source: draft-ietf-httpbis/latest/p3-payload.xml @ 136

Last change on this file since 136 was 136, checked in by julian.reschke@…, 12 years ago

Add missing whitespace in BNF (addresses #36)

  • Property svn:eol-style set to native
File size: 104.5 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 "January">
16  <!ENTITY ID-YEAR "2008">
17  <!ENTITY caching-neg-resp         "<xref target='Part6' x:rel='#caching.negotiated.responses' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
18  <!ENTITY header-transfer-encoding "<xref target='Part1' x:rel='#header.transfer-encoding' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
19  <!ENTITY header-allow             "<xref target='Part2' x:rel='#header.allow' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
20  <!ENTITY header-content-length    "<xref target='Part1' x:rel='#header.content-length' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
21  <!ENTITY header-content-range     "<xref target='Part5' x:rel='#header.content-range' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
22  <!ENTITY header-expires           "<xref target='Part6' x:rel='#header.expires' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
23  <!ENTITY header-last-modified     "<xref target='Part4' x:rel='#header.last-modified' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
24  <!ENTITY header-user-agent        "<xref target='Part2' x:rel='#header.user-agent' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
25  <!ENTITY header-vary              "<xref target='Part6' x:rel='#header.vary' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
26  <!ENTITY message-body             "<xref target='Part1' x:rel='#message.body' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
27  <!ENTITY message-length           "<xref target='Part1' x:rel='#message.length' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
28  <!ENTITY multipart-byteranges     "<xref target='Part5' x:rel='#internet.media.type.multipart.byteranges' xmlns:x='http://purl.org/net/xml2rfc/ext'/>">
29]>
30<?rfc toc="yes" ?>
31<?rfc symrefs="yes" ?>
32<?rfc sortrefs="yes" ?>
33<?rfc compact="yes"?>
34<?rfc subcompact="no" ?>
35<?rfc linkmailto="no" ?>
36<?rfc editing="no" ?>
37<?rfc-ext allow-markup-in-artwork="yes" ?>
38<?rfc-ext include-references-in-index="yes" ?>
39<rfc obsoletes="2068, 2616" category="std"
40     ipr="full3978" docName="draft-ietf-httpbis-p3-payload-&ID-VERSION;"
41     xmlns:x='http://purl.org/net/xml2rfc/ext' xmlns:ed="http://greenbytes.de/2002/rfcedit">
42<front>
43
44  <title abbrev="HTTP/1.1, Part 3">HTTP/1.1, part 3: Message Payload and Content Negotiation</title>
45
46  <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
47    <organization abbrev="Day Software">Day Software</organization>
48    <address>
49      <postal>
50        <street>23 Corporate Plaza DR, Suite 280</street>
51        <city>Newport Beach</city>
52        <region>CA</region>
53        <code>92660</code>
54        <country>USA</country>
55      </postal>
56      <phone>+1-949-706-5300</phone>
57      <facsimile>+1-949-706-5305</facsimile>
58      <email>fielding@gbiv.com</email>
59      <uri>http://roy.gbiv.com/</uri>
60    </address>
61  </author>
62
63  <author initials="J." surname="Gettys" fullname="Jim Gettys">
64    <organization>One Laptop per Child</organization>
65    <address>
66      <postal>
67        <street>21 Oak Knoll Road</street>
68        <city>Carlisle</city>
69        <region>MA</region>
70        <code>01741</code>
71        <country>USA</country>
72      </postal>
73      <email>jg@laptop.org</email>
74      <uri>http://www.laptop.org/</uri>
75    </address>
76  </author>
77 
78  <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
79    <organization abbrev="HP">Hewlett-Packard Company</organization>
80    <address>
81      <postal>
82        <street>HP Labs, Large Scale Systems Group</street>
83        <street>1501 Page Mill Road, MS 1177</street>
84        <city>Palo Alto</city>
85        <region>CA</region>
86        <code>94304</code>
87        <country>USA</country>
88      </postal>
89      <email>JeffMogul@acm.org</email>
90    </address>
91  </author>
92
93  <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
94    <organization abbrev="Microsoft">Microsoft Corporation</organization>
95    <address>
96      <postal>
97        <street>1 Microsoft Way</street>
98        <city>Redmond</city>
99        <region>WA</region>
100        <code>98052</code>
101        <country>USA</country>
102      </postal>
103      <email>henrikn@microsoft.com</email>
104    </address>
105  </author>
106
107  <author initials="L." surname="Masinter" fullname="Larry Masinter">
108    <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
109    <address>
110      <postal>
111        <street>345 Park Ave</street>
112        <city>San Jose</city>
113        <region>CA</region>
114        <code>95110</code>
115        <country>USA</country>
116      </postal>
117      <email>LMM@acm.org</email>
118      <uri>http://larry.masinter.net/</uri>
119    </address>
120  </author>
121 
122  <author initials="P." surname="Leach" fullname="Paul J. Leach">
123    <organization abbrev="Microsoft">Microsoft Corporation</organization>
124    <address>
125      <postal>
126        <street>1 Microsoft Way</street>
127        <city>Redmond</city>
128        <region>WA</region>
129        <code>98052</code>
130      </postal>
131      <email>paulle@microsoft.com</email>
132    </address>
133  </author>
134   
135  <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
136    <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
137    <address>
138      <postal>
139        <street>MIT Computer Science and Artificial Intelligence Laboratory</street>
140        <street>The Stata Center, Building 32</street>
141        <street>32 Vassar Street</street>
142        <city>Cambridge</city>
143        <region>MA</region>
144        <code>02139</code>
145        <country>USA</country>
146      </postal>
147      <email>timbl@w3.org</email>
148      <uri>http://www.w3.org/People/Berners-Lee/</uri>
149    </address>
150  </author>
151
152  <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
153    <organization abbrev="W3C">World Wide Web Consortium</organization>
154    <address>
155      <postal>
156        <street>W3C / ERCIM</street>
157        <street>2004, rte des Lucioles</street>
158        <city>Sophia-Antipolis</city>
159        <region>AM</region>
160        <code>06902</code>
161        <country>France</country>
162      </postal>
163      <email>ylafon@w3.org</email>
164      <uri>http://www.raubacapeu.net/people/yves/</uri>
165    </address>
166  </author>
167
168  <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
169    <organization abbrev="greenbytes">greenbytes GmbH</organization>
170    <address>
171      <postal>
172        <street>Hafenweg 16</street>
173        <city>Muenster</city><region>NW</region><code>48155</code>
174        <country>Germany</country>
175      </postal>
176      <phone>+49 251 2807760</phone>   
177      <facsimile>+49 251 2807761</facsimile>   
178      <email>julian.reschke@greenbytes.de</email>       
179      <uri>http://greenbytes.de/tech/webdav/</uri>     
180    </address>
181  </author>
182
183  <date month="&ID-MONTH;" year="&ID-YEAR;"/>
184
185<abstract>
186<t>
187   The Hypertext Transfer Protocol (HTTP) is an application-level
188   protocol for distributed, collaborative, hypermedia information
189   systems. HTTP has been in use by the World Wide Web global information
190   initiative since 1990. This document is Part 3 of the seven-part specification
191   that defines the protocol referred to as "HTTP/1.1" and, taken together,
192   obsoletes RFC 2616.  Part 3 defines HTTP message content,
193   metadata, and content negotiation.
194</t>
195</abstract>
196
197<note title="Editorial Note (To be removed by RFC Editor)">
198  <t>
199    Discussion of this draft should take place on the HTTPBIS working group
200    mailing list (ietf-http-wg@w3.org). The current issues list is
201    at <eref target="http://www.tools.ietf.org/wg/httpbis/trac/report/11"/>
202    and related documents (including fancy diffs) can be found at
203    <eref target="http://www.tools.ietf.org/wg/httpbis/"/>.
204  </t>
205</note>
206</front>
207<middle>
208<section title="Introduction" anchor="introduction">
209<t>
210   This document will define aspects of HTTP related to the payload of
211   messages (message content), including metadata and media types, along
212   with HTTP content negotiation.  Right now it only includes the extracted
213   relevant sections of RFC 2616 without edit.
214</t>
215
216<section title="Requirements" anchor="intro.requirements">
217<t>
218   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
219   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
220   document are to be interpreted as described in <xref target="RFC2119"/>.
221</t>
222<t>
223   An implementation is not compliant if it fails to satisfy one or more
224   of the &MUST; or &REQUIRED; level requirements for the protocols it
225   implements. An implementation that satisfies all the &MUST; or &REQUIRED;
226   level and all the &SHOULD; level requirements for its protocols is said
227   to be "unconditionally compliant"; one that satisfies all the &MUST;
228   level requirements but not all the &SHOULD; level requirements for its
229   protocols is said to be "conditionally compliant."
230</t>
231</section>
232</section>
233
234<section title="Protocol Parameters" anchor="protocol.parameters">
235
236<section title="Character Sets" anchor="character.sets">
237<t>
238   HTTP uses the same definition of the term "character set" as that
239   described for MIME:
240</t>
241<t>
242   The term "character set" is used in this document to refer to a
243   method used with one or more tables to convert a sequence of octets
244   into a sequence of characters. Note that unconditional conversion in
245   the other direction is not required, in that not all characters may
246   be available in a given character set and a character set may provide
247   more than one sequence of octets to represent a particular character.
248   This definition is intended to allow various kinds of character
249   encoding, from simple single-table mappings such as US-ASCII to
250   complex table switching methods such as those that use ISO-2022's
251   techniques. However, the definition associated with a MIME character
252   set name &MUST; fully specify the mapping to be performed from octets
253   to characters. In particular, use of external profiling information
254   to determine the exact mapping is not permitted.
255</t>
256<t><list><t>
257      <x:h>Note:</x:h> This use of the term "character set" is more commonly
258      referred to as a "character encoding." However, since HTTP and
259      MIME share the same registry, it is important that the terminology
260      also be shared.
261</t></list></t>
262<t>
263   HTTP character sets are identified by case-insensitive tokens. The
264   complete set of tokens is defined by the IANA Character Set registry
265   (<eref target="http://www.iana.org/assignments/character-sets"/>).
266</t>
267<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="charset"/>
268  charset = token
269</artwork></figure>
270<t>
271   Although HTTP allows an arbitrary token to be used as a charset
272   value, any token that has a predefined value within the IANA
273   Character Set registry &MUST; represent the character set defined
274   by that registry. Applications &SHOULD; limit their use of character
275   sets to those defined by the IANA registry.
276</t>
277<t>
278   HTTP uses charset in two contexts: within an Accept-Charset request
279   header (in which the charset value is an unquoted token) and as the
280   value of a parameter in a Content-Type header (within a request or
281   response), in which case the parameter value of the charset parameter
282   may be quoted.
283</t>
284<t>
285   Implementors should be aware of IETF character set requirements <xref target="RFC3629"/>
286   <xref target="RFC2277"/>.
287</t>
288
289<section title="Missing Charset" anchor="missing.charset">
290<t>
291   Some HTTP/1.0 software has interpreted a Content-Type header without
292   charset parameter incorrectly to mean "recipient should guess."
293   Senders wishing to defeat this behavior &MAY; include a charset
294   parameter even when the charset is ISO-8859-1 (<xref target="ISO-8859-1"/>) and &SHOULD; do so when
295   it is known that it will not confuse the recipient.
296</t>
297<t>
298   Unfortunately, some older HTTP/1.0 clients did not deal properly with
299   an explicit charset parameter. HTTP/1.1 recipients &MUST; respect the
300   charset label provided by the sender; and those user agents that have
301   a provision to "guess" a charset &MUST; use the charset from the
302   content-type field if they support that charset, rather than the
303   recipient's preference, when initially displaying a document. See
304   <xref target="canonicalization.and.text.defaults"/>.
305</t>
306</section>
307</section>
308
309<section title="Content Codings" anchor="content.codings">
310<t>
311   Content coding values indicate an encoding transformation that has
312   been or can be applied to an entity. Content codings are primarily
313   used to allow a document to be compressed or otherwise usefully
314   transformed without losing the identity of its underlying media type
315   and without loss of information. Frequently, the entity is stored in
316   coded form, transmitted directly, and only decoded by the recipient.
317</t>
318<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="content-coding"/>
319  content-coding   = token
320</artwork></figure>
321<t>
322   All content-coding values are case-insensitive. HTTP/1.1 uses
323   content-coding values in the Accept-Encoding (<xref target="header.accept-encoding"/>) and
324   Content-Encoding (<xref target="header.content-encoding"/>) header fields. Although the value
325   describes the content-coding, what is more important is that it
326   indicates what decoding mechanism will be required to remove the
327   encoding.
328</t>
329<t>
330   The Internet Assigned Numbers Authority (IANA) acts as a registry for
331   content-coding value tokens. Initially, the registry contains the
332   following tokens:
333</t>
334<t>
335   gzip<iref item="gzip"/>
336  <list>
337    <t>
338        An encoding format produced by the file compression program
339        "gzip" (GNU zip) as described in <xref target="RFC1952"/>. This format is a
340        Lempel-Ziv coding (LZ77) with a 32 bit CRC.
341    </t>
342  </list>
343</t>
344<t>
345   compress<iref item="compress"/>
346  <list><t>
347        The encoding format produced by the common UNIX file compression
348        program "compress". This format is an adaptive Lempel-Ziv-Welch
349        coding (LZW).
350</t><t>
351        Use of program names for the identification of encoding formats
352        is not desirable and is discouraged for future encodings. Their
353        use here is representative of historical practice, not good
354        design. For compatibility with previous implementations of HTTP,
355        applications &SHOULD; consider "x-gzip" and "x-compress" to be
356        equivalent to "gzip" and "compress" respectively.
357  </t></list>
358</t>
359<t>
360   deflate<iref item="deflate"/>
361  <list><t>
362        The "zlib" format defined in <xref target="RFC1950"/> in combination with
363        the "deflate" compression mechanism described in <xref target="RFC1951"/>.
364  </t></list>
365</t>
366<t>
367   identity<iref item="identity"/>
368  <list><t>
369        The default (identity) encoding; the use of no transformation
370        whatsoever. This content-coding is used only in the Accept-Encoding
371        header, and &SHOULD-NOT;  be used in the Content-Encoding
372        header.
373  </t></list>
374</t>
375<t>
376   New content-coding value tokens &SHOULD; be registered; to allow
377   interoperability between clients and servers, specifications of the
378   content coding algorithms needed to implement a new value &SHOULD; be
379   publicly available and adequate for independent implementation, and
380   conform to the purpose of content coding defined in this section.
381</t>
382</section>
383
384<section title="Media Types" anchor="media.types">
385<t>
386   HTTP uses Internet Media Types <xref target="RFC4288"/> in the Content-Type (<xref target="header.content-type"/>)
387   and Accept (<xref target="header.accept"/>) header fields in order to provide
388   open and extensible data typing and type negotiation.
389</t>
390<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"/>
391  media-type     = type "/" subtype *( ";" parameter )
392  type           = token
393  subtype        = token
394</artwork></figure>
395<t>
396   Parameters &MAY; follow the type/subtype in the form of attribute/value
397   pairs.
398</t>
399<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"/>
400  parameter               = attribute "=" value
401  attribute               = token
402  value                   = token | quoted-string
403</artwork></figure>
404<t>
405   The type, subtype, and parameter attribute names are case-insensitive.
406   Parameter values might or might not be case-sensitive,
407   depending on the semantics of the parameter name. Linear white space
408   (LWS) &MUST-NOT; be used between the type and subtype, nor between an
409   attribute and its value. The presence or absence of a parameter might
410   be significant to the processing of a media-type, depending on its
411   definition within the media type registry.
412</t>
413<t>
414   Note that some older HTTP applications do not recognize media type
415   parameters. When sending data to older HTTP applications,
416   implementations &SHOULD; only use media type parameters when they are
417   required by that type/subtype definition.
418</t>
419<t>
420   Media-type values are registered with the Internet Assigned Number
421   Authority (IANA). The media type registration process is
422   outlined in <xref target="RFC4288"/>. Use of non-registered media types is
423   discouraged.
424</t>
425
426<section title="Canonicalization and Text Defaults" anchor="canonicalization.and.text.defaults">
427<t>
428   Internet media types are registered with a canonical form. An
429   entity-body transferred via HTTP messages &MUST; be represented in the
430   appropriate canonical form prior to its transmission except for
431   "text" types, as defined in the next paragraph.
432</t>
433<t>
434   When in canonical form, media subtypes of the "text" type use CRLF as
435   the text line break. HTTP relaxes this requirement and allows the
436   transport of text media with plain CR or LF alone representing a line
437   break when it is done consistently for an entire entity-body. HTTP
438   applications &MUST; accept CRLF, bare CR, and bare LF as being
439   representative of a line break in text media received via HTTP. In
440   addition, if the text is represented in a character set that does not
441   use octets 13 and 10 for CR and LF respectively, as is the case for
442   some multi-byte character sets, HTTP allows the use of whatever octet
443   sequences are defined by that character set to represent the
444   equivalent of CR and LF for line breaks. This flexibility regarding
445   line breaks applies only to text media in the entity-body; a bare CR
446   or LF &MUST-NOT; be substituted for CRLF within any of the HTTP control
447   structures (such as header fields and multipart boundaries).
448</t>
449<t>
450   If an entity-body is encoded with a content-coding, the underlying
451   data &MUST; be in a form defined above prior to being encoded.
452</t>
453<t>
454   The "charset" parameter is used with some media types to define the
455   character set (<xref target="character.sets"/>) of the data. When no explicit charset
456   parameter is provided by the sender, media subtypes of the "text"
457   type are defined to have a default charset value of "ISO-8859-1" when
458   received via HTTP. Data in character sets other than "ISO-8859-1" or
459   its subsets &MUST; be labeled with an appropriate charset value. See
460   <xref target="missing.charset"/> for compatibility problems.
461</t>
462</section>
463
464<section title="Multipart Types" anchor="multipart.types">
465<t>
466   MIME provides for a number of "multipart" types -- encapsulations of
467   one or more entities within a single message-body. All multipart
468   types share a common syntax, as defined in <xref target="RFC2046" x:sec="5.1.1" x:fmt="of"/>,
469   and &MUST; include a boundary parameter as part of the media type
470   value. The message body is itself a protocol element and &MUST;
471   therefore use only CRLF to represent line breaks between body-parts.
472   Unlike in RFC 2046, the epilogue of any multipart message &MUST; be
473   empty; HTTP applications &MUST-NOT; transmit the epilogue (even if the
474   original multipart contains an epilogue). These restrictions exist in
475   order to preserve the self-delimiting nature of a multipart message-body,
476   wherein the "end" of the message-body is indicated by the
477   ending multipart boundary.
478</t>
479<t>
480   In general, HTTP treats a multipart message-body no differently than
481   any other media type: strictly as payload. The one exception is the
482   "multipart/byteranges" type (&multipart-byteranges;) when it appears in a 206
483   (Partial Content) response.
484   <!-- jre: re-insert removed text pointing to caching? -->
485   In all
486   other cases, an HTTP user agent &SHOULD; follow the same or similar
487   behavior as a MIME user agent would upon receipt of a multipart type.
488   The MIME header fields within each body-part of a multipart message-body
489   do not have any significance to HTTP beyond that defined by
490   their MIME semantics.
491</t>
492<t>
493   In general, an HTTP user agent &SHOULD; follow the same or similar
494   behavior as a MIME user agent would upon receipt of a multipart type.
495   If an application receives an unrecognized multipart subtype, the
496   application &MUST; treat it as being equivalent to "multipart/mixed".
497</t>
498<t><list><t>
499      <x:h>Note:</x:h> The "multipart/form-data" type has been specifically defined
500      for carrying form data suitable for processing via the POST
501      request method, as described in <xref target="RFC2388"/>.
502</t></list></t>
503</section>
504</section>
505
506<section title="Quality Values" anchor="quality.values">
507<t>
508   HTTP content negotiation (<xref target="content.negotiation"/>) uses short "floating point"
509   numbers to indicate the relative importance ("weight") of various
510   negotiable parameters.  A weight is normalized to a real number in
511   the range 0 through 1, where 0 is the minimum and 1 the maximum
512   value. If a parameter has a quality value of 0, then content with
513   this parameter is `not acceptable' for the client. HTTP/1.1
514   applications &MUST-NOT; generate more than three digits after the
515   decimal point. User configuration of these values &SHOULD; also be
516   limited in this fashion.
517</t>
518<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="qvalue"/>
519  qvalue         = ( "0" [ "." 0*3DIGIT ] )
520                 | ( "1" [ "." 0*3("0") ] )
521</artwork></figure>
522<t>
523   "Quality values" is a misnomer, since these values merely represent
524   relative degradation in desired quality.
525</t>
526</section>
527
528<section title="Language Tags" anchor="language.tags">
529<t>
530   A language tag identifies a natural language spoken, written, or
531   otherwise conveyed by human beings for communication of information
532   to other human beings. Computer languages are explicitly excluded.
533   HTTP uses language tags within the Accept-Language and Content-Language
534   fields.
535</t>
536<t>
537   The syntax and registry of HTTP language tags is the same as that
538   defined by <xref target="RFC1766"/>. In summary, a language tag is composed of 1
539   or more parts: A primary language tag and a possibly empty series of
540   subtags:
541</t>
542<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"/>
543  language-tag  = primary-tag *( "-" subtag )
544  primary-tag   = 1*8ALPHA
545  subtag        = 1*8ALPHA
546</artwork></figure>
547<t>
548   White space is not allowed within the tag and all tags are case-insensitive.
549   The name space of language tags is administered by the
550   IANA. Example tags include:
551</t>
552<figure><artwork type="example">
553    en, en-US, en-cockney, i-cherokee, x-pig-latin
554</artwork></figure>
555<t>
556   where any two-letter primary-tag is an ISO-639 language abbreviation
557   and any two-letter initial subtag is an ISO-3166 country code. (The
558   last three tags above are not registered tags; all but the last are
559   examples of tags which could be registered in future.)
560</t>
561</section>
562</section>
563
564<section title="Entity" anchor="entity">
565<t>
566   Request and Response messages &MAY; transfer an entity if not otherwise
567   restricted by the request method or response status code. An entity
568   consists of entity-header fields and an entity-body, although some
569   responses will only include the entity-headers.
570</t>
571<t>
572   In this section, both sender and recipient refer to either the client
573   or the server, depending on who sends and who receives the entity.
574</t>
575
576<section title="Entity Header Fields" anchor="entity.header.fields">
577<t>
578   Entity-header fields define metainformation about the entity-body or,
579   if no body is present, about the resource identified by the request.
580   Some of this metainformation is &OPTIONAL;; some might be &REQUIRED; by
581   portions of this specification.
582</t>
583<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="entity-header"/><iref primary="true" item="Grammar" subitem="extension-header"/>
584  entity-header  = Allow                    ; &header-allow;
585                 | Content-Encoding         ; <xref target="header.content-encoding"/>
586                 | Content-Language         ; <xref target="header.content-language"/>
587                 | Content-Length           ; &header-content-length;
588                 | Content-Location         ; <xref target="header.content-location"/>
589                 | Content-MD5              ; <xref target="header.content-md5"/>
590                 | Content-Range            ; &header-content-range;
591                 | Content-Type             ; <xref target="header.content-type"/>
592                 | Expires                  ; &header-expires;
593                 | Last-Modified            ; &header-last-modified;
594                 | extension-header
595 
596  extension-header = message-header
597</artwork></figure>
598<t>
599   The extension-header mechanism allows additional entity-header fields
600   to be defined without changing the protocol, but these fields cannot
601   be assumed to be recognizable by the recipient. Unrecognized header
602   fields &SHOULD; be ignored by the recipient and &MUST; be forwarded by
603   transparent proxies.
604</t>
605</section>
606
607<section title="Entity Body" anchor="entity.body">
608<t>
609   The entity-body (if any) sent with an HTTP request or response is in
610   a format and encoding defined by the entity-header fields.
611</t>
612<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="entity-body"/>
613  entity-body    = *OCTET
614</artwork></figure>
615<t>
616   An entity-body is only present in a message when a message-body is
617   present, as described in &message-body;. The entity-body is obtained
618   from the message-body by decoding any Transfer-Encoding that might
619   have been applied to ensure safe and proper transfer of the message.
620</t>
621
622<section title="Type" anchor="type">
623<t>
624   When an entity-body is included with a message, the data type of that
625   body is determined via the header fields Content-Type and Content-Encoding.
626   These define a two-layer, ordered encoding model:
627</t>
628<figure><artwork type="example">
629    entity-body := Content-Encoding( Content-Type( data ) )
630</artwork></figure>
631<t>
632   Content-Type specifies the media type of the underlying data.
633   Content-Encoding may be used to indicate any additional content
634   codings applied to the data, usually for the purpose of data
635   compression, that are a property of the requested resource. There is
636   no default encoding.
637</t>
638<t>
639   Any HTTP/1.1 message containing an entity-body &SHOULD; include a
640   Content-Type header field defining the media type of that body. If
641   and only if the media type is not given by a Content-Type field, the
642   recipient &MAY; attempt to guess the media type via inspection of its
643   content and/or the name extension(s) of the URI used to identify the
644   resource. If the media type remains unknown, the recipient &SHOULD;
645   treat it as type "application/octet-stream".
646</t>
647</section>
648   
649<section title="Entity Length" anchor="entity.length">
650<t>
651   The entity-length of a message is the length of the message-body
652   before any transfer-codings have been applied. &message-length; defines
653   how the transfer-length of a message-body is determined.
654</t>
655</section>
656</section>
657</section>
658
659<section title="Content Negotiation" anchor="content.negotiation">
660<t>
661   Most HTTP responses include an entity which contains information for
662   interpretation by a human user. Naturally, it is desirable to supply
663   the user with the "best available" entity corresponding to the
664   request. Unfortunately for servers and caches, not all users have the
665   same preferences for what is "best," and not all user agents are
666   equally capable of rendering all entity types. For that reason, HTTP
667   has provisions for several mechanisms for "content negotiation" --
668   the process of selecting the best representation for a given response
669   when there are multiple representations available.
670  <list><t>
671      <x:h>Note:</x:h> This is not called "format negotiation" because the
672      alternate representations may be of the same media type, but use
673      different capabilities of that type, be in different languages,
674      etc.
675  </t></list>
676</t>
677<t>
678   Any response containing an entity-body &MAY; be subject to negotiation,
679   including error responses.
680</t>
681<t>
682   There are two kinds of content negotiation which are possible in
683   HTTP: server-driven and agent-driven negotiation. These two kinds of
684   negotiation are orthogonal and thus may be used separately or in
685   combination. One method of combination, referred to as transparent
686   negotiation, occurs when a cache uses the agent-driven negotiation
687   information provided by the origin server in order to provide
688   server-driven negotiation for subsequent requests.
689</t>
690
691<section title="Server-driven Negotiation" anchor="server-driven.negotiation">
692<t>
693   If the selection of the best representation for a response is made by
694   an algorithm located at the server, it is called server-driven
695   negotiation. Selection is based on the available representations of
696   the response (the dimensions over which it can vary; e.g. language,
697   content-coding, etc.) and the contents of particular header fields in
698   the request message or on other information pertaining to the request
699   (such as the network address of the client).
700</t>
701<t>
702   Server-driven negotiation is advantageous when the algorithm for
703   selecting from among the available representations is difficult to
704   describe to the user agent, or when the server desires to send its
705   "best guess" to the client along with the first response (hoping to
706   avoid the round-trip delay of a subsequent request if the "best
707   guess" is good enough for the user). In order to improve the server's
708   guess, the user agent &MAY; include request header fields (Accept,
709   Accept-Language, Accept-Encoding, etc.) which describe its
710   preferences for such a response.
711</t>
712<t>
713   Server-driven negotiation has disadvantages:
714  <list style="numbers">
715    <t>
716         It is impossible for the server to accurately determine what
717         might be "best" for any given user, since that would require
718         complete knowledge of both the capabilities of the user agent
719         and the intended use for the response (e.g., does the user want
720         to view it on screen or print it on paper?).
721    </t>
722    <t>
723         Having the user agent describe its capabilities in every
724         request can be both very inefficient (given that only a small
725         percentage of responses have multiple representations) and a
726         potential violation of the user's privacy.
727    </t>
728    <t>
729         It complicates the implementation of an origin server and the
730         algorithms for generating responses to a request.
731    </t>
732    <t>
733         It may limit a public cache's ability to use the same response
734         for multiple user's requests.
735    </t>
736  </list>
737</t>
738<t>
739   HTTP/1.1 includes the following request-header fields for enabling
740   server-driven negotiation through description of user agent
741   capabilities and user preferences: Accept (<xref target="header.accept"/>), Accept-Charset
742   (<xref target="header.accept-charset"/>), Accept-Encoding (<xref target="header.accept-encoding"/>), Accept-Language
743   (<xref target="header.accept-language"/>), and User-Agent (&header-user-agent;). However, an
744   origin server is not limited to these dimensions and &MAY; vary the
745   response based on any aspect of the request, including information
746   outside the request-header fields or within extension header fields
747   not defined by this specification.
748</t>
749<t>
750   The Vary header field (&header-vary;) can be used to express the parameters the
751   server uses to select a representation that is subject to server-driven
752   negotiation.
753</t>
754</section>
755
756<section title="Agent-driven Negotiation" anchor="agent-driven.negotiation">
757<t>
758   With agent-driven negotiation, selection of the best representation
759   for a response is performed by the user agent after receiving an
760   initial response from the origin server. Selection is based on a list
761   of the available representations of the response included within the
762   header fields or entity-body of the initial response, with each
763   representation identified by its own URI. Selection from among the
764   representations may be performed automatically (if the user agent is
765   capable of doing so) or manually by the user selecting from a
766   generated (possibly hypertext) menu.
767</t>
768<t>
769   Agent-driven negotiation is advantageous when the response would vary
770   over commonly-used dimensions (such as type, language, or encoding),
771   when the origin server is unable to determine a user agent's
772   capabilities from examining the request, and generally when public
773   caches are used to distribute server load and reduce network usage.
774</t>
775<t>
776   Agent-driven negotiation suffers from the disadvantage of needing a
777   second request to obtain the best alternate representation. This
778   second request is only efficient when caching is used. In addition,
779   this specification does not define any mechanism for supporting
780   automatic selection, though it also does not prevent any such
781   mechanism from being developed as an extension and used within
782   HTTP/1.1.
783</t>
784<t>
785   HTTP/1.1 defines the 300 (Multiple Choices) and 406 (Not Acceptable)
786   status codes for enabling agent-driven negotiation when the server is
787   unwilling or unable to provide a varying response using server-driven
788   negotiation.
789</t>
790</section>
791
792<section title="Transparent Negotiation" anchor="transparent.negotiation">
793<t>
794   Transparent negotiation is a combination of both server-driven and
795   agent-driven negotiation. When a cache is supplied with a form of the
796   list of available representations of the response (as in agent-driven
797   negotiation) and the dimensions of variance are completely understood
798   by the cache, then the cache becomes capable of performing server-driven
799   negotiation on behalf of the origin server for subsequent
800   requests on that resource.
801</t>
802<t>
803   Transparent negotiation has the advantage of distributing the
804   negotiation work that would otherwise be required of the origin
805   server and also removing the second request delay of agent-driven
806   negotiation when the cache is able to correctly guess the right
807   response.
808</t>
809<t>
810   This specification does not define any mechanism for transparent
811   negotiation, though it also does not prevent any such mechanism from
812   being developed as an extension that could be used within HTTP/1.1.
813</t>
814</section>
815</section>
816
817<section title="Header Field Definitions" anchor="header.fields">
818<t>
819   This section defines the syntax and semantics of HTTP/1.1 header fields
820   related to the payload of messages.
821</t>
822<t>
823   For entity-header fields, both sender and recipient refer to either the
824   client or the server, depending on who sends and who receives the entity.
825</t>
826
827<section title="Accept" anchor="header.accept">
828  <iref primary="true" item="Accept header" x:for-anchor=""/>
829  <iref primary="true" item="Headers" subitem="Accept" x:for-anchor=""/>
830<t>
831   The Accept request-header field can be used to specify certain media
832   types which are acceptable for the response. Accept headers can be
833   used to indicate that the request is specifically limited to a small
834   set of desired types, as in the case of a request for an in-line
835   image.
836</t>
837<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"/>
838  Accept         = "Accept" ":"
839                   #( media-range [ accept-params ] )
840 
841  media-range    = ( "*/*"
842                   | ( type "/" "*" )
843                   | ( type "/" subtype )
844                   ) *( ";" parameter )
845  accept-params  = ";" "q" "=" qvalue *( accept-extension )
846  accept-extension = ";" token [ "=" ( token | quoted-string ) ]
847</artwork></figure>
848<t>
849   The asterisk "*" character is used to group media types into ranges,
850   with "*/*" indicating all media types and "type/*" indicating all
851   subtypes of that type. The media-range &MAY; include media type
852   parameters that are applicable to that range.
853</t>
854<t>
855   Each media-range &MAY; be followed by one or more accept-params,
856   beginning with the "q" parameter for indicating a relative quality
857   factor. The first "q" parameter (if any) separates the media-range
858   parameter(s) from the accept-params. Quality factors allow the user
859   or user agent to indicate the relative degree of preference for that
860   media-range, using the qvalue scale from 0 to 1 (<xref target="quality.values"/>). The
861   default value is q=1.
862  <list><t>
863      <x:h>Note:</x:h> Use of the "q" parameter name to separate media type
864      parameters from Accept extension parameters is due to historical
865      practice. Although this prevents any media type parameter named
866      "q" from being used with a media range, such an event is believed
867      to be unlikely given the lack of any "q" parameters in the IANA
868      media type registry and the rare usage of any media type
869      parameters in Accept. Future media types are discouraged from
870      registering any parameter named "q".
871  </t></list>
872</t>
873<t>
874   The example
875</t>
876<figure><artwork type="example">
877    Accept: audio/*; q=0.2, audio/basic
878</artwork></figure>
879<t>
880   &SHOULD; be interpreted as "I prefer audio/basic, but send me any audio
881   type if it is the best available after an 80% mark-down in quality."
882</t>
883<t>
884   If no Accept header field is present, then it is assumed that the
885   client accepts all media types. If an Accept header field is present,
886   and if the server cannot send a response which is acceptable
887   according to the combined Accept field value, then the server &SHOULD;
888   send a 406 (not acceptable) response.
889</t>
890<t>
891   A more elaborate example is
892</t>
893<figure><artwork type="example">
894    Accept: text/plain; q=0.5, text/html,
895            text/x-dvi; q=0.8, text/x-c
896</artwork></figure>
897<t>
898   Verbally, this would be interpreted as "text/html and text/x-c are
899   the preferred media types, but if they do not exist, then send the
900   text/x-dvi entity, and if that does not exist, send the text/plain
901   entity."
902</t>
903<t>
904   Media ranges can be overridden by more specific media ranges or
905   specific media types. If more than one media range applies to a given
906   type, the most specific reference has precedence. For example,
907</t>
908<figure><artwork type="example">
909    Accept: text/*, text/html, text/html;level=1, */*
910</artwork></figure>
911<t>
912   have the following precedence:
913</t>
914<figure><artwork type="example">
915    1) text/html;level=1
916    2) text/html
917    3) text/*
918    4) */*
919</artwork></figure>
920<t>
921   The media type quality factor associated with a given type is
922   determined by finding the media range with the highest precedence
923   which matches that type. For example,
924</t>
925<figure><artwork type="example">
926    Accept: text/*;q=0.3, text/html;q=0.7, text/html;level=1,
927            text/html;level=2;q=0.4, */*;q=0.5
928</artwork></figure>
929<t>
930   would cause the following values to be associated:
931</t>
932<figure><artwork type="example">
933    text/html;level=1         = 1
934    text/html                 = 0.7
935    text/plain                = 0.3
936    image/jpeg                = 0.5
937    text/html;level=2         = 0.4
938    text/html;level=3         = 0.7
939</artwork></figure>
940<t>
941      <x:h>Note:</x:h> A user agent might be provided with a default set of quality
942      values for certain media ranges. However, unless the user agent is
943      a closed system which cannot interact with other rendering agents,
944      this default set ought to be configurable by the user.
945</t>
946</section>
947
948<section title="Accept-Charset" anchor="header.accept-charset">
949  <iref primary="true" item="Accept-Charset header" x:for-anchor=""/>
950  <iref primary="true" item="Headers" subitem="Accept-Charset" x:for-anchor=""/>
951<t>
952   The Accept-Charset request-header field can be used to indicate what
953   character sets are acceptable for the response. This field allows
954   clients capable of understanding more comprehensive or special-purpose
955   character sets to signal that capability to a server which is
956   capable of representing documents in those character sets.
957</t>
958<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Accept-Charset"/>
959  Accept-Charset = "Accept-Charset" ":"
960          1#( ( charset | "*" ) [ ";" "q" "=" qvalue ] )
961</artwork></figure>
962<t>
963   Character set values are described in <xref target="character.sets"/>. Each charset &MAY;
964   be given an associated quality value which represents the user's
965   preference for that charset. The default value is q=1. An example is
966</t>
967<figure><artwork type="example">
968   Accept-Charset: iso-8859-5, unicode-1-1;q=0.8
969</artwork></figure>
970<t>
971   The special value "*", if present in the Accept-Charset field,
972   matches every character set (including ISO-8859-1) which is not
973   mentioned elsewhere in the Accept-Charset field. If no "*" is present
974   in an Accept-Charset field, then all character sets not explicitly
975   mentioned get a quality value of 0, except for ISO-8859-1, which gets
976   a quality value of 1 if not explicitly mentioned.
977</t>
978<t>
979   If no Accept-Charset header is present, the default is that any
980   character set is acceptable. If an Accept-Charset header is present,
981   and if the server cannot send a response which is acceptable
982   according to the Accept-Charset header, then the server &SHOULD; send
983   an error response with the 406 (not acceptable) status code, though
984   the sending of an unacceptable response is also allowed.
985</t>
986</section>
987
988<section title="Accept-Encoding" anchor="header.accept-encoding">
989  <iref primary="true" item="Accept-Encoding header" x:for-anchor=""/>
990  <iref primary="true" item="Headers" subitem="Accept-Encoding" x:for-anchor=""/>
991<t>
992   The Accept-Encoding request-header field is similar to Accept, but
993   restricts the content-codings (<xref target="content.codings"/>) that are acceptable in
994   the response.
995</t>
996<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Accept-Encoding"/><iref primary="true" item="Grammar" subitem="codings"/>
997  Accept-Encoding  = "Accept-Encoding" ":"
998                     #( codings [ ";" "q" "=" qvalue ] )
999  codings          = ( content-coding | "*" )
1000</artwork></figure>
1001<t>
1002   Examples of its use are:
1003</t>
1004<figure><artwork type="example">
1005    Accept-Encoding: compress, gzip
1006    Accept-Encoding:
1007    Accept-Encoding: *
1008    Accept-Encoding: compress;q=0.5, gzip;q=1.0
1009    Accept-Encoding: gzip;q=1.0, identity; q=0.5, *;q=0
1010</artwork></figure>
1011<t>
1012   A server tests whether a content-coding is acceptable, according to
1013   an Accept-Encoding field, using these rules:
1014  <list style="numbers">
1015      <t>If the content-coding is one of the content-codings listed in
1016         the Accept-Encoding field, then it is acceptable, unless it is
1017         accompanied by a qvalue of 0. (As defined in <xref target="quality.values"/>, a
1018         qvalue of 0 means "not acceptable.")</t>
1019
1020      <t>The special "*" symbol in an Accept-Encoding field matches any
1021         available content-coding not explicitly listed in the header
1022         field.</t>
1023
1024      <t>If multiple content-codings are acceptable, then the acceptable
1025         content-coding with the highest non-zero qvalue is preferred.</t>
1026
1027      <t>The "identity" content-coding is always acceptable, unless
1028         specifically refused because the Accept-Encoding field includes
1029         "identity;q=0", or because the field includes "*;q=0" and does
1030         not explicitly include the "identity" content-coding. If the
1031         Accept-Encoding field-value is empty, then only the "identity"
1032         encoding is acceptable.</t>
1033  </list>
1034</t>
1035<t>
1036   If an Accept-Encoding field is present in a request, and if the
1037   server cannot send a response which is acceptable according to the
1038   Accept-Encoding header, then the server &SHOULD; send an error response
1039   with the 406 (Not Acceptable) status code.
1040</t>
1041<t>
1042   If no Accept-Encoding field is present in a request, the server &MAY;
1043   assume that the client will accept any content coding. In this case,
1044   if "identity" is one of the available content-codings, then the
1045   server &SHOULD; use the "identity" content-coding, unless it has
1046   additional information that a different content-coding is meaningful
1047   to the client.
1048  <list><t>
1049      <x:h>Note:</x:h> If the request does not include an Accept-Encoding field,
1050      and if the "identity" content-coding is unavailable, then
1051      content-codings commonly understood by HTTP/1.0 clients (i.e.,
1052      "gzip" and "compress") are preferred; some older clients
1053      improperly display messages sent with other content-codings.  The
1054      server might also make this decision based on information about
1055      the particular user-agent or client.
1056    </t><t>
1057      <x:h>Note:</x:h> Most HTTP/1.0 applications do not recognize or obey qvalues
1058      associated with content-codings. This means that qvalues will not
1059      work and are not permitted with x-gzip or x-compress.
1060    </t></list>
1061</t>
1062</section>
1063
1064<section title="Accept-Language" anchor="header.accept-language">
1065  <iref primary="true" item="Accept-Language header" x:for-anchor=""/>
1066  <iref primary="true" item="Headers" subitem="Accept-Language" x:for-anchor=""/>
1067<t>
1068   The Accept-Language request-header field is similar to Accept, but
1069   restricts the set of natural languages that are preferred as a
1070   response to the request. Language tags are defined in <xref target="language.tags"/>.
1071</t>
1072<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Accept-Language"/><iref primary="true" item="Grammar" subitem="language-range"/>
1073  Accept-Language = "Accept-Language" ":"
1074                    1#( language-range [ ";" "q" "=" qvalue ] )
1075  language-range  = ( ( 1*8ALPHA *( "-" 1*8ALPHA ) ) | "*" )
1076</artwork></figure>
1077<t>
1078   Each language-range &MAY; be given an associated quality value which
1079   represents an estimate of the user's preference for the languages
1080   specified by that range. The quality value defaults to "q=1". For
1081   example,
1082</t>
1083<figure><artwork type="example">
1084    Accept-Language: da, en-gb;q=0.8, en;q=0.7
1085</artwork></figure>
1086<t>
1087   would mean: "I prefer Danish, but will accept British English and
1088   other types of English." A language-range matches a language-tag if
1089   it exactly equals the tag, or if it exactly equals a prefix of the
1090   tag such that the first tag character following the prefix is "-".
1091   The special range "*", if present in the Accept-Language field,
1092   matches every tag not matched by any other range present in the
1093   Accept-Language field.
1094  <list><t>
1095      <x:h>Note:</x:h> This use of a prefix matching rule does not imply that
1096      language tags are assigned to languages in such a way that it is
1097      always true that if a user understands a language with a certain
1098      tag, then this user will also understand all languages with tags
1099      for which this tag is a prefix. The prefix rule simply allows the
1100      use of prefix tags if this is the case.
1101  </t></list>
1102</t>
1103<t>
1104   The language quality factor assigned to a language-tag by the
1105   Accept-Language field is the quality value of the longest language-range
1106   in the field that matches the language-tag. If no language-range
1107   in the field matches the tag, the language quality factor
1108   assigned is 0. If no Accept-Language header is present in the
1109   request, the server
1110   &SHOULD; assume that all languages are equally acceptable. If an
1111   Accept-Language header is present, then all languages which are
1112   assigned a quality factor greater than 0 are acceptable.
1113</t>
1114<t>
1115   It might be contrary to the privacy expectations of the user to send
1116   an Accept-Language header with the complete linguistic preferences of
1117   the user in every request. For a discussion of this issue, see
1118   <xref target="privacy.issues.connected.to.accept.headers"/>.
1119</t>
1120<t>
1121   As intelligibility is highly dependent on the individual user, it is
1122   recommended that client applications make the choice of linguistic
1123   preference available to the user. If the choice is not made
1124   available, then the Accept-Language header field &MUST-NOT; be given in
1125   the request.
1126  <list><t>
1127      <x:h>Note:</x:h> When making the choice of linguistic preference available to
1128      the user, we remind implementors of  the fact that users are not
1129      familiar with the details of language matching as described above,
1130      and should provide appropriate guidance. As an example, users
1131      might assume that on selecting "en-gb", they will be served any
1132      kind of English document if British English is not available. A
1133      user agent might suggest in such a case to add "en" to get the
1134      best matching behavior.
1135  </t></list>
1136</t>
1137</section>
1138
1139<section title="Content-Encoding" anchor="header.content-encoding">
1140  <iref primary="true" item="Content-Encoding header" x:for-anchor=""/>
1141  <iref primary="true" item="Headers" subitem="Content-Encoding" x:for-anchor=""/>
1142<t>
1143   The Content-Encoding entity-header field is used as a modifier to the
1144   media-type. When present, its value indicates what additional content
1145   codings have been applied to the entity-body, and thus what decoding
1146   mechanisms must be applied in order to obtain the media-type
1147   referenced by the Content-Type header field. Content-Encoding is
1148   primarily used to allow a document to be compressed without losing
1149   the identity of its underlying media type.
1150</t>
1151<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Encoding"/>
1152  Content-Encoding  = "Content-Encoding" ":" 1#content-coding
1153</artwork></figure>
1154<t>
1155   Content codings are defined in <xref target="content.codings"/>. An example of its use is
1156</t>
1157<figure><artwork type="example">
1158    Content-Encoding: gzip
1159</artwork></figure>
1160<t>
1161   The content-coding is a characteristic of the entity identified by
1162   the Request-URI. Typically, the entity-body is stored with this
1163   encoding and is only decoded before rendering or analogous usage.
1164   However, a non-transparent proxy &MAY; modify the content-coding if the
1165   new coding is known to be acceptable to the recipient, unless the
1166   "no-transform" cache-control directive is present in the message.
1167</t>
1168<t>
1169   If the content-coding of an entity is not "identity", then the
1170   response &MUST; include a Content-Encoding entity-header (<xref target="header.content-encoding"/>)
1171   that lists the non-identity content-coding(s) used.
1172</t>
1173<t>
1174   If the content-coding of an entity in a request message is not
1175   acceptable to the origin server, the server &SHOULD; respond with a
1176   status code of 415 (Unsupported Media Type).
1177</t>
1178<t>
1179   If multiple encodings have been applied to an entity, the content
1180   codings &MUST; be listed in the order in which they were applied.
1181   Additional information about the encoding parameters &MAY; be provided
1182   by other entity-header fields not defined by this specification.
1183</t>
1184</section>
1185
1186<section title="Content-Language" anchor="header.content-language">
1187  <iref primary="true" item="Content-Language header" x:for-anchor=""/>
1188  <iref primary="true" item="Headers" subitem="Content-Language" x:for-anchor=""/>
1189<t>
1190   The Content-Language entity-header field describes the natural
1191   language(s) of the intended audience for the enclosed entity. Note
1192   that this might not be equivalent to all the languages used within
1193   the entity-body.
1194</t>
1195<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Language"/>
1196  Content-Language  = "Content-Language" ":" 1#language-tag
1197</artwork></figure>
1198<t>
1199   Language tags are defined in <xref target="language.tags"/>. The primary purpose of
1200   Content-Language is to allow a user to identify and differentiate
1201   entities according to the user's own preferred language. Thus, if the
1202   body content is intended only for a Danish-literate audience, the
1203   appropriate field is
1204</t>
1205<figure><artwork type="example">
1206    Content-Language: da
1207</artwork></figure>
1208<t>
1209   If no Content-Language is specified, the default is that the content
1210   is intended for all language audiences. This might mean that the
1211   sender does not consider it to be specific to any natural language,
1212   or that the sender does not know for which language it is intended.
1213</t>
1214<t>
1215   Multiple languages &MAY; be listed for content that is intended for
1216   multiple audiences. For example, a rendition of the "Treaty of
1217   Waitangi," presented simultaneously in the original Maori and English
1218   versions, would call for
1219</t>
1220<figure><artwork type="example">
1221    Content-Language: mi, en
1222</artwork></figure>
1223<t>
1224   However, just because multiple languages are present within an entity
1225   does not mean that it is intended for multiple linguistic audiences.
1226   An example would be a beginner's language primer, such as "A First
1227   Lesson in Latin," which is clearly intended to be used by an
1228   English-literate audience. In this case, the Content-Language would
1229   properly only include "en".
1230</t>
1231<t>
1232   Content-Language &MAY; be applied to any media type -- it is not
1233   limited to textual documents.
1234</t>
1235</section>
1236
1237<section title="Content-Location" anchor="header.content-location">
1238  <iref primary="true" item="Content-Location header" x:for-anchor=""/>
1239  <iref primary="true" item="Headers" subitem="Content-Location" x:for-anchor=""/>
1240<t>
1241   The Content-Location entity-header field &MAY; be used to supply the
1242   resource location for the entity enclosed in the message when that
1243   entity is accessible from a location separate from the requested
1244   resource's URI. A server &SHOULD; provide a Content-Location for the
1245   variant corresponding to the response entity; especially in the case
1246   where a resource has multiple entities associated with it, and those
1247   entities actually have separate locations by which they might be
1248   individually accessed, the server &SHOULD; provide a Content-Location
1249   for the particular variant which is returned.
1250</t>
1251<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Location"/>
1252  Content-Location = "Content-Location" ":"
1253                    ( absoluteURI | relativeURI )
1254</artwork></figure>
1255<t>
1256   The value of Content-Location also defines the base URI for the
1257   entity.
1258</t>
1259<t>
1260   The Content-Location value is not a replacement for the original
1261   requested URI; it is only a statement of the location of the resource
1262   corresponding to this particular entity at the time of the request.
1263   Future requests &MAY; specify the Content-Location URI as the request-URI
1264   if the desire is to identify the source of that particular
1265   entity.
1266</t>
1267<t>
1268   A cache cannot assume that an entity with a Content-Location
1269   different from the URI used to retrieve it can be used to respond to
1270   later requests on that Content-Location URI. However, the Content-Location
1271   can be used to differentiate between multiple entities
1272   retrieved from a single requested resource, as described in &caching-neg-resp;.
1273</t>
1274<t>
1275   If the Content-Location is a relative URI, the relative URI is
1276   interpreted relative to the Request-URI.
1277</t>
1278<t>
1279   The meaning of the Content-Location header in PUT or POST requests is
1280   undefined; servers are free to ignore it in those cases.
1281</t>
1282</section>
1283
1284<section title="Content-MD5" anchor="header.content-md5">
1285  <iref primary="true" item="Content-MD5 header" x:for-anchor=""/>
1286  <iref primary="true" item="Headers" subitem="Content-MD5" x:for-anchor=""/>
1287<t>
1288   The Content-MD5 entity-header field, as defined in <xref target="RFC1864"/>, is
1289   an MD5 digest of the entity-body for the purpose of providing an
1290   end-to-end message integrity check (MIC) of the entity-body. (Note: a
1291   MIC is good for detecting accidental modification of the entity-body
1292   in transit, but is not proof against malicious attacks.)
1293</t>
1294<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-MD5"/><iref primary="true" item="Grammar" subitem="md5-digest"/>
1295  Content-MD5   = "Content-MD5" ":" md5-digest
1296  md5-digest    = &lt;base64 of 128 bit MD5 digest as per <xref target="RFC1864"/>&gt;
1297</artwork></figure>
1298<t>
1299   The Content-MD5 header field &MAY; be generated by an origin server or
1300   client to function as an integrity check of the entity-body. Only
1301   origin servers or clients &MAY; generate the Content-MD5 header field;
1302   proxies and gateways &MUST-NOT; generate it, as this would defeat its
1303   value as an end-to-end integrity check. Any recipient of the entity-body,
1304   including gateways and proxies, &MAY; check that the digest value
1305   in this header field matches that of the entity-body as received.
1306</t>
1307<t>
1308   The MD5 digest is computed based on the content of the entity-body,
1309   including any content-coding that has been applied, but not including
1310   any transfer-encoding applied to the message-body. If the message is
1311   received with a transfer-encoding, that encoding &MUST; be removed
1312   prior to checking the Content-MD5 value against the received entity.
1313</t>
1314<t>
1315   This has the result that the digest is computed on the octets of the
1316   entity-body exactly as, and in the order that, they would be sent if
1317   no transfer-encoding were being applied.
1318</t>
1319<t>
1320   HTTP extends RFC 1864 to permit the digest to be computed for MIME
1321   composite media-types (e.g., multipart/* and message/rfc822), but
1322   this does not change how the digest is computed as defined in the
1323   preceding paragraph.
1324</t>
1325<t>
1326   There are several consequences of this. The entity-body for composite
1327   types &MAY; contain many body-parts, each with its own MIME and HTTP
1328   headers (including Content-MD5, Content-Transfer-Encoding, and
1329   Content-Encoding headers). If a body-part has a Content-Transfer-Encoding
1330   or Content-Encoding header, it is assumed that the content
1331   of the body-part has had the encoding applied, and the body-part is
1332   included in the Content-MD5 digest as is -- i.e., after the
1333   application. The Transfer-Encoding header field is not allowed within
1334   body-parts.
1335</t>
1336<t>
1337   Conversion of all line breaks to CRLF &MUST-NOT; be done before
1338   computing or checking the digest: the line break convention used in
1339   the text actually transmitted &MUST; be left unaltered when computing
1340   the digest.
1341  <list><t>
1342      <x:h>Note:</x:h> while the definition of Content-MD5 is exactly the same for
1343      HTTP as in RFC 1864 for MIME entity-bodies, there are several ways
1344      in which the application of Content-MD5 to HTTP entity-bodies
1345      differs from its application to MIME entity-bodies. One is that
1346      HTTP, unlike MIME, does not use Content-Transfer-Encoding, and
1347      does use Transfer-Encoding and Content-Encoding. Another is that
1348      HTTP more frequently uses binary content types than MIME, so it is
1349      worth noting that, in such cases, the byte order used to compute
1350      the digest is the transmission byte order defined for the type.
1351      Lastly, HTTP allows transmission of text types with any of several
1352      line break conventions and not just the canonical form using CRLF.
1353  </t></list>
1354</t>
1355</section>
1356
1357<section title="Content-Type" anchor="header.content-type">
1358  <iref primary="true" item="Content-Type header" x:for-anchor=""/>
1359  <iref primary="true" item="Headers" subitem="Content-Type" x:for-anchor=""/>
1360<t>
1361   The Content-Type entity-header field indicates the media type of the
1362   entity-body sent to the recipient or, in the case of the HEAD method,
1363   the media type that would have been sent had the request been a GET.
1364</t>
1365<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="Content-Type"/>
1366  Content-Type   = "Content-Type" ":" media-type
1367</artwork></figure>
1368<t>
1369   Media types are defined in <xref target="media.types"/>. An example of the field is
1370</t>
1371<figure><artwork type="example">
1372    Content-Type: text/html; charset=ISO-8859-4
1373</artwork></figure>
1374<t>
1375   Further discussion of methods for identifying the media type of an
1376   entity is provided in <xref target="type"/>.
1377</t>
1378</section>
1379
1380</section>
1381
1382<section title="IANA Considerations" anchor="IANA.considerations">
1383<t>
1384   TBD.
1385</t>
1386</section>
1387
1388<section title="Security Considerations" anchor="security.considerations">
1389<t>
1390   This section is meant to inform application developers, information
1391   providers, and users of the security limitations in HTTP/1.1 as
1392   described by this document. The discussion does not include
1393   definitive solutions to the problems revealed, though it does make
1394   some suggestions for reducing security risks.
1395</t>
1396
1397<section title="Privacy Issues Connected to Accept Headers" anchor="privacy.issues.connected.to.accept.headers">
1398<t>
1399   Accept request-headers can reveal information about the user to all
1400   servers which are accessed. The Accept-Language header in particular
1401   can reveal information the user would consider to be of a private
1402   nature, because the understanding of particular languages is often
1403   strongly correlated to the membership of a particular ethnic group.
1404   User agents which offer the option to configure the contents of an
1405   Accept-Language header to be sent in every request are strongly
1406   encouraged to let the configuration process include a message which
1407   makes the user aware of the loss of privacy involved.
1408</t>
1409<t>
1410   An approach that limits the loss of privacy would be for a user agent
1411   to omit the sending of Accept-Language headers by default, and to ask
1412   the user whether or not to start sending Accept-Language headers to a
1413   server if it detects, by looking for any Vary response-header fields
1414   generated by the server, that such sending could improve the quality
1415   of service.
1416</t>
1417<t>
1418   Elaborate user-customized accept header fields sent in every request,
1419   in particular if these include quality values, can be used by servers
1420   as relatively reliable and long-lived user identifiers. Such user
1421   identifiers would allow content providers to do click-trail tracking,
1422   and would allow collaborating content providers to match cross-server
1423   click-trails or form submissions of individual users. Note that for
1424   many users not behind a proxy, the network address of the host
1425   running the user agent will also serve as a long-lived user
1426   identifier. In environments where proxies are used to enhance
1427   privacy, user agents ought to be conservative in offering accept
1428   header configuration options to end users. As an extreme privacy
1429   measure, proxies could filter the accept headers in relayed requests.
1430   General purpose user agents which provide a high degree of header
1431   configurability &SHOULD; warn users about the loss of privacy which can
1432   be involved.
1433</t>
1434</section>
1435
1436<section title="Content-Disposition Issues" anchor="content-disposition.issues">
1437<t>
1438   <xref target="RFC1806"/>, from which the often implemented Content-Disposition
1439   (see <xref target="content-disposition"/>) header in HTTP is derived, has a number of very
1440   serious security considerations. Content-Disposition is not part of
1441   the HTTP standard, but since it is widely implemented, we are
1442   documenting its use and risks for implementors. See <xref target="RFC2183"/>
1443   (which updates <xref target="RFC1806"/>) for details.
1444</t>
1445</section>
1446
1447</section>
1448
1449<section title="Acknowledgments" anchor="ack">
1450</section>
1451</middle>
1452<back>
1453
1454<references title="Normative References">
1455
1456<reference anchor="ISO-8859-1">
1457  <front>
1458    <title>
1459     Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1
1460    </title>
1461    <author>
1462      <organization>International Organization for Standardization</organization>
1463    </author>
1464    <date year="1998"/>
1465  </front>
1466  <seriesInfo name="ISO/IEC" value="8859-1:1998"/>
1467</reference>
1468
1469<reference anchor="Part1">
1470  <front>
1471    <title abbrev="HTTP/1.1">HTTP/1.1, part 1: URIs, Connections, and Message Parsing</title>
1472    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1473      <organization abbrev="Day Software">Day Software</organization>
1474      <address><email>fielding@gbiv.com</email></address>
1475    </author>
1476    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1477      <organization>One Laptop per Child</organization>
1478      <address><email>jg@laptop.org</email></address>
1479    </author>
1480    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1481      <organization abbrev="HP">Hewlett-Packard Company</organization>
1482      <address><email>JeffMogul@acm.org</email></address>
1483    </author>
1484    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1485      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1486      <address><email>henrikn@microsoft.com</email></address>
1487    </author>
1488    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1489      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1490      <address><email>LMM@acm.org</email></address>
1491    </author>
1492    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1493      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1494      <address><email>paulle@microsoft.com</email></address>
1495    </author>
1496    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1497      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1498      <address><email>timbl@w3.org</email></address>
1499    </author>
1500    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1501      <organization abbrev="W3C">World Wide Web Consortium</organization>
1502      <address><email>ylafon@w3.org</email></address>
1503    </author>
1504    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1505      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1506      <address><email>julian.reschke@greenbytes.de</email></address>
1507    </author>
1508    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1509  </front>
1510  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p1-messaging-&ID-VERSION;"/>
1511  <x:source href="p1-messaging.xml" basename="p1-messaging"/>
1512</reference>
1513
1514<reference anchor="Part2">
1515  <front>
1516    <title abbrev="HTTP/1.1">HTTP/1.1, part 2: Message Semantics</title>
1517    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1518      <organization abbrev="Day Software">Day Software</organization>
1519      <address><email>fielding@gbiv.com</email></address>
1520    </author>
1521    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1522      <organization>One Laptop per Child</organization>
1523      <address><email>jg@laptop.org</email></address>
1524    </author>
1525    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1526      <organization abbrev="HP">Hewlett-Packard Company</organization>
1527      <address><email>JeffMogul@acm.org</email></address>
1528    </author>
1529    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1530      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1531      <address><email>henrikn@microsoft.com</email></address>
1532    </author>
1533    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1534      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1535      <address><email>LMM@acm.org</email></address>
1536    </author>
1537    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1538      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1539      <address><email>paulle@microsoft.com</email></address>
1540    </author>
1541    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1542      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1543      <address><email>timbl@w3.org</email></address>
1544    </author>
1545    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1546      <organization abbrev="W3C">World Wide Web Consortium</organization>
1547      <address><email>ylafon@w3.org</email></address>
1548    </author>
1549    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1550      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1551      <address><email>julian.reschke@greenbytes.de</email></address>
1552    </author>
1553    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1554  </front>
1555  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p2-semantics-&ID-VERSION;"/>
1556  <x:source href="p2-semantics.xml" basename="p2-semantics"/>
1557</reference>
1558
1559<reference anchor="Part4">
1560  <front>
1561    <title abbrev="HTTP/1.1">HTTP/1.1, part 4: Conditional Requests</title>
1562    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1563      <organization abbrev="Day Software">Day Software</organization>
1564      <address><email>fielding@gbiv.com</email></address>
1565    </author>
1566    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1567      <organization>One Laptop per Child</organization>
1568      <address><email>jg@laptop.org</email></address>
1569    </author>
1570    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1571      <organization abbrev="HP">Hewlett-Packard Company</organization>
1572      <address><email>JeffMogul@acm.org</email></address>
1573    </author>
1574    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1575      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1576      <address><email>henrikn@microsoft.com</email></address>
1577    </author>
1578    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1579      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1580      <address><email>LMM@acm.org</email></address>
1581    </author>
1582    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1583      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1584      <address><email>paulle@microsoft.com</email></address>
1585    </author>
1586    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1587      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1588      <address><email>timbl@w3.org</email></address>
1589    </author>
1590    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1591      <organization abbrev="W3C">World Wide Web Consortium</organization>
1592      <address><email>ylafon@w3.org</email></address>
1593    </author>
1594    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1595      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1596      <address><email>julian.reschke@greenbytes.de</email></address>
1597    </author>
1598    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1599  </front>
1600  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p4-conditional-&ID-VERSION;"/>
1601  <x:source href="p4-conditional.xml" basename="p4-conditional"/>
1602</reference>
1603
1604<reference anchor="Part5">
1605  <front>
1606    <title abbrev="HTTP/1.1">HTTP/1.1, part 5: Range Requests and Partial Responses</title>
1607    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1608      <organization abbrev="Day Software">Day Software</organization>
1609      <address><email>fielding@gbiv.com</email></address>
1610    </author>
1611    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1612      <organization>One Laptop per Child</organization>
1613      <address><email>jg@laptop.org</email></address>
1614    </author>
1615    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1616      <organization abbrev="HP">Hewlett-Packard Company</organization>
1617      <address><email>JeffMogul@acm.org</email></address>
1618    </author>
1619    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1620      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1621      <address><email>henrikn@microsoft.com</email></address>
1622    </author>
1623    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1624      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1625      <address><email>LMM@acm.org</email></address>
1626    </author>
1627    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1628      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1629      <address><email>paulle@microsoft.com</email></address>
1630    </author>
1631    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1632      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1633      <address><email>timbl@w3.org</email></address>
1634    </author>
1635    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1636      <organization abbrev="W3C">World Wide Web Consortium</organization>
1637      <address><email>ylafon@w3.org</email></address>
1638    </author>
1639    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1640      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1641      <address><email>julian.reschke@greenbytes.de</email></address>
1642    </author>
1643    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1644  </front>
1645  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p5-range-&ID-VERSION;"/>
1646  <x:source href="p5-range.xml" basename="p5-range"/>
1647</reference>
1648
1649<reference anchor="Part6">
1650  <front>
1651    <title abbrev="HTTP/1.1">HTTP/1.1, part 6: Caching</title>
1652    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor">
1653      <organization abbrev="Day Software">Day Software</organization>
1654      <address><email>fielding@gbiv.com</email></address>
1655    </author>
1656    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1657      <organization>One Laptop per Child</organization>
1658      <address><email>jg@laptop.org</email></address>
1659    </author>
1660    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1661      <organization abbrev="HP">Hewlett-Packard Company</organization>
1662      <address><email>JeffMogul@acm.org</email></address>
1663    </author>
1664    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen">
1665      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1666      <address><email>henrikn@microsoft.com</email></address>
1667    </author>
1668    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1669      <organization abbrev="Adobe Systems">Adobe Systems, Incorporated</organization>
1670      <address><email>LMM@acm.org</email></address>
1671    </author>
1672    <author initials="P." surname="Leach" fullname="Paul J. Leach">
1673      <organization abbrev="Microsoft">Microsoft Corporation</organization>
1674      <address><email>paulle@microsoft.com</email></address>
1675    </author>
1676    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1677      <organization abbrev="W3C/MIT">World Wide Web Consortium</organization>
1678      <address><email>timbl@w3.org</email></address>
1679    </author>
1680    <author initials="Y." surname="Lafon" fullname="Yves Lafon" role="editor">
1681      <organization abbrev="W3C">World Wide Web Consortium</organization>
1682      <address><email>ylafon@w3.org</email></address>
1683    </author>
1684    <author initials="J. F." surname="Reschke" fullname="Julian F. Reschke" role="editor">
1685      <organization abbrev="greenbytes">greenbytes GmbH</organization>
1686      <address><email>julian.reschke@greenbytes.de</email></address>
1687    </author>
1688    <date month="&ID-MONTH;" year="&ID-YEAR;"/>
1689  </front>
1690  <seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-p6-cache-&ID-VERSION;"/>
1691  <x:source href="p6-cache.xml" basename="p6-cache"/>
1692</reference>
1693
1694<reference anchor="RFC1766">
1695  <front>
1696    <title abbrev="Language Tag">Tags for the Identification of Languages</title>
1697    <author initials="H." surname="Alvestrand" fullname="Harald Tveit Alvestrand">
1698      <organization>UNINETT</organization>
1699      <address><email>Harald.T.Alvestrand@uninett.no</email></address>
1700    </author>
1701    <date month="March" year="1995"/>
1702  </front>
1703  <seriesInfo name="RFC" value="1766"/>
1704</reference>
1705
1706<reference anchor="RFC1864">
1707  <front>
1708    <title abbrev="Content-MD5 Header Field">The Content-MD5 Header Field</title>
1709    <author initials="J." surname="Myers" fullname="John G. Myers">
1710      <organization>Carnegie Mellon University</organization>
1711      <address><email>jgm+@cmu.edu</email></address>
1712    </author>
1713    <author initials="M." surname="Rose" fullname="Marshall T. Rose">
1714      <organization>Dover Beach Consulting, Inc.</organization>
1715      <address><email>mrose@dbc.mtview.ca.us</email></address>
1716    </author>
1717    <date month="October" year="1995"/>
1718  </front>
1719  <seriesInfo name="RFC" value="1864"/>
1720</reference>
1721
1722<reference anchor="RFC1950">
1723  <front>
1724    <title>ZLIB Compressed Data Format Specification version 3.3</title>
1725    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
1726      <organization>Aladdin Enterprises</organization>
1727      <address><email>ghost@aladdin.com</email></address>
1728    </author>
1729    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
1730      <organization/>
1731    </author>
1732    <date month="May" year="1996"/>
1733  </front>
1734  <seriesInfo name="RFC" value="1950"/>
1735  <annotation>
1736    RFC1950 is an Informational RFC, thus it may be less stable than
1737    this specification. On the other hand, this downward reference was
1738    present since <xref target="RFC2068"/> (published in 1997), therefore it is unlikely
1739    to cause problems in practice.
1740  </annotation>
1741</reference>
1742
1743<reference anchor="RFC1951">
1744  <front>
1745    <title>DEFLATE Compressed Data Format Specification version 1.3</title>
1746    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
1747      <organization>Aladdin Enterprises</organization>
1748      <address><email>ghost@aladdin.com</email></address>
1749    </author>
1750    <date month="May" year="1996"/>
1751  </front>
1752  <seriesInfo name="RFC" value="1951"/>
1753  <annotation>
1754    RFC1951 is an Informational RFC, thus it may be less stable than
1755    this specification. On the other hand, this downward reference was
1756    present since <xref target="RFC2068"/> (published in 1997), therefore it is unlikely
1757    to cause problems in practice.
1758  </annotation>
1759</reference>
1760
1761<reference anchor="RFC1952">
1762  <front>
1763    <title>GZIP file format specification version 4.3</title>
1764    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch">
1765      <organization>Aladdin Enterprises</organization>
1766      <address><email>ghost@aladdin.com</email></address>
1767    </author>
1768    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly">
1769      <organization/>
1770      <address><email>gzip@prep.ai.mit.edu</email></address>
1771    </author>
1772    <author initials="M." surname="Adler" fullname="Mark Adler">
1773      <organization/>
1774      <address><email>madler@alumni.caltech.edu</email></address>
1775    </author>
1776    <author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch">
1777      <organization/>
1778      <address><email>ghost@aladdin.com</email></address>
1779    </author>
1780    <author initials="G." surname="Randers-Pehrson" fullname="Glenn Randers-Pehrson">
1781      <organization/>
1782      <address><email>randeg@alumni.rpi.edu</email></address>
1783    </author>
1784    <date month="May" year="1996"/>
1785  </front>
1786  <seriesInfo name="RFC" value="1952"/>
1787  <annotation>
1788    RFC1952 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="RFC2045">
1796  <front>
1797    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
1798    <author initials="N." surname="Freed" fullname="Ned Freed">
1799      <organization>Innosoft International, Inc.</organization>
1800      <address><email>ned@innosoft.com</email></address>
1801    </author>
1802    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
1803      <organization>First Virtual Holdings</organization>
1804      <address><email>nsb@nsb.fv.com</email></address>
1805    </author>
1806    <date month="November" year="1996"/>
1807  </front>
1808  <seriesInfo name="RFC" value="2045"/>
1809</reference>
1810
1811<reference anchor="RFC2046">
1812  <front>
1813    <title abbrev="Media Types">Multipurpose Internet Mail Extensions (MIME) Part Two: Media Types</title>
1814    <author initials="N." surname="Freed" fullname="Ned Freed">
1815      <organization>Innosoft International, Inc.</organization>
1816      <address><email>ned@innosoft.com</email></address>
1817    </author>
1818    <author initials="N." surname="Borenstein" fullname="Nathaniel S. Borenstein">
1819      <organization>First Virtual Holdings</organization>
1820      <address><email>nsb@nsb.fv.com</email></address>
1821    </author>
1822    <date month="November" year="1996"/>
1823  </front>
1824  <seriesInfo name="RFC" value="2046"/>
1825</reference>
1826
1827<reference anchor="RFC2119">
1828  <front>
1829    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
1830    <author initials="S." surname="Bradner" fullname="Scott Bradner">
1831      <organization>Harvard University</organization>
1832      <address><email>sob@harvard.edu</email></address>
1833    </author>
1834    <date month="March" year="1997"/>
1835  </front>
1836  <seriesInfo name="BCP" value="14"/>
1837  <seriesInfo name="RFC" value="2119"/>
1838</reference>
1839
1840</references>
1841
1842<references title="Informative References">
1843
1844<reference anchor="RFC1806">
1845  <front>
1846    <title abbrev="Content-Disposition">Communicating Presentation Information in Internet Messages: The Content-Disposition Header</title>
1847    <author initials="R." surname="Troost" fullname="Rens Troost">
1848      <organization>New Century Systems</organization>
1849      <address><email>rens@century.com</email></address>
1850    </author>
1851    <author initials="S." surname="Dorner" fullname="Steve Dorner">
1852      <organization>QUALCOMM Incorporated</organization>
1853      <address><email>sdorner@qualcomm.com</email></address>
1854    </author>
1855    <date month="June" year="1995"/>
1856  </front>
1857  <seriesInfo name="RFC" value="1806"/>
1858</reference>
1859
1860<reference anchor="RFC1945">
1861  <front>
1862    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
1863    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee">
1864      <organization>MIT, Laboratory for Computer Science</organization>
1865      <address><email>timbl@w3.org</email></address>
1866    </author>
1867    <author initials="R.T." surname="Fielding" fullname="Roy T. Fielding">
1868      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
1869      <address><email>fielding@ics.uci.edu</email></address>
1870    </author>
1871    <author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
1872      <organization>W3 Consortium, MIT Laboratory for Computer Science</organization>
1873      <address><email>frystyk@w3.org</email></address>
1874    </author>
1875    <date month="May" year="1996"/>
1876  </front>
1877  <seriesInfo name="RFC" value="1945"/>
1878</reference>
1879
1880<reference anchor="RFC2049">
1881  <front>
1882    <title abbrev="MIME Conformance">Multipurpose Internet Mail Extensions (MIME) Part Five: Conformance Criteria and Examples</title>
1883    <author initials="N." surname="Freed" fullname="Ned Freed">
1884      <organization>Innosoft International, Inc.</organization>
1885      <address><email>ned@innosoft.com</email></address>
1886    </author>
1887    <author initials="N.S." surname="Borenstein" fullname="Nathaniel S. Borenstein">
1888      <organization>First Virtual Holdings</organization>
1889      <address><email>nsb@nsb.fv.com</email></address>
1890    </author>
1891    <date month="November" year="1996"/>
1892  </front>
1893  <seriesInfo name="RFC" value="2049"/>
1894</reference>
1895
1896<reference anchor="RFC2068">
1897  <front>
1898    <title abbrev="HTTP/1.1">Hypertext Transfer Protocol -- HTTP/1.1</title>
1899    <author initials="R." surname="Fielding" fullname="Roy T. Fielding">
1900      <organization>University of California, Irvine, Department of Information and Computer Science</organization>
1901      <address><email>fielding@ics.uci.edu</email></address>
1902    </author>
1903    <author initials="J." surname="Gettys" fullname="Jim Gettys">
1904      <organization>MIT Laboratory for Computer Science</organization>
1905      <address><email>jg@w3.org</email></address>
1906    </author>
1907    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul">
1908      <organization>Digital Equipment Corporation, Western Research Laboratory</organization>
1909      <address><email>mogul@wrl.dec.com</email></address>
1910    </author>
1911    <author initials="H." surname="Nielsen" fullname="Henrik Frystyk Nielsen">
1912      <organization>MIT Laboratory for Computer Science</organization>
1913      <address><email>frystyk@w3.org</email></address>
1914    </author>
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    <date month="January" year="1997"/>
1920  </front>
1921  <seriesInfo name="RFC" value="2068"/>
1922</reference>
1923
1924<reference anchor="RFC2076">
1925  <front>
1926    <title abbrev="Internet Message Headers">Common Internet Message Headers</title>
1927    <author initials="J." surname="Palme" fullname="Jacob Palme">
1928      <organization>Stockholm University/KTH</organization>
1929      <address><email>jpalme@dsv.su.se</email></address>
1930    </author>
1931    <date month="February" year="1997"/>
1932  </front>
1933  <seriesInfo name="RFC" value="2076"/>
1934</reference>
1935
1936<reference anchor="RFC2183">
1937  <front>
1938    <title abbrev="Content-Disposition">Communicating Presentation Information in Internet Messages: The Content-Disposition Header Field</title>
1939    <author initials="R." surname="Troost" fullname="Rens Troost">
1940      <organization>New Century Systems</organization>
1941      <address><email>rens@century.com</email></address>
1942    </author>
1943    <author initials="S." surname="Dorner" fullname="Steve Dorner">
1944      <organization>QUALCOMM Incorporated</organization>
1945      <address><email>sdorner@qualcomm.com</email></address>
1946    </author>
1947    <author initials="K." surname="Moore" fullname="Keith Moore">
1948      <organization>Department of Computer Science</organization>
1949      <address><email>moore@cs.utk.edu</email></address>
1950    </author>
1951    <date month="August" year="1997"/>
1952  </front>
1953  <seriesInfo name="RFC" value="2183"/>
1954</reference>
1955
1956<reference anchor="RFC2277">
1957  <front>
1958    <title abbrev="Charset Policy">IETF Policy on Character Sets and Languages</title>
1959    <author initials="H.T." surname="Alvestrand" fullname="Harald Tveit Alvestrand">
1960      <organization>UNINETT</organization>
1961      <address><email>Harald.T.Alvestrand@uninett.no</email></address>
1962    </author>
1963    <date month="January" year="1998"/>
1964  </front>
1965  <seriesInfo name="BCP" value="18"/>
1966  <seriesInfo name="RFC" value="2277"/>
1967</reference>
1968
1969<reference anchor="RFC2388">
1970  <front>
1971    <title abbrev="multipart/form-data">Returning Values from Forms:  multipart/form-data</title>
1972    <author initials="L." surname="Masinter" fullname="Larry Masinter">
1973      <organization>Xerox Palo Alto Research Center</organization>
1974      <address><email>masinter@parc.xerox.com</email></address>
1975    </author>
1976    <date year="1998" month="August"/>
1977  </front>
1978  <seriesInfo name="RFC" value="2388"/>
1979</reference>
1980
1981<reference anchor="RFC2557">
1982  <front>
1983    <title abbrev="MIME Encapsulation of Aggregate Documents">MIME Encapsulation of Aggregate Documents, such as HTML (MHTML)</title>
1984    <author initials="F." surname="Palme" fullname="Jacob Palme">
1985      <organization>Stockholm University and KTH</organization>
1986      <address><email>jpalme@dsv.su.se</email></address>
1987    </author>
1988    <author initials="A." surname="Hopmann" fullname="Alex Hopmann">
1989      <organization>Microsoft Corporation</organization>
1990      <address><email>alexhop@microsoft.com</email></address>
1991    </author>
1992    <author initials="N." surname="Shelness" fullname="Nick Shelness">
1993      <organization>Lotus Development Corporation</organization>
1994      <address><email>Shelness@lotus.com</email></address>
1995    </author>
1996    <author initials="E." surname="Stefferud" fullname="Einar Stefferud">
1997      <organization/>
1998      <address><email>stef@nma.com</email></address>
1999    </author>
2000    <date year="1999" month="March"/>
2001  </front>
2002  <seriesInfo name="RFC" value="2557"/>
2003</reference>
2004
2005<reference anchor="RFC2616">
2006  <front>
2007    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
2008    <author initials="R." surname="Fielding" fullname="R. Fielding">
2009      <organization>University of California, Irvine</organization>
2010      <address><email>fielding@ics.uci.edu</email></address>
2011    </author>
2012    <author initials="J." surname="Gettys" fullname="J. Gettys">
2013      <organization>W3C</organization>
2014      <address><email>jg@w3.org</email></address>
2015    </author>
2016    <author initials="J." surname="Mogul" fullname="J. Mogul">
2017      <organization>Compaq Computer Corporation</organization>
2018      <address><email>mogul@wrl.dec.com</email></address>
2019    </author>
2020    <author initials="H." surname="Frystyk" fullname="H. Frystyk">
2021      <organization>MIT Laboratory for Computer Science</organization>
2022      <address><email>frystyk@w3.org</email></address>
2023    </author>
2024    <author initials="L." surname="Masinter" fullname="L. Masinter">
2025      <organization>Xerox Corporation</organization>
2026      <address><email>masinter@parc.xerox.com</email></address>
2027    </author>
2028    <author initials="P." surname="Leach" fullname="P. Leach">
2029      <organization>Microsoft Corporation</organization>
2030      <address><email>paulle@microsoft.com</email></address>
2031    </author>
2032    <author initials="T." surname="Berners-Lee" fullname="T. Berners-Lee">
2033      <organization>W3C</organization>
2034      <address><email>timbl@w3.org</email></address>
2035    </author>
2036    <date month="June" year="1999"/>
2037  </front>
2038  <seriesInfo name="RFC" value="2616"/>
2039</reference>
2040
2041<reference anchor="RFC2822">
2042  <front>
2043    <title>Internet Message Format</title>
2044    <author initials="P." surname="Resnick" fullname="P. Resnick">
2045      <organization>QUALCOMM Incorporated</organization>
2046    </author>
2047    <date year="2001" month="April"/>
2048  </front> 
2049  <seriesInfo name="RFC" value="2822"/>
2050</reference>
2051
2052<reference anchor="RFC3629">
2053  <front>
2054    <title>UTF-8, a transformation format of ISO 10646</title>
2055    <author initials="F." surname="Yergeau" fullname="F. Yergeau">
2056      <organization>Alis Technologies</organization>
2057      <address><email>fyergeau@alis.com</email></address>
2058    </author>
2059    <date month="November" year="2003"/>
2060  </front>
2061  <seriesInfo name="RFC" value="3629"/>
2062  <seriesInfo name="STD" value="63"/>
2063</reference>
2064
2065</references>
2066
2067<references title="References (to be categorized)">
2068
2069<reference anchor="RFC4288">
2070  <front>
2071    <title>Media Type Specifications and Registration Procedures</title>
2072    <author initials="N." surname="Freed" fullname="N. Freed">
2073      <organization>Sun Microsystems</organization>
2074      <address>
2075        <email>ned.freed@mrochek.com</email>
2076      </address>
2077    </author>
2078    <author initials="J." surname="Klensin" fullname="J. Klensin">
2079      <organization/>
2080      <address>
2081        <email>klensin+ietf@jck.com</email>
2082      </address>
2083    </author>
2084    <date year="2005" month="December"/>
2085  </front>
2086  <seriesInfo name="BCP" value="13"/>
2087  <seriesInfo name="RFC" value="4288"/>
2088</reference>
2089
2090</references>
2091
2092<section title="Differences Between HTTP Entities and RFC 2045 Entities" anchor="differences.between.http.entities.and.rfc.2045.entities">
2093<t>
2094   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
2095   allow entities to be transmitted in an open variety of
2096   representations and with extensible mechanisms. However, RFC 2045
2097   discusses mail, and HTTP has a few features that are different from
2098   those described in RFC 2045. These differences were carefully chosen
2099   to optimize performance over binary connections, to allow greater
2100   freedom in the use of new media types, to make date comparisons
2101   easier, and to acknowledge the practice of some early HTTP servers
2102   and clients.
2103</t>
2104<t>
2105   This appendix describes specific areas where HTTP differs from RFC
2106   2045. Proxies and gateways to strict MIME environments &SHOULD; be
2107   aware of these differences and provide the appropriate conversions
2108   where necessary. Proxies and gateways from MIME environments to HTTP
2109   also need to be aware of the differences because some conversions
2110   might be required.
2111</t>
2112<section title="MIME-Version" anchor="mime-version">
2113<t>
2114   HTTP is not a MIME-compliant protocol. However, HTTP/1.1 messages &MAY;
2115   include a single MIME-Version general-header field to indicate what
2116   version of the MIME protocol was used to construct the message. Use
2117   of the MIME-Version header field indicates that the message is in
2118   full compliance with the MIME protocol (as defined in <xref target="RFC2045"/>).
2119   Proxies/gateways are responsible for ensuring full compliance (where
2120   possible) when exporting HTTP messages to strict MIME environments.
2121</t>
2122<figure><artwork type="abnf2616"><iref primary="true" item="Grammar" subitem="MIME-Version"/>
2123  MIME-Version   = "MIME-Version" ":" 1*DIGIT "." 1*DIGIT
2124</artwork></figure>
2125<t>
2126   MIME version "1.0" is the default for use in HTTP/1.1. However,
2127   HTTP/1.1 message parsing and semantics are defined by this document
2128   and not the MIME specification.
2129</t>
2130</section>
2131
2132<section title="Conversion to Canonical Form" anchor="conversion.to.canonical.form">
2133<t>
2134   <xref target="RFC2045"/> requires that an Internet mail entity be converted to
2135   canonical form prior to being transferred, as described in <xref target="RFC2049" x:fmt="of" x:sec="4"/>.
2136   <xref target="canonicalization.and.text.defaults"/> of this document describes the forms
2137   allowed for subtypes of the "text" media type when transmitted over
2138   HTTP. <xref target="RFC2046"/> requires that content with a type of "text" represent
2139   line breaks as CRLF and forbids the use of CR or LF outside of line
2140   break sequences. HTTP allows CRLF, bare CR, and bare LF to indicate a
2141   line break within text content when a message is transmitted over
2142   HTTP.
2143</t>
2144<t>
2145   Where it is possible, a proxy or gateway from HTTP to a strict MIME
2146   environment &SHOULD; translate all line breaks within the text media
2147   types described in <xref target="canonicalization.and.text.defaults"/> of this document to the RFC 2049
2148   canonical form of CRLF. Note, however, that this might be complicated
2149   by the presence of a Content-Encoding and by the fact that HTTP
2150   allows the use of some character sets which do not use octets 13 and
2151   10 to represent CR and LF, as is the case for some multi-byte
2152   character sets.
2153</t>
2154<t>
2155   Implementors should note that conversion will break any cryptographic
2156   checksums applied to the original content unless the original content
2157   is already in canonical form. Therefore, the canonical form is
2158   recommended for any content that uses such checksums in HTTP.
2159</t>
2160</section>
2161
2162<section title="Introduction of Content-Encoding" anchor="introduction.of.content-encoding">
2163<t>
2164   RFC 2045 does not include any concept equivalent to HTTP/1.1's
2165   Content-Encoding header field. Since this acts as a modifier on the
2166   media type, proxies and gateways from HTTP to MIME-compliant
2167   protocols &MUST; either change the value of the Content-Type header
2168   field or decode the entity-body before forwarding the message. (Some
2169   experimental applications of Content-Type for Internet mail have used
2170   a media-type parameter of ";conversions=&lt;content-coding&gt;" to perform
2171   a function equivalent to Content-Encoding. However, this parameter is
2172   not part of RFC 2045).
2173</t>
2174</section>
2175
2176<section title="No Content-Transfer-Encoding" anchor="no.content-transfer-encoding">
2177<t>
2178   HTTP does not use the Content-Transfer-Encoding field of RFC
2179   2045. Proxies and gateways from MIME-compliant protocols to HTTP &MUST;
2180   remove any Content-Transfer-Encoding
2181   prior to delivering the response message to an HTTP client.
2182</t>
2183<t>
2184   Proxies and gateways from HTTP to MIME-compliant protocols are
2185   responsible for ensuring that the message is in the correct format
2186   and encoding for safe transport on that protocol, where "safe
2187   transport" is defined by the limitations of the protocol being used.
2188   Such a proxy or gateway &SHOULD; label the data with an appropriate
2189   Content-Transfer-Encoding if doing so will improve the likelihood of
2190   safe transport over the destination protocol.
2191</t>
2192</section>
2193
2194<section title="Introduction of Transfer-Encoding" anchor="introduction.of.transfer-encoding">
2195<t>
2196   HTTP/1.1 introduces the Transfer-Encoding header field (&header-transfer-encoding;).
2197   Proxies/gateways &MUST; remove any transfer-coding prior to
2198   forwarding a message via a MIME-compliant protocol.
2199</t>
2200</section>
2201
2202<section title="MHTML and Line Length Limitations" anchor="mhtml.line.length">
2203<t>
2204   HTTP implementations which share code with MHTML <xref target="RFC2557"/> implementations
2205   need to be aware of MIME line length limitations. Since HTTP does not
2206   have this limitation, HTTP does not fold long lines. MHTML messages
2207   being transported by HTTP follow all conventions of MHTML, including
2208   line length limitations and folding, canonicalization, etc., since
2209   HTTP transports all message-bodies as payload (see <xref target="multipart.types"/>) and
2210   does not interpret the content or any MIME header lines that might be
2211   contained therein.
2212</t>
2213</section>
2214</section>
2215
2216<section title="Additional Features" anchor="additional.features">
2217<t>
2218   <xref target="RFC1945"/> and <xref target="RFC2068"/> document protocol elements used by some
2219   existing HTTP implementations, but not consistently and correctly
2220   across most HTTP/1.1 applications. Implementors are advised to be
2221   aware of these features, but cannot rely upon their presence in, or
2222   interoperability with, other HTTP/1.1 applications. Some of these
2223   describe proposed experimental features, and some describe features
2224   that experimental deployment found lacking that are now addressed in
2225   the base HTTP/1.1 specification.
2226</t>
2227<t>
2228   A number of other headers, such as Content-Disposition and Title,
2229   from SMTP and MIME are also often implemented (see <xref target="RFC2076"/>).
2230</t>
2231
2232<section title="Content-Disposition" anchor="content-disposition">
2233<iref item="Headers" subitem="Content-Disposition" primary="true" x:for-anchor=""/>
2234<iref item="Content-Disposition header" primary="true" x:for-anchor=""/>
2235<t>
2236   The Content-Disposition response-header field has been proposed as a
2237   means for the origin server to suggest a default filename if the user
2238   requests that the content is saved to a file. This usage is derived
2239   from the definition of Content-Disposition in <xref target="RFC1806"/>.
2240</t>
2241<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"/>
2242  content-disposition = "Content-Disposition" ":"
2243                        disposition-type *( ";" disposition-parm )
2244  disposition-type = "attachment" | disp-extension-token
2245  disposition-parm = filename-parm | disp-extension-parm
2246  filename-parm = "filename" "=" quoted-string
2247  disp-extension-token = token
2248  disp-extension-parm = token "=" ( token | quoted-string )
2249</artwork></figure>
2250<t>
2251   An example is
2252</t>
2253<figure><artwork type="example">
2254     Content-Disposition: attachment; filename="fname.ext"
2255</artwork></figure>
2256<t>
2257   The receiving user agent &SHOULD-NOT;  respect any directory path
2258   information present in the filename-parm parameter, which is the only
2259   parameter believed to apply to HTTP implementations at this time. The
2260   filename &SHOULD; be treated as a terminal component only.
2261</t>
2262<t>
2263   If this header is used in a response with the application/octet-stream
2264   content-type, the implied suggestion is that the user agent
2265   should not display the response, but directly enter a `save response
2266   as...' dialog.
2267</t>
2268<t>
2269   See <xref target="content-disposition.issues"/> for Content-Disposition security issues.
2270</t>
2271</section>
2272</section>
2273
2274<section title="Compatibility with Previous Versions" anchor="compatibility">
2275<section title="Changes from RFC 2068" anchor="changes.from.rfc.2068">
2276<t>
2277   Charset wildcarding is introduced to avoid explosion of character set
2278   names in accept headers. (<xref target="header.accept-charset"/>)
2279</t>
2280<t>
2281   Content-Base was deleted from the specification: it was not
2282   implemented widely, and there is no simple, safe way to introduce it
2283   without a robust extension mechanism. In addition, it is used in a
2284   similar, but not identical fashion in MHTML <xref target="RFC2557"/>.
2285</t>
2286<t>
2287   A content-coding of "identity" was introduced, to solve problems
2288   discovered in caching. (<xref target="content.codings"/>)
2289</t>
2290<t>
2291   Quality Values of zero should indicate that "I don't want something"
2292   to allow clients to refuse a representation. (<xref target="quality.values"/>)
2293</t>
2294<t>
2295   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
2296   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
2297   specification, but not commonly implemented. See <xref target="RFC2068"/>.
2298</t>
2299</section>
2300
2301<section title="Changes from RFC 2616" anchor="changes.from.rfc.2616">
2302<t>
2303  Clarify contexts that charset is used in.
2304  (<xref target="character.sets"/>)
2305</t>
2306<t>
2307  Remove reference to non-existant identity transfer-coding value tokens.
2308  (<xref target="no.content-transfer-encoding"/>)
2309</t>
2310</section>
2311
2312</section>
2313
2314<section title="Change Log (to be removed by RFC Editor before publication)">
2315
2316<section title="Since RFC2616">
2317<t>
2318  Extracted relevant partitions from <xref target="RFC2616"/>.
2319</t>
2320</section>
2321
2322<section title="Since draft-ietf-httpbis-p3-payload-00">
2323<t>
2324  Closed issues:
2325  <list style="symbols"> 
2326    <t>
2327      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/8"/>:
2328      "Media Type Registrations"
2329      (<eref target="http://purl.org/NET/http-errata#media-reg"/>)
2330    </t>
2331    <t>
2332      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/14"/>:
2333      "Clarification regarding quoting of charset values"
2334      (<eref target="http://purl.org/NET/http-errata#charactersets"/>)
2335    </t>
2336    <t>
2337      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/16"/>:
2338      "Remove 'identity' token references"
2339      (<eref target="http://purl.org/NET/http-errata#identity"/>)
2340    </t>
2341    <t>
2342      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/25"/>:
2343      "Accept-Encoding BNF"
2344    </t>
2345    <t>
2346      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/46"/>:
2347      "RFC1700 references"
2348    </t>
2349    <t>
2350      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/65"/>:
2351      "Informative references"
2352    </t>
2353    <t>
2354      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/66"/>:
2355      "ISO-8859-1 Reference"
2356    </t>
2357    <t>
2358      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/68"/>:
2359      "Encoding References Normative"
2360    </t>
2361    <t>
2362      <eref target="http://www3.tools.ietf.org/wg/httpbis/trac/ticket/86"/>:
2363      "Normative up-to-date references"
2364    </t>
2365  </list>
2366</t>
2367<t>
2368  Other changes:
2369  <list style="symbols">
2370    <t>
2371      Start work on categorizing references as "Normative" or "Informative".
2372    </t>
2373  </list>
2374</t>
2375</section>
2376
2377</section>
2378
2379</back>
2380</rfc>
Note: See TracBrowser for help on using the repository browser.