primary="true" item="Grammar" subitem="type"/><iref primary="true" item="Grammar" subitem="subtype"/>
445  media-type     = type "/" subtype *( ";" parameter )
446  type           = token
447  subtype        = token
448</artwork></figure>
449<t>
450   Parameters &MAY; follow the type/subtype in the form of attribute/value
451   pairs.
452</t>
453<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="parameter"/><iref primary="true" item="Grammar" subitem="attribute"/><iref primary="true" item="Grammar" subitem="value"/>
454  parameter               = attribute "=" value
455  attribute               = token
456  value                   = token | quoted-string
457</artwork></figure>
458<t>
459   The type, subtype, and parameter attribute names are case-insensitive.
460   Parameter values might or might not be case-sensitive,
461   depending on the semantics of the parameter name. Linear white space
462   (LWS) &MUST-NOT; be used between the type and subtype, nor between an
463   attribute and its value. The presence or absence of a parameter might
464   be significant to the processing of a media-type, depending on its
465   definition within the media type registry.
466</t>
467<t>
468   Note that some older HTTP applications do not recognize media type
469   parameters. When sending data to older HTTP applications,
470   implementations &SHOULD; only use media type parameters when they are
471   required by that type/subtype definition.
472</t>
473<t>
474   Media-type values are registered with the Internet Assigned Number
475   Authority (IANA). The media type registration process is
476   outlined in <xref target="RFC4288"/>. Use of non-registered media types is
477   discouraged.
478</t>
479
480<section title="Canonicalization and Text Defaults" anchor="canonicalization.and.text.defaults">
481<t>
482   Internet media types are registered with a canonical form. An
483   entity-body transferred via HTTP messages &MUST; be represented in the
484   appropriate canonical form prior to its transmission except for
485   "text" types, as defined in the next paragraph.
486</t>
487<t>
488   When in canonical form, media subtypes of the "text" type use CRLF as
489   the text line break. HTTP relaxes this requirement and allows the
490   transport of text media with plain CR or LF alone representing a line
491   break when it is done consistently for an entire entity-body. HTTP
492   applications &MUST; accept CRLF, bare CR, and bare LF as being
493   representative of a line break in text media received via HTTP. In
494   addition, if the text is represented in a character set that does not
495   use octets 13 and 10 for CR and LF respectively, as is the case for
496   some multi-byte character sets, HTTP allows the use of whatever octet
497   sequences are defined by that character set to represent the
498   equivalent of CR and LF for line breaks. This flexibility regarding
499   line breaks applies only to text media in the entity-body; a bare CR
500   or LF &MUST-NOT; be substituted for CRLF within any of the HTTP control
501   structures (such as header fields and multipart boundaries).
502</t>
503<t>
504   If an entity-body is encoded with a content-coding, the underlying
505   data &MUST; be in a form defined above prior to being encoded.
506</t>
507<t>
508   The "charset" parameter is used with some media types to define the
509   character set (<xref target="character.sets"/>) of the data. When no explicit charset
510   parameter is provided by the sender, media subtypes of the "text"
511   type are defined to have a default charset value of "ISO-8859-1" when
512   received via HTTP. Data in character sets other than "ISO-8859-1" or
513   its subsets &MUST; be labeled with an appropriate charset value. See
514   <xref target="missing.charset"/> for compatibility problems.
515</t>
516</section>
517
518<section title="Multipart Types" anchor="multipart.types">
519<t>
520   MIME provides for a number of "multipart" types -- encapsulations of
521   one or more entities within a single message-body. All multipart
522   types share a common syntax, as defined in <xref target="RFC2046" x:sec="5.1.1" x:fmt="of"/>,
523   and &MUST; include a boundary parameter as part of the media type
524   value. The message body is itself a protocol element and &MUST;
525   therefore use only CRLF to represent line breaks between body-parts.
526   Unlike in RFC 2046, the epilogue of any multipart message &MUST; be
527   empty; HTTP applications &MUST-NOT; transmit the epilogue (even if the
528   original multipart contains an epilogue). These restrictions exist in
529   order to preserve the self-delimiting nature of a multipart message-body,
530   wherein the "end" of the message-body is indicated by the
531   ending multipart boundary.
532</t>
533<t>
534   In general, HTTP treats a multipart message-body no differently than
535   any other media type: strictly as payload. The one exception is the
536   "multipart/byteranges" type (&multipart-byteranges;) when it appears in a 206
537   (Partial Content) response.
538   <!-- jre: re-insert removed text pointing to caching? -->
539   In all
540   other cases, an HTTP user agent &SHOULD; follow the same or similar
541   behavior as a MIME user agent would upon receipt of a multipart type.
542   The MIME header fields within each body-part of a multipart message-body
543   do not have any significance to HTTP beyond that defined by
544   their MIME semantics.
545</t>
546<t>
547   In general, an HTTP user agent &SHOULD; follow the same or similar
548   behavior as a MIME user agent would upon receipt of a multipart type.
549   If an application receives an unrecognized multipart subtype, the
550   application &MUST; treat it as being equivalent to "multipart/mixed".
551</t>
552<t><list><t>
553      <x:h>Note:</x:h> The "multipart/form-data" type has been specifically defined
554      for carrying form data suitable for processing via the POST
555      request method, as described in <xref target="RFC2388"/>.
556</t></list></t>
557</section>
558</section>
559
560<section title="Quality Values" anchor="quality.values">
561<t>
562   HTTP content negotiation (<xref target="content.negotiation"/>) uses short "floating point"
563   numbers to indicate the relative importance ("weight") of various
564   negotiable parameters.  A weight is normalized to a real number in
565   the range 0 through 1, where 0 is the minimum and 1 the maximum
566   value. If a parameter has a quality value of 0, then content with
567   this parameter is `not acceptable' for the client. HTTP/1.1
568   applications &MUST-NOT; generate more than three digits after the
569   decimal point. User configuration of these values &SHOULD; also be
570   limited in this fashion.
571</t>
572<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="qvalue"/>
573  qvalue         = ( "0" [ "." 0*3DIGIT ] )
574                 | ( "1" [ "." 0*3("0") ] )
575</artwork></figure>
576<t>
577   "Quality values" is a misnomer, since these values merely represent
578   relative degradation in desired quality.
579</t>
580</section>
581
582<section title="Language Tags" anchor="language.tags">
583<t>
584   A language tag identifies a natural language spoken, written, or
585   otherwise conveyed by human beings for communication of information
586   to other human beings. Computer languages are explicitly excluded.
587   HTTP uses language tags within the Accept-Language and Content-Language
588   fields.
589</t>
590<t>
591   The syntax and registry of HTTP language tags is the same as that
592   defined by <xref target="RFC1766"/>. In summary, a language tag is composed of 1
593   or more parts: A primary language tag and a possibly empty series of
594   subtags:
595</t>
596<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="language-tag"/><iref primary="true" item="Grammar" subitem="primary-tag"/><iref primary="true" item="Grammar" subitem="subtag"/>
597  language-tag  = primary-tag *( "-" subtag )
598  primary-tag   = 1*8ALPHA
599  subtag        = 1*8ALPHA
600</artwork></figure>
601<t>
602   White space is not allowed within the tag and all tags are case-insensitive.
603   The name space of language tags is administered by the
604   IANA. Example tags include:
605</t>
606<figure><artwork type="example">
607    en, en-US, en-cockney, i-cherokee, x-pig-latin
608</artwork></figure>
609<t>
610   where any two-letter primary-tag is an ISO-639 language abbreviation
611   and any two-letter initial subtag is an ISO-3166 country code. (The
612   last three tags above are not registered tags; all but the last are
613   examples of tags which could be registered in future.)
614</t>
615</section>
616</section>
617
618<section title="Entity" anchor="entity">
619<t>
620   Request and Response messages &MAY; transfer an entity if not otherwise
621   restricted by the request method or response status code. An entity
622   consists of entity-header fields and an entity-body, although some
623   responses will only include the entity-headers.
624</t>
625<t>
626   In this section, both sender and recipient refer to either the client
627   or the server, depending on who sends and who receives the entity.
628</t>
629
630<section title="Entity Header Fields" anchor="entity.header.fields">
631<t>
632   Entity-header fields define metainformation about the entity-body or,
633   if no body is present, about the resource identified by the request.
634</t>
635<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="entity-header"/><iref primary="true" item="Grammar" subitem="extension-header"/>
636  entity-header  = Allow                    ; &header-allow;
637                 | Content-Encoding         ; <xref target="header.content-encoding"/>
638                 | Content-Language         ; <xref target="header.content-language"/>
639                 | Content-Length           ; &header-content-length;
640                 | Content-Location         ; <xref target="header.content-location"/>
641                 | Content-MD5              ; <xref target="header.content-md5"/>
642                 | Content-Range            ; &header-content-range;
643                 | Content-Type             ; <xref target="header.content-type"/>
644                 | Expires                  ; &header-expires;
645                 | Last-Modified            ; &header-last-modified;
646                 | extension-header
647 
648  extension-header = message-header
649</artwork></figure>
650<t>
651   The extension-header mechanism allows additional entity-header fields
652   to be defined without changing the protocol, but these fields cannot
653   be assumed to be recognizable by the recipient. Unrecognized header
654   fields &SHOULD; be ignored by the recipient and &MUST; be forwarded by
655   transparent proxies.
656</t>
657</section>
658
659<section title="Entity Body" anchor="entity.body">
660<t>
661   The entity-body (if any) sent with an HTTP request or response is in
662   a format and encoding defined by the entity-header fields.
663</t>
664<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="entity-body"/>
665  entity-body    = *OCTET
666</artwork></figure>
667<t>
668   An entity-body is only present in a message when a message-body is
669   present, as described in &message-body;. The entity-body is obtained
670   from the message-body by decoding any Transfer-Encoding that might
671   have been applied to ensure safe and proper transfer of the message.
672</t>
673
674<section title="Type" anchor="type">
675<t>
676   When an entity-body is included with a message, the data type of that
677   body is determined via the header fields Content-Type and Content-Encoding.
678   These define a two-layer, ordered encoding model:
679</t>
680<figure><artwork type="example">
681    entity-body := Content-Encoding( Content-Type( data ) )
682</artwork></figure>
683<t>
684   Content-Type specifies the media type of the underlying data.
685   Content-Encoding may be used to indicate any additional content
686   codings applied to the data, usually for the purpose of data
687   compression, that are a property of the requested resource. There is
688   no default encoding.
689</t>
690<t>
691   Any HTTP/1.1 message containing an entity-body &SHOULD; include a
692   Content-Type header field defining the media type of that body. If
693   and only if the media type is not given by a Content-Type field, the
694   recipient &MAY; attempt to guess the media type via inspection of its
695   content and/or the name extension(s) of the URI used to identify the
696   resource. If the media type remains unknown, the recipient &SHOULD;
697   treat it as type "application/octet-stream".
698</t>
699</section>
700   
701<section title="Entity Length" anchor="entity.length">
702<t>
703   The entity-length of a message is the length of the message-body
704   before any transfer-codings have been applied. &message-length; defines
705   how the transfer-length of a message-body is determined.
706</t>
707</section>
708</section>
709</section>
710
711<section title="Content Negotiation" anchor="content.negotiation">
712<t>
713   Most HTTP responses include an entity which contains information for
714   interpretation by a human user. Naturally, it is desirable to supply
715   the user with the "best available" entity corresponding to the
716   request. Unfortunately for servers and caches, not all users have the
717   same preferences for what is "best," and not all user agents are
718   equally capable of rendering all entity types. For that reason, HTTP
719   has provisions for several mechanisms for "content negotiation" --
720   the process of selecting the best representation for a given response
721   when there are multiple representations available.
722  <list><t>
723      <x:h>Note:</x:h> This is not called "format negotiation" because the
724      alternate representations may be of the same media type, but use
725      different capabilities of that type, be in different languages,
726      etc.
727  </t></list>
728</t>
729<t>
730   Any response containing an entity-body &MAY; be subject to negotiation,
731   including error responses.
732</t>
733<t>
734   There are two kinds of content negotiation which are possible in
735   HTTP: server-driven and agent-driven negotiation. These two kinds of
736   negotiation are orthogonal and thus may be used separately or in
737   combination. One method of combination, referred to as transparent
738   negotiation, occurs when a cache uses the agent-driven negotiation
739   information provided by the origin server in order to provide
740   server-driven negotiation for subsequent requests.
741</t>
742
743<section title="Server-driven Negotiation" anchor="server-driven.negotiation">
744<t>
745   If the selection of the best representation for a response is made by
746   an algorithm located at the server, it is called server-driven
747   negotiation. Selection is based on the available representations of
748   the response (the dimensions over which it can vary; e.g. language,
749   content-coding, etc.) and the contents of particular header fields in
750   the request message or on other information pertaining to the request
751   (such as the network address of the client).
752</t>
753<t>
754   Server-driven negotiation is advantageous when the algorithm for
755   selecting from among the available representations is difficult to
756   describe to the user agent, or when the server desires to send its
757   "best guess" to the client along with the first response (hoping to
758   avoid the round-trip delay of a subsequent request if the "best
759   guess" is good enough for the user). In order to improve the server's
760   guess, the user agent &MAY; include request header fields (Accept,
761   Accept-Language, Accept-Encoding, etc.) which describe its
762   preferences for such a response.
763</t>
764<t>
765   Server-driven negotiation has disadvantages:
766  <list style="numbers">
767    <t>
768         It is impossible for the server to accurately determine what
769         might be "best" for any given user, since that would require
770         complete knowledge of both the capabilities of the user agent
771         and the intended use for the response (e.g., does the user want
772         to view it on screen or print it on paper?).
773    </t>
774    <t>
775         Having the user agent describe its capabilities in every
776         request can be both very inefficient (given that only a small
777         percentage of responses have multiple representations) and a
778         potential violation of the user's privacy.
779    </t>
780    <t>
781         It complicates the implementation of an origin server and the
782         algorithms for generating responses to a request.
783    </t>
784    <t>
785         It may limit a public cache's ability to use the same response
786         for multiple user's requests.
787    </t>
788  </list>
789</t>
790<t>
791   HTTP/1.1 includes the following request-header fields for enabling
792   server-driven negotiation through description of user agent
793   capabilities and user preferences: Accept (<xref target="header.accept"/>), Accept-Charset
794   (<xref target="header.accept-charset"/>), Accept-Encoding (<xref target="header.accept-encoding"/>), Accept-Language
795   (<xref target="header.accept-language"/>), and User-Agent (&header-user-agent;). However, an
796   origin server is not limited to these dimensions and &MAY; vary the
797   response based on any aspect of the request, including information
798   outside the request-header fields or within extension header fields
799   not defined by this specification.
800</t>
801<t>
802   The Vary header field (&header-vary;) can be used to express the parameters the
803   server uses to select a representation that is subject to server-driven
804   negotiation.
805</t>
806</section>
807
808<section title="Agent-driven Negotiation" anchor="agent-driven.negotiation">
809<t>
810   With agent-driven negotiation, selection of the best representation
811   for a response is performed by the user agent after receiving an
812   initial response from the origin server. Selection is based on a list
813   of the available representations of the response included within the
814   header fields or entity-body of the initial response, with each
815   representation identified by its own URI. Selection from among the
816   representations may be performed automatically (if the user agent is
817   capable of doing so) or manually by the user selecting from a
818   generated (possibly hypertext) menu.
819</t>
820<t>
821   Agent-driven negotiation is advantageous when the response would vary
822   over commonly-used dimensions (such as type, language, or encoding),
823   when the origin server is unable to determine a user agent's
824   capabilities from examining the request, and generally when public
825   caches are used to distribute server load and reduce network usage.
826</t>
827<t>
828   Agent-driven negotiation suffers from the disadvantage of needing a
829   second request to obtain the best alternate representation. This
830   second request is only efficient when caching is used. In addition,
831   this specification does not define any mechanism for supporting
832   automatic selection, though it also does not prevent any such
833   mechanism from being developed as an extension and used within
834   HTTP/1.1.
835</t>
836<t>
837   HTTP/1.1 defines the 300 (Multiple Choices) and 406 (Not Acceptable)
838   status codes for enabling agent-driven negotiation when the server is
839   unwilling or unable to provide a varying response using server-driven
840   negotiation.
841</t>
842</section>
843
844<section title="Transparent Negotiation" anchor="transparent.negotiation">
845<t>
846   Transparent negotiation is a combination of both server-driven and
847   agent-driven negotiation. When a cache is supplied with a form of the
848   list of available representations of the response (as in agent-driven
849   negotiation) and the dimensions of variance are completely understood
850   by the cache, then the cache becomes capable of performing server-driven
851   negotiation on behalf of the origin server for subsequent
852   requests on that resource.
853</t>
854<t>
855   Transparent negotiation has the advantage of distributing the
856   negotiation work that would otherwise be required of the origin
857   server and also removing the second request delay of agent-driven
858   negotiation when the cache is able to correctly guess the right
859   response.
860</t>
861<t>
862   This specification does not define any mechanism for transparent
863   negotiation, though it also does not prevent any such mechanism from
864   being developed as an extension that could be used within HTTP/1.1.
865</t>
866</section>
867</section>
868
869<section title="Header Field Definitions" anchor="header.fields">
870<t>
871   This section defines the syntax and semantics of HTTP/1.1 header fields
872   related to the payload of messages.
873</t>
874<t>
875   For entity-header fields, both sender and recipient refer to either the
876   client or the server, depending on who sends and who receives the entity.
877</t>
878
879<section title="Accept" anchor="header.accept">
880  <iref primary="true" item="Accept header" x:for-anchor=""/>
881  <iref primary="true" item="Headers" subitem="Accept" x:for-anchor=""/>
882<t>
883   The Accept request-header field can be used to specify certain media
884   types which are acceptable for the response. Accept headers can be
885   used to indicate that the request is specifically limited to a small
886   set of desired types, as in the case of a request for an in-line
887   image.
888</t>
889<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Accept"/><iref primary="true" item="Grammar" subitem="media-range"/><iref primary="true" item="Grammar" subitem="accept-params"/><iref primary="true" item="Grammar" subitem="accept-extension"/>
890  Accept         = "Accept" ":"
891                   #( media-range [ accept-params ] )
892 
893  media-range    = ( "*/*"
894                   | ( type "/" "*" )
895                   | ( type "/" subtype )
896                   ) *( ";" parameter )
897  accept-params  = ";" "q" "=" qvalue *( accept-extension )
898  accept-extension = ";" token [ "=" ( token | quoted-string ) ]
899</artwork></figure>
900<t>
901   The asterisk "*" character is used to group media types into ranges,
902   with "*/*" indicating all media types and "type/*" indicating all
903   subtypes of that type. The media-range &MAY; include media type
904   parameters that are applicable to that range.
905</t>
906<t>
907   Each media-range &MAY; be followed by one or more accept-params,
908   beginning with the "q" parameter for indicating a relative quality
909   factor. The first "q" parameter (if any) separates the media-range
910   parameter(s) from the accept-params. Quality factors allow the user
911   or user agent to indicate the relative degree of preference for that
912   media-range, using the qvalue scale from 0 to 1 (<xref target="quality.values"/>). The
913   default value is q=1.
914  <list><t>
915      <x:h>Note:</x:h> Use of the "q" parameter name to separate media type
916      parameters from Accept extension parameters is due to historical
917      practice. Although this prevents any media type parameter named
918      "q" from being used with a media range, such an event is believed
919      to be unlikely given the lack of any "q" parameters in the IANA
920      media type registry and the rare usage of any media type
921      parameters in Accept. Future media types are discouraged from
922      registering any parameter named "q".
923  </t></list>
924</t>
925<t>
926   The example
927</t>
928<figure><artwork type="example">
929    Accept: audio/*; q=0.2, audio/basic
930</artwork></figure>
931<t>
932   &SHOULD; be interpreted as "I prefer audio/basic, but send me any audio
933   type if it is the best available after an 80% mark-down in quality."
934</t>
935<t>
936   If no Accept header field is present, then it is assumed that the
937   client accepts all media types. If an Accept header field is present,
938   and if the server cannot send a response which is acceptable
939   according to the combined Accept field value, then the server &SHOULD;
940   send a 406 (Not Acceptable) response.
941</t>
942<t>
943   A more elaborate example is
944</t>
945<figure><artwork type="example">
946    Accept: text/plain; q=0.5, text/html,
947            text/x-dvi; q=0.8, text/x-c
948</artwork></figure>
949<t>
950   Verbally, this would be interpreted as "text/html and text/x-c are
951   the preferred media types, but if they do not exist, then send the
952   text/x-dvi entity, and if that does not exist, send the text/plain
953   entity."
954</t>
955<t>
956   Media ranges can be overridden by more specific media ranges or
957   specific media types. If more than one media range applies to a given
958   type, the most specific reference has precedence. For example,
959</t>
960<figure><artwork type="example">
961    Accept: text/*, text/html, text/html;level=1, */*
962</artwork></figure>
963<t>
964   have the following precedence:
965</t>
966<figure><artwork type="example">
967    1) text/html;level=1
968    2) text/html
969    3) text/*
970    4) */*
971</artwork></figure>
972<t>
973   The media type quality factor associated with a given type is
974   determined by finding the media range with the highest precedence
975   which matches that type. For example,
976</t>
977<figure><artwork type="example">
978    Accept: text/*;q=0.3, text/html;q=0.7, text/html;level=1,
979            text/html;level=2;q=0.4, */*;q=0.5
980</artwork></figure>
981<t>
982   would cause the following values to be associated:
983</t>
984<figure><artwork type="example">
985    text/html;level=1         = 1
986    text/html                 = 0.7
987    text/plain                = 0.3
988    image/jpeg                = 0.5
989    text/html;level=2         = 0.4
990    text/html;level=3         = 0.7
991</artwork></figure>
992<t>
993      <x:h>Note:</x:h> A user agent might be provided with a default set of quality
994      values for certain media ranges. However, unless the user agent is
995      a closed system which cannot interact with other rendering agents,
996      this default set ought to be configurable by the user.
997</t>
998</section>
999
1000<section title="Accept-Charset" anchor="header.accept-charset">
1001  <iref primary="true" item="Accept-Charset header" x:for-anchor=""/>
1002  <iref primary="true" item="Headers" subitem="Accept-Charset" x:for-anchor=""/>
1003<t>
1004   The Accept-Charset request-header field can be used to indicate what
1005   character sets are acceptable for the response. This field allows
1006   clients capable of understanding more comprehensive or special-purpose
1007   character sets to signal that capability to a server which is
1008   capable of representing documents in those character sets.
1009</t>
1010<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Accept-Charset"/>
1011  Accept-Charset = "Accept-Charset" ":"
1012          1#( ( charset | "*" ) [ ";" "q" "=" qvalue ] )
1013</artwork></figure>
1014<t>
1015   Character set values are described in <xref target="character.sets"/>. Each charset &MAY;
1016   be given an associated quality value which represents the user's
1017   preference for that charset. The default value is q=1. An example is
1018</t>
1019<figure><artwork type="example">
1020   Accept-Charset: iso-8859-5, unicode-1-1;q=0.8
1021</artwork></figure>
1022<t>
1023   The special value "*", if present in the Accept-Charset field,
1024   matches every character set (including ISO-8859-1) which is not
1025   mentioned elsewhere in the Accept-Charset field. If no "*" is present
1026   in an Accept-Charset field, then all character sets not explicitly
1027   mentioned get a quality value of 0, except for ISO-8859-1, which gets
1028   a quality value of 1 if not explicitly mentioned.
1029</t>
1030<t>
1031   If no Accept-Charset header is present, the default is that any
1032   character set is acceptable. If an Accept-Charset header is present,
1033   and if the server cannot send a response which is acceptable
1034   according to the Accept-Charset header, then the server &SHOULD; send
1035   an error response with the 406 (Not Acceptable) status code, though
1036   the sending of an unacceptable response is also allowed.
1037</t>
1038</section>
1039
1040<section title="Accept-Encoding" anchor="header.accept-encoding">
1041  <iref primary="true" item="Accept-Encoding header" x:for-anchor=""/>
1042  <iref primary="true" item="Headers" subitem="Accept-Encoding" x:for-anchor=""/>
1043<t>
1044   The Accept-Encoding request-header field is similar to Accept, but
1045   restricts the content-codings (<xref target="content.codings"/>) that are acceptable in
1046   the response.
1047</t>
1048<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Accept-Encoding"/><iref primary="true" item="Grammar" subitem="codings"/>
1049  Accept-Encoding  = "Accept-Encoding" ":"
1050                     #( codings [ ";" "q" "=" qvalue ] )
1051  codings          = ( content-coding | "*" )
1052</artwork></figure>
1053<t>
1054   Examples of its use are:
1055</t>
1056<figure><artwork type="example">
1057    Accept-Encoding: compress, gzip
1058    Accept-Encoding:
1059    Accept-Encoding: *
1060    Accept-Encoding: compress;q=0.5, gzip;q=1.0
1061    Accept-Encoding: gzip;q=1.0, identity; q=0.5, *;q=0
1062</artwork></figure>
1063<t>
1064   A server tests whether a content-coding is acceptable, according to
1065   an Accept-Encoding field, using these rules:
1066  <list style="numbers">
1067      <t>If the content-coding is one of the content-codings listed in
1068         the Accept-Encoding field, then it is acceptable, unless it is
1069         accompanied by a qvalue of 0. (As defined in <xref target="quality.values"/>, a
1070         qvalue of 0 means "not acceptable.")</t>
1071
1072      <t>The special "*" symbol in an Accept-Encoding field matches any
1073         available content-coding not explicitly listed in the header
1074         field.</t>
1075
1076      <t>If multiple content-codings are acceptable, then the acceptable
1077         content-coding with the highest non-zero qvalue is preferred.</t>
1078
1079      <t>The "identity" content-coding is always acceptable, unless
1080         specifically refused because the Accept-Encoding field includes
1081         "identity;q=0", or because the field includes "*;q=0" and does
1082         not explicitly include the "identity" content-coding. If the
1083         Accept-Encoding field-value is empty, then only the "identity"
1084         encoding is acceptable.</t>
1085  </list>
1086</t>
1087<t>
1088   If an Accept-Encoding field is present in a request, and if the
1089   server cannot send a response which is acceptable according to the
1090   Accept-Encoding header, then the server &SHOULD; send an error response
1091   with the 406 (Not Acceptable) status code.
1092</t>
1093<t>
1094   If no Accept-Encoding field is present in a request, the server &MAY;
1095   assume that the client will accept any content coding. In this case,
1096   if "identity" is one of the available content-codings, then the
1097   server &SHOULD; use the "identity" content-coding, unless it has
1098   additional information that a different content-coding is meaningful
1099   to the client.
1100  <list><t>
1101      <x:h>Note:</x:h> If the request does not include an Accept-Encoding field,
1102      and if the "identity" content-coding is unavailable, then
1103      content-codings commonly understood by HTTP/1.0 clients (i.e.,
1104      "gzip" and "compress") are preferred; some older clients
1105      improperly display messages sent with other content-codings.  The
1106      server might also make this decision based on information about
1107      the particular user-agent or client.
1108    </t><t>
1109      <x:h>Note:</x:h> Most HTTP/1.0 applications do not recognize or obey qvalues
1110      associated with content-codings. This means that qvalues will not
1111      work and are not permitted with x-gzip or x-compress.
1112    </t></list>
1113</t>
1114</section>
1115
1116<section title="Accept-Language" anchor="header.accept-language">
1117  <iref primary="true" item="Accept-Language header" x:for-anchor=""/>
1118  <iref primary="true" item="Headers" subitem="Accept-Language" x:for-anchor=""/>
1119<t>
1120   The Accept-Language request-header field is similar to Accept, but
1121   restricts the set of natural languages that are preferred as a
1122   response to the request. Language tags are defined in <xref target="language.tags"/>.
1123</t>
1124<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Accept-Language"/><iref primary="true" item="Grammar" subitem="language-range"/>
1125  Accept-Language = "Accept-Language" ":"
1126                    1#( language-range [ ";" "q" "=" qvalue ] )
1127  language-range  = ( ( 1*8ALPHA *( "-" 1*8ALPHA ) ) | "*" )
1128</artwork></figure>
1129<t>
1130   Each language-range &MAY; be given an associated quality value which
1131   represents an estimate of the user's preference for the languages
1132   specified by that range. The quality value defaults to "q=1". For
1133   example,
1134</t>
1135<figure><artwork type="example">
1136    Accept-Language: da, en-gb;q=0.8, en;q=0.7
1137</artwork></figure>
1138<t>
1139   would mean: "I prefer Danish, but will accept British English and
1140   other types of English." A language-range matches a language-tag if
1141   it exactly equals the tag, or if it exactly equals a prefix of the
1142   tag such that the first tag character following the prefix is "-".
1143   The special range "*", if present in the Accept-Language field,
1144   matches every tag not matched by any other range present in the
1145   Accept-Language field.
1146  <list><t>
1147      <x:h>Note:</x:h> This use of a prefix matching rule does not imply that
1148      language tags are assigned to languages in such a way that it is
1149      always true that if a user understands a language with a certain
1150      tag, then this user will also understand all languages with tags
1151      for which this tag is a prefix. The prefix rule simply allows the
1152      use of prefix tags if this is the case.
1153  </t></list>
1154</t>
1155<t>
1156   The language quality factor assigned to a language-tag by the
1157   Accept-Language field is the quality value of the longest language-range
1158   in the field that matches the language-tag. If no language-range
1159   in the field matches the tag, the language quality factor
1160   assigned is 0. If no Accept-Language header is present in the
1161   request, the server
1162   &SHOULD; assume that all languages are equally acceptable. If an
1163   Accept-Language header is present, then all languages which are
1164   assigned a quality factor greater than 0 are acceptable.
1165</t>
1166<t>
1167   It might be contrary to the privacy expectations of the user to send
1168   an Accept-Language header with the complete linguistic preferences of
1169   the user in every request. For a discussion of this issue, see
1170   <xref target="privacy.issues.connected.to.accept.headers"/>.
1171</t>
1172<t>
1173   As intelligibility is highly dependent on the individual user, it is
1174   recommended that client applications make the choice of linguistic
1175   preference available to the user. If the choice is not made
1176   available, then the Accept-Language header field &MUST-NOT; be given in
1177   the request.
1178  <list><t>
1179      <x:h>Note:</x:h> When making the choice of linguistic preference available to
1180      the user, we remind implementors of  the fact that users are not
1181      familiar with the details of language matching as described above,
1182      and should provide appropriate guidance. As an example, users
1183      might assume that on selecting "en-gb", they will be served any
1184      kind of English document if British English is not available. A
1185      user agent might suggest in such a case to add "en" to get the
1186      best matching behavior.
1187  </t></list>
1188</t>
1189</section>
1190
1191<section title="Content-Encoding" anchor="header.content-encoding">
1192  <iref primary="true" item="Content-Encoding header" x:for-anchor=""/>
1193  <iref primary="true" item="Headers" subitem="Content-Encoding" x:for-anchor=""/>
1194<t>
1195   The Content-Encoding entity-header field is used as a modifier to the
1196   media-type. When present, its value indicates what additional content
1197   codings have been applied to the entity-body, and thus what decoding
1198   mechanisms must be applied in order to obtain the media-type
1199   referenced by the Content-Type header field. Content-Encoding is
1200   primarily used to allow a document to be compressed without losing
1201   the identity of its underlying media type.
1202</t>
1203<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Encoding"/>
1204  Content-Encoding  = "Content-Encoding" ":" 1#content-coding
1205</artwork></figure>
1206<t>
1207   Content codings are defined in <xref target="content.codings"/>. An example of its use is
1208</t>
1209<figure><artwork type="example">
1210    Content-Encoding: gzip
1211</artwork></figure>
1212<t>
1213   The content-coding is a characteristic of the entity identified by
1214   the Request-URI. Typically, the entity-body is stored with this
1215   encoding and is only decoded before rendering or analogous usage.
1216   However, a non-transparent proxy &MAY; modify the content-coding if the
1217   new coding is known to be acceptable to the recipient, unless the
1218   "no-transform" cache-control directive is present in the message.
1219</t>
1220<t>
1221   If the content-coding of an entity is not "identity", then the
1222   response &MUST; include a Content-Encoding entity-header (<xref target="header.content-encoding"/>)
1223   that lists the non-identity content-coding(s) used.
1224</t>
1225<t>
1226   If the content-coding of an entity in a request message is not
1227   acceptable to the origin server, the server &SHOULD; respond with a
1228   status code of 415 (Unsupported Media Type).
1229</t>
1230<t>
1231   If multiple encodings have been applied to an entity, the content
1232   codings &MUST; be listed in the order in which they were applied.
1233   Additional information about the encoding parameters &MAY; be provided
1234   by other entity-header fields not defined by this specification.
1235</t>
1236</section>
1237
1238<section title="Content-Language" anchor="header.content-language">
1239  <iref primary="true" item="Content-Language header" x:for-anchor=""/>
1240  <iref primary="true" item="Headers" subitem="Content-Language" x:for-anchor=""/>
1241<t>
1242   The Content-Language entity-header field describes the natural
1243   language(s) of the intended audience for the enclosed entity. Note
1244   that this might not be equivalent to all the languages used within
1245   the entity-body.
1246</t>
1247<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Language"/>
1248  Content-Language  = "Content-Language" ":" 1#language-tag
1249</artwork></figure>
1250<t>
1251   Language tags are defined in <xref target="language.tags"/>. The primary purpose of
1252   Content-Language is to allow a user to identify and differentiate
1253   entities according to the user's own preferred language. Thus, if the
1254   body content is intended only for a Danish-literate audience, the
1255   appropriate field is
1256</t>
1257<figure><artwork type="example">
1258    Content-Language: da
1259</artwork></figure>
1260<t>
1261   If no Content-Language is specified, the default is that the content
1262   is intended for all language audiences. This might mean that the
1263   sender does not consider it to be specific to any natural language,
1264   or that the sender does not know for which language it is intended.
1265</t>
1266<t>
1267   Multiple languages &MAY; be listed for content that is intended for
1268   multiple audiences. For example, a rendition of the "Treaty of
1269   Waitangi," presented simultaneously in the original Maori and English
1270   versions, would call for
1271</t>
1272<figure><artwork type="example">
1273    Content-Language: mi, en
1274</artwork></figure>
1275<t>
1276   However, just because multiple languages are present within an entity
1277   does not mean that it is intended for multiple linguistic audiences.
1278   An example would be a beginner's language primer, such as "A First
1279   Lesson in Latin," which is clearly intended to be used by an
1280   English-literate audience. In this case, the Content-Language would
1281   properly only include "en".
1282</t>
1283<t>
1284   Content-Language &MAY; be applied to any media type -- it is not
1285   limited to textual documents.
1286</t>
1287</section>
1288
1289<section title="Content-Location" anchor="header.content-location">
1290  <iref primary="true" item="Content-Location header" x:for-anchor=""/>
1291  <iref primary="true" item="Headers" subitem="Content-Location" x:for-anchor=""/>
1292<t>
1293   The Content-Location entity-header field &MAY; be used to supply the
1294   resource location for the entity enclosed in the message when that
1295   entity is accessible from a location separate from the requested
1296   resource's URI. A server &SHOULD; provide a Content-Location for the
1297   variant corresponding to the response entity; especially in the case
1298   where a resource has multiple entities associated with it, and those
1299   entities actually have separate locations by which they might be
1300   individually accessed, the server &SHOULD; provide a Content-Location
1301   for the particular variant which is returned.
1302</t>
1303<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Location"/>
1304  Content-Location = "Content-Location" ":"
1305                    ( absoluteURI | relativeURI )
1306</artwork></figure>
1307<t>
1308   The value of Content-Location also defines the base URI for the
1309   entity.
1310</t>
1311<t>
1312   The Content-Location value is not a replacement for the original
1313   requested URI; it is only a statement of the location of the resource
1314   corresponding to this particular entity at the time of the request.
1315   Future requests &MAY; specify the Content-Location URI as the request-URI
1316   if the desire is to identify the source of that particular
1317   entity.
1318</t>
1319<t>
1320   A cache cannot assume that an entity with a Content-Location
1321   different from the URI used to retrieve it can be used to respond to
1322   later requests on that Content-Location URI. However, the Content-Location
1323   can be used to differentiate between multiple entities
1324   retrieved from a single requested resource, as described in &caching-neg-resp;.
1325</t>
1326<t>
1327   If the Content-Location is a relative URI, the relative URI is
1328   interpreted relative to the Request-URI.
1329</t>
1330<t>
1331   The meaning of the Content-Location header in PUT or POST requests is
1332   undefined; servers are free to ignore it in those cases.
1333</t>
1334</section>
1335
1336<section title="Content-MD5" anchor="header.content-md5">
1337  <iref primary="true" item="Content-MD5 header" x:for-anchor=""/>
1338  <iref primary="true" item="Headers" subitem="Content-MD5" x:for-anchor=""/>
1339<t>
1340   The Content-MD5 entity-header field, as defined in <xref target="RFC1864"/>, is
1341   an MD5 digest of the entity-body for the purpose of providing an
1342   end-to-end message integrity check (MIC) of the entity-body. (Note: a
1343   MIC is good for detecting accidental modification of the entity-body
1344   in transit, but is not proof against malicious attacks.)
1345</t>
1346<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-MD5"/><iref primary="true" item="Grammar" subitem="md5-digest"/>
1347  Content-MD5   = "Content-MD5" ":" md5-digest
1348  md5-digest    = &lt;base64 of 128 bit MD5 digest as per <xref target="RFC1864"/>&gt;
1349</artwork></figure>
1350<t>
1351   The Content-MD5 header field &MAY; be generated by an origin server or
1352   client to function as an integrity check of the entity-body. Only
1353   origin servers or clients &MAY; generate the Content-MD5 header field;
1354   proxies and gateways &MUST-NOT; generate it, as this would defeat its
1355   value as an end-to-end integrity check. Any recipient of the entity-body,
1356   including gateways and proxies, &MAY; check that the digest value
1357   in this header field matches that of the entity-body as received.
1358</t>
1359<t>
1360   The MD5 digest is computed based on the content of the entity-body,
1361   including any content-coding that has been applied, but not including
1362   any transfer-encoding applied to the message-body. If the message is
1363   received with a transfer-encoding, that encoding &MUST; be removed
1364   prior to checking the Content-MD5 value against the received entity.
1365</t>
1366<t>
1367   This has the result that the digest is computed on the octets of the
1368   entity-body exactly as, and in the order that, they would be sent if
1369   no transfer-encoding were being applied.
1370</t>
1371<t>
1372   HTTP extends RFC 1864 to permit the digest to be computed for MIME
1373   composite media-types (e.g., multipart/* and message/rfc822), but
1374   this does not change how the digest is computed as defined in the
1375   preceding paragraph.
1376</t>
1377<t>
1378   There are several consequences of this. The entity-body for composite
1379   types &MAY; contain many body-parts, each with its own MIME and HTTP
1380   headers (including Content-MD5, Content-Transfer-Encoding, and
1381   Content-Encoding headers). If a body-part has a Content-Transfer-Encoding
1382   or Content-Encoding header, it is assumed that the content
1383   of the body-part has had the encoding applied, and the body-part is
1384   included in the Content-MD5 digest as is -- i.e., after the
1385   application. The Transfer-Encoding header field is not allowed within
1386   body-parts.
1387</t>
1388<t>
1389   Conversion of all line breaks to CRLF &MUST-NOT; be done before
1390   computing or checking the digest: the line break convention used in
1391   the text actually transmitted &MUST; be left unaltered when computing
1392   the digest.
1393  <list><t>
1394      <x:h>Note:</x:h> while the definition of Content-MD5 is exactly the same for
1395      HTTP as in RFC 1864 for MIME entity-bodies, there are several ways
1396      in which the application of Content-MD5 to HTTP entity-bodies
1397      differs from its application to MIME entity-bodies. One is that
1398      HTTP, unlike MIME, does not use Content-Transfer-Encoding, and
1399      does use Transfer-Encoding and Content-Encoding. Another is that
1400      HTTP more frequently uses binary content types than MIME, so it is
1401      worth noting that, in such cases, the byte order used to compute
1402      the digest is the transmission byte order defined for the type.
1403      Lastly, HTTP allows transmission of text types with any of several
1404      line break conventions and not just the canonical form using CRLF.
1405  </t></list>
1406</t>
1407</section>
1408
1409<section title="Content-Type" anchor="header.content-type">
1410  <iref primary="true" item="Content-Type header" x:for-anchor=""/>
1411  <iref primary="true" item="Headers" subitem="Content-Type" x:for-anchor=""/>
1412<t>
1413   The Content-Type entity-header field indicates the media type of the
1414   entity-body sent to the recipient or, in the case of the HEAD method,
1415   the media type that would have been sent had the request been a GET.
1416</t>
1417<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Type"/>
1418  Content-Type   = "Content-Type" ":" media-type
1419</artwork></figure>
1420<t>
1421   Media types are defined in <xref target="media.types"/>. An example of the field is
1422</t>
1423<figure><artwork type="example">
1424    Content-Type: text/html; charset=ISO-8859-4
1425</artwork></figure>
1426<t>
1427   Further discussion of methods for identifying the media type of an
1428   entity is provided in <xref target="type"/>.
1429</t>
1430</section>
1431
1432</section>
1433
1434<section title="IANA Considerations" anchor="IANA.considerations">
1435<t>
1436   <cref>TBD.</cref>
1437</t>
1438</section>
1439
1440<section title="Security Considerations" anchor="security.considerations">
1441<t>
1442   This section is meant to inform application developers, information
1443   providers, and users of the security limitations in HTTP/1.1 as
1444   described by this document. The discussion does not include
1445   definitive solutions to the problems revealed, though it does make
1446   some suggestions for reducing security risks.
1447</t>
1448
1449<section title="Privacy Issues Connected to Accept Headers" anchor="privacy.issues.connected.to.accept.headers">
1450<t>
1451   Accept request-headers can reveal information about the user to all
1452   servers which are accessed. The Accept-Language header in particular
1453   can reveal information the user would consider to be of a private
1454   nature, because the understanding of particular languages is often
1455   strongly correlated to the membership of a particular ethnic group.
1456   User agents which offer the option to configure the contents of an
1457   Accept-Language header to be sent in every request are strongly
1458   encouraged to let the configuration process include a message which
1459   makes the user aware of the loss of privacy involved.
1460</t>
1461<t>
1462   An approach that limits the loss of privacy would be for a user agent
1463   to omit the sending of Accept-Language headers by default, and to ask
1464   the user whether or not to start sending Accept-Language headers to a
1465   server if it detects, by looking for any Vary response-header fields
1466   generated by the server, that such sending could improve the quality
1467   of service.
1468</t>
1469<t>
1470   Elaborate user-customized accept header fields sent in every request,
1471   in particular if these include quality values, can be used by servers
1472   as relatively reliable and long-lived user identifiers. Such user
1473   identifiers would allow content providers to do click-trail tracking,
1474   and would allow collaborating content providers to match cross-server
1475   click-trails or form submissions of individual users. Note that for
1476   many users not behind a proxy, the network address of the host
1477   running the user agent will also serve as a long-lived user
1478   identifier. In environments where proxies are used to enhance
1479   privacy, user agents ought to be conservative in offering accept
1480   header configuration options to end users. As an extreme privacy
1481   measure, proxies could filter the accept headers in relayed requests.
1482   General purpose user agents which provide a high degree of header
1483   configurability &SHOULD; warn users about the loss of privacy which can
1484   be involved.
1485</t>
1486</section>
1487
1488<section title="Content-Disposition Issues" anchor="content-disposition.issues">
1489<t>
1490   <xref target="RFC1806"/>, from which the often implemented Content-Disposition
1491   (see <xref target="content-disposition"/>) header in HTTP is derived, has a number of very
1492   serious security considerations. Content-Disposition is not part of
1493   the HTTP standard, but since it is widely implemented, we are
1494   documenting its use and risks for implementors. See <xref target="RFC2183"/>
1495   (which updates <xref target="RFC1806"/>) for details.
1496</t>
1497</section>
1498
1499</section>
1500
1501<section title="Acknowledgments" anchor="ack">
1502</section>
1503</middle>
1504<back>
1505
1506<references title="Normative References">
1507
1508<reference anchor="ISO-8859-1">
1509  <front>
1510    <title>
1511     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
1512    </title>
1513    <author>
1514      <organization>International Organization for Standardization</organization>
1515    </author>
1516    <date year="1998"/>
1517  </front>
1518  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
1519</reference>
1520
1521<reference anchor="Part1">
1522  <front>
1523    <title abbrev="HTTP/1.1">HTTP/1.1, part 1: URIs, Connections, and Message Parsing</title>
1524    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1525      <organization abbrev="Day Software">Day Software</organization>
1526      <address><email>fielding@gbiv.com</email></address>
1527    </author>
1528    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1529      <organization>One Laptop per Child</organization>
1530      <address><email>jg@laptop.org</email></address>
1531    </author>
1532    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1533      <organization abbrev="HP">Hewlett-Packard Company</organization>
1534      <address><email>JeffMogul@acm.org</email></address>
1535    </author>
1536    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1537      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1538      <address><email>henrikn@microsoft.com</email></address>
1539    </author>
1540    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1541      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1542      <address><email>LMM@acm.org</email></address>
1543    </author>
1544    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1545      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1546      <address><email>paulle@microsoft.com</email></address>
1547    </author>
1548    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1549      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1550      <address><email>timbl@w3.org</email></address>
1551    </author>
1552    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1553      <organization abbrev="W3C">World Wide Web Consortium</organization>
1554      <address><email>ylafon@w3.org</email></address>
1555    </author>
1556    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1557      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1558      <address><email>julian.reschke@greenbytes.de</email></address>
1559    </author>
1560    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1561  </front>
1562  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"/>
1563  <x:source href="p1-messaging.xml" basename="p1-messaging"/>
1564</reference>
1565
1566<reference anchor="Part2">
1567  <front>
1568    <title abbrev="HTTP/1.1">HTTP/1.1, part 2: Message Semantics</title>
1569    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1570      <organization abbrev="Day Software">Day Software</organization>
1571      <address><email>fielding@gbiv.com</email></address>
1572    </author>
1573    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1574      <organization>One Laptop per Child</organization>
1575      <address><email>jg@laptop.org</email></address>
1576    </author>
1577    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1578      <organization abbrev="HP">Hewlett-Packard Company</organization>
1579      <address><email>JeffMogul@acm.org</email></address>
1580    </author>
1581    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1582      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1583      <address><email>henrikn@microsoft.com</email></address>
1584    </author>
1585    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1586      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1587      <address><email>LMM@acm.org</email></address>
1588    </author>
1589    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1590      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1591      <address><email>paulle@microsoft.com</email></address>
1592    </author>
1593    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1594      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1595      <address><email>timbl@w3.org</email></address>
1596    </author>
1597    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1598      <organization abbrev="W3C">World Wide Web Consortium</organization>
1599      <address><email>ylafon@w3.org</email></address>
1600    </author>
1601    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1602      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1603      <address><email>julian.reschke@greenbytes.de</email></address>
1604    </author>
1605    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1606  </front>
1607  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-&ID-VERSION;"/>
1608  <x:source href="p2-semantics.xml" basename="p2-semantics"/>
1609</reference>
1610
1611<reference anchor="Part4">
1612  <front>
1613    <title abbrev="HTTP/1.1">HTTP/1.1, part 4: Conditional Requests</title>
1614    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1615      <organization abbrev="Day Software">Day Software</organization>
1616      <address><email>fielding@gbiv.com</email></address>
1617    </author>
1618    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1619      <organization>One Laptop per Child</organization>
1620      <address><email>jg@laptop.org</email></address>
1621    </author>
1622    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1623      <organization abbrev="HP">Hewlett-Packard Company</organization>
1624      <address><email>JeffMogul@acm.org</email></address>
1625    </author>
1626    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1627      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1628      <address><email>henrikn@microsoft.com</email></address>
1629    </author>
1630    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1631      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1632      <address><email>LMM@acm.org</email></address>
1633    </author>
1634    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1635      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1636      <address><email>paulle@microsoft.com</email></address>
1637    </author>
1638    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1639      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1640      <address><email>timbl@w3.org</email></address>
1641    </author>
1642    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1643      <organization abbrev="W3C">World Wide Web Consortium</organization>
1644      <address><email>ylafon@w3.org</email></address>
1645    </author>
1646    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1647      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1648      <address><email>julian.reschke@greenbytes.de</email></address>
1649    </author>
1650    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1651  </front>
1652  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p4-conditional-&ID-VERSION;"/>
1653  <x:source href="p4-conditional.xml" basename="p4-conditional"/>
1654</reference>
1655
1656<reference anchor="Part5">
1657  <front>
1658    <title abbrev="HTTP/1.1">HTTP/1.1, part 5: Range Requests and Partial Responses</title>
1659    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1660      <organization abbrev="Day Software">Day Software</organization>
1661      <address><email>fielding@gbiv.com</email></address>
1662    </author>
1663    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1664      <organization>One Laptop per Child</organization>
1665      <address><email>jg@laptop.org</email></address>
1666    </author>
1667    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1668      <organization abbrev="HP">Hewlett-Packard Company</organization>
1669      <address><email>JeffMogul@acm.org</email></address>
1670    </author>
1671    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1672      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1673      <address><email>henrikn@microsoft.com</email></address>
1674    </author>
1675    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1676      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1677      <address><email>LMM@acm.org</email></address>
1678    </author>
1679    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1680      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1681      <address><email>paulle@microsoft.com</email></address>
1682    </author>
1683    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1684      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1685      <address><email>timbl@w3.org</email></address>
1686    </author>
1687    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1688      <organization abbrev="W3C">World Wide Web Consortium</organization>
1689      <address><email>ylafon@w3.org</email></address>
1690    </author>
1691    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1692      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1693      <address><email>julian.reschke@greenbytes.de</email></address>
1694    </author>
1695    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1696  </front>
1697  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-&ID-VERSION;"/>
1698  <x:source href="p5-range.xml" basename="p5-range"/>
1699</reference>
1700
1701<reference anchor="Part6">
1702  <front>
1703    <title abbrev="HTTP/1.1">HTTP/1.1, part 6: Caching</title>
1704    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1705      <organization abbrev="Day Software">Day Software</organization>
1706      <address><email>fielding@gbiv.com</email></address>
1707    </author>
1708    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1709      <organization>One Laptop per Child</organization>
1710      <address><email>jg@laptop.org</email></address>
1711    </author>
1712    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1713      <organization abbrev="HP">Hewlett-Packard Company</organization>
1714      <address><email>JeffMogul@acm.org</email></address>
1715    </author>
1716    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1717      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1718      <address><email>henrikn@microsoft.com</email></address>
1719    </author>
1720    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1721      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1722      <address><email>LMM@acm.org</email></address>
1723    </author>
1724    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1725      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1726      <address><email>paulle@microsoft.com</email></address>
1727    </author>
1728    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1729      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1730      <address><email>timbl@w3.org</email></address>
1731    </author>
1732    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1733      <organization abbrev="W3C">World Wide Web Consortium</organization>
1734      <address><email>ylafon@w3.org</email></address>
1735    </author>
1736    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1737      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1738      <address><email>julian.reschke@greenbytes.de</email></address>
1739    </author>
1740    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1741  </front>
1742  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-&ID-VERSION;"/>
1743  <x:source href="p6-cache.xml" basename="p6-cache"/>
1744</reference>
1745
1746<reference anchor="RFC1766">
1747  <front>
1748    <title abbrev="Language Tag">Tags for the Identification of Languages</title>
1749    <author initials="H." surname="Alvestrand" fullname="Harald Tveit Alvestrand">
1750      <organization>UNINETT</organization>
1751      <address><email>Harald.T.Alvestrand@uninett.no</email></address>
1752    </author>
1753    <date month="March" year="1995"/>
1754  </front>
1755  <seriesInfo name="RFC" value="1766"/>
1756</reference>
1757
1758<reference anchor="RFC1864">
1759  <front>
1760    <title abbrev="Content-MD5 Header Field">The Content-MD5 Header Field</title>
1761    <author initials="J." surname="Myers" fullname="John G. Myers">
1762      <organization>Carnegie Mellon University</organization>
1763      <address><email>jgm+@cmu.edu</email></address>
1764    </author>
1765    <author initials="M." surname="Rose" fullname="Marshall T. Rose">
1766      <organization>Dover Beach Consulting, Inc.</organization>
1767      <address><email>mrose@dbc.mtview.ca.us</email></address>
1768    </author>
1769    <date month="October" year="1995"/>
1770  </front>
1771  <seriesInfo name="RFC" value="1864"/>
1772</reference>
1773
1774<reference anchor="RFC1950">
1775  <front>
1776    <title>ZLIB Compressed Data Format Specification version 3.3</title>
1777    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
1778      <organization>Aladdin Enterprises</organization>
1779      <address><email>ghost@aladdin.com</email></address>
1780    </author>
1781    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
1782      <organization/>
1783    </author>
1784    <date month="May" year="1996"/>
1785  </front>
1786  <seriesInfo name="RFC" value="1950"/>
1787  <annotation>
1788    RFC1950 is an Informational RFC, thus it may be less stable than
1789    this specification. On the other hand, this downward reference was
1790    present since <xref target="RFC2068"/> (published in 1997), therefore it is unlikely
1791    to cause problems in practice.
1792  </annotation>
1793</reference>
1794
1795<reference anchor="RFC1951">
1796  <front>
1797    <title>DEFLATE Compressed Data Format Specification version 1.3</title>
1798    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
1799      <organization>Aladdin Enterprises</organization>
1800      <address><email>ghost@aladdin.com</email></address>
1801    </author>
1802    <date month="May" year="1996"/>
1803  </front>
1804  <seriesInfo name="RFC" value="1951"/>
1805  <annotation>
1806    RFC1951 is an Informational RFC, thus it may be less stable than
1807    this specification. On the other hand, this downward reference was
1808    present since <xref target="RFC2068"/> (published in 1997), therefore it is unlikely
1809    to cause problems in practice.
1810  </annotation>
1811</reference>
1812
1813<reference anchor="RFC1952">
1814  <front>
1815    <title>GZIP file format specification version 4.3</title>
1816    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
1817      <organization>Aladdin Enterprises</organization>
1818      <address><email>ghost@aladdin.com</email></address>
1819    </author>
1820    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
1821      <organization/>
1822      <address><email>gzip@prep.ai.mit.edu</email></address>
1823    </author>
1824    <author initials="M." surname="Adler" fullname="Mark Adler">
1825      <organization/>
1826      <address><email>madler@alumni.caltech.edu</email></address>
1827    </author>
1828    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
1829      <organization/>
1830      <address><email>ghost@aladdin.com</email></address>
1831    </author>
1832    <author initials="G." surname="Randers-Pehrson" fullname="Glenn Randers-Pehrson">
1833      <organization/>
1834      <address><email>randeg@alumni.rpi.edu</email></address>
1835    </author>
1836    <date month="May" year="1996"/>
1837  </front>
1838  <seriesInfo name="RFC" value="1952"/>
1839  <annotation>
1840    RFC1952 is an Informational RFC, thus it may be less stable than
1841    this specification. On the other hand, this downward reference was
1842    present since <xref target="RFC2068"/> (published in 1997), therefore it is unlikely
1843    to cause problems in practice.
1844  </annotation>
1845</reference>
1846
1847<reference anchor="RFC2045">
1848  <front>
1849    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
1850    <author initials="N." surname="Freed" fullname="Ned Freed">
1851      <organization>Innosoft International, Inc.</organization>
1852      <address><email>ned@innosoft.com</email></address>
1853    </author>
1854    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
1855      <organization>First Virtual Holdings</organization>
1856      <address><email>nsb@nsb.fv.com</email></address>
1857    </author>
1858    <date month="November" year="1996"/>
1859  </front>
1860  <seriesInfo name="RFC" value="2045"/>
1861</reference>
1862
1863<reference anchor="RFC2046">
1864  <front>
1865    <title abbrev="Media Types">Multipurpose Internet Mail Extensions (MIME) Part Two: Media Types</title>
1866    <author initials="N." surname="Freed" fullname="Ned Freed">
1867      <organization>Innosoft International, Inc.</organization>
1868      <address><email>ned@innosoft.com</email></address>
1869    </author>
1870    <author initials="N." surname="Borenstein" fullname="Nathaniel S. Borenstein">
1871      <organization>First Virtual Holdings</organization>
1872      <address><email>nsb@nsb.fv.com</email></address>
1873    </author>
1874    <date month="November" year="1996"/>
1875  </front>
1876  <seriesInfo name="RFC" value="2046"/>
1877</reference>
1878
1879<reference anchor="RFC2119">
1880  <front>
1881    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
1882    <author initials="S." surname="Bradner" fullname="Scott Bradner">
1883      <organization>Harvard University</organization>
1884      <address><email>sob@harvard.edu</email></address>
1885    </author>
1886    <date month="March" year="1997"/>
1887  </front>
1888  <seriesInfo name="BCP" value="14"/>
1889  <seriesInfo name="RFC" value="2119"/>
1890</reference>
1891
1892</references>
1893
1894<references title="Informative References">
1895
1896<reference anchor="RFC1806">
1897  <front>
1898    <title abbrev="Content-Disposition">Communicating Presentation Information in Internet Messages: The Content-Disposition Header</title>
1899    <author initials="R." surname="Troost" fullname="Rens Troost">
1900      <organization>New Century Systems</organization>
1901      <address><email>rens@century.com</email></address>
1902    </author>
1903    <author initials="S." surname="Dorner" fullname="Steve Dorner">
1904      <organization>QUALCOMM Incorporated</organization>
1905      <address><email>sdorner@qualcomm.com</email></address>
1906    </author>
1907    <date month="June" year="1995"/>
1908  </front>
1909  <seriesInfo name="RFC" value="1806"/>
1910</reference>
1911
1912<reference anchor="RFC1945">
1913  <front>
1914    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
1915    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1916      <organization>MIT, Laboratory for Computer Science</organization>
1917      <address><email>timbl@w3.org</email></address>
1918    </author>
1919    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
1920      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
1921      <address><email>fielding@ics.uci.edu</email></address>
1922    </author>
1923    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
1924      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
1925      <address><email>frystyk@w3.org</email></address>
1926    </author>
1927    <date month="May" year="1996"/>
1928  </front>
1929  <seriesInfo name="RFC" value="1945"/>
1930</reference>
1931
1932<reference anchor="RFC2049">
1933  <front>
1934    <title abbrev="MIME Conformance">Multipurpose Internet Mail Extensions (MIME) Part Five: Conformance Criteria and Examples</title>
1935    <author initials="N." surname="Freed" fullname="Ned Freed">
1936      <organization>Innosoft International, Inc.</organization>
1937      <address><email>ned@innosoft.com</email></address>
1938    </author>
1939    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
1940      <organization>First Virtual Holdings</organization>
1941      <address><email>nsb@nsb.fv.com</email></address>
1942    </author>
1943    <date month="November" year="1996"/>
1944  </front>
1945  <seriesInfo name="RFC" value="2049"/>
1946</reference>
1947
1948<reference anchor="RFC2068">
1949  <front>
1950    <title abbrev="HTTP/1.1">Hypertext Transfer Protocol -- HTTP/1.1</title>
1951    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
1952      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
1953      <address><email>fielding@ics.uci.edu</email></address>
1954    </author>
1955    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1956      <organization>MIT Laboratory for Computer Science</organization>
1957      <address><email>jg@w3.org</email></address>
1958    </author>
1959    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1960      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
1961      <address><email>mogul@wrl.dec.com</email></address>
1962    </author>
1963    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
1964      <organization>MIT Laboratory for Computer Science</organization>
1965      <address><email>frystyk@w3.org</email></address>
1966    </author>
1967    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1968      <organization>MIT Laboratory for Computer Science</organization>
1969      <address><email>timbl@w3.org</email></address>
1970    </author>
1971    <date month="January" year="1997"/>
1972  </front>
1973  <seriesInfo name="RFC" value="2068"/>
1974</reference>
1975
1976<reference anchor="RFC2076">
1977  <front>
1978    <title abbrev="Internet Message Headers">Common Internet Message Headers</title>
1979    <author initials="J." surname="Palme" fullname="Jacob Palme">
1980      <organization>Stockholm University/KTH</organization>
1981      <address><email>jpalme@dsv.su.se</email></address>
1982    </author>
1983    <date month="February" year="1997"/>
1984  </front>
1985  <seriesInfo name="RFC" value="2076"/>
1986</reference>
1987
1988<reference anchor="RFC2183">
1989  <front>
1990    <title abbrev="Content-Disposition">Communicating Presentation Information in Internet Messages: The Content-Disposition Header Field</title>
1991    <author initials="R." surname="Troost" fullname="Rens Troost">
1992      <organization>New Century Systems</organization>
1993      <address><email>rens@century.com</email></address>
1994    </author>
1995    <author initials="S." surname="Dorner" fullname="Steve Dorner">
1996      <organization>QUALCOMM Incorporated</organization>
1997      <address><email>sdorner@qualcomm.com</email></address>
1998    </author>
1999    <author initials="K." surname="Moore" fullname="Keith Moore">
2000      <organization>Department of Computer Science</organization>
2001      <address><email>moore@cs.utk.edu</email></address>
2002    </author>
2003    <date month="August" year="1997"/>
2004  </front>
2005  <seriesInfo name="RFC" value="2183"/>
2006</reference>
2007
2008<reference anchor="RFC2277">
2009  <front>
2010    <title abbrev="Charset Policy">IETF Policy on Character Sets and Languages</title>
2011    <author initials="H.T." surname="Alvestrand" fullname="Harald Tveit Alvestrand">
2012      <organization>UNINETT</organization>
2013      <address><email>Harald.T.Alvestrand@uninett.no</email></address>
2014    </author>
2015    <date month="January" year="1998"/>
2016  </front>
2017  <seriesInfo name="BCP" value="18"/>
2018  <seriesInfo name="RFC" value="2277"/>
2019</reference>
2020
2021<reference anchor="RFC2388">
2022  <front>
2023    <title abbrev="multipart/form-data">Returning Values from Forms:  multipart/form-data</title>
2024    <author initials="L." surname="Masinter" fullname="Larry Masinter">
2025      <organization>Xerox Palo Alto Research Center</organization>
2026      <address><email>masinter@parc.xerox.com</email></address>
2027    </author>
2028    <date year="1998" month="August"/>
2029  </front>
2030  <seriesInfo name="RFC" value="2388"/>
2031</reference>
2032
2033<reference anchor="RFC2557">
2034  <front>
2035    <title abbrev="MIME Encapsulation of Aggregate Documents">MIME Encapsulation of Aggregate Documents, such as HTML (MHTML)</title>
2036    <author initials="F." surname="Palme" fullname="Jacob Palme">
2037      <organization>Stockholm University and KTH</organization>
2038      <address><email>jpalme@dsv.su.se</email></address>
2039    </author>
2040    <author initials="A." surname="Hopmann" fullname="Alex Hopmann">
2041      <organization>Microsoft Corporation</organization>
2042      <address><email>alexhop@microsoft.com</email></address>
2043    </author>
2044    <author initials="N." surname="Shelness" fullname="Nick Shelness">
2045      <organization>Lotus Development Corporation</organization>
2046      <address><email>Shelness@lotus.com</email></address>
2047    </author>
2048    <author initials="E." surname="Stefferud" fullname="Einar Stefferud">
2049      <organization/>
2050      <address><email>stef@nma.com</email></address>
2051    </author>
2052    <date year="1999" month="March"/>
2053  </front>
2054  <seriesInfo name="RFC" value="2557"/>
2055</reference>
2056
2057<reference anchor="RFC2616">
2058  <front>
2059    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
2060    <author initials="R." surname="Fielding" fullname="R. Fielding">
2061      <organization>University of California, Irvine</organization>
2062      <address><email>fielding@ics.uci.edu</email></address>
2063    </author>
2064    <author initials="J." surname="Gettys" fullname="J. Gettys">
2065      <organization>W3C</organization>
2066      <address><email>jg@w3.org</email></address>
2067    </author>
2068    <author initials="J." surname="Mogul" fullname="J. Mogul">
2069      <organization>Compaq Computer Corporation</organization>
2070      <address><email>mogul@wrl.dec.com</email></address>
2071    </author>
2072    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
2073      <organization>MIT Laboratory for Computer Science</organization>
2074      <address><email>frystyk@w3.org</email></address>
2075    </author>
2076    <author initials="L." surname="Masinter" fullname="L. Masinter">
2077      <organization>Xerox Corporation</organization>
2078      <address><email>masinter@parc.xerox.com</email></address>
2079    </author>
2080    <author initials="P." surname="Leach" fullname="P. Leach">
2081      <organization>Microsoft Corporation</organization>
2082      <address><email>paulle@microsoft.com</email></address>
2083    </author>
2084    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
2085      <organization>W3C</organization>
2086      <address><email>timbl@w3.org</email></address>
2087    </author>
2088    <date month="June" year="1999"/>
2089  </front>
2090  <seriesInfo name="RFC" value="2616"/>
2091</reference>
2092
2093<reference anchor="RFC2822">
2094  <front>
2095    <title>Internet Message Format</title>
2096    <author initials="P." surname="Resnick" fullname="P. Resnick">
2097      <organization>QUALCOMM Incorporated</organization>
2098    </author>
2099    <date year="2001" month="April"/>
2100  </front> 
2101  <seriesInfo name="RFC" value="2822"/>
2102</reference>
2103
2104<reference anchor="RFC3629">
2105  <front>
2106    <title>UTF-8, a transformation format of ISO 10646</title>
2107    <author initials="F." surname="Yergeau" fullname="F. Yergeau">
2108      <organization>Alis Technologies</organization>
2109      <address><email>fyergeau@alis.com</email></address>
2110    </author>
2111    <date month="November" year="2003"/>
2112  </front>
2113  <seriesInfo name="RFC" value="3629"/>
2114  <seriesInfo name="STD" value="63"/>
2115</reference>
2116
2117<reference anchor="RFC4288">
2118  <front>
2119    <title>Media Type Specifications and Registration Procedures</title>
2120    <author initials="N." surname="Freed" fullname="N. Freed">
2121      <organization>Sun Microsystems</organization>
2122      <address>
2123        <email>ned.freed@mrochek.com</email>
2124      </address>
2125    </author>
2126    <author initials="J." surname="Klensin" fullname="J. Klensin">
2127      <organization/>
2128      <address>
2129        <email>klensin+ietf@jck.com</email>
2130      </address>
2131    </author>
2132    <date year="2005" month="December"/>
2133  </front>
2134  <seriesInfo name="BCP" value="13"/>
2135  <seriesInfo name="RFC" value="4288"/>
2136</reference>
2137
2138</references>
2139
2140<section title="Differences Between HTTP Entities and RFC 2045 Entities" anchor="differences.between.http.entities.and.rfc.2045.entities">
2141<t>
2142   HTTP/1.1 uses many of the constructs defined for Internet Mail (<xref target="RFC2822"/>) and the Multipurpose Internet Mail Extensions (MIME <xref target="RFC2045"/>) to
2143   allow entities to be transmitted in an open variety of
2144   representations and with extensible mechanisms. However, RFC 2045
2145   discusses mail, and HTTP has a few features that are different from
2146   those described in RFC 2045. These differences were carefully chosen
2147   to optimize performance over binary connections, to allow greater
2148   freedom in the use of new media types, to make date comparisons
2149   easier, and to acknowledge the practice of some early HTTP servers
2150   and clients.
2151</t>
2152<t>
2153   This appendix describes specific areas where HTTP differs from RFC
2154   2045. Proxies and gateways to strict MIME environments &SHOULD; be
2155   aware of these differences and provide the appropriate conversions
2156   where necessary. Proxies and gateways from MIME environments to HTTP
2157   also need to be aware of the differences because some conversions
2158   might be required.
2159</t>
2160<section title="MIME-Version" anchor="mime-version">
2161<t>
2162   HTTP is not a MIME-compliant protocol. However, HTTP/1.1 messages &MAY;
2163   include a single MIME-Version general-header field to indicate what
2164   version of the MIME protocol was used to construct the message. Use
2165   of the MIME-Version header field indicates that the message is in
2166   full compliance with the MIME protocol (as defined in <xref target="RFC2045"/>).
2167   Proxies/gateways are responsible for ensuring full compliance (where
2168   possible) when exporting HTTP messages to strict MIME environments.
2169</t>
2170<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="MIME-Version"/>
2171  MIME-Version   = "MIME-Version" ":" 1*DIGIT "." 1*DIGIT
2172</artwork></figure>
2173<t>
2174   MIME version "1.0" is the default for use in HTTP/1.1. However,
2175   HTTP/1.1 message parsing and semantics are defined by this document
2176   and not the MIME specification.
2177</t>
2178</section>
2179
2180<section title="Conversion to Canonical Form" anchor="conversion.to.canonical.form">
2181<t>
2182   <xref target="RFC2045"/> requires that an Internet mail entity be converted to
2183   canonical form prior to being transferred, as described in <xref target="RFC2049" x:fmt="of" x:sec="4"/>.
2184   <xref target="canonicalization.and.text.defaults"/> of this document describes the forms
2185   allowed for subtypes of the "text" media type when transmitted over
2186   HTTP. <xref target="RFC2046"/> requires that content with a type of "text" represent
2187   line breaks as CRLF and forbids the use of CR or LF outside of line
2188   break sequences. HTTP allows CRLF, bare CR, and bare LF to indicate a
2189   line break within text content when a message is transmitted over
2190   HTTP.
2191</t>
2192<t>
2193   Where it is possible, a proxy or gateway from HTTP to a strict MIME
2194   environment &SHOULD; translate all line breaks within the text media
2195   types described in <xref target="canonicalization.and.text.defaults"/> of this document to the RFC 2049
2196   canonical form of CRLF. Note, however, that this might be complicated
2197   by the presence of a Content-Encoding and by the fact that HTTP
2198   allows the use of some character sets which do not use octets 13 and
2199   10 to represent CR and LF, as is the case for some multi-byte
2200   character sets.
2201</t>
2202<t>
2203   Implementors should note that conversion will break any cryptographic
2204   checksums applied to the original content unless the original content
2205   is already in canonical form. Therefore, the canonical form is
2206   recommended for any content that uses such checksums in HTTP.
2207</t>
2208</section>
2209
2210<section title="Introduction of Content-Encoding" anchor="introduction.of.content-encoding">
2211<t>
2212   RFC 2045 does not include any concept equivalent to HTTP/1.1's
2213   Content-Encoding header field. Since this acts as a modifier on the
2214   media type, proxies and gateways from HTTP to MIME-compliant
2215   protocols &MUST; either change the value of the Content-Type header
2216   field or decode the entity-body before forwarding the message. (Some
2217   experimental applications of Content-Type for Internet mail have used
2218   a media-type parameter of ";conversions=&lt;content-coding&gt;" to perform
2219   a function equivalent to Content-Encoding. However, this parameter is
2220   not part of RFC 2045).
2221</t>
2222</section>
2223
2224<section title="No Content-Transfer-Encoding" anchor="no.content-transfer-encoding">
2225<t>
2226   HTTP does not use the Content-Transfer-Encoding field of RFC
2227   2045. Proxies and gateways from MIME-compliant protocols to HTTP &MUST;
2228   remove any Content-Transfer-Encoding
2229   prior to delivering the response message to an HTTP client.
2230</t>
2231<t>
2232   Proxies and gateways from HTTP to MIME-compliant protocols are
2233   responsible for ensuring that the message is in the correct format
2234   and encoding for safe transport on that protocol, where "safe
2235   transport" is defined by the limitations of the protocol being used.
2236   Such a proxy or gateway &SHOULD; label the data with an appropriate
2237   Content-Transfer-Encoding if doing so will improve the likelihood of
2238   safe transport over the destination protocol.
2239</t>
2240</section>
2241
2242<section title="Introduction of Transfer-Encoding" anchor="introduction.of.transfer-encoding">
2243<t>
2244   HTTP/1.1 introduces the Transfer-Encoding header field (&header-transfer-encoding;).
2245   Proxies/gateways &MUST; remove any transfer-coding prior to
2246   forwarding a message via a MIME-compliant protocol.
2247</t>
2248</section>
2249
2250<section title="MHTML and Line Length Limitations" anchor="mhtml.line.length">
2251<t>
2252   HTTP implementations which share code with MHTML <xref target="RFC2557"/> implementations
2253   need to be aware of MIME line length limitations. Since HTTP does not
2254   have this limitation, HTTP does not fold long lines. MHTML messages
2255   being transported by HTTP follow all conventions of MHTML, including
2256   line length limitations and folding, canonicalization, etc., since
2257   HTTP transports all message-bodies as payload (see <xref target="multipart.types"/>) and
2258   does not interpret the content or any MIME header lines that might be
2259   contained therein.
2260</t>
2261</section>
2262</section>
2263
2264<section title="Additional Features" anchor="additional.features">
2265<t>
2266   <xref target="RFC1945"/> and <xref target="RFC2068"/> document protocol elements used by some
2267   existing HTTP implementations, but not consistently and correctly
2268   across most HTTP/1.1 applications. Implementors are advised to be
2269   aware of these features, but cannot rely upon their presence in, or
2270   interoperability with, other HTTP/1.1 applications. Some of these
2271   describe proposed experimental features, and some describe features
2272   that experimental deployment found lacking that are now addressed in
2273   the base HTTP/1.1 specification.
2274</t>
2275<t>
2276   A number of other headers, such as Content-Disposition and Title,
2277   from SMTP and MIME are also often implemented (see <xref target="RFC2076"/>).
2278</t>
2279
2280<section title="Content-Disposition" anchor="content-disposition">
2281<iref item="Headers" subitem="Content-Disposition" primary="true" x:for-anchor=""/>
2282<iref item="Content-Disposition header" primary="true" x:for-anchor=""/>
2283<t>
2284   The Content-Disposition response-header field has been proposed as a
2285   means for the origin server to suggest a default filename if the user
2286   requests that the content is saved to a file. This usage is derived
2287   from the definition of Content-Disposition in <xref target="RFC1806"/>.
2288</t>
2289<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="content-disposition"/><iref primary="true" item="Grammar" subitem="disposition-type"/><iref primary="true" item="Grammar" subitem="disposition-parm"/><iref primary="true" item="Grammar" subitem="filename-parm"/><iref primary="true" item="Grammar" subitem="disp-extension-token"/><iref primary="true" item="Grammar" subitem="disp-extension-parm"/>
2290  content-disposition = "Content-Disposition" ":"
2291                        disposition-type *( ";" disposition-parm )
2292  disposition-type = "attachment" | disp-extension-token
2293  disposition-parm = filename-parm | disp-extension-parm
2294  filename-parm = "filename" "=" quoted-string
2295  disp-extension-token = token
2296  disp-extension-parm = token "=" ( token | quoted-string )
2297</artwork></figure>
2298<t>
2299   An example is
2300</t>
2301<figure><artwork type="example">
2302     Content-Disposition: attachment; filename="fname.ext"
2303</artwork></figure>
2304<t>
2305   The receiving user agent &SHOULD-NOT;  respect any directory path
2306   information present in the filename-parm parameter, which is the only
2307   parameter believed to apply to HTTP implementations at this time. The
2308   filename &SHOULD; be treated as a terminal component only.
2309</t>
2310<t>
2311   If this header is used in a response with the application/octet-stream
2312   content-type, the implied suggestion is that the user agent
2313   should not display the response, but directly enter a `save response
2314   as...' dialog.
2315</t>
2316<t>
2317   See <xref target="content-disposition.issues"/> for Content-Disposition security issues.
2318</t>
2319</section>
2320</section>
2321
2322<section title="Compatibility with Previous Versions" anchor="compatibility">
2323<section title="Changes from RFC 2068" anchor="changes.from.rfc.2068">
2324<t>
2325   Transfer-coding and message lengths all interact in ways that
2326   required fixing exactly when chunked encoding is used (to allow for
2327   transfer encoding that may not be self delimiting); it was important
2328   to straighten out exactly how message lengths are computed.
2329   (<xref target="entity.length"/>, see also <xref target="Part1"/>,
2330   <xref target="Part5"/> and <xref target="Part6"/>).
2331</t>
2332<t>
2333   Charset wildcarding is introduced to avoid explosion of character set
2334   names in accept headers. (<xref target="header.accept-charset"/>)
2335</t>
2336<t>
2337   Content-Base was deleted from the specification: it was not
2338   implemented widely, and there is no simple, safe way to introduce it
2339   without a robust extension mechanism. In addition, it is used in a
2340   similar, but not identical fashion in MHTML <xref target="RFC2557"/>.
2341</t>
2342<t>
2343   A content-coding of "identity" was introduced, to solve problems
2344   discovered in caching. (<xref target="content.codings"/>)
2345</t>
2346<t>
2347   Quality Values of zero should indicate that "I don't want something"
2348   to allow clients to refuse a representation. (<xref target="quality.values"/>)
2349</t>
2350<t>
2351   The Alternates<iref item="Alternates header" primary="true"/><iref item="Headers" subitem="Alternate" primary="true"/>, Content-Version<iref item="Content-Version header" primary="true"/><iref item="Headers" subitem="Content-Version" primary="true"/>, Derived-From<iref item="Derived-From header" primary="true"/><iref item="Headers" subitem="Derived-From" primary="true"/>, Link<iref item="Link header" primary="true"/><iref item="Headers" subitem="Link" primary="true"/>, URI<iref item="URI header" primary="true"/><iref item="Headers" subitem="URI" primary="true"/>, Public<iref item="Public header" primary="true"/><iref item="Headers" subitem="Public" primary="true"/> and
2352   Content-Base<iref item="Content-Base header" primary="true"/><iref item="Headers" subitem="Content-Base" primary="true"/> header fields were defined in previous versions of this
2353   specification, but not commonly implemented. See <xref target="RFC2068"/>.
2354</t>
2355</section>
2356
2357<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
2358<t>
2359  Clarify contexts that charset is used in.
2360  (<xref target="character.sets"/>)
2361</t>
2362<t>
2363  Remove reference to non-existant identity transfer-coding value tokens.
2364  (<xref target="no.content-transfer-encoding"/>)
2365</t>
2366</section>
2367
2368</section>
2369
2370<section title="Change Log (to be removed by RFC Editor before publication)">
2371
2372<section title="Since RFC2616">
2373<t>
2374  Extracted relevant partitions from <xref target="RFC2616"/>.
2375</t>
2376</section>
2377
2378<section title="Since draft-ietf-httpbis-p3-payload-00">
2379<t>
2380  Closed issues:
2381  <list style="symbols"> 
2382    <t>
2383      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/8"/>:
2384      "Media Type Registrations"
2385      (<eref target="http://purl.org/NET/http-errata#media-reg"/>)
2386    </t>
2387    <t>
2388      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/14"/>:
2389      "Clarification regarding quoting of charset values"
2390      (<eref target="http://purl.org/NET/http-errata#charactersets"/>)
2391    </t>
2392    <t>
2393      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/16"/>:
2394      "Remove 'identity' token references"
2395      (<eref target="http://purl.org/NET/http-errata#identity"/>)
2396    </t>
2397    <t>
2398      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/25"/>:
2399      "Accept-Encoding BNF"
2400    </t>
2401    <t>
2402      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/35"/>:
2403      "Normative and Informative references"
2404    </t>
2405    <t>
2406      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/46"/>:
2407      "RFC1700 references"
2408    </t>
2409    <t>
2410      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/55"/>:
2411      "Updating to RFC4288"
2412    </t>
2413    <t>
2414      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/65"/>:
2415      "Informative references"
2416    </t>
2417    <t>
2418      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/66"/>:
2419      "ISO-8859-1 Reference"
2420    </t>
2421    <t>
2422      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/68"/>:
2423      "Encoding References Normative"
2424    </t>
2425    <t>
2426      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/86"/>:
2427      "Normative up-to-date references"
2428    </t>
2429  </list>
2430</t>
2431</section>
2432
2433<section title="Since draft-ietf-httpbis-p3-payload-01">
2434<t>
2435  Ongoing work on ABNF conversion (<eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/36"/>):
2436  <list style="symbols"> 
2437    <t>
2438      Add explicit references to BNF syntax and rules imported from other parts of the specification.
2439    </t>
2440  </list>
2441</t>
2442</section>
2443
2444</section>
2445
2446</back>
2447</rfc>
Note: See TracBrowser for help on using the repository browser.